MATH+Logic: An engineering approach in search for a perfect TURBO. (1 Viewer)

This site may earn a commission from merchant affiliate
links, including eBay, Amazon, Skimlinks, and others.

Magic Spool Bus

SILVER Star
Joined
Jan 23, 2018
Threads
13
Messages
217
Location
Houston, TX
*** Apologies for such a long post but there is much to discuss ***:bang:

I am hoping to get some brilliant minds in here to help facilitate a scientific approach I suppose in exploring different turbo options that may have not been tried yet. My criteria for searching for a turbo is to find a large enough turbo to lower backpressure and get as close to 1:1 EMP:IMP ratios or lower but not to exceed +1.5:1 and still be fun to drive in the lower RPMs (1500-2000rpm) which we know high EMP:IMP (exhaust : intake manifold pressure ratio) say 1.7-2 and higher is not efficient for an engine, I realize some may not care about this as much as others. Going by what I see on the forums this may be considered TOO big by most here for our engines to spool up as most want all the power as early as possible which usually leads to a turbo so small it chokes on top end, different strokes for different folks. To me, there are a lot of different options out there for finding a modern efficient turbo that could work well on these engines that may have not been attempted out of either lack of knowledge in turbo selection, or if it has, not in a way where it is public knowledge on some of these forums we have or even because the cost will be higher and more complicated to fit rather than just bolt in turbos based off the CT26. When I say different options specifically I have been interested in Borg Warner turbos. As most will know they offer some rather advanced technology when it comes to turbos. The EFR line has things like twin-scroll exhaust housings, Ti-Gamma extremely lightweight turbine wheels, and even very well designed internal wastegate ports among other things. Twin scroll housings incorporated with a true split manifold is proven technology that works decreasing spool time while maintaining high RPM power. The diesel guys in USA are pretty sold on this in the Cummins community but we don't see it much here for whatever reason of which I can't figure out, which leads me down a path to try it. As we know the EFR got a bad wrap with some turbine wheel failures early on but that can be a conversation for later, my own conclusion was these turbos did exceed their shaft limit, just my well thought out opinion.

BW also has a very neat calculator that helps you select a turbo which I have used quite a bit and I believe I'm in a good ballpark with it. It is very in-depth and needs basic values such as engine displacement, fuel type, altitude, but it has more complicated values of things such as Volumetric Efficiency, AFR, turbine and compressor efficiency, intercooler efficiency, and many others all broken down at 6 RPM points. The values I struggle to calculate accurately is mainly Volumetric Efficiency at given rpm points. This is a major contributor to matching a turbo to an engine for the desired HP goal. I have come very near to what I thought was a way to estimate VE through calculation but feel I'm missing a piece to the puzzle somewhere. This is where I am and looking for help...

My truck is a 1HDT 5 speed, G turbo Green, front mount intercooler, big airbox, 3.5in exhaust with 1 muffler. Using a Garret tech link to learn to plot a point on a compressor map of a turbo in choice for the desired HP. In my case, I'm using 290 crank hp to hopefully get me in the ballpark of 250WHP. This is a rough estimate calculated by multiplying 250 and 1.15 (typical drivetrain loss for rear-wheel drive). Which gives me 287 and I rounded to 290.

Step 1) we need the Pressure Ratio (PR) for our desired target Boost level. PR=(PSIg + absolute atmospheric pressure)/(atmospheric pressure(14.7) - 1 system depression or air ducting restriction). My values would be 25 PSIg + 14.7 psi being at sea level divided by 13.7 = 2.9 Pressure Ratio. PSIg is the gauge pressure above ambient.

Step 2) Calculate Air Flow Demand we will call it AD to make it simple. AD=(target hp)x(air-fuel ratio)x(BSFC/60). For BSFC garret gives an estimation for diesel at (.36). This could be inaccurate.

So AD = (290) (20) (.36)
AD = 35lb/min to make 290 crank hp


Step3) Calculate Manifold Absolute Pressure (MAP/req). MAP/req=(airflow demands from above x gas constant x 460+Intake air tempt)/(VE
x Max engine rpm/2 x engine displacement in CID).
This is where it gets tricky for me. The gas constant is just a value garret gives as 639.6 and I may be confused but wonder if there is a diesel constant value but my logic may be wrong. VE is also an estimation of .80 from garret for a 2 valve per cylinder engine. For intake air temp I am using 150F as I am intercooled. These values could be inaccurate. For VE I used 75% so .75 for the value of VE in the equation. I'll explain why I used 75% later.

MAP/req=(35lb/min x 639.6 x 610)/(.75 x 2100 x 256) = 33.87
MAP/req=33.87


Subtracting 14.7 from this value also gives us our Gauge PSI to make 290HP which is 19.17psi on gauge. This value seems off to me compared to what most run here to get 200+hp.

Then we correct for pressure loss in the IC by adding 2 as a value of typical loss from Garret. MAP + 2 = 35.87 MAP/req.

If we calculate the boost gauge now after the new MAP of 35.87 we get 21.17psi on the gauge which is making more sense.

Now we calculate a new PR based on these values we have found. PR = MAP / 13.7(system depression from before)
=2.6 pressure ratio.


Last we can now start plotting different rpm points below max rpm. The Equations is (MAP x VE x RPM/2 x Discplacement) / (Gas constant x (460 + IAT)

Lets plot 3100 rpm because this seems to be our peak power RPM on a 1hdt. (35.87 x .75 x 3100/2 x 256) / (639.6 x 610) = 27.36 LB/MIN PR still 2.6

Next 2100 RPM as this seems to be peak torque. = 18.53 LB/MIN PR 2.2?

Last point of 1800 rpm just to see air flow at lower rpm. = 15.88 LB/MIN I used PR of 2 as just a guess?

Garret states after plotting peak power and torque we can use less Pressure Ratio values to see how the turbo will act with normal driving at lower RPMs.


This is all from How To Select A Turbo Part 2: Calculations - Garrett Motion if you want to check my work.

Now we can plot a point on a compressor map to see if it will flow our required air at a given pressure ratio and at what efficiency. This is a compressor map of an EFR 7064 and below that a Garret GT3071. EFR 7064 70/52 exducer/inducer and a 64mm turbine wheel lightweight Ti Gamma. To plot this point we find 2.6 on the Y or left axis and MAP/req of 35.87 on bottom X-axis. You will see we are nearly directly in the middle of the efficiency islands but this is just a consideration of a turbo I have been looking at. There are many more.

borgwarner-efr-7064-turbo-2-content-9.jpg

Comp-Map-GT-3071R.jpg

Ignore the black lines. I suck at editing.

Using the Match Bot on BW I have dialed in the 7064 and it seems quite a good fit. It would have a .92 A/R twin scroll rear housing. BW advises raising turbine efficiency in the values set by 15% for point 1 10% for point 2 and 5% for point three because of the added efficiency of the twin-scroll housing so this is reflected on match bot. I will upload the link here so other more intelligent minds can play with it. I struggle to figure out all of the values but have made very educated guesses on certain values of pressure losses or gains in piping, intake, and exhaust with the BW tutorial videos and hints in the menus. Volumetric Efficiency being the most elusive. I have a theory though.

In the MAP/req equation (airflow x gas constant x 460 + IAT) / (VE? x RPM/2 x Displacement). Well, we discover through this equation what MAP is which is 33.87. So if we set the equation equal to this and replace the estimated value of .80 VE with X and then solve for X wouldn't this be our Volumetric Efficiency? Doing this I come up with 75% VE. My logic could be inaccurate.

What I hope to achieve with this post is to gain and spread knowledge about a more in-depth look at selecting an alternate turbo for our engines so it will help me make a decision on a unique turbo setup. Hopefully, someone more intelligent than me can offer help or just people out there using what would be considered an ODD turbo selection outside the norm with real-world experience. Even if it's just correcting my math or logic please do I'm here to learn. One concept I really struggle with is mapping a turbine map:crybaby:. Luckily the Match Bot does that for us.

Here is the Match Bot link I have completed to the best of my current abilities. My engine RPMs reflect the ranges I'm more concerned about but anything can be adjusted and if you do you just need to copy and paste the link that you made adjustments on. VE starts at 75% tapers up to 80% for peak power and back down to 75% in the higher rpm. The EFR7064 is more of a middle of the pack between low down response and not choking on the top end. Also, it seems backpressure is near always less than 1:1 EMP vs IMP which is great. There are non EFR options like the S252SXE its a 70/52 compressor and a 61/70 turbine. For all of these, I'm trying my best to run it with a twin-scroll housing. The issue with the SXE line is you lose the internal wastegate and you get a traditional Inconel turbine wheel but it's much much cheaper than the EFR 7064. So sorting out twin external wastagtes for the SXE is a problem if we want to run a true divided setup and all the money you saved goes to fabbing a manifold and what not. Open to single scroll but then it becomes much harder to get a turbo that I believe will give me the characteristics I'm looking for. A properly sized single with decent bottom end and great top end. Compounds would also solve this but sizing those correctly is even more difficult.

 
Last edited:
My HDJ80 on 285/76/16 made 76 rwkw with front shaft out and CDL locked, pretty much dead stock. That's approx 65% of flywheel power. I think you're gonna need to make circa 390 at the fly. I made 160 rwkw and 600 rwnm at 3600 and 1800 respectively, 25 pounds of boost with redwheel, moonlight snorkel and airbox, worked pump, 600x400x76 cooler and a 3" exhaust
 
Last edited:
My HDJ80 on 285/76/16 made 76 rwkw with front shaft out and CDL locked, pretty much dead stock. That's approx 65% of flywheel power. I think you're gonna need to make circa 390 at the fly. I made 160 rwkw and 600 rwnm at 3600 and 1800 respectively, 25 pounds of boost with redwheel, moonlight snorkel and airbox, worked pump, 600x400x76 cooler and a 3" exhaust
While you are correct in probably needing to make 390 at the flywheel to get 250whp. I’m also not entirely concerned about a hard number from a dyno as I am with just increasing the crank hp to my goal of around 300hp regardless of what drivetrain loss equates too. 390 crank might not last long :rofl:
 
While you are correct in probably needing to make 390 at the flywheel to get 250whp. I’m also not entirely concerned about a hard number from a dyno as I am with just increasing the crank hp to my goal of around 300hp regardless of what drivetrain loss equates too. 390 crank might not last long :rofl:
Nah she'll be right at 390 crank...I used to tow with mine for ~ 10 hour trips and never any sign of mechanical distress.
 
Nah she'll be right at 390 crank...I used to tow with mine for ~ 10 hour trips and never any sign of mechanical distress.
I am going to play with some numbers now with a higher drivetrain loss like you pointed out just to see how much the values change.
 
I am going to play with some numbers now with a higher drivetrain loss like you pointed out just to see how much the values change.
I'd take a stab that if you had stock tyres you'd probably only lose 30%
 
Yeah I've been sizing turbos like that for about 15 years now. The matchbot came along later and the EFR series have produced a few unexplained turbo failures. Often involving the wheel coming off the shaft.

Total drive pressure depends on total turbo efficiency, how hot your EGT is and where you are on the turbine map. Hotter EGT gives you lower drive pressures and low engine rpm is required too.

Even my tiny garrett T25 on a 4BD1T can do better than 1:1 if rpm is below 2000rpm and EGT hotter than 650C.
 
Yeah I've been sizing turbos like that for about 15 years now. The matchbot came along later and the EFR series have produced a few unexplained turbo failures. Often involving the wheel coming off the shaft.

Total drive pressure depends on total turbo efficiency, how hot your EGT is and where you are on the turbine map. Hotter EGT gives you lower drive pressures and low engine rpm is required too.

Even my tiny garrett T25 on a 4BD1T can do better than 1:1 if rpm is below 2000rpm and EGT hotter than 650C.
Dougal, I have gained quite a bit of knowledge from a lot of the posts you are involved in with the 4bt, and patrol forums, and even here. I know you have done some calculating on Volumetric Efficiency and I'm curious to know your thought on this and our 12v engines. Is there even any way to calculate this roughly without the use of MAP and MAF sensors? Did you see any major discrepancies in my math?

When it comes to drive pressure I know many things affect this, rpm, restrictions in the exhaust, to small of a turbine side. My goal is to never exceed 1.2-1.5:1 at any given rpm. If BW match bot is correct some of these turbos I'm looking at will absolutely do this but I don't know the real-world results or how can calculate them myself to compensate for any shortcomings in the calculator.
 
Last edited:
Dougal, I have gained quite a bit of knowledge from a lot of the posts you are involved in with the 4bt, and patrol forums, and even here. I know you have done some calculating on Volumetric Efficiency and I'm curious to know your thought on this and our 12v engines. Is there even any way to calculate this roughly without the use of MAP and MAF sensors? Did you see any major discrepancies in my math?

When it comes to drive pressure I know many things affect this, rpm, restrictions in the exhaust, to small of a turbine side. My goal is to never exceed 1.2-1.5:1 at any given rpm. If BW match bot is correct some of these turbos I'm looking at will absolutely do this but I don't know the real-world results or how can calculate them myself to compensate for any shortcomings in the calculator.

I haven't had the time to check your numbers. The only robust way for me to do that is to setup a side by side calculation example. That takes about 2 hours and spare time isn't something I have any more.

The two biggest factors in emp/boost ratio are the efficiencies of the compressor and turbine. If they are either inefficient designs or running on the edges the results will be poor. Turbine size isn't the factor that most think it is. It changes your operating rpm range more than your boost/drive ratio.

Your turbine isn't undersize until you're wastegating flow around it. That's when total turbine efficiency takes a dive because you're only extracting power from part of the flow.

It has been years since I had drive and boost gauges running. Back then I had a lot of ugly intake and cross-over piping and was still able to do better than 1:1 at best point and about 1.5:1 at 2000rpm cruise. This was with a turbo design that is about 30 years old now. I think the better piping I have now would just make those ratios a little more favourable and available over a wider range.

For VE I have done it two ways.
For engines with all the factory info available (fuel injection calibration figures, boost numbers, torque curves etc) I can come up with a calculation set that can get airflow at max torque and max power to within a few percent.
I've done enough of those that I can just throw similar numbers at similar engines. Ball park figures. At max torque I use 95% for a 4 valve engine and 90% for a 2 valve engine. At max power I use 5-10% less depending on rpm and how good the engine is. For an old engine with less sophisticated intake design (i.e. Toyota 3B) It's more like 75% at max power.

Stock 1HD-T I have VE at 90% until 2000rpm. Down to 80% at 3500rpm.
BSFC at 2000rpm is 260 g/kwh. at 3500rpm that's almost 290 g/kwh.

But. Those BSFC numbers can be better with a better turbo.
 
I haven't had the time to check your numbers. The only robust way for me to do that is to setup a side by side calculation example. That takes about 2 hours and spare time isn't something I have any more.

The two biggest factors in emp/boost ratio are the efficiencies of the compressor and turbine. If they are either inefficient designs or running on the edges the results will be poor. Turbine size isn't the factor that most think it is. It changes your operating rpm range more than your boost/drive ratio.

Your turbine isn't undersize until you're wastegating flow around it. That's when total turbine efficiency takes a dive because you're only extracting power from part of the flow.

It has been years since I had drive and boost gauges running. Back then I had a lot of ugly intake and cross-over piping and was still able to do better than 1:1 at best point and about 1.5:1 at 2000rpm cruise. This was with a turbo design that is about 30 years old now. I think the better piping I have now would just make those ratios a little more favourable and available over a wider range.

For VE I have done it two ways.
For engines with all the factory info available (fuel injection calibration figures, boost numbers, torque curves etc) I can come up with a calculation set that can get airflow at max torque and max power to within a few percent.
I've done enough of those that I can just throw similar numbers at similar engines. Ball park figures. At max torque I use 95% for a 4 valve engine and 90% for a 2 valve engine. At max power I use 5-10% less depending on rpm and how good the engine is. For an old engine with less sophisticated intake design (i.e. Toyota 3B) It's more like 75% at max power.

Stock 1HD-T I have VE at 90% until 2000rpm. Down to 80% at 3500rpm.
BSFC at 2000rpm is 260 g/kwh. at 3500rpm that's almost 290 g/kwh.

But. Those BSFC numbers can be better with a better turbo.
Thanks for this info, Dougal. At least now I know I was in the ballpark for VE. This change in VE dramatically changes the power output and puts the turbo in a very good efficiency island in the turbo map. I'm hesitant to mess with the BSFC values but I did do a comparison of what the matchbot suggests as BSFC vs what you suggest and output is affected dramatically for the same turbo. On one the BSFC is 206 g/kwh rising up to 231g/kwh in higher rpm as BW suggests for diesel. On the other, I put in 231 g/kwh rising to 255 g/kwh for the same turbo. In doing this with the higher BSFC value it lost 31 peak hp and 78 ft/lbs of torque. I finally came across a wastegate solution for running a twin-scroll turbine and manifold and only one external wastegate. Originally I thought I would have to run two external gates to keep the scrolls separate. I found this...

unnamed-2.jpg


This would allow me to run a single gate off the twin-scroll turbo housing. Its sandwiched between the turbo and manifold where a tube is fitted to the side ports for the wastegate. So now my focus has shifted to the much cheaper and probably more durable S252SXE. Here is the updated compressor map after adjusting VE from dougals input with points plotted from 1500 rpm to 4000rpm

e70s75.jpg


The calculator lets me get 25psi at 2000 rpm but I don't know how realistic that is with this turbo being a .83 A/R twin-scroll and a 70mm turbine wheel. Either way im starting to convince myself to just commit to this route and see what happens.
 
Great turbo choice! @Socal81 that's the exact one I was getting until I got my modified CT26 to perform so well it made me put plans on hold for a while. It actually runs the very same comp wheel that's in the EFR 7064 you were looking at getting. Obviously the turbine is much larger though being 70/61mm vs the 64/56mm of the EFR 7064.
 
Great turbo choice! @Socal81 that's the exact one I was getting until I got my modified CT26 to perform so well it made me put plans on hold for a while. It actually runs the very same comp wheel that's in the EFR 7064 you were looking at getting. Obviously the turbine is much larger though being 70/61mm vs the 64/56mm of the EFR 7064.
So many options :worms:
 
Another question that has plagued my mind is how to predict compressor surge for low load cruising. In match bot or when you manually calculate Pressure Ratios, it seems that low-speed or low load situations aren't taken into consideration. Meaning if I calculate a pressure ratio based on what boost I want to run let's say 25psi its roughly a PR of 2.8. This will give me two points on the graph for peak power and peak torque and you will either be on the map or not depending on the size of the turbo compressor. It seems with the calculator and most other sites that help you calculate, that the only consideration is high load max output situations. In match bot, for example, my first rpm point is 1500rpm, AFR is 15:1(this is just a value I see on my own gauge briefly before the turbo leans it out if I floor it at 1500rpm) so at this point, it's assuming I'm taking off like a rocket ship. So to simulate a low load cruising situation I bumped AFR to 25:1 EGT lowered to 600F and lowered boost to 5psi. This will plot a point very low off the map. Shown below...
e70s75-2.jpg


We know the blue line is the surge line but nowhere does it mention the red line at the bottom of the map that I have drawn, so it could be unimportant. What does it mean for a turbo to operate in this region? I could be asking the wrong questions. Maybe by calculating a turbo for max output and staying within the map sorts out all the low load operating points but this is an assumption as nothing that I find about sizing a turbo does it every mention these normal driving situations and instead its all max output and input values.
 
Thanks for this info, Dougal. At least now I know I was in the ballpark for VE. This change in VE dramatically changes the power output and puts the turbo in a very good efficiency island in the turbo map. I'm hesitant to mess with the BSFC values but I did do a comparison of what the matchbot suggests as BSFC vs what you suggest and output is affected dramatically for the same turbo. On one the BSFC is 206 g/kwh rising up to 231g/kwh in higher rpm as BW suggests for diesel. On the other, I put in 231 g/kwh rising to 255 g/kwh for the same turbo. In doing this with the higher BSFC value it lost 31 peak hp and 78 ft/lbs of torque. I finally came across a wastegate solution for running a twin-scroll turbine and manifold and only one external wastegate. Originally I thought I would have to run two external gates to keep the scrolls separate. I found this...

View attachment 2445447

This would allow me to run a single gate off the twin-scroll turbo housing. Its sandwiched between the turbo and manifold where a tube is fitted to the side ports for the wastegate. So now my focus has shifted to the much cheaper and probably more durable S252SXE. Here is the updated compressor map after adjusting VE from dougals input with points plotted from 1500 rpm to 4000rpm

View attachment 2445515

The calculator lets me get 25psi at 2000 rpm but I don't know how realistic that is with this turbo being a .83 A/R twin-scroll and a 70mm turbine wheel. Either way im starting to convince myself to just commit to this route and see what happens.

BSFC of 206g/kwh is a diesel of very high efficiency operating at or very near to best point near max torque. No way a 1HD-T is going to get there. I'd expect 230 g/kwh will be more realistic for max torque and 250-260 g/kwh for max power when you've got a very efficient turbo bolted on.

I've got TD05H-16G with a 7cm turbine housing and TD06H with an 8cm turbine housing mapped out. Your compressor response looks similar to the 18G with 8cm turbine housing.
But you are spending a lot of your operating range in areas of low compressor efficiency. I would look for a smaller compressor so the ramp up to full boost can happen closer to the middle of the map. This will give you a lot better spool and better part load efficiency where fuel economy is made or lost.

Don't worry about the bottom end of the map. It all tapers down to meet at 1 and 0. I do map out a 2000rpm point at low load cruise. On my 4BD1T that's about 35:1 AFR, 380C EGT and about 8psi boost.

Post up your turbine map and EGT.
 
BSFC of 206g/kwh is a diesel of very high efficiency operating at or very near to best point near max torque. No way a 1HD-T is going to get there. I'd expect 230 g/kwh will be more realistic for max torque and 250-260 g/kwh for max power when you've got a very efficient turbo bolted on.

I've got TD05H-16G with a 7cm turbine housing and TD06H with an 8cm turbine housing mapped out. Your compressor response looks similar to the 18G with 8cm turbine housing.
But you are spending a lot of your operating range in areas of low compressor efficiency. I would look for a smaller compressor so the ramp up to full boost can happen closer to the middle of the map. This will give you a lot better spool and better part load efficiency where fuel economy is made or lost.

Don't worry about the bottom end of the map. It all tapers down to meet at 1 and 0. I do map out a 2000rpm point at low load cruise. On my 4BD1T that's about 35:1 AFR, 380C EGT and about 8psi boost.

Post up your turbine map and EGT.
I should say my main goal with this turbo is slightly more top-end power knowing and being fine with sacrificing a little low-end response to achieve this. So when I read this compressor map from up a few posts on the 252SXE I see that at 2000 rpm I am at 71% efficiency and from 2500-3500 rpm I am at 75% percent efficiency. I have run every turbo from the EFR 6258 to the EFR 7670 through match bot and no other turbo gets me to this high of efficiency except the EFR 7064 comes in a close second. As far as a turbine map I do not have one. I can not find one for this turbo nor can I find a good source of information yet on how to graph one. Here is the most up to date Match Bot after changing a lot of the values you suggested.


EDIT: Dougal, I think I was missing the point of your post. So far when I have been calculating using 1350F across the board for egt by what BW suggest in the tutorials. I realize that this isn't a realistic figure as I would never be at 1350F below 2000 rpm unless I'm in a very specific high load low rpm situation for example say if I floor it at 1500rpm in 4th to test my bottom end AFR which I never drive this way any other time. Now when I lower the EGT values to something more realistic I can see what you mean about rising through the center of the map and this also forces me to run a smaller turbo potentially than the 252sxe. So am I correct in saying when I use this calculator I should try and simulate a normal driving condition as well as a max power condition with the EGT values to get a realistic idea of how the turbo will behave?
 
Last edited:
So after playing with match bot some more trying to simulate spool up response from 1500-3500rpm I ended up needing to step down a turbo size as the S252SXE just wasn't capable of making more than 5psi by 2500 rpm. EGT is 1500rpm=700F - 2000rpm=900F - 2500rpm=1000F - 3000rpm=1200F 3500rpm=1350. These are very rough values as I don't stare at my EGT constantly. In this, I'm trying to simulate sort of like a dyno pull in 4th gear at 1500rpm and beginning a hard pull. I know EGT effects turbo response dramatically but never considered it when using the match bot for whatever reason. Here is the turbine map from BW and the plotted compressor map on the EFR 7163 71/57 compressor wheel and 63mm turbine .80A/R twin-scroll hot side. I still couldn't get the map rising through the center quite as you suggested. EMP:IMP is around 1.2:1 with this turbo which is acceptable I suppose.

EDIT: 7163 is not a good match. Kiwidingo caught that I had an issue with intercooler pressure drop and muffler back pressure and this changed a lot. The 7163 is just not near its efficient island. The 7064 EFR really seems like the only viable option from what I'm trying to achieve and even then it's close to being too laggy. 11psi at 2000rpm and 25psi at 2500RPM.
IMG_3529 2.JPG
IMG_3530.JPG
 
Last edited:
Just a heads up on the 7163 if that's what you're now looking at. It's the only one out of the EFR range which has a different turbine wheel to all the others. The 7163 uses a mixed flow/axial type turbine where all the other EFR models use a more traditional radial flow type. It doesn't simulated it very well in matchbot but the mixed flow turbine has much higher EMP in the higher rpm and produces much more turbine drive in the lower rpm than what it appears in the turbine sizing map in the simulation.

Info below from the guys developing the EFR turbos;

"The mixed flow wheel in the 7163 has its efficiency peak at lower Uc/Uo ratios(0.5-0.6). This translates into better shaft torque when the wheel is moving slower relative to the exhaust gas speed than a normal radial flow designed wheel that will typically peak the efficiency in the Uc/Uo range of about 0.7-0.8.

The tradeoff is you'll have a bit more backpressure at high RPM/high boost situations as the wheel falls out of its ideal Uc/Uo range vs. the radial flow wheel that's really in its sweet spot."



Also a bit more info here on turbine types in a previous thread a while ago in case you haven't seen it. radial vs axial turbine tech
 
Last edited:
So after playing with match bot some more trying to simulate spool up response from 1500-3500rpm I ended up needing to step down a turbo size as the S252SXE just wasn't capable of making more than 5psi by 2500 rpm. EGT is 1500rpm=700F - 2000rpm=900F - 2500rpm=1000F - 3000rpm=1200F 3500rpm=1350. These are very rough values as I don't stare at my EGT constantly. In this, I'm trying to simulate sort of like a dyno pull in 4th gear at 1500rpm and beginning a hard pull. I know EGT effects turbo response dramatically but never considered it when using the match bot for whatever reason. Here is the turbine map from BW and the plotted compressor map on the EFR 7163 71/57 compressor wheel and 63mm turbine .80A/R twin-scroll hot side. I still couldn't get the map rising through the center quite as you suggested. EMP:IMP is around 1.2:1 with this turbo which is acceptable I suppose.

EDIT: 7163 is not a good match. Kiwidingo caught that I had an issue with intercooler pressure drop and muffler back pressure and this changed a lot. The 7163 is just not near its efficient island. The 7064 EFR really seems like the only viable option from what I'm trying to achieve and even then it's close to being too laggy. 11psi at 2000rpm and 25psi at 2500RPM.
View attachment 2448220View attachment 2448221

So that turbine map you're on is equivalen to ~ 20lb/min corrected choke flow (CCF) which is a quick way to compare turbine sizes. Take a look at some Garrett turbine maps using that number for the turbine map to level out at.
By my calculations a CT26 from a 1HD-T has a CCF of 14 lb/min. A CT20B from a 1HD-FTE is more like 12 lb/min. My old T25 is 11.5 lb/min.
More efficient turbos can be bigger CCF and still perform.
Hotter running turbos can be bigger CCF and still perform.
High altitude means you need a smaller CCF to perform.

Regarding EGT at lower rpm. That's exactly right. My old work car had an EGT gauge before it got a turbo and I couldn't get EGT hotter than about 550C and that took a while to heat-soak to that point.
EGT is strongly correlated with boost and drive pressure. When a turbo was added the EGT had no problem hitting 700-750C.

But also on that car. The turbo responded a lot better when the turbine was heat-soaked. A turbine that is a shade too big can really suck on cold acceleration.
 
So that turbine map you're on is equivalen to ~ 20lb/min corrected choke flow (CCF) which is a quick way to compare turbine sizes. Take a look at some Garrett turbine maps using that number for the turbine map to level out at.
By my calculations a CT26 from a 1HD-T has a CCF of 14 lb/min. A CT20B from a 1HD-FTE is more like 12 lb/min. My old T25 is 11.5 lb/min.
More efficient turbos can be bigger CCF and still perform.
Hotter running turbos can be bigger CCF and still perform.
High altitude means you need a smaller CCF to perform.

Regarding EGT at lower rpm. That's exactly right. My old work car had an EGT gauge before it got a turbo and I couldn't get EGT hotter than about 550C and that took a while to heat-soak to that point.
EGT is strongly correlated with boost and drive pressure. When a turbo was added the EGT had no problem hitting 700-750C.

But also on that car. The turbo responded a lot better when the turbine was heat-soaked. A turbine that is a shade too big can really suck on cold acceleration.
Dougal, Great info! This definitely helps me understand Turbine A/Rs and matching a bit better. One thing I cant figure out is how to convert Phi value on the BW maps to LBS/min that is on say a Garret map. So could you share how you achieved the 20lb/min from that graph? Just looking I obviously see .028 so is it technically 20.8 lbs/min or am I off?

It's also interesting what you point about CCF and turbos performing hot or cold based off design and efficiency. Just yesterday I was browsing a 90 page forum over on patrol4x4. Dudes over there doing some interesting stuff with the Patrols and EFR 6758 and 7163

Quote from OldMav "I just changed the turbo and a few adjustments to the gate to stop it humping, on my sons 2004 T ute. On the 6758 it was set to 204rwkw and 785Nm same fuel no changes it did 217rwkw and 815Nm. but all 100 to 150 rpm later. Driving it i couldn't tell the difference even off idle spool and driveability. It didn't blow off the recirc valve or i didn't notice it still made a racket from the snorkel which is normal from a efr. I have my doubts if it would work on less fuel but for this power tune its pretty good and would be a lot better than this had i played with pump tune. i am positive i could get it to spool the same as a 6758.with more early torque. AFR on the 6758 is at 16 to 17 highest and never see 650C pre turbo. The 7163 picked up the afr's to 17 to 18.5 highest. I would chuck a bit of smoke off idle with 14 to 15 to get 6758 off idle spool. Also emp:imp was all lower than 1.1:1 btw and egt was never above 600C after each dyno run pre turbo."

One thing interesting about the EFR is the lower inertia shaft as a product of the light Ti-Gamma turbine wheel. This paired with BW very well designed exhaust housing and efficient internal wastegate abilities it creates a more sophisticated way to tune this turbo. It's observed in these turbos that backpressure and EGT are lower down compared to say... a Garret equivalent. So not being able to drive the turbo with what's typically done by driving a turbine with EGT and backpressure being higher they had to come up with a way to drive the turbo to get the response they wanted in lower RPM cooler EGT areas with an electronic boost controller, specifically the Eboost street 2 from turbo smart. From what I could gather they use this controller to absolutely make sure the wastegate doesn't crack open for any reason other than the controller telling it to at a specific boost pressure which I believe for them was right at 24psi, Target boost was 28-30psi and they also had a 30psi wastegate can on the turbo making sure that the wastegate doesn't crack open with backpressure too early. This created a scenario where backpressure was reaching 1:1 right around 28psi on the gauge the wastegate would now swing open to control boost at a very specific moment that EMP was at a great drive ratio and EGT high enough to continue driving the turbo.

I may have butchered this a bit but the idea here I think is because of the efficient turbo and having a higher CCF they have to drive the turbo by making sure the gate doesn't open until a very specific moment 3-5psi before the target boost to get the turbo to respond very fast without sufficient EGT. With a manual controller set to 28 psi with the medium wastegate can that BW comes with it was causing the turbo to get near peak boost but then EMP was pushing the wastegate open early and the turbo would momentarily drop boost before rising again. My logic may be off a bit but this is currently how I interpret it.

Another interesting quote from OldMav explaining it

"The standard actuator high boost is a quality actuator but like all cans it is a compromise. The 30 psi can is full open at 30 psi and it cracks at about 16 to 18 psi as i remember (would need to check again). Ideally on a 11mm IP we don't want the actuator opening before 22 to 25 psi then we want the actuator controlling EMP from then on. In the real world we just don't have boost only or IMP holding the waste gate flap closed we also have exhaust pressure acting on the flap from the other side pushing it open. From Maths we can calculate that there is 1:1 emp:imp ratio at 2000 rpm (yet to be confirmed) at full fuel from a adjusted std 11mm at 28 psi with zero waste gate movement. So we have 28psi acting on the actuator from both directions. Ideally we want the actuator to only open from here on to control boost pressure but also to hold EMP at 28 psi also for an ideal 1:1 emp:imp ratio. A manual controller and single can has no reference to EMP or what the can is doing, all it can do is move the crack open point further up the boost graph. For our 30 psi can and with pre load we can set boost for 28 psi but what we are doing is setting a equilibrium between boost and exhaust pressure. So it will crack open about 20psi to maintain a 28psi pressure. But will not have the boost pressure to move the waste gate flap further than a few mm or about 10 degrees of flap of the 50 degrees possible needed for full gate open position. Thinking this logic i am sure you can see the problems with slap together can gate systems with manual controllers restricting full gate open condition on a gate hole too small for the engine to start with.

What happens here is we loose boost and we see boost drop off because the manual actuator has no control over EMP. So EMP builds up due to the can has run out boost control movement ability loosing lbs/min pumped but not necessarily Boost psi. This effect sees torque and Kw's drop off dramatically after 2000rpm which is very common to see on dyno runs. The electronic controller set up as normal has no direct control over EMP either but has the ability to hold 28psi then it has the ability to move the dual port actuator to full open if need be to control EMP and still keep 28 psi desired boost pressure from 2000 rpm onward. This effect has the ability to move the PR point across the compressor graph because we don't have a too high an EMP acting on the wheel loosing turbo shaft speed. Hence we get more Lbs/min pumped which equals torque holding ability, by product is Kw's."
 
Last edited:

Users who are viewing this thread

Back
Top Bottom