Yellow Bullet Forums banner

1561 - 1580 of 1589 Posts

·
Registered
Joined
·
245 Posts
Come on man... "work" does not mean making the shaft spin. Go drop your turbo in an oven or furnace and let me know how fast it spins from the heat alone.
For some reason you are trying to make this way too difficult.

In your drawing and description, with a 3-way routing the small turbo exit to the big turbo inlet, the small turbo would have effectively zero force being applied to the shaft. Path of least resistance.. both turbo inlets are on the same manifold...
Work (is actually the technical name) is extracted from a pressure temperature/differential across the turbine wheel. Temperature/pressure differences is the reason why a small gasoline motor can spool up a large diesel turbo with relative ease in spite of having 1/3 the displacement of the diesel motor. This is literally one of the fundamental principles of thermodynamics. I'm not saying it's the only factor, just that it IS a factor. If you don't believe that extracting work from the exhaust gas reduces exhaust gas temperature, go through an EGT in the manifold and in the downpipe and test it yourself. It's also the reason leaner exhaust mixtures will spool the turbo up faster. Mass flow is similar to a stoic burn, but egt is higher.

In the drawing, the gate leading in to the large turbo from the manifold could be modulated similar to the small turbo bypass valve on a traditional compound set up. It doesn't have to be an all or nothing.

Plus this whole conversation is a hypothetical. CAN the exhaust routing be optimized farther than it already has been? By rerouting the small turbos exhaust away from the large turbo AFTER the large turbo is online and system mass flow flow and back pressure might be able to be improved through the entire system.
 

·
Registered
Joined
·
1,580 Posts
Discussion Starter #1,563
Why exactly do you guys run the exhaust from the small turbo into the larger turbo? That will reduce the big turbos exhaust energy. It also results in backpressure. If you have separate exhausts for the small and large turbo you will have the same exhaust response but with less backpressure??
Spoiler alert: I miss the point and go on a tangent, read to the end to find out where I realize I'm a dipshit and get back on track.

You definitely need the fresh unmolested exhaust passed by the wastegate to get the big turbine working. Consider this simple fact. The small turbine needs a given amount of exhaust passed through it, all excess is dumped by the WG (to atmosphere in your suggestion and wasted). The large turbine is certainly going to need more than the small turbine does, because it's bigger. Then factor in the fact that the small turbine is cutting EGT nearly in half and there is way less energy in that small amount of exhaust, it's not going to spool the big turbine any time today. Sending that wastegated exhaust to the large turbine increases exhaust energy to the big turbine.

To illustrate this, simply keep your small turbine bypass WG shut and see if the big turbine spools. I've done this, and it won't. In some diesel discussion I've read they refer to "early wastegating" to get the big turbine online quicker. I can definitely tell you this is important. This is the reason I keep coming back to my old boost control strategy that I've outlined several times in this thread. It opens that bypass WG early and gets the big turbo online quickly. I'm trying to go from idle to 60 psi boost and 7k rpm in 3-4 seconds at sea level, both turbos need all the exhaust they can get, you can't "waste" gate any of it. Everything else I try spools slower.

Regarding this causing increased back pressure, I would argue that it doesn't. Any addition exhaust sent to the big turbine that it does not need, wll be wastegated. There is no free lunch here however, the back pressures compound just like the boost pressures do. The best compound turbo setups I have worked with will achieve a 1:1 ratio, but no better. I still argue that this is stellar performance for something that spools the way it does, particular for a 2 liter engine.

-Ok, now having said all that I realize you meant turbine discharge, not WG discharge (which is another common question that always comes up). I'm going to leave the above paragraphs there because they may help someone else wondering about the WG side.

I have NOT tested dumping the turbine discharge and trying to run the big turbine on the WG only, which I believe is effectively what you're describing. I understand what you're saying in that the spent exhaust from the small turbine is much cooler and is reducing total EGT to the big turbine. I think there is some merit to the basic idea, but there are huge unintended consequences not being given due credit, and now everyone else's posts make more sense to me. :D

The main downside I see is the loss of mass flow. Mass flow is critical. I may attempt to enter that separate conversation in another post and provide examples, but for now I will say mass flow is critical even if EGT drops. Now, if the small turbo is doing little work, there will be little drop in mass flow. But there would also be little drop in EGT, so I'm not sure the benefit is there. To say that mixing the two exhaust streams reduces the big turbine's exhaust energy assumes that the loss in EGT outweighs the loss in mass flow and I don't expect you'll find that to be the case. There are exponents involved etc that make this a little less simple than that. I'll attempt to add to that discussion in another post.

There is also the effect of small turbine discharge preheating the large turbine, which is very important, and would be lost if not connected. Using valves to switch this in and out would solve that part of it, but every single attempt at a sequential/compound hybrid I've seen has performed pretty poorly. To the point that I will likely never try any such thing myself with my own dollars. Forget about packaging, it gets exponentially worse.
 

·
Registered
Joined
·
1,580 Posts
Discussion Starter #1,564
I think you guys are agreeing on what is driving the turbine but explaining it such different ways you feel like you're disagreeing. For example, you're agreeing that the turbine is pulling heat out of the exhaust, and a lot of it. Otherwise the original question wouldn't exist. The question seems be about the relative weight of each factor, mass flow and heat.

Mass flow is critical. Consider that the turbine housing is there solely for the purpose of directing hot exhaust gasses into the turbine through a NOZZLE, aimed at an appropriate angle to impart maximum energy to the wheel's blades. The nozzle creates the back pressure, which provides the pressure differential to increase exhaust gas velocity through that nozzle (this is the definition of lag, you need to get the nozzle working before you can get the turbine working). Smaller nozzle, higher back pressure sooner, higher velocity sooner, spool sooner. And vice versa. This is a mass flow driven system looking at it from this angle.

Consider also what happens when switching fuels, from those with higher stoichs to those with lower stoichs. When going from gas to E85, turbos spool much quicker. The effect really is quite pronounced. To the point that my current setup (with a ~62mm small turbo) will NOT spool at all on gas, ever. But spools very quickly on E85 (and even quicker still on methanol). In each case EGT usually goes down, particularly if you go to richer mixtures, but spool gets better. The difference is mass flow. With fuels that require more mass fuel flow, total exhaust mass flow (air mass flow plus fuel mass flow) goes up.

Looking at the formula for CTGF, Correct Turbine Gas Flow, there is an exponent that gives more weight to mass flow than to EGT. I'm going from memory here. I won't embarrass myself by trying to go above my paygrade on the thermodynamics and I'll leave people to explore that for themselves.

Now, that said, EGT does have an effect, obviously. And it's very obvious when you do stupid things like run your wastegates to control back pressure instead of boost pressure, or run a back pressure limited turbine side instead of boost control, etc. I've found some interesting things that are obvious in hindsight, but cool to see actually happen. For example, our 2g with it's WG control issue is effectively back pressure limited on the small turbo. I can't make more than 40 psi with it. Yet it would launch with up to 45 psi or so, dropping to the "set" 40 psi a couple seconds into the run which it would hold steady to the end of the run. The reason is the added EGT from the 2 step heating up all of the hotside parts. At the same back pressure, boost goes up a lot just from that extra heat. Enough that I started to preheat the exhaust manifold before the water box and really take advantage of this, leaving with up to 50 psi. I'm tempted to coat and wrap all the hot parts on my car.

The effect of EGT is also obvious when mass flow is held constant and ignition advance is retarded. Boost goes up.


Back to the original assumption, which if I'm finally understanding correctly, is that the loss in EGT, due to cooler small turbine exhaust discharge dilution, completely cancels out the benefits of its mass flow and then some. I don't think you'll find this to be the case. It could be estimated by running both scenarios through the CTGF formula and see which one comes out on top. It would require a few assumptions, but nothing that can't be reasonably estimated from data we have.

I don't know if I understand this well enough to explain it more simply or more correctly than has already been done, but there's my two cents as an untrained layman, for better or worse. :D
 

·
Registered
Joined
·
1,580 Posts
Discussion Starter #1,565
Exhaust velocity is dictated by pressure differential. The hotter the temp the higher the pressure = the higher the velocity
This is a tricky one. On the surface it is correct. But the WG throws a wrench into that. Higher EGT requires LESS pressure to do the same shaft work. Lower temp requires more pressure, which is what you stated in your concern for cooling the exhaust to the large turbo and that adding back pressure. But I thought this was an important distinction to make.
 

·
Registered
Joined
·
94 Posts
I think you guys are agreeing on what is driving the turbine but explaining it such different ways you feel like you're disagreeing. For example, you're agreeing that the turbine is pulling heat out of the exhaust, and a lot of it. Otherwise the original question wouldn't exist. The question seems be about the relative weight of each factor, mass flow and heat.

Mass flow is critical. Consider that the turbine housing is there solely for the purpose of directing hot exhaust gasses into the turbine through a NOZZLE, aimed at an appropriate angle to impart maximum energy to the wheel's blades. The nozzle creates the back pressure, which provides the pressure differential to increase exhaust gas velocity through that nozzle (this is the definition of lag, you need to get the nozzle working before you can get the turbine working). Smaller nozzle, higher back pressure sooner, higher velocity sooner, spool sooner. And vice versa. This is a mass flow driven system looking at it from this angle.

Consider also what happens when switching fuels, from those with higher stoichs to those with lower stoichs. When going from gas to E85, turbos spool much quicker. The effect really is quite pronounced. To the point that my current setup (with a ~62mm small turbo) will NOT spool at all on gas, ever. But spools very quickly on E85 (and even quicker still on methanol). In each case EGT usually goes down, particularly if you go to richer mixtures, but spool gets better. The difference is mass flow. With fuels that require more mass fuel flow, total exhaust mass flow (air mass flow plus fuel mass flow) goes up.

Looking at the formula for CTGF, Correct Turbine Gas Flow, there is an exponent that gives more weight to mass flow than to EGT. I'm going from memory here. I won't embarrass myself by trying to go above my paygrade on the thermodynamics and I'll leave people to explore that for themselves.

Now, that said, EGT does have an effect, obviously. And it's very obvious when you do stupid things like run your wastegates to control back pressure instead of boost pressure, or run a back pressure limited turbine side instead of boost control, etc. I've found some interesting things that are obvious in hindsight, but cool to see actually happen. For example, our 2g with it's WG control issue is effectively back pressure limited on the small turbo. I can't make more than 40 psi with it. Yet it would launch with up to 45 psi or so, dropping to the "set" 40 psi a couple seconds into the run which it would hold steady to the end of the run. The reason is the added EGT from the 2 step heating up all of the hotside parts. At the same back pressure, boost goes up a lot just from that extra heat. Enough that I started to preheat the exhaust manifold before the water box and really take advantage of this, leaving with up to 50 psi. I'm tempted to coat and wrap all the hot parts on my car.

The effect of EGT is also obvious when mass flow is held constant and ignition advance is retarded. Boost goes up.


Back to the original assumption, which if I'm finally understanding correctly, is that the loss in EGT, due to cooler small turbine exhaust discharge dilution, completely cancels out the benefits of its mass flow and then some. I don't think you'll find this to be the case. It could be estimated by running both scenarios through the CTGF formula and see which one comes out on top. It would require a few assumptions, but nothing that can't be reasonably estimated from data we have.

I don't know if I understand this well enough to explain it more simply or more correctly than has already been done, but there's my two cents as an untrained layman, for better or worse. :D
That was much more clear and on-target than you thought it was.
 

·
Registered
Joined
·
94 Posts
Work (is actually the technical name) is extracted from a pressure temperature/differential across the turbine wheel. Temperature/pressure differences is the reason why a small gasoline motor can spool up a large diesel turbo with relative ease in spite of having 1/3 the displacement of the diesel motor. This is literally one of the fundamental principles of thermodynamics. I'm not saying it's the only factor, just that it IS a factor. If you don't believe that extracting work from the exhaust gas reduces exhaust gas temperature, go through an EGT in the manifold and in the downpipe and test it yourself. It's also the reason leaner exhaust mixtures will spool the turbo up faster. Mass flow is similar to a stoic burn, but egt is higher.
In practice this is not the case. Especially with e85 and methanol, richer can and often will spool faster, up to the point that you're hurting combustion.

ALL expanding gasses are going to drop in temperature... This is a basic fundamental you learn before leaving gradeschool. The temperature differential across the turbine wheel is an effect, not a cause.

If there was ANY percent of shaft rotation whatsoever that was NOT caused by kenetic energy transfer, then you would be able to make the shaft spin in a vacuum. or with zero air movement. I encourage you to go try this.

In the drawing, the gate leading in to the large turbo from the manifold could be modulated similar to the small turbo bypass valve on a traditional compound set up. It doesn't have to be an all or nothing.

Plus this whole conversation is a hypothetical. CAN the exhaust routing be optimized farther than it already has been? By rerouting the small turbos exhaust away from the large turbo AFTER the large turbo is online and system mass flow flow and back pressure might be able to be improved through the entire system.
Some of what you're getting at is already being tested with Kevin's current WG setup, and the system that i'm running. It does not require any special valves though. All we're doing is rerouting the gates to control the parts of the system as a whole, NOT a specific gate dedicated to each turbo.

The problem with mounting both turbos on the exhaust manifold in parallel (as in the earlier drawing) is that the small turbo would take so long to come online that our entire purpose of compounding (fast spool) would be wasted. You must force ALL exhaust through the small turbo on spoolup in order for it to spool it's fastest, which would not happen if they were mounted in parallel.
 

·
Registered
Joined
·
94 Posts
I'm assuming in this drawing, the smallest black squares are both supposed to be giant valves?

If that's the case, i missed the far right one earlier, and all you did was redraw a standard compound setup with extra steps. The end result is the same IF WE'RE STAYING COMPOUND. Sequential is a different story, but can also be done with just gates, as kevin has done on the 2g.

The end result is that we must have a certain amount of energy put into the shaft in order to get each compressor stage to do what we want. Nothing is going to change that.
 

·
Registered
Joined
·
179 Posts
Kevin yes I followed your posts perfectly. That answers my question. It is more about the massflow than anything else. Apparently the cooler exhaust gives a significant boost in exhaust density and massflow for the E85 and Methanol allowing the turbo to spool up.
 

·
Registered
Joined
·
6,853 Posts
When it comes to spooling turbos in general, I'm a two stage guy. I do what the n/a engine likes to get the engine speed up within the confines of the stall of the converter. Popping that engine speed up quickly is the fastest way to increase exhaust gas velocity, because you produce so much more of it. So I go not overly rich or lean. Plenty of timing, which makes the na engine responsive to gain engine speed. Make the engine happy first. THEN, when I hit the two step rpm, I pull some timing out, and sometimes will add SOME fuel. The later firing tends to leave additional heat in the exhaust as the valve opens. The extra fuel can help with additional heat to a point. After that point it can actually cool the exhaust charge and even drop cylinders. But some setups seem to be able to light the fuel so late that the flame is still active with the exhaust valve open. That's what it seems at least. I have not been able to accomplish this. The fuel can even pop in the exhaust, which can really accelerate the turbo. Again, I have no experience doing this with methanol.
 

·
Registered
Joined
·
94 Posts
Kevin yes I followed your posts perfectly. That answers my question. It is more about the massflow than anything else. Apparently the cooler exhaust gives a significant boost in exhaust density and massflow for the E85 and Methanol allowing the turbo to spool up.
This is pretty much what i was trying to say, kevin just did a way better job of it.

One of the points you were hitting on was theoretically solved by relocating the "total system boost" wastegate (what you would normally label the big turbo's wastegate); taking it off of the inter-stage exhaust pipe, and putting it directly on the exhaust manifold.
I only say "theoretically" because i don't believe he has a/b numbers yet. But it all makes sense conceptually.
 

·
Registered
Joined
·
1,580 Posts
Discussion Starter #1,575
Varying EGT doesn't change the mass flow of the exhaust of course, that is predetermined by how much air and fuel you put through the engine. In any series system the mass flow can not be different at different points in that system. So 100 lbs/min air in plus 15 lbs/min fuel in, results in 115 lbs/min exhaust mass flow out. But EGT does affect the CTGF (which is a good way of seeing mathematically how mass flow and EGT interrelate). To look at it another way, with higher EGT, we can wastegate more of that 115 lbs/min, reducing the turbine drive pressure yet doing the same work on the turbine wheel. If all mass flow goes through the turbine, that is what it is (on a given fuel type), EGT becomes the main thing we can vary.

To put an example to the timing reduction, at 6500 rpm or so on an ignition cut 2 step I can easily vary boost between 35 and 60 psi with a 10-12 degree range in timing. It works the same on methanol and ethanol, I don't find much difference there. With timing reduced past TDC boost will likely go up even further, but I tend to find on both of my compound setups on 2 liters that I'm able to make boost a 1/100th of RPM. 40 psi at 4000 rpm (3500 rpm on the 2g if I preheat the exhaust side right before staging), 50 psi at 5000 rpm, 60 psi at 6000 rpm, without retarding timing excessively or adding any fuel. I do the same thing that someone described above, leave normal timing to speed spool/stall up, the pull the timing under load on the 2 step. I've been doing this since 2008 on this car with no issues.
 

·
Registered
Joined
·
245 Posts
Its simply pv=nrt. N, R, & V are constants. therefore: [Pressure : temperature] (are relative). Kevin, as you said, the nozzle shape and orifice in the turbine housing controls the manifold pressure. As for mass flow relating to spool up, mass flow is the N value. Inject more fuel mass, N value increases.

In the steam turbine world we use multi stage turbines and use interstate bypass lines for service equipment operating at lower pressures than the main feed steam line. Each turbine stage lowers pressure proportionally to it's work extraction percentage and temperature is reduced relatively.

I'm assuming in this drawing, the smallest black squares are both supposed to be giant valves?

If that's the case, i missed the far right one earlier, and all you did was redraw a standard compound setup with extra steps. The end result is the same IF WE'RE STAYING COMPOUND. Sequential is a different story, but can also be done with just gates, as kevin has done on the 2g.

The end result is that we must have a certain amount of energy put into the shaft in order to get each compressor stage to do what we want. Nothing is going to change that.
Not quite.

The rough idea of that operation is it's a traditional compound set up until the large turbo begins to make boost. In no boost situations the valve between the large turbo and manifold is closed and the exhaust path is manifold -> small turbo -> large turbo. As the large turbo spools up, the valves change position so that the exhaust streams are isolated and do not interfere with each other. The gate leading into the large turbine would be modulated to control how much exhaust gas is diverted in to the small turbine: it's acting as the primary bypass valve. It would need one more wastegate that I forgot to put on there to control large turbo exhaust bypass. But no flow would pass directly from the manifold in to the large turbo without first passing through the small turbo if the big turbo is not spun up.

The benefit is there is no cross-contamination between the small turbine discharge and large turbine inlet. The small turbine becomes MORE efficient. Back pressure can be lower since the small turbine will not have to fight the pressure from the primary bypass valve - the same way an externally dumped wastegate makes more power than an internal gate in every situation.

It's just an idea. Like I said, it would require a 3-way wastegate discharge valve that is very large for the automotive world. Like 3" inlet/out with proper anti-swirl internal geometry. Perhaps something line flapper valve. A butterfly or gate valve would never work in this situation.

Anyway, it's just an idea expanding off what Chuck was thinking.
 

·
Registered
Joined
·
179 Posts
I know the theory that is taught about massflow being constant but you do have a combustion process in the middle of it. Therefore I think the lower combustion temps of the alcohol leads to a greater thermal efficiency so in the end your massflow goes up on the exhaust side. How else to you account for it. If massflow does stay constant as is taught in school then the alcohol would spool up much worse since it is at a lower temp??
 

·
Registered
Joined
·
1,580 Posts
Discussion Starter #1,578
Internal combustion engines can do amazing things, but they can not create or destroy mass.

As I covered earlier, with alky fuels mass flow goes UP, temp often goes DOWN (this is totally adjustable with timing), and spool improves dramatically.
 

·
Registered
Joined
·
1 Posts
Work(aka HP) is equal to moving 550lbs one foot in one second. So yes MOTION is most definitely a requirement.
Motion is not a requirement for work. You can put 745.7 watt (1 HP) into a light bulb and there is no motion. You could also heat up your leftover dinner in the microwave, the food is hot, yet no motion. Work is being done.
 
1561 - 1580 of 1589 Posts
Top