What do turbine maps tell us about a turbo?
09-09-2007, 11:42 PM
#1
Registered User
Thread Starter
Join Date: Apr 2003
Location: Cummins Technical Center, IN
Posts: 6,564
Likes: 0
Received 6 Likes
on
5 Posts
What do turbine maps tell us about a turbo?
I've reading a lot lately about turbine maps and turbine theory in an effort to increase my understanding of how to take what a turbine map says and translate that into practical knowledge- like what kind of spoolup a turbo would have and such.
First, let's looks at an example turbine map of a GT4088/4094:
This map has four separate curves, each representing the performance of the same turbine wheel with progressively larger turbine housings.
Across the left side we have numbers representing the corrected mass flow. We'll come back to the "corrected" part. Across the bottom, we have the pressure ratio across the turbine (keep in mind this is all in absolute pressures and temps and such).
So what do these curves on the map tell us?
As we increase the mass flow to the turbine, at first it can "absorb" the increase almost completely. It does this by spinning faster. But as you continue to increase the mass flow rate, you eventually hit a point where the turbine starts to pose a restriction to the flow. Increasing the mass flow past this point you will now see a definite restriction from the turbine, and the pressure drop across the turbine will increase. Looking at the level part of each curve, you can see that you get to a point where you cannot force the turbine to flow any more, no matter how much you increase the pressure ratio across the turbine. For example, the curve climbs pretty far from a 1.5PR to a 2 PR, but after 2 it's almost level up to a PR of 3.
It just so happens that this restriction is a good thing, because unless the turbine is restricting flow, it cannot extract any power from the gas to power the compressor.
Said another way, once the curve starts to level out, the turbo is fully spooled-- it has reached it's maximum efficiency, or ability to get work from the available gas energy. Also note that you'll sometimes see the right hand side labelled as a secondary Y axis showing "efficiency". It just so happens that efficiency increases with a larger housing.
Here's what I'm talking about:
See how the efficiency on the right side increases with A/R? The tightest housing here is 86% or 87% efficient, but the largest housing is approaching 97%. (dubious numbers? but the point is illustrated).
So as you step on the pedal, the mass flow delivered to the turbo increases until the turbine start to choke things a bit, and then you'll see the initial spoolup. Once you hit this point, you'll see the first stages of positive manifold pressure (boost) because you're actually starting to build a little drive pressure. (remember, without a pressure drop across the turbine, there's no work be extracted).
So, what does this tell us about spoolup??
This is where we get into the "corrected" mass flow. The turbine map mass flow figures are "Corrected" to represent sea level pressure on a 59º day. So we have to "correct" our EGTs (or Turbine Inlet Temp, and I won't abbreviate that
) so they correlate to the turbine map.
This can get into a chicken-or-the-egg scenario, so you have to fix a variable and proceed from there.
For example, let's say that I have my CTD spinning 1800rpm with no turbo attached and EGTs are 1000º. What's my mass flow? Would this spool a Garrett GT4088 or GT4094?
At 1800rpm, my engine is swallowing 187cfm of air. At "standard pressure and temp" of 70ºF and 14.7psia atmosperic pressure, this 187cfm converts to about 14 lb/min. This is the air going into the engine. What goes in must come out, but we're also burning fuel. So, depending on air fuel ratio, we'll have anywhere from another .5lb to 1lb more per min from the fuel. Let's say we're running lean and it only took .5LB of fuel in that minute.
So now we have about 14.5lb/min coming out of the engine, but at 1000ºF temperature. Since we have no turbo attached, let's say that there's no backpressure in the manifold or that it's negligible.
When we "correct" to atmospheric conditions, this 14.5 lb/min at 1000º becomes 23.8 lb/min! This is how the hotter drive temps show up as more energy to drive the turbine.
So let's say we want to slap on a turbo. Will the Garrett GT4088 be spooled up with this amount of drive energy if we use the tightest housing?
Referring back to the first pic (turbine map of said GT4088), we see that the "corrected" 23.8 lb/min falls on the lowermost curve around a PR of 1.25. Unfortunately, this isn't enough PR to really extract much useful work to the compressor. Our "drive pressure" in the manifold is only about 8psi on the gauge. The turbine just isn't restrictive enough at this point, and you'll have no boost.
But what if we increased both RPM and EGT of our fictional turboless engine? What if we went up to 2000 rpm and EGT of 1200º? Now the "corrected" mass flow goes up to almost 29 lb/min. Plotting this point on the turbine map shows that we are now at a PR of right around 1.5, and the curve is starting to level out. This represents the beginning of the operating range for this turbine, and you'd just be seeing the first stages of actual boost to your engine.
Another useful application of a turbine map is predicting the performance of a turbo based on the known performance of another one-- you'll need the maps for both turbos.
I've lately been trading PMs with a member here trying to get info on how the Garrett Stg 3 (GT3788 with .99A/R housing) is performing for him. (Turbine map in the garrett PDF catalog from their site). He says he loves it, and he has about 4-6psi cruising on the hwy and gets instant response from it.
How would the GT4088 respond by comparison? Looking at the turbine maps for the the two turbos, let's select a reference point of 1.5PR, since this roughly corresponds to the threshold of the "operating range" of the turbo. The GT3788 at this point is flowing about 26lb/min "corrected." The 4088 with the tightest housing is right around 29lb/min at the same point. In other words, it takes 3lb/min more to put the 4088's turbine where the 3788's turbine is. In other words, it's about 12% laggier to get to that point of its operating range. At a PR of 2, the difference is about 2lb/min, or about 7% more.
Keep in mind this is only just the THRESHOLD of boost, not how fast it will accelerate from 10-40psi of boost or anything. That has little to do with the turbine map, and more to do with wheel sizing and rotational inertia.
Speaking of that: note that the GT4094 and the 4088 use the same turbine, and for the response penalty outlined above, you can step up to a legitimate 550hp turbo in teh GT4094. Now, to make the big top end you'll want to run a housing the next step up from the .85 and the lag penalty will be another 2-3 lb/min mass flow. Doesn't sound like much,but that can be pretty extensive in terms of EGT.
For example, let's say you have a 3788 turbine at a pressure ratio of 2:1 and EGT of 1000º. Your drive pressure is around 15psi.
Now, all else being equal, how much hotter would the EGT have to get to power the next larger size turbo (HT4088/4094) to the same point? Answer: about 250º hotter EGT. The next larger size housing for the bigger turbo is an ADDITIONAL 200º or so over this!
Now, keep in mind that everything you change affects something else, and in this case, you are running hotter pushing a larger turbine, but this is in turn operating a larger, higher flowing compressor, so you won't see the full temperature increase that turbine calculations alone would predict.
Hopefully this will help some of you make a little more sense of what's happening on the hot side of the turbo. Garrett is kind enough to provide all the data we need to compare any of their turbos to each other.
Now, if only we could get the major diesel turbocharger vendors to demonstrate that they aren't afraid of an informed customer, we could make meaningful turbo comparisons. Too bad all we have to go on with most diesel turbochargers is "such and such did this with that, and trust us our turbo kicks butt."
Anyone have turbine maps for the "diesel" turbos? Ha, we can't even get compressor maps from them! The people that have the compressor maps got them because they know a little something about the turbo and looked up the compressor wheels in some B-W catalog or database or something.
OK-- end minirant. I hope that this helps you understand a little about what's going on with the hot side of the turbo. There's more to it as far as outlet restriction and such like that, but this is a good starting point at how you can compare one turbo to another IF you have the data to do so.
Justin
First, let's looks at an example turbine map of a GT4088/4094:
This map has four separate curves, each representing the performance of the same turbine wheel with progressively larger turbine housings.
Across the left side we have numbers representing the corrected mass flow. We'll come back to the "corrected" part. Across the bottom, we have the pressure ratio across the turbine (keep in mind this is all in absolute pressures and temps and such).
So what do these curves on the map tell us?
As we increase the mass flow to the turbine, at first it can "absorb" the increase almost completely. It does this by spinning faster. But as you continue to increase the mass flow rate, you eventually hit a point where the turbine starts to pose a restriction to the flow. Increasing the mass flow past this point you will now see a definite restriction from the turbine, and the pressure drop across the turbine will increase. Looking at the level part of each curve, you can see that you get to a point where you cannot force the turbine to flow any more, no matter how much you increase the pressure ratio across the turbine. For example, the curve climbs pretty far from a 1.5PR to a 2 PR, but after 2 it's almost level up to a PR of 3.
It just so happens that this restriction is a good thing, because unless the turbine is restricting flow, it cannot extract any power from the gas to power the compressor.
Said another way, once the curve starts to level out, the turbo is fully spooled-- it has reached it's maximum efficiency, or ability to get work from the available gas energy. Also note that you'll sometimes see the right hand side labelled as a secondary Y axis showing "efficiency". It just so happens that efficiency increases with a larger housing.
Here's what I'm talking about:
See how the efficiency on the right side increases with A/R? The tightest housing here is 86% or 87% efficient, but the largest housing is approaching 97%. (dubious numbers? but the point is illustrated).
So as you step on the pedal, the mass flow delivered to the turbo increases until the turbine start to choke things a bit, and then you'll see the initial spoolup. Once you hit this point, you'll see the first stages of positive manifold pressure (boost) because you're actually starting to build a little drive pressure. (remember, without a pressure drop across the turbine, there's no work be extracted).
So, what does this tell us about spoolup??
This is where we get into the "corrected" mass flow. The turbine map mass flow figures are "Corrected" to represent sea level pressure on a 59º day. So we have to "correct" our EGTs (or Turbine Inlet Temp, and I won't abbreviate that
) so they correlate to the turbine map.This can get into a chicken-or-the-egg scenario, so you have to fix a variable and proceed from there.
For example, let's say that I have my CTD spinning 1800rpm with no turbo attached and EGTs are 1000º. What's my mass flow? Would this spool a Garrett GT4088 or GT4094?
At 1800rpm, my engine is swallowing 187cfm of air. At "standard pressure and temp" of 70ºF and 14.7psia atmosperic pressure, this 187cfm converts to about 14 lb/min. This is the air going into the engine. What goes in must come out, but we're also burning fuel. So, depending on air fuel ratio, we'll have anywhere from another .5lb to 1lb more per min from the fuel. Let's say we're running lean and it only took .5LB of fuel in that minute.
So now we have about 14.5lb/min coming out of the engine, but at 1000ºF temperature. Since we have no turbo attached, let's say that there's no backpressure in the manifold or that it's negligible.
When we "correct" to atmospheric conditions, this 14.5 lb/min at 1000º becomes 23.8 lb/min! This is how the hotter drive temps show up as more energy to drive the turbine.
So let's say we want to slap on a turbo. Will the Garrett GT4088 be spooled up with this amount of drive energy if we use the tightest housing?
Referring back to the first pic (turbine map of said GT4088), we see that the "corrected" 23.8 lb/min falls on the lowermost curve around a PR of 1.25. Unfortunately, this isn't enough PR to really extract much useful work to the compressor. Our "drive pressure" in the manifold is only about 8psi on the gauge. The turbine just isn't restrictive enough at this point, and you'll have no boost.
But what if we increased both RPM and EGT of our fictional turboless engine? What if we went up to 2000 rpm and EGT of 1200º? Now the "corrected" mass flow goes up to almost 29 lb/min. Plotting this point on the turbine map shows that we are now at a PR of right around 1.5, and the curve is starting to level out. This represents the beginning of the operating range for this turbine, and you'd just be seeing the first stages of actual boost to your engine.
Another useful application of a turbine map is predicting the performance of a turbo based on the known performance of another one-- you'll need the maps for both turbos.
I've lately been trading PMs with a member here trying to get info on how the Garrett Stg 3 (GT3788 with .99A/R housing) is performing for him. (Turbine map in the garrett PDF catalog from their site). He says he loves it, and he has about 4-6psi cruising on the hwy and gets instant response from it.
How would the GT4088 respond by comparison? Looking at the turbine maps for the the two turbos, let's select a reference point of 1.5PR, since this roughly corresponds to the threshold of the "operating range" of the turbo. The GT3788 at this point is flowing about 26lb/min "corrected." The 4088 with the tightest housing is right around 29lb/min at the same point. In other words, it takes 3lb/min more to put the 4088's turbine where the 3788's turbine is. In other words, it's about 12% laggier to get to that point of its operating range. At a PR of 2, the difference is about 2lb/min, or about 7% more.
Keep in mind this is only just the THRESHOLD of boost, not how fast it will accelerate from 10-40psi of boost or anything. That has little to do with the turbine map, and more to do with wheel sizing and rotational inertia.
Speaking of that: note that the GT4094 and the 4088 use the same turbine, and for the response penalty outlined above, you can step up to a legitimate 550hp turbo in teh GT4094. Now, to make the big top end you'll want to run a housing the next step up from the .85 and the lag penalty will be another 2-3 lb/min mass flow. Doesn't sound like much,but that can be pretty extensive in terms of EGT.
For example, let's say you have a 3788 turbine at a pressure ratio of 2:1 and EGT of 1000º. Your drive pressure is around 15psi.
Now, all else being equal, how much hotter would the EGT have to get to power the next larger size turbo (HT4088/4094) to the same point? Answer: about 250º hotter EGT. The next larger size housing for the bigger turbo is an ADDITIONAL 200º or so over this!
Now, keep in mind that everything you change affects something else, and in this case, you are running hotter pushing a larger turbine, but this is in turn operating a larger, higher flowing compressor, so you won't see the full temperature increase that turbine calculations alone would predict.
Hopefully this will help some of you make a little more sense of what's happening on the hot side of the turbo. Garrett is kind enough to provide all the data we need to compare any of their turbos to each other.
Now, if only we could get the major diesel turbocharger vendors to demonstrate that they aren't afraid of an informed customer, we could make meaningful turbo comparisons. Too bad all we have to go on with most diesel turbochargers is "such and such did this with that, and trust us our turbo kicks butt."
Anyone have turbine maps for the "diesel" turbos? Ha, we can't even get compressor maps from them! The people that have the compressor maps got them because they know a little something about the turbo and looked up the compressor wheels in some B-W catalog or database or something.
OK-- end minirant. I hope that this helps you understand a little about what's going on with the hot side of the turbo. There's more to it as far as outlet restriction and such like that, but this is a good starting point at how you can compare one turbo to another IF you have the data to do so.
Justin
.
09-10-2007, 12:58 PM
#4
Registered User
Join Date: Nov 2005
Location: Springfield Va.
Posts: 451
Likes: 0
Received 0 Likes
on
0 Posts
Really cool of you to take the time to explain all this to us (I need it more than most but I'm no dummy 
) but alas I'm terrible at new languages 
and while it's well written I was lost after the statement: First, let's looks at an example turbine map of a GT4088/4094: 
I'm a pretty good wrench but I think I'll leave theory and selection to the experts...
) but alas I'm terrible at new languages and while it's well written I was lost after the statement: First, let's looks at an example turbine map of a GT4088/4094: I'm a pretty good wrench but I think I'll leave theory and selection to the experts...
09-10-2007, 07:16 PM
#6
Registered User
Thread Starter
Join Date: Apr 2003
Location: Cummins Technical Center, IN
Posts: 6,564
Likes: 0
Received 6 Likes
on
5 Posts
Really cool of you to take the time to explain all this to us (I need it more than most but I'm no dummy ) but alas I'm terrible at new languages and while it's well written I was lost after the statement: First, let's looks at an example turbine map of a GT4088/4094: I'm a pretty good wrench but I think I'll leave theory and selection to the experts...
) but alas I'm terrible at new languages and while it's well written I was lost after the statement: First, let's looks at an example turbine map of a GT4088/4094: I'm a pretty good wrench but I think I'll leave theory and selection to the experts...My point is that even a typical shlep like me can learn this stuff and benefit from it.
Justin
10-01-2007, 06:19 PM
#7
Registered User
Join Date: Feb 2003
Location: Let'r Buck!
Posts: 137
Likes: 0
Received 0 Likes
on
0 Posts
Justin -
Thanks for taking the time and thought you put into all of the information you share here. Lot's of guys on here, like yourself, really push me to learn and apply what I do know to this 'hobby.' I had a conversation with an old friend of mine regarding some of the information here. I just thought I would add some questions/discussion to this thread.
The turbine is ALWAYS a flow restriction regardless of mass flow rate. Are you implying that flow restriction due to the turbine is a bad thing? Becasue if it is not resulting in any pressure drop, then it isn't providing work to the compressor that results in boost, right?
Help me understand the comparision you make regarding EGTs. Are you saying that you would like to control the EGTs and increase them to increase boost?
What is the OPTIMUM effeciency range of the cummins? 1900-2100 RPM?
I'll have to get my turbines book out and touch up on some of this stuff as the ole noodle is pretty rusty. Maybe too much
+ 
.
Thanks for taking the time and thought you put into all of the information you share here. Lot's of guys on here, like yourself, really push me to learn and apply what I do know to this 'hobby.' I had a conversation with an old friend of mine regarding some of the information here. I just thought I would add some questions/discussion to this thread.
Now, all else being equal, how much hotter would the EGT have to get to power the next larger size turbo (HT4088/4094) to the same point? Answer: about 250º hotter EGT. The next larger size housing for the bigger turbo is an ADDITIONAL 200º or so over this!
Now, keep in mind that everything you change affects something else, and in this case, you are running hotter pushing a larger turbine, but this is in turn operating a larger, higher flowing compressor, so you won't see the full temperature increase that turbine calculations alone would predict.
Now, keep in mind that everything you change affects something else, and in this case, you are running hotter pushing a larger turbine, but this is in turn operating a larger, higher flowing compressor, so you won't see the full temperature increase that turbine calculations alone would predict.
What is the OPTIMUM effeciency range of the cummins? 1900-2100 RPM?
I'll have to get my turbines book out and touch up on some of this stuff as the ole noodle is pretty rusty. Maybe too much
+ .
Trending Topics
10-01-2007, 07:24 PM
#8
Registered User
Thread Starter
Join Date: Apr 2003
Location: Cummins Technical Center, IN
Posts: 6,564
Likes: 0
Received 6 Likes
on
5 Posts
Justin -
The turbine is ALWAYS a flow restriction regardless of mass flow rate. Are you implying that flow restriction due to the turbine is a bad thing? Becasue if it is not resulting in any pressure drop, then it isn't providing work to the compressor that results in boost, right?
The turbine is ALWAYS a flow restriction regardless of mass flow rate. Are you implying that flow restriction due to the turbine is a bad thing? Becasue if it is not resulting in any pressure drop, then it isn't providing work to the compressor that results in boost, right?
So think of the "lb/min" as being like gph. At 1 or 2 lb/min of mass flow, the turbine is essentially zero restriction. As mass flow increases, eventually you get to where the turbine is restrictive enough to build up pressure in front of it. The turbine lists "pressure ratio" which is because you are comparing the pressure on one side of the turbine to the pressure onthe other. On the backside of the turbine, you have atmospheric pressure pushing back, and on the frontside of the turbine you have drive pressure. So at PR of 1, you have no boost, because there's no restriction. At a PR of 1.5, you have about 7psi more pressure upstream of the turbine (and since most turbos have drive pressure and boost pressure fairly comparable, this is where you'll see a couple PSI of boost).
As the mass flow in lb/min increases more, it gets "converted" to higher PRs by the restriction of the turbine.
Help me understand the comparision you make regarding EGTs. Are you saying that you would like to control the EGTs and increase them to increase boost?
What is the OPTIMUM effeciency range of the cummins? 1900-2100 RPM?
What is the OPTIMUM effeciency range of the cummins? 1900-2100 RPM?
The optimum efficiency range of a 24V is 2000rpm, based on BSFC (point of lowest BSFC). For a 12V, it's 1600rpm.
Hope this doesn't make things more confusing than they already are. Sometimes my wording is pretty
.
10-01-2007, 07:44 PM
#9
Registered User
Join Date: Jun 2007
Location: Schaefferstown, PA
Posts: 743
Likes: 0
Received 0 Likes
on
0 Posts
thanks a lot justin, this really helped me come closer to understanding turbo's. Are there any posts on here explaining now to read a compressor map? I've seen them before, but could never make heads of tails of them?
10-01-2007, 07:48 PM
#10
Registered User
Thread Starter
Join Date: Apr 2003
Location: Cummins Technical Center, IN
Posts: 6,564
Likes: 0
Received 6 Likes
on
5 Posts
I could make one for you.... You want one? It would take a little while because you end up having to crunch some numbers to get values that mean something to a compressor map. None of the math is beyond 6th grade level though.
One you get the numbers you need, reading a comp map is easy.
justin
One you get the numbers you need, reading a comp map is easy.
justin
10-01-2007, 10:03 PM
#13
Banned
Join Date: Aug 2005
Location: WV
Posts: 408
Likes: 0
Received 0 Likes
on
0 Posts
1600 for a 12v?? I know the torque numbers for a stock 12v were max at 1600 but for the life of me I can't see how. My truck, when it was stock would not even start to pull until 1800. Under that the afc would not give fuel and no boost. At 1800 it was like flipping a switch and off you'd go. I can't believe I can remember that far back!!!!!!
10-01-2007, 10:10 PM
#14
DTR 1st Sergeant
10-01-2007, 10:20 PM
#15
Registered User
Thread Starter
Join Date: Apr 2003
Location: Cummins Technical Center, IN
Posts: 6,564
Likes: 0
Received 6 Likes
on
5 Posts
1600 for a 12v?? I know the torque numbers for a stock 12v were max at 1600 but for the life of me I can't see how. My truck, when it was stock would not even start to pull until 1800. Under that the afc would not give fuel and no boost. At 1800 it was like flipping a switch and off you'd go. I can't believe I can remember that far back!!!!!!
It has almost nothing to do with power production.
Justin