 Good morning we started to look at air assist atomization towards the end of the last class okay we looked at three different configurations of air assist atomizers essentially varying in the angle at which the air is introduced towards the liquid sheet and the advantage of the third design we said is that I can by varying this angle theta independently also control independently also control the shape of the spray meaning like a spray angle okay. So let us before we go much further make a list of a few different things that are different from spray nozzle to spray nozzle okay and we will then use that qualitatively to differentiate the nozzles that are that we are going to talk about today. So spray parameters the first end one of the most widely used is called the spray angle. So if I take a typical spray this sort of an angle I can measure that by taking by coming some distance downstream and looking at the diameter of the disc that I make I can define a tan alpha to be equal to h over d by 2 or rather other way around d by 2h okay this is one of the simplest ways of measuring this angle another one is to just bring a couple of blades of a protractor towards them and when these appear parallel to the human eye the angle between the protractor blades is equal to the angle over which the spray happens. So these are sort of common ways of measuring the spray angle now you have to understand that there is no such thing as a spray angle like a geometric angle between two lines it is what is visible to the human eye and what is visible to the human eye could be slightly different from person to person. So it is usually very hard to define a spray angle you know any more accurate than let us say 3, 4, 5 degrees that is about as good as the human eye can get. Another parameter that we want to talk about is called spray pattern. So if I take the same spray nozzle and at this distance I look at what this disc looks like. So I look at like there is a diameter to this disc we know we saw that but again it is not like a real hard geometric disc so the edges are fairly blurred out. Now if I look on the inside at every point here if I take a small cross section let us say about 1 mm squared of this disc and if I have a way of sampling all the liquid flowing through that 1 mm squared cross sectional area there is drops flowing through there and over time I can define a volume flux which is of let us just say ml of liquid per mm squared of area per second. So if I sample for 1 second how much volume of liquid flows through this 1 mm squared of area clearly that would be different over here and over here in a general spray we do not know how much is flowing through here here and here. Now so if I assume axis symmetry for a moment and I will see what this means in a moment I will show you what it means if I assume axis symmetry and draw this volume flux versus radial position and this happens to be my center line. If I get a pattern that looks like this that is as I move away from the center line the center line happens to be over here as I move away from the center line the volume flux first increases at some radial distance the volume flux achieves a maximum and then drops off as you go towards the edge. This would be an example of a hollow cone spray there is very low volume flux near the center line and most of the volume is concentrated around the around some thin rim inside the spray. If I make the same volume flux measurements in a similar axis symmetric spray and it if it happens to show this kind of a pattern this is more reflective of it being a solid cone spray. Ideally we want it to be something like this the volume flux is at some value for a little while and then it drops off both of these are possible choices for it to be called a solid cone spray. Now I said I will assume axis symmetry if I take if I take this spray disc let us just draw a disc this is my so called center line but remember it is not really a spray center line it is my nozzle center line that I am extending into the spray what is the chance that my spray is also symmetric about my nozzle geometric center line that nozzle has a nice geometric center line because that was somehow somewhere drawn on autocad. So it is a geometric object it has a center line possibly or at least an axis I should not say center line an axis but if I can I when I extend that axis into the real spray how much of an axis is it really the answer to that lies in taking one circle inside the spray typically this could be the point where you have this maximum volume flux and plotting the same volume flux as a function of phi the azimuthal angle around the circle in the most ideal perfect spray this would be a constant that is walking around a certain circle symmetric around the center the volume flux at all the points on that circle are the same but in reality I could get something that is slightly different but this point and this point have the same volume flux essentially because of circular symmetry but say if this is my real spray what it is saying is that if I now divide this into four quadrants I have somehow more flowing through this quadrant and less flowing through the other quadrant I have a higher volume flux in one quadrant in relation to the other. Now in most instances this may not be desirable this is what causes non-uniformity in your deposition if you are spraying paint for example or non-uniformity in the fuel vapor distribution if you are in a combustion system it is extremely bad because you create where you have this high volume flux you will end up creating a hot spot which could burn through the combustor lining material if you are looking at some sort of a land based power generation. So I want to avoid these kinds of hot spots and cool and low flux regions but before I go much further I need a way to quantify both of these spray patterns the simplest way to make these measurements are using what are called patternators there are two kinds of patternators one is called an in line patternator and the other is called a sector patternator we will look at the in line patternator first so let us say I have this spray the in line patternator is just a series of tubes when I expose the spray to these tubes this is sort of like a rain gauge essentially right I end up collecting some volume something like that let us say and these heights are an indication of the spray pattern. So I take a row of tubes move them into the spray and then let volume be collected in the tubes for a fixed period of time over the entire tubes since the tubes are all of the same cross sectional area what I have measured is the volume per unit cross sectional area per second if I take the volume divided by the cross sectional area of each tube and then divided by the time over which I collected all this liquid. So I can now plot this and that looks like volume flux in units of let us say cc of liquid or ml of liquid per mm squared per second as a function of radial position. So in this case I am going to draw this slightly differently I am going to make the center line be somewhere in the middle and I could do this I have intentionally drawn this peak to be higher than this just to show you that this kind of a patternator can detect these kinds of non uniformities in one diameter. So I take one diameter of the spray and see if the uniform how much of a mirror image is one half of the spray in comparison to the other. So this is the simplest kind of a patternator called the in line patternator the other is called a sector patternator as you can now begin to see from the name if I take a spray I insert a patternator that are divided into 8 sectors let us say or I could do more and each of these sectors drains into one test tube. So typically this is the patternator with the top view shown here and each sector drains into one test tube. So I have 8 such test tubes for the 8 sectors and if this is the cross sectional area of each sector and they are the same for all the 8 sectors I can plot the volume flux in each sector. In a perfectly axisymmetric spray I expect that all the 8 sectors catch the same volume. This is a perfect place to talk of the spray center line say if I take this sector patternator and displace it slightly. So I am going to move the spray part out slightly just about that far. So this is my spray right now where the center line of the patternator the patternator is also a geometric object. So I can define an axis for it it is only the spray that does not have a line that I can see right. So if I now take the spray patternator the sector patternator and align it and misalign it slightly with the nozzle geometry center line what sort of a volume flux do I will I expect. So we will take the if so if I take the volume flux and plot it as a function of phi I am expecting it to be a dead straight line if it is a perfectly ideal spray. Now if I took an ideal spray but place the sector patternator slightly off axis then what I will expect is that this that diametrically opposite sectors get a minimum and a maximum of the volume flux. So essentially if I take the sector patternator and displace it slightly I am going to expect the spray volume flux to look like that. And this displace patternator so if I know if I take the volume flux measurement and what is usually done is if I take the Fourier components of that volume flux in this phi variable any variation that is either like sin phi or cos phi is essentially due to misalignment between the nozzle geometric center line and the spray center line nozzle geometric center line and the patternator center line. Any other variation that is outside of sin phi or cos phi is due to real non uniformity in the volume itself. So I can by looking at this magnitude sort of arrive at a degree of eccentricity of the spray itself to the nozzle geometry center line. So the sin phi or sin 1 phi and cosine 1 phi the magnitude of that is an indication of the eccentricity of the spray with respect to the nozzle geometry center line. That is assuming I had the patternator I use the example where the patternator was misaligned. But if the patternator was aligned to the nozzle center line and I still saw a pattern like this it is essentially that the spray is slightly displace in relation to the nozzle center line. So if you think of a simple example where I am expecting this to be alpha and this to be alpha. But if the whole spray was shifted to one side sort of like that nozzle was nice and straight. But the spray was shifted slightly to the to one side that would give me something that looks like this. So these are all just simple diagnostic tools to help us arrive at different metrics that we could use to measure sprays. So in most axis symmetric sprays I am expecting the sector patternator to give me the same amount of fluid in all the sector tubes and in the in line patternator measurement I am expecting whatever my design of design objective is to be achieved. So if I am looking for a solid cone I want something that looks like this for the volume captured by each tube. Now these are all intrusive measurement techniques there are non intrusive equivalents of these we will talk about towards the end of the semester. So essentially like optical ways of measuring some of these variables. So we looked at spray angle spray pattern of course we talked about drop size in some detail earlier on these are all indicators of what the spray what are all the different variables that we used to characterize the spray. So we looked at hollow cone sprays coming mostly from simplex nozzles and we have solid cone sprays arising mostly out of pressure jet nozzles and we looked at combinations of these we looked at then when we get to air assist nozzles the two or three designs the three designs that we saw have independent control over the spray pattern by controlling the angle over which the air is introduced like the third design here produces more of a hollow cone spray in comparison to these two these two will generally produce more of a solid cone spray because you are intentionally pushing the liquid towards the center line you will most probably produce more of a solid cone spray in designs one and two design three because of the nature of the air flow will most likely produce a hollow cone spray. Now these are all different designs of external mixing air assist atomizers. So we are introducing the air and liquid into the nozzle and they are allowed to come in contact with each other outside the spray nozzle essentially. We want to look at another extension of these external air assist atomizers called the pre-filming air blast atomizer. This pre-filming air blast atomizer is again one of the most widely used spray nozzles that this is essentially a design that is used in most aircrafts now that this is a liquid passage and the fuel goes through this passage it is got a swirler slot in it somewhere here those are swirlers but the purpose of the swirler is only to gently push the liquid outwards. So what you see here is a liquid film that is spilling out from this tip essentially that is impacted by a swirling air stream on the inside often called the inner swirling air and then this is called the outer swirl air. So this part these veins are also at an angle so the air at in this part is swirling and the air coming out of these passages here is also swirling by controlling the ratio of the air coming in on the inside and outside. So I will call this m dot air inside call this m dot air 1 m dot 2 of the air by controlling the ratio of this m dot air 1 and m dot air 2 we can essentially shape this spray. So if I want a spray that goes very far out a wide spray angle I am going to increase m dot air 1 so it is essentially going to push the spray further out. If I want a narrower spray I can decrease m dot air 1 and increase m dot air 2 and in addition I can also increase or decrease the atomization quality the drop size by essentially varying the total mass flow rate that I introduce. So if I increase m dot air 1 plus m dot air 2 I have put in more air per unit mass flow rate of the liquid and that generally means I get a finer spray or a lower drop size. So I have a way in this design of independently controlling spray drop size spray cone angle and spray pattern pattern is whether it is hollow or solid by these three different handles by essentially controlling the mass flow rates of the two air streams in relation to the liquid flow rate. So this is m dot liquid as a result this is one of the most widely used nozzle designs for aircraft applications. The air coming in is essentially the compressor exit the compressor stage exit so it is at a very low pressure on the order of like maybe like compressor may be at like 5 times 2 to 3 times may be the increase in the pressure the pressure ratio at the compressor stage. So we are looking at fairly low pressure air and the liquid is also operated by a pump that is on the same shaft typically. So that is also a low pressure liquid but the flow rates of the air is very high and that is what that is what gives us good atomization. This is also an example we are still in the realm of external mix air assist atomizers although this is this particular design is patented by Parker and it goes by the name pre-filming air blast atomizer because you are essentially creating a film in fact the specific design looks at there is a little lip that extends out on which the film is it is like a little annular lip and the liquid film is sticking to the wall as it comes out. This is a design from the I think early 50s so just to give you some chronology of where these were developed. Parker is a company that used to be it is based out of the US. So now there are other ways that we want to look at some of the more newer designs we look at one of those newer designs here called the effervescent atomizer. This is in fact a design from my masters work where we looked at the spray performance on this particular atomizer. The design itself is where you have a liquid coming in through the side of a nozzle and it flows in through this annular passage and at the exit here you have a gas coming in and is and being injected into this liquid through a series of holes. In fact another design could be extend this all the way down to the nearly the exit but this piece is a porous piece. This porous piece allows so it is like a porous it is made out of a porous metal sintered metal let us just say that allows this air flowing in to be injected into the liquid. So what you essentially create in the liquid are these tiny bubbles and that bubble laden liquid now flows through this atomizer comes out in the form of a spray here. There are two modes of operation of this one where due to this convergent nature of this passage some or all of these bubbles may coalesce into one big bubble which is usually in the middle of this injector. So if I take zoomed out view of this I could have essentially created a thin liquid film with an air bubble in the middle. But the liquid film itself may have some air still trapped in it in the form of small bubbles but most of the big bubbles could have coalesced to create one big air bubble it is sort of like a core annular flow in two phase flows this would be called a core annular flow. So I have an annular liquid film flowing down the walls in the middle is a core of air. This core annular flow design is like our old pre-filming air blast because I now have an inner air except in this case I do not have an outer air but I can always introduce another outer air stream to this. But what people find is that is not required because you are mixing the air internal to the spray nozzle it is much more efficient at using the air for atomization. In the previous examples where we were using where we were looking at external air mix the only place where the air and the liquid come in contact is at that interface that cross sectional area where the two are allowed to exchange momentum is very small in comparison to the total surface area of the liquid. Whereas in the case of an effervescent atomizer you have created a spray system that can be that where the air and the liquid are internally mixed and they are very intimately mixed as a result giving you much more efficient transfer of momentum or and energy between the air and the liquid. Because the air could be at a slightly higher pressure the air and the liquid coming in contact gives rise to this expansion of the air and that is essentially what gives rise to atomization. This is the energy source for atomization in this particular design. Now this is it is really where it was first proposed in the in the eighties it came in with a lot of promise because it uses about a tenth of the air that is required for the air blast or air assist designs tenth of the air by mass but it requires slightly higher supply pressures for the air. So, there is a slight overhead but the problem is not that the problem is that it is very prone to instabilities. So, I could for example ideally I want this nice air flow in the middle and liquid film sticking to the walls and getting a nice steady spray. But what I actually end up seeing is that liquid may come flood this orifice so I get a big slug of only pure liquid that is being pushed out by some accumulated gas behind there and this big slug of poorly atomized liquid is followed by a slug of air with hardly any liquid in it. So, it is just like a big puff of air that comes out. So, this alternating slug of liquid poorly atomized and a puff of air is not good because it gives rise to intermittency that is one of the biggest challenges with this design. If there is a way to overcome that there are some advantages of this design even for diesel injectors the big problem though is this intermittency. But this is one of the only concepts of internal mix air nozzles internal mix air assist atomizer. So, the class if I if I need to classify this nozzle it would be as an internal mix everything else is external mix. The advantage of the external mix is that I get to control the liquid flow rate because that is a separate pump and that is not as much affected by the air in this case if I increase the air flow rate for a moment keeping the fuel liquid supply pressure the same the liquid flow rate is going to go down slightly because the pressure in this chamber will go up. So, the back pressure against which the liquid is being supplied into the nozzle goes up as a result the liquid flow rate goes down. So, there the two flow rates are more coupled in this problem in comparison to the external mix design. So, as far as robustness in an application is concerned this is not as good as the other for that reason. We will look at may be one or two more other designs now we are moving away from we looked at liquid inertia as being a source of energy those were the simplex and the spill return designs we looked at air as being the source of energy those are the external mix and the internal mix air assist designs. Here is a case where the source of energy is a mechanical object. So, if you will imagine an ultrasonic nozzle that have a set of PSA electric discs around a supply tube. So, essentially what I am doing is I could take this atomizing this horn as it is called and have it oscillate back and forth in this fashion and what I end up doing on the end here is I am imparting that liquid, but by creating this transient motion of the liquid I can have liquid bump into itself sort of it is a very sort of a crude way of saying it, but I think that describes the situation. I have fast moving liquid behind slow moving liquid behind fast moving liquid because of this oscillatory motion of the horn because of which I end up creating a spray here. It is not really a spray in the sense of the previous sprays it is more like a drip, but a controlled drip. This is in fact some of the very early ink jet printers out of IBM use a design like this. So, if you will imagine the ink jet nozzle just oscillating it creates a nice steady stream of drops one behind the other and that steady stream of drops then goes hits your paper and gives you print. If I take it further if I do not look at a steady single stream of drops, but I do not mind them dispersed over a certain area then I end up getting what looks like a spray. Now, this is very efficient in the way it uses the energy because all the energy that we put in the form of the ultrasonic nozzle is entirely transferred to the liquid and I the liquid inlet has a pump that drives the liquid through this nozzle that is independent of the energy source. The energy source in this case is this piezoelectric discs that cause this mechanical movement of the horn and the liquid flowing through the nozzle is due to some other source of push coming from the liquid inlet side. So, these two being decoupled is a good thing because I can put in whatever energy density that I need to get my desired spray quality because I can put in as much ultrasonic energy as I need to per unit mass flow rate of the liquid because I have independent controls over the two. A disadvantage of this practically is that it is not prone to go up to you know any high flow rates. So, typically design like this can put out you know less than about 0.1 liter per hour very very low flows. One of the places where you require low flows, but good atomization quality is in humidification applications. So, this is in some humidifiers you see a design like this being used where I just want to in a very dry environment if I need to humidify the air in the room I want to spray water into the air and allow the water to evaporate. So, as to saturate the air and that process requires some sort of a spray and in that application this kind of a design is often used. Another mechanical design where the source of energy is mechanical in nature is what is called a rotating cup atomizer. We will look at this design essentially if you will imagine a spinning cup the spinning cup has a liquid feet. So, just a little imagine a just like is shown in a schematic it does not matter do not have to get very fancy with this. If I keep dripping water into the cup and the cup itself is spinning at some angular velocity omega then water is going to climb up this lip and come out in the form of a spray the form of a sheet spilling over the edge. So, it is just like a simplex nozzle except the swirl for the liquid and the centrifugal action of the liquid is not due to the liquid swirl, but due to a wall swirl due to wall motion that is the only fundamental difference. Now, if I actually start to use this this is now the top view of the rotating cup except what is shown here is where one half of this cup in this part has serrations. So, if you imagine little grooves like that. So, it is like a I have grooves on the edge on this rim of the cup when I put those grooves what I end up doing is the liquid flows only through those grooves as it spills over the cup and each of the liquid that is spilling in through one groove is like a little jet that breaks up into drops. Now, the size of the drops are comparable to the size of the liquid jet. So, liquid this liquid this diameter d there. So, essentially by controlling the diameter or the depth the width of the grooves I can control the drop size by controlling the rpm I can control the cone angle of this spray. So, the angle over which this you know the angle over which the conical sheet that is spilled over the lip is at. Now, what is also shown in the schematic is where if I take the cup and do not put these grooves what I end up getting is actual film spilling over that looks like is shown in the schematic. So, it is an actual it is like a liquid film itself that spills over the edge of the cup that is not atomized at the point where it spills over the problem with that is now I have no control over the drop size as much as I had in the other design. So, the more. So, this sort of serrations on the grooves does serrations or grooves does two things one it allows me to control the drop size and two I have control over the width of the drop size because of the nature of the atomization process the primary atomization process. The fact that I have each individual jet spilling over through the grooves and each jet spilling over through each groove is going to be responsible for break up. And if I make sure that all the grooves are about the same width I have the same fluid mechanic process happening in all the grooves. So, that particles that I get the drops that I get are all going to be comparably very similar in size right. So, this gives me much better control over the width of the distribution than any other design of spray nozzles. Consequently this design is used where width control is very important much more than drop size control you know I can live with the size being off a little bit I do not want the sizes to be far apart in their own width. In those kinds of applications this is used one example is powdered food manufacturing tide in the spray drying process in tide or any of these drying processes you see rotary cup atomizers used. There is actually another reason why rotary cup atomizers are used in drying processes because drying by nature involves slurries. So, these are particle laden fluids and you do not want the particles to be clogging fine features on a spray nozzle. Here is a design where the finest feature is the groove otherwise everything else is big right. So, essentially the groove size is sufficiently large that it does not clog up and so it is very resistant to clogging and gives you good control over the drop size distribution width. So, this is another mechanical source of energy based design of spray nozzles. Now again this produces essentially a hollow cone spray there is really no way to produce a solid cone spray from a design like this, if I am only interested in drying particles a solid cone is actually bad because I have to get unsaturated air to the middle of the spray if I have a solid cone spray. If I have a hollow cone spray I have to get unsaturated air only to near the drops which is only on the periphery. So, it is in a way good. So, let us quickly recap we looked at performance measures performance matrix the one I did not mention, but I know we did we talked about it earlier is the actual flow rate this is very important. The second is spray cone angle or the third is spray pattern we also have drop size as one of the matrix and the cost to incur all this is in some sense power input or supply pressure. So, at what supply pressure do I have to operate a certain nozzle in order to get these desire satisfied. We will continue this discussion in the next class.