 Hello, my name is Nicola Tecli and I'm a field application engineer at the Steam Microelectronics. In this video, I'm going to talk about the Digital Power Controller ST-Energy 011. I'll start with the high-level features of the ST-Energy 011, Digital Combo PFC and LSC controller, and then I'll describe more in detail what there is inside the ST-Energy 011 and why it is an excellent choice for high-efficiency power suppliers. In the second part of the presentation, I'll review the complete evaluation and prototyping ecosystem available from ST-Energy 011, and I'll show you how to use the graphical user interface to configure the numerous controller parameters. ST-Energy 011 is a digital combo IC that integrates both PFC and LSC controllers. It combines key features of typical analog controllers like the integration of high-voltage startup, integration of the gate drivers for both PFC and half-bridge LSC MOSFETs. It doesn't require to write any firmware, and it can achieve very low standby power. On the other side, it has key benefits of digital controllers like high-level of configurability, numerous fully configurable protections, UART and I2C communication interfaces for remote monitoring and black box recording. Just FYI, ST also offers pure analog and pure digital power controllers, and you can find them going on our website at sd.com. Before I move forward with more details about the ST-Energy 011 controller, I want to give you some information about the typical applications for this combo controller. The typical characteristics of the target applications are a power level from 90 to 300W, the presence of a main CPU that will communicate with the ST-Energy 011 via I2C or UART, the requirement of meeting stringent energy saving regulations. Based on those characteristics, the typical applications are power adapters for computers and TVs, power adapters for industrial and medical equipment, LED drivers that require connectivity for remote monitoring and control like straight lighting. Now let's go more in detail on what there is inside this device. In the block diagram, you can see the main building blocks of ST-Energy 011. On the top right, you can see the built-in high-voltage startup circuit that allows this controller to start without the need of any additional housekeeping power supply. On the top left, you can see the gate drivers for PFC and LFC, with one amp peak current for the most demanding of MOSFETs. These two features combined are key to reduce the bomb for the power supplier. Most digital controllers do not have these blocks and require external components to take care of these functions. On the bottom, you can see the 8-bit microcontroller running at 60MHz, with 10-bit high-speed ADC to allow for precise control and monitoring of the digital power supply. The device also integrates a nonvolatile memory that stores the numerous configuration parameters. I'll cover this more in detail later. Finally, this device has UART and I2C communication interfaces that allow the configuration of the digital power supply as well as remote monitoring or black box logging. Let's take a look now at the high-level functionality of the two logical macro blocks present in the STNRG-011. The first block is the front-end PFC controller that is implemented with the digital PI and two SMEDs. SMED stands for State Machine Event Driven and it is a ST proprietary IP block that is present in all STNRG devices. The inputs to this block are the output voltage and the PFC current information coming from the auxiliary winding of the PFC choke. The mode of operation of this PFC is based on ramp-enounced constant on-time, which is a modified version of the standard constant on-time with a proprietary ST algorithm to compensate for the current caused by the input filter capacitors. This algorithm improves the power factor and the THD. In order to optimize efficiency, the PFC switches from transition mode at high loads to valence keeping at low loads and burst mode at very low loads. In addition, it also implements an optional skipping area mode that essentially stops the switching activity when the input voltage is around the zero crossing. The second macro block in STNRG-011 is the LSE controller. Here, two SMEDs are utilized to manage the MOSFETs in the half-bridge. The LSE controller is based on one more ST proprietary IP called the timeshift control. Timeshift control provides two major advantages compared to a direct frequency control. First, dynamic load performance is greatly improved. Second, it provides anti-capacity mode protection. As an option to further increase power supply efficiency, the rectifier diodes on the secondary side can be upgraded to MOSFET-based synchronous rectification controlled by ST devices SRK-2000 or SRK-2001. ST provides a fully integrated evaluation platform for testing of STNRG-011 controller. In particular, we offer a demo board named EVAL-STNRG-011-150 with a GUI that allows customers to easily evaluate the controller on a real working solution. This demo board is a complete power supply that includes also synchronous rectification using SRK-2001 for optimal efficiency. It delivers 150W at 12V and it accepts wide-range AC voltage from 90V to 264V. The two major components of the ecosystem are the interface board and the PC graphical user interface. The interface board provides electrical insulation between the PC and the STNRG-011 board, as well as the ability to utilize the USB port to communicate with the STNRG-011 via the built-in UART. This board also provides the correct voltages and signals to program and read the E-square problem inside the device. The graphical user interface gives easy access to all the parameters of the device and it also provides real-time status of the power supply. The graphical user interface provides control of 85 parameters to tailor the power supply performance to suit the application. These parameters include the protection behaviors, filtering PFC control, LSC control and Burks mode parameters. This graphical user interface and evaluation toolkit can serve as the launch pad for your own high-performance digital power supply using STNRG-011. To close the overview of the STNRG-011 evaluation board, I want to quickly mention its performance. As you can see in the main table, the efficiency of the board is good across the full range of output power, but the most impressive performance is at low and no load. The no load consumption of 17 milliwatts at 115 volt AC and 93 milliwatts at 230 volt AC is really low and it is very difficult to achieve with any other digital controllers. In order to help anybody who gets this evaluation board, I'm going to show step-by-step how to access and modify the configuration parameters of STNRG-011. Later, I'll also show you a few live examples using the actual GUI, showing the impact of the changes on an oscilloscope. Let's start with the step-by-step description. First basic step is to connect the interface board to the PC using the USB cable and to the evaluation board using the six-wire ribbon cable present in the kit. Second, you need to make sure that the evaluation board is not connected to the AC line, otherwise you will not be able to proceed. Now go to the GUI and click on enter ATE mode present in the ATE mode menu. After this command, you will see on the interface board a red LED turning on, which means that the interface board is supplying power to the STNRG device. At this point, we are ready to access the parameters. Click on NVMe operations in the tools menu and then click on read from IC on the NVMe operations window that just opened. Please note that if you don't click on read from IC, the GUI will show you its own default values for each parameter, but they don't have anything to do with the values in the device. Now you can modify any of the parameters of your interest. When you're done with the changes, you need to write the changes by clicking on write to IC. After you do this, the parameters are written into the non-volatile memory and at the next power up they will be used by the device. In order to do the test with this new configuration, you need to exit ATE mode, which is done by clicking on go run mode in the ATE mode menu present on the main graphical user interface window. Now you can connect the AC line and test the board with the new configuration. As briefly mentioned earlier, I will also show you a few examples of parameter changes and verify within a oscilloscope what is the impact of such changes. The first parameter that I will test is related to a protection and I will show you the different behavior for latched versus non-latched protection. I selected LSE OLP that means overload protection as it is one of the simplest and safest to test. The second parameter that I will test is the burst mode entering threshold, which modifies the power level at which the board will enter in burst mode. The last parameter that I will test is the skip area parameter that defines the power level threshold to apply a power saving technique that stops PFC switching activity when the AC voltage is low. As I mentioned, I'm going to cover a few changes of parameters and verify how this will impact the behavior of the evaluation board. So we're going to start with the latched versus non-latched behavior of the protections. We'll use the overload protection of the LSE first. As we saw earlier, in order to change any parameters, we first need to enter in ATE mode and then modify the parameter, but the power needs to be disconnected. So the board is now running and switching off the power, the AC line, then I enter in ATE mode. It is now in ATE mode. I see the VCC for the STNFG and I see the red LED on the interface board. Also the oscilloscope you see that there is no activity anymore. Second, I go to NVM operations to access the parameters. Here we can see the LSE OLP behavior that is a parameter that we're going to change. Let's start with the non-latched mode of this protection. So I changed it. I click OK to make it effective and then I write the parameter into the non-volatile memory. It asks if I'm sure to write this parameter and I say yes. The device is now updated and I can close the parameter window and I can go back into normal mode of operation. So we go around mode. I can now give back power to the board and I can see that it starts working at 80 W. I now increase the power consumption, the load and I exceed basically the maximum for the protection. When this happens, as you see, the board stops. But after some seconds you can see the activity on the oscilloscope. You see that it tries to restart. We can actually trigger on the current so that we see this behavior better. As you see for several seconds there is no activity. It then starts sometimes the voltage, which is the 12 volt. You can see here the yellow line is the 12 volt line. It runs, gets almost to 12 volt. Here there is a zoom around the 12 volt. But then it stops almost immediately. So in this way it is actually much better and much more clear the behavior right now for the board. You can see that every several seconds the ST energy board tries to restart. It says that the power is exceeding the maximum power set on the protection, on the OLP protection for the LLC. And at that point it stops. Let's try to reduce the load and verify that actually it does work. It did work for a little bit earlier. Okay, it was just because the power was still excessive but not much. And so it did start, but then it stopped. Now it is actually working fine. It is below the 150 watt that is the maximum. The protection is actually a little higher. As you can see, it tried to restart multiple times until the protection was not triggered anymore and now the board is working. Let's now try with the latch mode and see how it behaves. I switched off the power supply. I go back to ATE mode, go to the parameters, change to latched the OLP. Now I didn't read back from the device because I already did it earlier but otherwise you should read from the device. I might not have done it also earlier because I had it already opened. So I already had the values that actually were coming from the device. I did write the new configuration to the device and now I go back to RAM mode and power on the board again. It is now working at 140 watt and it is working fine. Let's now exceed the maximum power. Now it did stop. As you can see, there was an increase on the input current, the red line, and there was a slight decrease of the output voltage because of the overload. Now, even if we wait forever, it will not try to restart. In this case, the latch protection will keep the device stopped until it is powered off. In order to power off the ST energy device, in this case, it means that you need to remove the AC line otherwise it will continue to stay powered. Let's switch off the AC line, go back to normal level of output power, switch on again, and now it does restart. So the same behavior, latch versus non-latched, can be set for basically all the protections. Let's now work on the second test related to the Bose mode entry threshold. Let's check the configuration right now and then see with the oscilloscope and the actual test on the board what is the behavior. So we enter in ATE mode, we open the parameter window. The entry level for the Bose mode is defined by the level of the feedback to the LLC. So this depends essentially on the actual implementation on the board. So you will need to read more in detail the user manual and also the NVM parameter document to better understand the details of the configuration. But we'll basically test this configuration and then change it and see the different behavior just to give you a feeling of what this means and what is the different behavior at the level of the device. So the level right now is this 314 digital 767 millivolt on the feedback. So we keep this value for now and we check the behavior on the device. So we go back to RAM mode, we switch on the board, see that it starts right now. The power consumption is about 14 watt or so, 15 watt. Please do not consider what you see here on the window on the GUI. When you are at low power consumption there is a significant error on this because this is just an estimate and it is pretty accurate at higher power but very low power becomes really not very accurate. So let's try to decrease the current. Right now it's not working in Bose mode. You will see the different behavior on the oscilloscope. So we are now decreasing the current. It is still in continuous mode. I am at about 8 watt right now. I'm decreasing more and now it enters in Bose mode. So right now it is about around 7 watt and I can then decrease even more the power so you see even better the Bose mode behavior. Essentially when it is in Bose mode right now we are around 1 watt of consumption. When it is in Bose mode the activity of the LSE and PFC is not continuous. It is only working when it's needed. Basically the LSE and the PFC are giving power to the load to the output capacitor. Let's say for some time very brief and then when it goes above a certain threshold the voltage threshold it stops and then it restarts and then stops again. So in this way even though the output voltage is not perfectly stable it is actually still pretty good. So right now I'm zooming at 0.5 volt per division so it is really not much the ripple. It is not perfectly stable and also the input current is not a sinusoid but it is very low power and the Bose mode allows to optimize the quite significantly the efficiency at very low power. So we have seen earlier in the presentation that this device is very efficient at low power and it's super low power and this is one of the ways that it can achieve that high efficiency. Let's try now to change the configuration of the Bose mode and see at what level it will enter in Bose mode. I switched off the power supply. I'm now going back to ATE mode. Let's see the parameter. Let's change the feedback level to about 282 say 688 millivolts. Okay we're right into the device and now we go back to run mode. Switch on again. Let me restart from the maximum power that is about 14 watt 14 or 15 watt and let's try to decrease slowly. Right now I'm already at around 25 watt and it is still not in Bose mode. So it now decreased the level of power at which it will enter in Bose mode. Now the benefit is that you see the current the input current and also the voltage are very nice but it is consuming quite a bit of power just because the PFC is switching and also the LSE is switching to deliver very low power to the output. Let's decrease more. Right now we are just below 2 watt and it now entered in Bose mode. So you can see that you can select trade-offs. You can define at what level of power you go in this deep power saving mode. There is some ripple on the output voltage and the input current is not perfect but there is a significant advantage in the efficiency of the power supply. In the last test I'm going to show the parameter area skip that is another power saving technique that you will fully understand in a moment. Basically the PFC switching activity is stopped when the AC input voltage is around zero volt. So let's check the configuration right now for this parameter. The configuration right now is zero. Zero means disabled basically. Let's see the behavior first with the current configuration. Let's switch on. Now it's zero load. Right now we are around seven watt also. This is about 20 ish watts and as you can see at both 25 20 ish watts or seven watt the input waveform the current waveform is the same. Basically the switching activity is going all the time. It is never stopped. And the only noise that you see is probably where there is maybe some more current that is the one by the PFC. But in any case you can see also on the graphical user interface here in orange you can see the value of zero for skip area regardless of the power level. So now I increase the power level I will decrease it and it always remains to zero. You will see later instead that this will change and you will see on the oscilloscope the different behavior. Let's switch off the board and increase the value of the parameter. Go to about 512. Need to write it switch on again. Now go to seven watt. You can see the different behavior basically on the graphical user interface you see the skip area at level two this point and on the oscilloscope you can see the input current that is is basically zero is the input capacitor that and maybe also the measure that are showing current value that is different than zero around the zero voltage crossing of the AC line. There is no activity from the PFC at that point then it starts when the voltage is higher and then it stops when the voltage decreases again. So in this way basically you can make the PFC work more efficiently because it only switches when there is a higher input voltage. If I increase the current more the current then you can see that it goes to zero on the graphical user interface and it becomes a sinusoid the input of the actual board. Let's try now to increase more the value of that parameter. Let's test 1000. So right now it is actually working at 25 watt roughly and you can see that now it is in area skip area or mode and even though I am now working at the power at the higher power level that I was testing earlier and with the previous configuration it was completely out of this mode of operation while now I increased the level and it is still working area skip. So with this parameter basically you can select the power level at which you want to use this type of approach to improve the efficiency. Now the drawback is obviously on the THD. Now at very low power there is no big deal if the input current is distorted is not a sinusoid because there is no regulation and actually in general there is no regulation up to 75 watt. So depending on your application depending on the regulations for your application you can decide whether to have a nice and clean input current so that it has a nice THD it doesn't generate any noise that can impact other applications connected to the line or you can decide to improve the efficiency so that you get some benefit from that point of view. Let's increase now the power more so we'll see that it will exit from the skip area mode right now we are around 70 watt and you can see that at this point it is completely out of the skip area mode and you can see the input current that is now again a nice sinusoid following the input voltage. That's it for this video. If you have any questions I recommend you to contact your STE technical support. We have several offices around the US. You can also go on the website and contact us to the online technical support.