 Good afternoon. Welcome to this public meeting of the United States Consumer Product Safety Commission. We have one item on the agenda for this afternoon. CPSC staff will brief the Commission on the proposed rule safety standard for portable generators. CPSC staff members briefing us today are Ms. Janet Beyer, Mechanical Engineer in the Office of Hazard Identification and Reduction, and Ms. Barbara Little, Attorney in the Office of General Counsel. At the conclusion of staff briefing, we will turn to questions from the commissioners. We're now going to start with the staff briefing, Ms. Beyer. Thank you, Chairman. Good afternoon, Chairman and Commissioners. As the Chairman just mentioned, today we will be briefing you on the draft NPR for portable generators. I will begin with the legal requirements and then turn it over to Janet to continue the briefing. This is a rulemaking under the CPSA. Section 7 of the CPSA authorizes the Commission to issue consumer product safety standards that consist of performance requirements and requirements for warnings or instructions. The requirements must be reasonably necessary to prevent or reduce an unreasonable risk of injury associated with the product. Section 9 of the CPSA specifies the procedure the Commission must follow and findings the Commission must make to issue a consumer product safety standard under Section 7. Section 9 now provides that the Commission may initiate rulemaking through publication of an A&PR in the Federal Register. At the time of the publication of the portable generator A&PR, however, the A&PR was not optional, but was required under Section 9A of the CPSA. The Commission published an A&PR in the Federal Register on December 12th, 2006. Section 9 requires the Commission in issuing an NPR to identify the product and the risk of injury, summarize the regulatory alternatives considered by the Commission, present a preliminary regulatory analysis which includes the text of the proposed rule, respond to comments received on the A&PR and invite comments from the public on the NPR. Section 9 also requires the Commission following publication of an NPR to give interested persons an opportunity for the oral presentation of data, views or arguments in addition to an opportunity to make written submissions. Commission must make certain findings to issue a rule, a final rule. The Commission makes these findings preliminarily in the proposed rule. These findings include the degree and nature of the risk of injury the rule is designed to eliminate or reduce, the approximate number of consumer products subject to the rule, the need of the public for the product subject to the rule, and the probable effect the rule will have on the cost, availability and utility of such products, and other means to achieve the objective of the rule while minimizing adverse effects on competition and manufacturing. The Commission must also find that the rule is reasonably necessary to eliminate or reduce an unreasonable risk of injury associated with the product, and that issuing the rules in the public interest. Additionally, the Commission must find that the expected benefits of the rule bear a reasonable relationship to the rule's costs, and the rule imposes the least burdensome requirements that would adequately reduce the risk of injury. The Commission makes these findings on a preliminary basis in the NPR. In addition, if a voluntary standard addressing the risk of injury has been adopted and implemented, the Commission must find that the voluntary standard is not likely to eliminate or adequately reduce the risk of injury, or that substantial compliance with the voluntary standard is unlikely. And once again, the Commission makes these findings on a preliminary basis in the NPR. I'll turn it over to Janet now to continue with the presentation. Thank you, Barbara. Good afternoon, Chairman and Commissioners. We'll begin by answering the question, what is a portable generator? To put it simply, it is an engine-driven machine that converts fuel to electricity. It has a receptacle panel so that appliances and tools can be plugged directly into it, or by use of extension cords. It is designed for portability, and for the portable generators we're talking about today, we mean generators that can be carried, pulled, or pushed by a single person. They can provide as little as a few hundred watts of power for the smallest generators to more than 15,000 watts for the largest. You'll see here pictures of generators that span the full-size range with the smaller models on the top and the larger ones on the bottom. The prices can vary quite substantially depending on the size of the generator and its features. The primary fatal hazard associated with portable generators is carbon monoxide poisoning. Carbon monoxide, commonly known as CO for its chemical name, is produced by the incomplete combustion of the fuel in the engine. Engines-powering portable generators typically have very high CO emission rates, and to give you an example a five kilowatt generator operating outdoors produces about the same amount of CO as 450 idling mid-sized cars. Furthermore, when a current generator is operated in an enclosed space, we have found that its CO emission rate nominally triples. In the nine-year period of 2004 to 2012, our databases have records of 659 CO poisoning deaths caused by portable generators, which averages to about 73 deaths per year. For the same period, we also estimate that there were more than 25,000 medically treated CO injuries. This is likely a conservative estimate because medical staff often fail to identify CO poisoning as being responsible for victim symptoms. 25% of the fatal incidents involved multiple fatalities, so these account for 44% of all the deaths. In fact, we have incidents where entire families were killed. In 2015, a father and his seven children were killed in a single incident along the eastern shore of Maryland when using a generator due to terminated service. In 2016, a couple and their four children in Michigan were killed in a single incident when using a generator due to a weather-related power outage. Most recently, in the aftermath of Hurricane Matthew, a nine-year-old boy in Daytona Beach died when a family brought a generator into the house due to loss of power. Last I heard, his father was in critical condition. In Port St. Lucie County, in two separate incidents, one elderly couple died after running a generator in their garage, and another elderly couple was hospitalized. Also, there were 12 people from separate incidents in Charleston, South Carolina who were non-fatally poisoned with various degrees of severity caused by generators. From our analysis of the incident data, we found that three-quarters of all deaths occurred when generators were used at a fixed structure home, and this includes detached and attached homes, apartments, fixed mobile homes, and cabins used as permanent residences. 25% of the deaths occurred at other locations. For the 75% of the deaths that occurred at a fixed structure home, we found that 45% of the incidents involved a generator located in a living space inside the home, 25% had a generator in a basement or crawl space, 25% had a generator in an attached garage or enclosed carport, 3% had the generator outside, and 2% had the generator at home, but in an unknown location. We started bringing our concerns about the CO hazard with generators to stakeholders in 2002, when Underwriters' Laboratories formed a Standards Technical Panel, or STP, to develop the first US voluntary standard dedicated to portable generators, UL-2201. We joined the STP at its inception and have been an active participant with a long record of advocating that the standard addressed the CO hazard through technical means. UL-2201 addresses safety requirements and tests related to electrical, fire, and mechanical hazards, but only addresses the CO hazard by way of incorporating CPSC's mandatory hazard labeling requirements. UL published the first edition of UL-2201 as a UL standard in 2009. We are not aware of any portable generators that have been certified to UL-2201. Nearly three years ago, in 2014, at staff's request, UL formed a task group to develop performance and test requirements to address the CO hazard that the STP could consider for adopting into UL-2201. The task group and subgroups have met 27 times to date, but have not developed a proposal of requirements for the STP to consider. We have also met with the Portable Generators Manufacturers Association, PGMA, on numerous occasions since they formed in 2010, and subsequently developed a standard dedicated to portable generators, G300, in 2014. Again, we have advocated that their standard address the CO hazard through technical requirements. G300 includes requirements and tests related to electrical, fire, and mechanical hazards, but like the UL standard only addresses the CO hazard by referencing CPSC's mandatory hazard label. PGMA's G300 standard obtained ANSI recognition in 2015. We are not aware of any portable generators that have been certified to G300. The only other applicable voluntary standard we found is an international standard, ISO 8528-13. Although this standard is not specific to portable generators, some parts of it pertain to these products. However, it also addresses the CO hazard only through labeling requirements. Because we believe that labeling requirements alone are not sufficient to address the risk, we have concluded that none of these voluntary standards adequately address the CO hazard. We received comments on the ANPR and two other reports that staff released relative to our prototype development work, which I will discuss shortly. The comments covered a variety of topics and issues, but the primary ones were these. We received comments both in favor of and in opposition to shut off concepts. We were asked to review the effectiveness of the label in reducing CO deaths since it was introduced in 2007. It was suggested that CO alarms are the solution to addressing not only CO deaths from portable generators, but all other CO producing products. We received comments that proof does not exist that reduced CO emissions will result in reduced deaths and that reduced CO emissions will negatively impact generator cost, utility, availability, reliability, and durability. We looked at different ways that the CO hazard might possibly be mitigated. The first is CPSC's mandatory hazard label about the CO hazard. In 2007, CPSC adopted a mandatory CO hazard label that describes the hazard, its consequences, and the steps that consumers should take to avoid the hazard. While we believe the warning label is important, we do not think the label alone adequately addresses the hazard. We also looked at several automatic generator shut off concepts, but we did not find any to be technically feasible at this time due to a number of issues. These include concerns about how a system could shut the generator off quickly enough without having negative impact on generator utility when used in proper locations, concerns about reliability in the myriad of fatal scenarios and weather conditions generators have been used, concerns about the environments that they are stored in between uses, and other concerns as well. We then looked at a CO emission rate reduction strategy which would reduce the hazard at its source. Based on our investigation into this strategy, we consider it the best most feasible option. I will also add that the source reduction strategy in general has been found to be effective in reducing deaths and injuries elsewhere. For instance, a retrospective analysis of CO deaths involving automobiles showed that in the first 20 years after vehicles were required to have reduced emissions, CO deaths decreased by 81%. CO deaths caused by cars most often happen as a result of a car left idling in an attached garage. So when looking to address the CO hazard and to have a good measure of confidence that the CO emission rates from portable generators could be substantially lowered using proven existing emission control technology, we investigated the technical feasibility in multiple ways. First off, we contracted with the University of Alabama or UA to develop a durability test prototype low CO emission portable generator. UA made the prototype by taking a commercially available generator and replacing the carburetor with an electronic fuel injection system or EFI, tuning it to operate in closed loop and adding a catalyst. Then we contracted with the National Institute for Standards and Technology or NIST to test it in the common fatal scenario of the generator operating in an attached garage. I'll show you some of the results from that testing in the next couple of slides. We also considered the feasibility work that the EPA did to examine how other exhaust emissions, namely hydrocarbons and oxides of nitrogen, could be lowered using open loop EFI and a catalyst, which also resulted in substantially lowered CO emission rates. And finally, we also tested three generators made by different manufacturers that have fuel injection and evaluated their emission rates. Also, I'm going to just add one late addition to this slide that just about an hour or two ago I received a press release from Kohler. Today they announced they're rolling out a new low emission engine that they say reduces CO emissions without sacrificing engine performance, which they say it is important for the health and safety of equipment operators. They're showcasing this at an industry expo in Louisville, Kentucky, and they just announced that today. This slide shows the test house on the NIST campus that was used in the generator testing. It's a double wide manufactured house with an attached garage shown here on the right in this floor plan. The generators in both the unmodified carbureted and prototype configurations were tested in the garage to compare the difference of the resulting CO concentrations that each produced in the garage and throughout the house. And here are the comparative results from one test of the test house configurations with the bay door fully closed and the connecting door between the garage and the house closed. This top solid green trace is the CO concentration in the garage over the course of the two and a quarter hour period that the generator was operating. The garage CO concentration reached over 21,000 ppm. At the end of that two and a quarter hours the test operator shut off the generator and turned on an exhaust fan to vent the house and garage. And so the CO concentration dropped very quickly compared to how it would if we had allowed it to decay naturally. The dotted green trace near the bottom is the CO concentration in the family room of the house. By the end of the two and a quarter hours of run time the CO concentration at that location was about 1800 ppm. In stark contrast the prototype generator was run for six hours in the garage. In the garage CO concentration reached about 900 ppm and the family room CO concentration reached a peak of about 145 ppm at the end of that six hours. The purpose of this slide is just to give you an example of how quickly a current generator can create an unsafe exposure and that it is possible to significantly reduce CO emission rates to levels that will significantly reduce the exposure and are expected to save lives. So we applied what we learned from all the evaluations I just mentioned previously by considering generators as fitting into one of these four different generator categories. These categories which we call hand held class one, class two single cylinder and class two twin cylinder are largely based on the EPA's classifications of the different engine sizes which the EPA distinguishes by engine displacement in cubic centimeters or CCs. But we also made an additional distinction of whether the largest of these engine classes, the class two engines, has just one cylinder or two cylinders up to a maximum engine power of 25 kilowatts. We looked at generators in these different categories so that we could take into consideration the different hazard patterns, costs and technically feasible CO emission rates associated with each. This table presents what we believe to be technically feasible CO emission rates for the four generator categories when used outdoors shown in the middle column and the CO emission rates we assumed each category would increase to when used in enclosed space shown in the right hand column. We used the CO rates in the right hand column in our benefits analysis to estimate the deaths that would have been averted by the draft proposed standard and I'll talk more about that analysis in the next slides. But first I want to point out that the rates in this right hand column are three times the rates shown in the middle column. As I mentioned earlier, the CO emission rate for current generators nominally triples when operated in enclosed space so we are assuming that the same would happen for generators that meet the draft proposed standard. These rates are weighted rates meaning that they are calculated from the emission rates that are measured when each of six particular loads are applied to the generator. So then to assess the benefits that would be derived from these CO rates for each of the four generator categories we contracted with NIST to perform modeling for us. The first step in that process was the selection of 40 different structures that best fit within reason many of the houses and garages in which fatal incidents occurred and which broadly represent where 76% of the deaths occurred. The other 24% of the deaths occurred with the generator operating inside structures that NIST did not have existing models for or were caused by the generator operating outside. In those 40 structures NIST modeled the resulting CO distribution from a generator operating within it for 8 to 10 hours to simulate the run time on a full tank of gas. This was done for both the estimated CO rates of current generators in each of the four categories as well as the reduced CO emission rates in the range of those we estimated would be technically feasible. The 14 to 16 hours after the 8 to 10 hour run time were also modeled to complete a full 24 hour modeled period. NIST's modeling also used different generator locations and different seasonal weather data to have the modeled scenarios replicate the fatal incident data as much as reasonably possible. The second step in the process involved predicting the home's CO exposure profiles that would result from the predicted CO distribution throughout the house. This was done using a physiological model to estimate the carboxy hemoglobin level or COHB for someone in each room of the house. COHB is a measure of the body's exposure to CO. We then took that modeling output and then the third step matched it to reflect the frequency of occurrence of those modeled scenarios in the incident data. Finally in the fourth step we estimated the number of deaths that could have been averted with generators if they had emitted the reduced CO rates. I'll explain this fourth step in a later slide. I'll add here though that we purposely used conservative estimates and assumptions in the modeling such as occupants air intake rates, generator CO emission rates when used in an enclosed space, generator run times and other aspects so that we would likely derive an underestimate of the benefits rather than risk overstating the benefits. Now just to give you an example of the modeled CO distribution output here are some results of the modeling for a midsize house on a winter day with a class one generator running in a garage that has a living space above it. The blue trace is the CO distribution resulting in the living room and garage from a current generator and the lower yellow trace is that from one of the reduced CO emission rates that is in the range of what we think is technically feasible for this generator. This generator category when operated in an enclosed space. The hump in both traces is due to the generator running out of fuel after 10 hours of operation. Now to make estimates of the deaths for each CO rate that NIST modeled we applied these four criteria to the modeled COHB output. In other words these were the criteria we used to make a determination or assumption of death or survival. If the peak level was below 40% COHB we assumed that person survived. If the peak level was at or above 60% COHB we assumed that person died. And if the peak was somewhere in between these two levels then we took into account not only what the peak level was but how long the person would have been exposed at that level to determine survival or death. And just as a quick example of this before I go on referring back to that mid-sized house I spoke of a moment ago. These are the COHB levels that were calculated from the modeled CO distribution in the living room and garage. Again the resulting trace from a current class one generator is the top blue trace and that resulting from the reduced CO rate is the lower trace. By comparing them you can see that the occupants peak COHB level is much lower and the amount of time it took to get to that peak level is increased. We then estimated the deaths averted by determining the difference between the number of known fatalities involving current generator models versus those predicted based on reduced CO emission rates. In doing so we assumed an equal probability of intervention over 24 hours and we made this assumption because we don't have enough complete information to know exactly how long people were exposed when they died and when they were discovered. Furthermore the high ratio of injuries to deaths in the incident data suggests that occupants often do leave the exposure or are removed before death can occur. From our analysis and given our assumptions we estimate that 208 out of the 503 deaths that were modeled could have been averted by generators that emitted CO rates based on the draft proposed standards limit for each category. Further we expect an unquantified number of the 156 deaths from scenarios that were not modeled to also have been averted especially since some of these involved the generator operating outdoors or inside large structures such as churches and apartment buildings. The benefits that we calculated then were then used as one of the inputs into staff's preliminary regulatory analysis. This analysis included an assessment of the portable generator market. In our market analysis we identified about 80 manufacturers and importers of portable generators. We observed that sales fluctuate fairly significantly with major power outages. We estimate that annual shipments in recent years ranged from 1.2 million to 1.6 million units. We estimate that there were more than 11 million units in use during the period considered by our benefits analysis which again is from 2004 through 2012. And we also observed a recent trend in the market of increased sales of the smaller handheld and class 1 generators compared to earlier years. This last point is important because our hazard analysis indicates that smaller generators are more likely to be brought inside the home compared to larger generators which are most often used in attached garages. Two-thirds of residential CO fatalities occurred with the generator brought into the home so an increase in smaller generated use might increase the incidence of people bringing the generator into the home which is already the most deadly usage pattern. This slide summarizes the major findings of the preliminary regulatory analysis. The base analysis considers the estimated impact of reduced CO emissions on deaths and injuries in scenarios modeled by NIST and it suggests substantial gross benefits for most generators. The estimated gross benefits per generator shown in the first row ranged from $214 to $254 for handheld class 1 and single cylinder class 2 generators. However, the expected gross benefits for class 2 cylinder generators are only about $4 per unit. The estimated cost of the draft proposed standard shown in the second row were generally similar across the generator types ranging from $112 to $139. Given these findings and with the exception of the class 2 twin cylinder models, estimated net benefits per generator meaning the benefits minus the costs would be about $122 for handheld generators, $136 for class 1 generators and $101 for class 2 single cylinder generators. Estimated net benefits were a negative of $135 for class 2 twin cylinder generators. Total costs, benefits and net benefits were estimated by multiplying per unit estimates by the projected annual unit sales of each generator type. The aggregate estimated net benefits are $144.6 million annually. These net benefits do not include the unquantified additional benefits staff expects from the 156 deaths from scenarios that were not modeled. The preliminary regulatory analysis indicates that excluding class 2 twin cylinder generators from the draft proposed rule would increase aggregate net benefits to $153.2 million. The main findings of the base analysis were not altered by the results of our sensitivity analysis which considered variations in products expected product life, discount rate, higher compliance costs, the value of statistical life applied and the estimated effectiveness in reducing CO emissions for each generator category. For each variation that we analyzed, the overall estimated net benefits of the draft proposed standard were positive and as with the base analysis, class 2 twin cylinder generators were found to have estimated costs that were greater than the present value of the projected benefits. As part of our regulatory analysis, we also performed an initial regulatory flexibility analysis to assess the impact of the proposed rule on small businesses. We've identified 13 domestic manufacturers of portable generators which could be considered small based on the size guidelines of the Small Business Administration. For nine of these firms, portable generators could comprise a significant part of their business. We've also identified 20 small importers of generators that could also be affected by the draft proposed rule. Small manufacturers could face higher compliance costs per unit if they're fixed costs are spread over lower production volume and if they have to pay higher per unit costs for components that they purchase in lower quantities. Even all the analyses I just stepped through, we are recommending a proposed rule for portable generators that includes performance requirements and effective dates and compliance dates for four generator categories. Our recommendation for the scope of the draft proposed rule is to apply these requirements to portable generators powered by small spark ignited engines and exclude portable generators that are utility types such as trailer and truck mounted generators and combined generator welding machines. We also recommend an anti-stockpiling provision that is intended to allow manufacturers to continue to meet market demand in the period between when the rule is promulgated and when manufacturers would have to comply without allowing too much of a surplus of non-compliant generators after that period. And these are the performance requirements we recommend. The CO rates that are in this table are emission rates again that are calculated from CO emission rates emitted by a generator when operating in normal oxygen when each of six particular loads applied to the generator. These rates assume that manufacturers will have 50% manufacturing variation around the target production rate based on technical feasibility. For example, to meet a rate of 75 grams per hour, we think firms will set a manufacturing target rate of 50 grams per hour so that they can have as much as 25 grams per hour margin to account for production variation. Manufacturers can meet the proposed standard in any number of ways. It is a performance requirement and is not prescriptive. We are also not prescribing a particular test that manufacturers must follow to demonstrate compliance. We recommend this so as to lessen the burden on manufacturers. We have a list of options you could consider as alternatives to staff's recommendation. These include less stringent CO emission rates. Regarding this, when we looked at increasing the CO emission rates to levels that would reduce the cost, the benefits decreased greater than the cost decreased. So increased CO emission rates actually resulted in decreased net benefits. Exempting the Class 2 twin-cylinder generators. We offer this as an option because as we already said, this category of generators has a negative net benefit. However, we recommend keeping them in the proposed rule because we are concerned about the possibility for creating a market for smaller twin-cylinder generators. We would like to get comments on whether there is a real potential for that or not. Allowing the use of automatic shut-off systems as either a supplement to limits on the CO emission rate or as an alternative shut-off standard instead of reduced CO emission rates. We do not believe shut-off concepts are feasible at this time, but if we get comments that demonstrate otherwise, along with recommended requirements, we'll certainly consider that. Allowing later effective dates. Later effective dates and compliance dates would give more time for manufacturers to comply. We believe, however, that given the emission control technology we expect manufacturers would use to comply with the standard or already in the marketplace, we believe the dates we recommend are reasonable. Informational measures. The commission could decide to take additional informational measures beyond what we're already doing, such as telephone hotlines or public interest broadcast announcements. However, staff believes that informational measures alone would not adequately address the risks presented by these products. Or the commission could choose to take no action to establish a mandatory standard. In that case, manufacturers could market low CO emitting generators if they believe there would be a market for such products. However, generators with CO emission rates proposed by the draft standard probably would not be marketed in significant numbers voluntarily, at least in the short run. So now I'll close with staff's recommendations. CPSC staff recommends that the commission publish an NPR as drafted by the Office of General Counsel to reduce CO poisoning deaths and injuries associated with portable generators. That includes specific performance requirements for the CO emission limits of four different generator categories. Staff recommends an effective date after one year after publication of a final rule for manufacturers to comply with the requirements for class two single and twin cylinder generators. Staff recommends a compliance date of three years after publication of a final rule for manufacturers to comply with the requirements for handheld and class one generators. Thank you. We can now field any questions you might have. Thank you, Ms. Little. Thank you, Ms. Byer. We'll now turn to questions from the commission. It'll be 10 minutes per commissioner. I'm just going to start with a couple of comments. I've seen a lot of great packages since I've been at the commission, but I don't think I've seen any package that better exemplifies the best of what this agency is and what it stands for. It's so well done. It's so thorough. It's so technically sound, from a lay person's perspective at least. It's so clearly written. It addresses such a critical issue, and it does it in such a thoughtful manner. And I know that while the two of you are here presenting, I can see you have everyone or many of the key members of the team have your back. Mr. Hanway, Mr. Smith, Mr. Brooklyn, Mr. Natov, Dr. Inkster, and so many more who have worked on this. It's a really proud day for the agency. And I know in particular Ms. Byer, because you and I have had many discussions visiting many different manufacturers of how much you have put into this package and how much of your life you have given to it. So bravo for the work that was done. I think it's phenomenal. A couple of questions that I'll start out with, and I don't think I'll use my full round, but we'll see how it goes. You mentioned in the package four different auto shut-off technologies that the staff pursued. And I anticipate that one of the themes that we'll hear probably today and in the coming days and months is that even though we call this a performance standard, that it's actually very prescriptive because there are other potential technologies out there that staff, and I'm not saying I believe this, but I think we'll be told that staff is not accommodating in particular auto shut-off technology. So can you give us a fuller sense, please, as to what kind of work went into looking at those four different areas and why at the end of the day the staff did not feel comfortable recommending with all the work that you've done and all the different things you've looked at, why did that not make the cut? Sure, I'd be happy to do that. If I could just take a moment, I really would like to acknowledge the team first because much of what you heard today was a phenomenal team effort, and I would like to. I know you rattled off almost all the names, but I don't think it was all. It wasn't, and please use my time to credit this one. I'm sorry to cut into your time. This is the best use of my time possible. Please go. Okay, so of course Barbara Little and General Counsel. We have Susan Bathalon, Matt Brookman, Steve Hanway, Matt Natov, Sandy Inkster, Charo Krishnan, Han Lim, Barbara Little, Chuck Smith, Tim Smith, Robert Squibb, Andrew Trada, Troy Whitfield, and last but not least but many, many field investigators who involved, investigated these tragic incidents. Thank you for that. Thank you for allowing me to say that. Okay, so two year question. Yes, we did look at four different types of shut off concepts that we would automatically, the intent was to automatically shut off a generator before creating an unsafe exposure. And these included a global positioning system mounted system on the generator to try and infer indoor operation when it received, didn't receive signal from the outside, from the satellites. We also looked at CO sensors mounted on a generator to detect elevated CO emission concentrations around the generator. We also looked at a remote concept where there's a CO sensor located away from the generator. Ideally the consumer would place that where it's appropriate within their house. So if there were high levels detected in those locations it would shut the generator off by way of a wireless communication to the generator. And lastly as part of our contract with University of Alabama we tasked them to try and develop a sensor based algorithm that made use of just the sensors that were part of the EFI system. And with each one of those we had different concerns, sometimes overlapping concerns that amounted to negative impact on shutting the generator off when it is used in a proper location. We don't want to negatively impact the proper use of a generator tamper resistance. The concept of the remote CO sensor, first off you're relying on the consumer to find the most appropriate place to put it and he can easily choose to put it out the window if he thinks it's shutting the generator off without feeling any symptoms. The ability to shut off quickly enough, that was something we observed with the generator mounted CO sensing concept. We measured CO concentrations on the far side of the room that were 1000 ppm before the sensor that was mounted on the generator actually shut it off. So we had a number of concerns like that and so we pursued other paths. But at the same time we also in response to our ANPR we did get comments from generator manufacturers who were also opposed to the shut off concept. And we concurred with all their comments. We did get comments in favor of and we considered all that as well in addressing those comments. So the team decided to move forward with reducing the CO emission rate because it does address the hazard at the source. And like I say it's all, it is a proven strategy to help reduce the hazard. Thank you for that. Is it fair to say as we consider the safety hierarchy where you first try to design out a hazard and if that doesn't work you try to guard against it and if that doesn't work you try to warn against it. In simple terms is what staff proposed really at the top of that hierarchy where it's a design change and a CO sensor or an auto shut off. Is that really more of a guard? That's correct. So from a staff perspective at least for addressing hazards the staff approach is at the highest level of safety in terms of what you would prefer to see. Is that correct? That's correct. So recently we're all aware of this. I think everyone at least at the commission level and I know that you've seen this too that the portable generator manufacturers association the standards body that has been working on this or at least has the most of industry involved in it has recently pledged that they are reopening the standard and they are very aggressively moving toward their own CO reduction or their own way of measuring or meeting the issue. What did they know that we don't know in terms of auto shut off technology because they believe that they that that's a viable path. Is there any technical information that we are not aware of that you think that they possess that would lead you to believe that there's that is a viable path. We haven't we haven't seen that. We we have not seen that. They we did receive the same letters you and we did meet with the PGMA staff and representatives. But they spoke of and included in their letter only broad generalities without any specifics really about what they're thinking to offer in the way of an alternative standard. So given that that broad general framework we do think that it would take a long time to evolve those generalities into clear concepts before you could even begin to develop requirements performance and test requirements that we could then have to take time to evaluate would take time to evaluate the effectiveness of those. So we haven't seen any data to you know or or really anything beyond this broad generalities. Got it. And fair to say that you'll continue to participate in PGMA and UL's voluntary standards efforts even while the room making moves through the process if the commission were to approve the. Absolutely. Like I say we've been engaged since four years and I care to admit that we would not stop that. Great. Thank you. Well I think the chairman said it best but I'm still going to try to add a few thoughts about the quality of this briefing package and waxing philosophical just for a second. This is an indication of government at its best. We set up governments to protect citizens. I can't think of any higher calling for government than taking steps to protect especially vulnerable consumers. I've been following the commission as I'm sure everybody knows for over 40 years I cannot think of a better drafted package than I've seen here. This is the platinum standard for packages and I hope everybody here knows how much we appreciate it but not just us consumers at large. And this is also an indication that the free market can fail. We we all appreciate support and advocate for free market approaches but this is one where for a variety of reasons without government involved. You would not have industry taking action at least not in any sort of expeditious way. And on that very point I wanted to go back to the NPR which was published back in was it December of 2006. So that's roughly 10 years ago. And when we publish an NPR we are directed in the NPR to invite persons to submit a notice of an intention to submit or to draft a voluntary standard. A statement of intention to modify or develop a voluntary consumer products safety standard. Now it's been 10 years. Can you tell me has anybody before this September letter from PMA walked in and offered a statement of intention a plan to develop a voluntary standard to address the risks associated with this product. No we have not. So what we have basically is talk about a midnight submission. What we have is a midnight submission from the industry. I don't doubt their sincerity but I do think this is reflection of the fact that the free market doesn't always provide a response that we would need. And I would also make the observation that this is to me a classic hidden hazard. I remember when I had to go to Chicago to talk about the hazards of portable generators and staff said something to me on the order of one generator produces CO at rate. And I think they said of dozens of idling cars and I took that out of this statement because I didn't believe it. I said this can't be right. It can't be that these things are so hazardous that they actually would match or equal so many cars idling and now to hear 450 idling cars. It just tells us how little we all appreciate the hazards associated with this product. And the thing that's also chilling me is every time I read the news now and I see a storm coming and I storm a snow storm a hurricane. I have trouble sleeping because I know people are going to die and there's almost no way to reach out to them and say please look at the warning label please really believe it. Please don't put it in your garage please move it outside but we see even a small percentage but a real percentage of people who moved it outside the house structure. And it's still produced so much CO that it's caused fatalities. And so I guess one thing that I did want to ask you to delve into a little bit more length. We have been working with the industry through UL. Can you just give a rough estimate is it been that same 10 year period that we've been working with them. Yes I joined the we joined the standards technical panel when they first formed it in 2002. And we were still involved in it. The STP itself is not particularly active right now but the task group that we asked UL to form to develop requirements to feed to the standards technical panel has been very active. We sent that letter to UL in January of 2014. We kicked off meetings we do it all by teleconference meetings starting in May and we've had 27 meetings to date. And can you tell me how many consumer groups or consumers were involved in the development of the voluntary standard over this period of time. I'm guessing none. We as I said we what we do is UL and I'm not picking them out other than their convenient example. They try to give us balance but what they do is they set up a category of user producer general interest. And then when you look at the people who are in the general interest they are all retirees from the industry their consultants to the industry their expert witnesses for the industry. And dare I say their professors legitimately so but who are working for financial gain for the industry. Question is are there any independent technical is there any independent technical input from consumers or consumer groups in this process. And my sense is no. Yeah I don't believe so. Yes. I guess the other question this is just a pet peeve and whenever the I've talked to the portable generator manufacturers and I've always asked this question and I think I know the answer but to your knowledge no one who sells portable generators actually includes a little cheap CO alarm when they sell the portable generator. Is that correct. Yes OK. And this is one where I blame the lawyers in part because every time I've suggested it they say the lawyers would kill us if we did that. Excuse me but this is one of those situations where it seems to me even if it wouldn't be a perfect solution it might be some measure along the spectrum of safety that would reduce deaths and injuries. But unfortunately we do have a lawyer infested society of which I am a proud member. And I also want to commend you for putting in any stockpiling provision in previous standards we've we've not done that. And I think this is a truly important step forward because otherwise we're going to find for years afterwards legally sold inventory and that then undermines the ability to take steps towards safety. So again I thank you and commend you for a job extremely well done. Mr. Robinson. I have been privileged in my very long career to work in a variety of environments where excellence has really been something that was for and achieve. I have never seen anything out of any of those environments that is any better than what you guys have put together. It just was outstanding and I know that it's a huge team. You know I'm one of those that's always complaining about the inertia of government. This is not one of those times when you look at this package that was put together. It's just so impressive so thoughtful so clear and I thank every single one of you who were involved in putting it together. It really helps us do our jobs so much better. I was involved years ago in a case where I was representing Whirlpool in a furnace case that had caused injury and death. A husband and wife and even though I was representing the manufacturer in that lawsuit I became intimately familiar with the injuries that not just deaths because that that's something we all understand. But I think that some some do not understand the incredibly grave injuries. In this case it was a woman who was going to have a normal life expectancy but she was severely brain injured for her life from the carbon monoxide. So this is a problem that I'm so glad that we are are addressing and I have to say that and looking at your package just I start with the incident data that you discussed. And I can't help but think that even those these numbers that are in your package are pretty astronomical that they're pretty low. Given the fact that there are as you point out so many instances of misdiagnosis because the symptoms mirror other other illnesses and also the ones that are treated outside of emergency departments. And so I these these numbers are are really impressive in in the most serious way. The first time the first time I met you Ms. Byer was when I came out to the lab and it was after I can't honestly remember if it was just before just after a visit from PGMA and other industry people who came in to talk to me about how we needed to do something about that. The labels and I hadn't really focused on what the work was here at the agency until until that time. And I have to say that I was really I walked away so impressed with the passion with which you were going you and your team were going after trying to get a better solution than than just the labels. And I Chairman Kay just brought up your and you answer questions about the shutoff switch and why that wouldn't work. Commissioner Adler said that asked you whether you'd seen anything before the September letter that shoulder the industry was trying to address the amount of CO admissions. I'm going back a little a little more in depth than that just because of our conversation at the time of the visit. Could you just sort of describe for us the efforts that you have made. And I know it's been years but the efforts that you and your team have made to try to get industry to do something to lower CO emissions portable generators. Well and and I will say at the outset you know we were just looking for CO mitigation strategies requirements. Much better phrase. Right. Yes. So that's what we were advocating for beyond the labeling requirements a technical solution. And we started having that conversation back in 2002. So like I say when we first joined the standards technical panel and that's when the project first entered the CPSC operating plan where we were monitoring these deaths that were occurring in increasing numbers as the years went went on. So we did bring it to you else and we had you know as Commissioner Adler pointed out we had well representation of our manufacturing community on the STP hearing hearing us bring this in repeated meetings to the standards technical panel at you well. Outside of that in 2004 we held a public forum and we invited many stakeholders to attend to attend that meeting where we presented hazard patterns and hazard data that we had death data at that point in time. So these are just some examples we met with PGMA for the first time when they formed in 2010 and they formed to develop a safety standard and we asked them to address the hazards. So that was back in 2010 and like I say three years ago we've had active you know encouraging industry through our task group meetings with you well same thing. We have well representation and a lot of contribution good contribution from industry in those meetings. So that's been it has been very helpful and PGMA hosted a technical summit in March of this year. So we presented at that and participated and then also I traveled with Chairman back in 2013 when he was in then Chairman Tenenbaum's office. And he personally wanted to reach out to manufacturers to see if if they were going to help help us find the technical solution and we found from some of those meetings that some of them are doing. So the information that you gave us in the package in terms of the technological developments that came through University of Alabama through what we learned from this as these things were going on. I assume we were informing industry of what we were finding out and what they might be able to do. Yes absolutely. In fact we voluntarily solicited comments after we completed our technology demonstration program with our prototype at UA and then at NIST. We put that on our website and notified many stakeholders so we could get their feedback on that and and we did we got we got comments on that. So we've yeah we've we've really doubled down our efforts to keep the industry informed of of of our activities. And we've seen and you just brought up color which I'm going to ask you a follow up question on in a second but we've seen industry I certainly don't want to paint a broad brush about industry. But in terms of the standards committees what you just you just said in response to Commissioner Adler's question that we haven't seen any indication that they wanted to do something with the standards such as you're advocating in this NPR today. Is that correct. Yes. Now I know it's going to be more expensive to make these generators in a way that's safe. That's for sure. We all know that and you've done it. Your team has done an excellent job with the analysis on that. But have you seen any other I was going to say legitimate and that's a pejorative term. But have you seen any other basis other than the cost that industry has told you. I don't want to say industry. I'm sorry that the standards committees have told you is a reason that they will not do something or haven't until expressing this desire a couple months a few weeks ago. Why they won't do anything with respect to addressing the omission. I think it's a number of things. They've the comments against it or we haven't we haven't proven that reducing emission rates will actually reduce deaths. So we have done the analysis to support that we do believe that it will save lives. That's been one argument that you know that proof doesn't exist. There have been some complaints some comments on the NPR that we would negatively impact engine durability and reliability. But the analysis that we've we've looked at and others as well does not bear that out to be the case. In fact if you're looking for applying fuel injection as a technology to meet the standard that actually improves reliability. And durability. There are many benefits associated with that particular emission control strategy to meet the standard and manufacturers do tout those benefits as part of advertising engines that have that emission control technology. If we proceed with this NPR is there any reason that your work with the voluntary standards committees can't continue. Oh no we would we would continue working with them. Yes. I just wanted to ask a follow up question about the color press release. I'm almost out of time for this round. But you said that it was a low CO engine that was available. Can this this can this technology that they're talking about be used on all portable generators. They're they're they're advertising it as an engine so it can go in multiple products. And although color is now in the portable generator market as well. I lost my train of thought. Let me go back because I asked the question. Is it what this can this engine go in all portable generators. Yeah. It's it's a it's a single cylinder class two engine. So it would nominally power a generator that's in the five to seven kilowatt range from what I read about it. And yeah. And just one more because I know my time's up. Do you know if the color engine that they're advertising would meet the performance requirements in our NPR. I don't know that yet. Yeah. Thank you. Sorry. Thank you. Commissioner Robinson. My understanding from Commissioner Berkele is joining us by phone today is that she would like for this round for Commissioner Mojavec to go next. So Commissioner Mojavec. Thank you Mr. Chairman and congratulations to the staff in this package. Like the rest of the commission I was waiting in great anticipation for this and it absolutely did not disappoint. So congratulations on all the work to the entire team. And it was said much better by my colleagues. I do have some specific questions that I'll do my best. We've had a long day to get to the first have to do with handhelds in particular with regards to utility. So with the in regards to the weight. Now we acknowledge that the models with handheld engines often are valued and promoted for the compactness and light weight. And I spent a lot of time reading a lot of customer reviews and a lot of websites. So I could personally confirm that at least from what I've seen from my own study of consumer feedback. The package does talk about utility as it relates to EFI and the market the marketable benefits of EFI. And I think from my from my own perspective I guess lay perspective that with class ones and class two generators I would imagine they could more reasonably carry increased weights without a utility loss. But the testimony suggests that there is an increased weight and if the average weight might increase by 25 percent with the handhelds in particular knowing that they're handhelds so they're meant to be carried as opposed to on a dolly like situation where what might be an incremental amount of weight on a much heavier unit might be more impactful on a handheld. Do you have any comment on how increased weight for the reasonable ways that manufacturers might be able to meet the proposed performance standards outlined might impact utility specifically for handhelds. Right. We've been we've been mindful of that impact because there is there are additional components that have to go if they choose the EFI route for meeting the emission standard. And so oh my gosh. Let me just general utility. Oh OK. I know where I was going with this. Is it a different distinction with the difference. Sorry it's OK. We do. We are mindful of that. And that was one reason that we are suggesting a longer compliance date for the smaller generators because we we expect that manufacturers I mean we're seeing largely this emission controlled technology coming into the marketplace on the larger engines. And so we figured they can use they can leverage what they learn from applying it to the larger generators in that learning curve and then later apply it to the smaller generators. So that 25 percent weight that that you just mentioned that was discussed during one of the task group meetings and I think that's largely the bulk of that weight comes from the battery. And so there has been a look at battery less EFI. And so I think that needs to mature more and that would help address that concern. Thank you. And also specifically with handhelds and I recognize what you emphasize in the in the presentation with regards to their their increased market share or what we think is an increased market share specifically with the smaller. And I think you mentioned handhelds as well as regards to also our concern that there may be a greater propensity to bring that into into the home may maybe even by just by virtual looking at it because they're mostly an enclosed. It's an enclosed product that just quite frankly doesn't look as dangerous as a running engine with the open frame. So our proposed regulation aspires to reduce the deaths from foreseeable misuse around the dwellings. And though our staff says we can't quantify the safety benefits from the outdoor non dwelling use. We do think that there are ancillary benefits to be achieved there too. But with the handhelds in over 15 years we've still only seen two incidents related to fatalities around and in the dwelling. So I'm looking at the at the at the I should reference it specifically for our friends playing at home tab L of the package. This is the the the preliminary regulatory package the RIA. Anyway there's table two that has the that has some of the average engine displacements as well as the power ratings and watts. So bear with me with the handheld and in table two we identified that the average rated power wattage is is a thousand ninety four with the median of a thousand fifty. And I'm not a mechanical engineer but what I really wanted to spend time is it's really understand whether or not is a handheld generator likely to be used in that scenario of a power outage to run the kind of units that are in and around the dwelling as opposed to the campsite or the you know the the the tailgate party where it's a TV and speakers but you know AC units and refrigerators so I went through countless testimonials and got a bit a bit of a education whether or not it's the right education to have by reading that from consumers with regards to what the smaller rated products might be able to reasonably power. So I'm just wondering trying to understand if it's really not fit for purpose for that use but but would it really work in that scenario as I understand that AC units and refrigerators when they kick on they have as much of a six times power surge. So if you could explain that dynamic in terms of what you think that a handheld at those averages might be able to power and can we really see consumers wanting to use a handheld when they've got you know critical elements. Of their home they need to power. Sure. I would say that in an emergency situation if there's a power outage at somebody's home and they have one of these handheld portable generators because of they take it tailgating or camping recreational uses. They're going to use it. They may use it and we've seen it in the incident data to charge a cell phone. And so if they have it they'll use whatever generator they have. And the thing about these smaller handhelds is they can also be connected in parallel so you can if you have multiple ones you can actually up the the voltage the wattage to power larger appliances. Now granted yes how many people are going to have multiple ones of those around but people will use what they have if it is an emergency situation. So they may not be able to you know keep their refrigerator running but if they want to keep fire up their laptop and or charge their cell phone they bring it inside and they do that. The other thing with regards to the two fatalities you're right that three I think the two incidents. Okay that's right. The thing to bear in mind with that is there was a very low population of those handhelds in that benefits analysis period. I think they covered a very small fraction a much smaller fraction in the fraction of the fatalities associated with them and that's why they do have a net positive benefit associated with them. We do have concerns that if those become the more particularly if they are excluded from the rule that those will become very popular. And we we would expect to see not only more deaths but more if they're excluded from the rule. So that would be a real concern. Thank you. Thank you. I think I'll have time for just one other set of questions around the the benefit I guess that directly towards the benefit side of the benefit cost analysis and the premise of the three times carbon monoxide emissions at low C at low oxygen levels. We have been provided data whereby at 17% oxygen at what we recognize to be low 02. While performing the six stage test a manufacturer has experienced actually lower CO emissions than normal at lower loads for carbureted engines. And then lower CO emissions at higher loads for proposed EFI engine. So I recognize that we're saying that the lower oxygen levels we should be seeing three times the emissions of carbon monoxide. Why do you imagine in those tests that we we saw actually lower lower CO given those scenarios. Are you referring to data from our report or something data that was provided to the commission. OK. I know if I've gotten it you've gotten it. But I don't. Well have you have you done. Have you done that in running the the six stage test test with either prototypes or carbureted engines and experienced actually at some of those six stages. Lower CO or I was just surprised to see that because it's it's thematic throughout the package that lower at lower oxygen we're going to see three times CO emission. And I was expecting to see that as they were showing output by stage stage one stage two. Right. So the factor of three that we're referring to is the the calculated weighted rate. So it's not we're not referring to the factor of three at any one of those individual six modes. Six loads. It's it's the you calculate the weighted rate across all six loads. I see. I can I can help explain that maybe. Offline but it's the nation. But but I appreciate where where that's go because I have seen that some of the other stages. Yes. Significantly higher CO. So to be able to get a better appreciation for that will be something for later. So thank you. Yeah. Thank you Chairman. Commissioner Berkel I think we have you for one round. Is that correct. Yes. Thank you very much Mr. Chairman. And let me begin by echoing my colleagues really praise to both of you and all of the staff that have been involved in putting this package together. It's a remarkable package. And certainly the most complicated one I've seen and I think my colleagues agree with me that's probably the most complicated one ever. We thought that ROVs was a was a difficult one. But I think you put together a remarkable package and I thank you for all of your efforts and all of the work that's gone into this. This morning in the ops plan we talked about lithium ion batteries and over the course of the last few months we talked about nanotechnology and recycled rubber and how various arms of the government are doing different things. And so with regards to portable generators I'd like to focus on another government agency the EPA and just get some clarification with regards to what their jurisdiction is versus ours. I kind of telegraphed this morning in my closing statement. And so I just want to spend a little bit of time with this and I'm not sure this may be Ms. Little. Yes. This may be your problem. So let me just begin with regards to the EPA. Are they currently regulating carbon monoxide emissions from portable generators. EPA is focusing on large scale ambient air pollution with regard to CO emissions. Whereas in this draft proposed rule. EPA staff is intends to or seeking to address the acute poisoning hazard to an individual consumer from carbon and monoxide emissions. EPA does regulate on a broad national scale and they have national air quality achievement standards that different regional areas across the country must meet in terms of CO emissions. But they're not regulating from portable generators that we're attempting or seeking to regulate today. They're not regulating acute CO emissions from these generators. So they're only regulating the emissions from engines used in portable generators that are larger than what we're talking about today? No. Their focus is on the large scale sort of ambient air pollution in contrast to the exposure to individual consumers. And if I could, you know, the section 31 of the CPSA and I think this is maybe where your question is originating from. It provides that the Commission lacks authority to regulate a risk of injury associated with a consumer product. If that risk could be eliminated or reduced to a sufficient extent through actions taken under a number of enumerated statutes, one of which is the Clean Air Act. But the legislative history and the case law indicate that the CPSC is to consider all aspects of the risk as well as the remedial powers available to it. And the agency administering the alternate statute and make a judgment of whether that statute can reduce the risk of injury to a sufficient extent. And the Commission has authority if there has not been sufficient reduction or elimination of the risk of injury. And here, you know, as I was mentioning, CPSC would be seeking to address a different risk of injury from CO emissions than EPA. And staff believes that this risk hasn't been sufficiently reduced. You know, again, the EPA is focusing and their regulations are sort of designed to address the large-scale ambient air pollution, whereas ours would be seeking to address the localized risk to consumers from acute CO poisoning. And as Janet mentioned, there have been at least 751 generator CO-related fatalities between 2004 and 2014. So staff believes that the EPA regulations have not sufficiently reduced the risk of injury to individual consumers from CO poisoning. And we do address this in greater detail in our legal memo, and I'd be happy to discuss that with you separately. Good. And I just have a couple other questions along the same line. And as I mentioned, too, and I believe that the chairman is aware of this, but I can only stand for this first round of questioning, but I'll submit additional questions for the record following the hearing or the briefing. So just talking about EPA and their standards, do their existing CO standards that differ from what the staff is proposing? They have a different emission rate, and I think Janet could better speak to that. Okay. That would be helpful. Thank you. Yes, Commissioner Buerkle. So the EPA CO emission standards for these small spark ignited engines is 610 grams per kilowatt hour. So they have a sliding scale, if you will, that scales with the amount of grams per hour of CO that can be emitted as a function of the kilowatt rating, the max engine power of the engine. So that's 610 grams per kilowatt hour. We're proposing a standard that is in grams per hour and it's product specific. The EPA standards are based on engine testing on a specific test platform, and ours is what we are concerned about is consumer exposure to tailpipe emissions when that engine is operating in the generator. So that's a distinct difference because we don't care what the engine does when it's on a dynamometer. What we care about is when that engine is loaded in the generator. And I've heard, Commissioner Murovic, a little bit about the prescriptive nature of our standard. And I'll just offer that it is a performance requirement that we are proposing. Manufacturers don't have to use EFI. That's how we chose to do our technology demonstration. That's how manufacturers who are working towards the same end as us are planning to meet the standard or are working towards meeting the standard. But because we have a standard that we're proposing that is with the engine in the product, we don't care what the emission rate is at max engine power on a dynamometer. And so manufacturers could choose to use current existing carbureted engines, an oversized engine for the current size generator they put it in, and they just have to design it such that it doesn't operate at those high CO loads. So they can meet the standard any number of ways. I went with the way that we think the technology is headed, but that is a fine distinction. I'm sorry, I'm straying away from your question, Commissioner Buerkle, but that is a fine distinction between what the EPA does and what we're doing. The EPA does not have product specific emission rates. They just care about the engine, not the product it goes in. We're looking at the product it goes in. Okay, so I guess one of my concerns and why I mentioned some of the other issues that we're seeing being shared across government is making sure what we do or what they do doesn't affect or have some unintended consequence. And so if we're going to have a tighter CO emission standard like we're proposing in this NPR, how does that affect the ability to comply with the EPA regulations, limiting other pollutants that the EPA regulates such as hydrocarbons and the rest? That's a very fair question and concern. And I will say that we went to great lengths in our durability and demonstration program to make sure we were not going to be looking at lowering CO emissions such that we would drive the engines out of compliance with EPA standards. And so we've been cognizant of that all along. And in fact, our work as well as that of contributing industry members as well as the EPA data that I mentioned before that we looked at, when you significantly lower CO emissions, you actually also lower the hydrocarbons and NOx. So you do have more of a benefit than what we're concerned about. And so we are not, with our emissions standard that we're proposing, is not going to drive engines out of compliance. In fact, they'll drive the HC plus NOx lower, which is what the EPA emission standards primarily care about. Okay. So just so I understand, so tightening up our CO standard for emissions is not going to make it harder for others to comply with the EPA standards at the same time? No, that's correct. Okay. And I just have one other question. So I'm wondering if we promulgate the standard, the CPSC emission standard for carbon monoxide from portable generators, how does that fit into, does it preempt California and or other state emission standards? I think that's something that we'd better address at a separate date. Okay. So is that required that we'd have a closed discussion on that? I guess. Commissioner Verco, I think OGC staff can follow up with you directly in your office if that works for you. Okay. That's fine. I appreciate that. And I appreciate the consideration and going out of order and all the accommodations for today. I thank you all very much. And Commissioner Berkel, do we have it right that you'll not be joining us any longer and you'll submit questions? I think you said for the record. Yes, unfortunately, I have to drop off here. I've got some responsibilities, but yes, so I will submit any additional questions for the record. Understood. Well, we wish you the best. And again, we continue to think about you and your mom. I so appreciate that. Thank you very much. So we'll go to the next round of questions. I'm going to pick up, I think somewhere where Commissioner Morović was with the smaller size, the handhelds in the class one. And one of the pieces in the package was that you do have staggered, effective dates that you propose. And so I'm curious to know your thoughts from a technological feasibility standpoint, if you think that we could potentially bring that three-year date closer, because even if today were the date that the Commission were to approve a final role and it were three years out for those to comply, it seems to me that from what we've heard from some companies that they are moving that much more rapidly to be able to incorporate some type of compliant product, probably through EFI and a catalyst to meet these emission requirements. And I do want to call out in particular three individuals, Lee Sol, Michael Gardner and Mark Rowe from TTI and the work that they've done. They are PGMA members, TTI, but I think that they've played a tremendous leadership role in showing that this could be done. And I believe that it is feasible, maybe I'm wrong, but I believe it's feasible that we could potentially see compliance earlier than three years. And if we build in getting back to that this is an NPR and not a final role, it's probably going to be another year or two before the Commission is able to turn around a final role and vote it out. That's potentially four or five years from today's date. Do you think it's technologically feasible that if the Commission were to consider bringing that compliance date forward by a year that that could be met without undue hardship? I think that's a reasonable possibility. I do know that the companies who've been working towards this end have started with the larger generators because it does seem to be a more natural fit and that's where you do see this technology applied to these engines in the marketplace. But I do think there are rapid advances happening. Like I say, in the package there is already a handheld engine with fuel injected. It's not for low CO, but it's a concrete cutter made by steel that has. So there's one engine in the marketplace already with handheld engine with that technology. So I would say that's a real possibility. And certainly, and this obviously we can get into later with the General Counsel's office, but it seems to me that it's easier for us to have a tighter requirement on the effective date now in the NPR and then if we get comments we can loosen that as opposed to trying to rein it in a final role. That's just me speculating nobody needs to comment. One of the issues that was brought up, and I don't know if this is still an issue when we traveled and I don't want to name the company, but one of the companies talked about that they felt through the University of Alabama work that we would be requiring them to tune these too lean as close to stoke as possible and that from a functionality standpoint that was not realistic. Has that borne out? Do you feel like that's a requirement to meet the performance standard or can it be tuned more richly and can you deal with it either through the catalyst or in some other way to meet the performance requirement as Commissioner Buerkle mentioned not have EPA issues as well? Yes I do believe that there may be some engines that just cannot operate at the fuel control strategy needed to get the emissions lowest at the highest loads. There may be engines that due to their design that would be a challenge without doing engine design modification. Again going back to what we care about is the emission rate when that engine is operating in the product is what matters to us so if they can't operate at wide open throttle which the engine should not be operating there in the product if the circuit breaker and the circuitry is operating properly then you just have to take that into consideration and your engine integration strategy into the product so that it doesn't operate at those high CO high fuel flow load points. So I think that's a strategy to avoid that concern. But you don't feel like for somebody who has to for a performance standpoint if it has to be tuned more richly you do feel like that can still comply? That there are different strategies to comply with the performance standard? Yes. And one of the areas in the package that talked about the staff seeking more information on is the oxygen level and whether it's at normal range or at 17%. Can you talk about why staff might be interested in 17% what kind of data you're looking for and ultimately how you would choose between the two of them as to which one you'd be more comfortable with? Well first off the reason the whole oxygen level comes into the picture is because when you operate what we found from our testing with NIST when we did that house garage house testing that I spoke about earlier we noticed that where the generator is operating it starts to deplete the oxygen in that space. It's an air breathing engine, it's consuming the oxygen faster than the space can replenish it so the oxygen level starts to drop and in that physical testing in the garage we noticed that the oxygen did drop down for the carbureted units even as low as like 16% oxygen for the prototype unit the engine I'm sorry the garage oxygen level dropped somewhere in the 17% range. Then as part of that NIST testing to characterize what those emission rates were that were creating the CO distribution profiles that ran emission testing that's entirely different than how the EPA does it with running the generator in an enclosed single zone space and calculating an emission rate as it's operating there and in the course of that testing we noticed that the emission rate as soon as the oxygen starts to drop the emission rate starts to climb and it reaches the peak around 17% oxygen based on the testing that NIST did as well as Matt Brookman did here in our combustion lab on separate generators. So we felt 17% is a reasonable oxygen level to expect in these incident scenarios to occur so that's where the 17% came in but what we also found with our prototype and prototype testing at NIST was that originally and a lot of the work with the UL task group was focused on it appeared that the prototype generator emission rate did not climb as the oxygen dropped but the more we analyzed other data that we had on different generators that are in the marketplace now with fuel injection on an enclosed loop and we analyzed the modal data that Commissioner Mo Rovick was talking about we saw CO emission rate increase even on generators that were operating in closed loop so it kind of rattled our philosophy that the emission rate would not climb for a closed loop fuel injected generator. The whole intent behind having originally trying to work with the task group to develop a test method for a low oxygen environment is because we wanted to make sure that if we have a performance standard for generators to meet to reduce deaths and injuries and most of our deaths are incurring in reduced oxygen environments we didn't want to have few different emission control technologies that could meet it in normal oxygen and then you go and put it in an incident scenario and one of them has the CO flight as we were calling it in the task group and the other one doesn't, one's going to address the hazard and one's not so that was the whole intent behind spending a lot of time in the UL task group to develop a reduced oxygen emission rate test but the more we looked at the data we couldn't really rationalize the burden for doing that although Matt Brookman did a wonderful job in streamlining that test process but we felt we didn't really have the support needed to justify it so we right now are proposing, and again we are proposing how we would test for compliance we're not requiring manufacturers to do that but if they wanted to do it it would just be a normal oxygen test but if people come to different conclusions than what we have we want to hear about that because we do want a standard that is going to address the hazard that's our primary goal Thank you that was very helpful and in my remaining time I just want to ask from what you're seeing on the marketplace with EFI do you see mostly closed loop and I realize that they're not being put on the market for the purpose of reduced CO there's other reasons why it's closed loop fuel efficiency or something else but are you seeing more open loop or more closed loop EFI engines just as a matter of course? Well it's hard to say because that's not usually a feature that they are going to advertise in the, that's not something that the typical consumer is going to understand what that feature is so and they may say closed loop but it's only for certain, certain mode points so it's a little bit of a mixed bag Okay, thank you very much Commissioner Adler Thank you very much and I'm glad to hear the explanation that you gave to Commissioner Berkel about the relative approach that EPA adopts versus what the staff is doing because when you're talking to Ambient Air you're talking about when I step outside and take a deep breath what's the degree of CO in the air and that comes from a variety of sources but they're really not looking at the acute hazard in closed spaces in people's homes and even if they were they'd still have to convince us that they're addressing it to a sufficient extent and I think 750S is a pretty good indicator that it's not but just down the broader level we have, have we not consulted with EPA have they ever voiced an objection to what the commission is doing with respect to addressing CO from portable generators? No they have not they assisted us with the development early in the University of Alabama program they actually helped us write the statement of work and evaluate the proposals that we got in and after the contract was in place they provided technical advice along the way and consulted with them a number of times so they do well know what we've been doing and we have good communication throughout the development of the NPR package which is yet another indication of how effective the staff has been and how comprehensive the approach staff has taken I did want to talk a little bit about a point that Commissioner Robinson raised because it's something we've heard and we're going to hear again and that is you're raising the cost of these generators but one thing I want everybody to understand is these generators are imposing a societal cost now by your calculation if I read it correctly it's $820 million and that's from fatalities and non-fatal injuries that's a societal cost that is being incurred now and so what we're saying is some of those costs actually ought to be borne by the people who are making the product and profiting from it and also from consumers who are getting the benefit of a safer product so it's not as though there aren't costs already and when we're done with this I'm guessing that those costs which are being externalized will now be appropriately, rationally internalized with respect to the price of the product so that's point one the second point I would make and this is just the general observation requires no response and that is we've seen the cost at their highest just as a standard is developed and this is a tribute to manufacturers in the industry they're almost always clever enough especially when there's mass production to come up with much cheaper ways of making the product so although the staff properly and conservatively is estimating cost of these products I would guess within a year even you're going to see the price of these products drop dramatically that's certainly been the experience we've had with almost every product that commission has regulated but I did have one question and that is it's my understanding that EFI and catalysts have been around a fair amount of time especially with respect to cars is this an expanding technology through engines because I see that in ROVs we're getting electronic fuel injection is this a technology that's well known and trying to think of one that even manufacturers aren't using it would have access to fairly readily yes absolutely yes and I guess one other question is when we're talking about the efficiency of a portable gas generator would the technology that is likely to be adopted would that have a negative neutral or positive effect in general on the efficiency of this product have a positive effect they claim higher fuel efficiency so there are actual benefits immediate tangible benefits other than not dying to consumers from purchasing these products right okay thank you very much Commissioner Robinson I think I just have a couple questions in looking at the cost benefit analysis that Commissioner Adler was obviously referencing at some level the I note that with the well let me just ask it this way the class 2 twin-cylinder generator twin-cell generators or twin-cylinder generators if we were to exclude those what effect do you think that would have in the marketplace well we do have a concern and we don't know how real it is but we do have a concern that we might see right now twin-cylinder class 2 engines powering generators are nominally rated 9 kilowatts and higher they're fairly large units we have a concern that maybe we would start to see generators more down in the in what we call the storm generator which is in the 5 to 6 kilowatt you know just in the smaller than below 9 kilowatts than what we're seeing now possibly coming on the marketplace with twin-cylinder engines in them to avoid having to comply if twin-cylinder class 2 engines were excluded from the role because then they would be less expensive to purchase well I don't when we don't know that actually I don't we don't it may not be that that's an advantageous strategy to take so we'd like to we'd like to receive comments on that I'm commissioner Adler in the last round referred to CO alarms and my reading of the package made me think that CO alarms were not the answer to this problem could you address that CO alarms we definitely CPSC promotes and encourages use of CO alarms to address any source of CO poisoning around the home so we definitely agree on that point that CO alarms are important but we we think the most effective way to address the hazard and I think this gets back to chairman K's point is design the hazard out of the product if you can do that that's the first strategy the CO alarm is the is the backup if you have you have nothing else to do you know to address it and I was looking at the tab a page 76 you don't need to refer to it but it just was talking about the incident data with respect to the analysis of CO alarm usage and it like it was the answer right right I guess I focused on that because it came along with the as the only person on the commission who's who devoted her his or her life to practicing law in the courtroom I take umbrage to comments like lawyer infested we don't dismiss all doctors because they're contributing to an opiate opioid epidemic in the country and I don't think we should dismiss all lawyers as being bad so with that I have nothing further thank you commissioner my roving thank you mr. chairman commissioner Robinson is a former member of the testing industry we're also um much maligned for many many years based on CPSIA imposed mandatory testing okay sorry I don't want to exhaust my colleagues and staffs thank you thank you mr. chairman one question all right this year in met in March we actually April we at the CPSC passed a retrospective review plan and in article five of that retrospective review plan one of my favorites at the agency it speaks to providing for retrospective review provisions and new rule makings ex ante retrospective review that is in consideration of future rule makings will conceive in the front end how we're going to measure the effectiveness down the road what will be those metrics incorporated so I will note that we pass this on April Fool's Day so I hope the answer isn't the jokes on me in terms of whether or not I should or should not have expected to see a retrospective review plan in the rule can you describe whether or not the team considered it rejected it or if or why there isn't a retrospective review plan in this new rule pursuant to the commission's wishes not sure we're prepared to speak to that not sure if it's appropriate directed yeah well we've got a proposed rule in front of us we have a plan the commission voted five oh saying in new perspective rule makings we're going to embed retrospective review as a matter of policy for new reviews and it doesn't exist so that I guess there's no answer why that doesn't exist not at the moment all right I've stumped you with that I apologize maybe on more mundane matters the cost benefit analysis so you might need one of your experts for this Miss Byer but if you can field them and have at it with regards to the the cost and tab L of the preliminary regulatory analysis we recognize that some manufacturers that we discussed just at our preliminary direct costs were understated and we've I've heard personally in this meantime that they might be in fact as much as double what we have identified as direct costs did we use the EPA estimates or in fact did we reach out directly to component manufacturers themselves to understand and get an identification for the direct cost that we incorporated in the BCS yes commissioner we did use the EPA estimates we but we also reached out to firms this summer as part of the effort with a contractor analysis staff participated in interviews with manufacturers trying to get their estimates of cost of these components and in general they ranged quite widely from the four firms that provided information but I'd say that probably in at least two of the manufacturers for the individual components that we provided the cost estimates for they were sort of in the same ballpark as ours so I gave us some confidence that that our estimates were still reasonable I'd say that the EPA estimates their result of extensive analysis and research and and widely reviewed and because it appears that the costs and to the industry haven't changed all that much the EPA estimates were provided and we've also taken steps to update those costs based on producer price index that we think is as appropriate for this product so ideally we would have more recent estimates to base our analysis on but we think that with the adjustments we've made that are and the information we've considered by the firms that are provided by the firms that our preliminary cost estimates are reasonable and supportable thank you Mr. Smith and I recognize also you have a sensitivity analysis in here as well yes in general the firms that did provide information somewhat provided somewhat overall higher costs than we have estimated I'm sorry I don't mean to interrupt you but we'll have noticing comment too our sensitivity analysis is going to get real annoyed with me if I don't start moving out if you don't mind we get a count for a 50% increase in costs I do recognize that one of the other assumptions or one of the presumptions that were made to us from manufacturers was that there would be a disproportionate cost impact on handhelds and the smaller versus the larger engines to incorporate EFI as well as the catalyst because of the reasonable limitations with regards to size and weight yet I recognize in our preliminary analysis the costs don't reflect that do we just not agree with that point of view that or are we surprised that there are actually lower costs to incorporate the same technology and smaller units with reduced housing and geography to work with I think we would benefit from more specific information on the handhelds it's a very small part of the market right now and although it is growing somewhat still only about 1% of the market in recent years sales and but we do recognize that there are some features of those generators that create problems additional problems for incorporating these features so okay Miss Byer this is a specific question with regards to well I've got a couple with the potential for loss consumer surplus per unit and it gets to a point that you're making I think with the chairman and I'm not a mechanical expert or a mechanical engineer but I recognize in the program stated objectives that we sought out a prototype that we quote not negatively impact the engine's power output durability maintainability fuel economy and risk of fire and burn yet I notice in the package there are several places where we would expect that generators in order to perform at the proposed performance elements would need to run at stoichiometric and I was I could only partially understand some of the dialogue you had with chairman Kay but is do in fact we expect that to achieve the performance outputs that we're recommending that engines would need to run at stoichiometric and if so how would that or would that impact any of the stated objectives particular with durability or power anything with with the fuel injection operate if close loop you're talking about stoichiometric fuel control that's where stoichiometry is where you obtain your lowest CO emission rate in the spectrum of air to fuel ratio so it doesn't our findings are that it doesn't impact power it improves fuel economy and again we were acknowledging that maybe not all engines not all engines are built the same so some engines may not be able to be modified enough in design either in the combustion chamber or the valve seats to tolerate operation at stoichiometric again if you can have have the engine operate in the generator around those load points where you need the fuel enrichment due to concerns about engine durability and temperatures then you just need to take that into account as you integrate that engine into the generator under the same idea of the objectives provided some have indicated a lack of power loss of power as a result of achieving the performance elements for lower CO as identified in the package do you agree with that and if so from the economist point of view should that have been incorporated into loss consumer surplus per unit if in fact these new performance elements will result in a loss of power by the units well I'll just the first part of that we haven't heard that specific concern from manufacturers or at least I'm not recalling it at the moment I can go back and check on our prototype we achieve nominally slightly less than the max engine power when it was in the carbureted configuration but it was it was very close to what it had been and I built we haven't heard that from the manufacturers who've been working we've gotten a lot of proprietary stuff I know you have too so that's what forgive me for the lack of transparency but we'll be able to discuss that offline but from the economist point of view if it was in fact determined that there was a loss of power we would incorporate that logically into loss consumer surplus I'd say we'd make an effort to do that yes I'm not sure if we'd have adequate information to make to monetize that thank you thank you Mr. Chairman I'm going to go ahead and yield if you'd like my 10 minutes to you since you're on a roll thank you I'll go as fast as I can thank you for your patience I have another question we talked about the increased weight of the handhelds and what might be a 25% increase weight would that also be something with regards to the cost analysis that can be analyzed as impacting loss consumer surplus and factored into the benefit cost analysis for the handhelds in particular could that be monetized I have my doubts that it could be to tell you the truth but we could at least make reference to it in the final analysis I would hope we might I mean if you think of it's a handheld product and you know it's meant to be carried and carried by one's hand and if they just can't make it to their tailgate party or out to the campsite from car camping out to the tents that's as Janet mentioned earlier I think during this process of adjusting to this regulation should it become a regulation I think there could be features that will become available that would mitigate those problems for the smaller generators such as the battery less EFI technology that's being used in some products such as that would not add additional weight to the product that would reduce the amount of additional weight that would and the reduction of no long without a carburetor as well I imagine too okay thank you now we Mr. Smith we talked a little bit about price elasticity privately in advance of the meeting and I recognize that especially with the handhelds many are marketing these as for recreational use and that's the main marketability of these particular products and I do recognize in our preliminary benefit cost analysis we used a surrogate by using household appliances with the assumption that we'll have inelasticity with regards to price demand for the product that's based on I think the assumption that these products are being used generally across all generators for a power outage situation but for the handhelds where the marketing is somewhat distinctly different with regards to not DIY projects not construction purposes but specifically for camping and recreational usages would you recommend a price elasticity figure that is more that is more properly associated with recreational use products as opposed to the other because it was broken down by class and I'm just questioning whether or not there should be greater elasticity to price for the handhelds versus the other particular classes of generators at this point I'd say that it's possible that given the different nature of the smaller generators that their use patterns might involve different price elasticities and you might say that they might be more elastic than we've assumed for generators as a whole and the result of that if we increase the price elasticity that would result in greater reduction in purchases relative to the increase in prices so we could attempt to account for that thank you very much and on the other side of the ledger Chairman Kedeman continue I've yielded my ten minutes to you with regards to the potential for increased consumer surplus per unit with fuel efficiency we've heard that the fuel efficiency savings is a recognized benefit for the commercial users but with an estimated 11 year product life cycle I wonder why we can't monetize the increased consumer surplus from increased fuel economy I think we might be able to do that if we have more information about usage patterns although the price is a little bit lower but I don't think it's a good idea because I think it's a good idea but I don't think it's a good idea to have more information about usage patterns although the products might last 11 years on average if consumers purchase them for power outages and the like it might be that they might not get much use over the course of a couple of years they might so in terms of fuel efficiency might be 20% greater efficiency we need more information on usage to calculate what that might be in the course of the life of the product if the commission does move forward perhaps you might want to consider asking or soliciting increased questions we've identified that as an ancillary benefit but I think it's something we should try to monetize but it might be monetized and perhaps we're undervaluing the benefits by not really being able to look at the reduced cost of the consumer by using a more fuel efficient generator I'd agree with that but I think it's a good idea to try to get the noise produced by the generator now it does and this may be for Janet first excuse me Ms. Byer is first does the EFI and catalyst increase or suppress noise I think the noise is mostly related to the muffler that you use primarily the engine noise and I can't say I understand thank you the last question I have is with regards to generator location so Ms. Byer we recognize that 25% of the 565 deaths that occurred out of fixed structure home occurred when the generator was in the attached garage or the enclosed carport now when a generator is located in a garage do we know how often the garage door was found open or partially open or closed I understand our models and the rationale behind those models but that's a pretty big sample there 25% of 565 should have been paying closer attention around the 140 mark or so do we know that I think I could defer to Matt Natov here but my recollection is it's not frequently reported you can correct me if I'm wrong Matt it's not frequently reported okay very well then that does it for me so they're saying that it's frequently it is a fully closed or when that information is available thank you very much I appreciate it and thank you Mr. Chairman for yielding to my colleagues hope I didn't exhaust your patience I have no further questions Commissioner Adler no further questions Commissioner Robinson anything else great having heard no further questions this concludes this public meeting of the United States Consumer Product Safety Commission thank you