How To Choose The Right Turbo
Page 1 of 2

Author:  Grashopr [ Fri Apr 21, 2006 6:08 pm ]
Post subject:  How To Choose The Right Turbo

Not my writing. Found HERE


I decided now would be the perfect time for everyone to learn how to properly read a Turbo Compressor map, and how to choose a proper Turbo for your vehicle.

Here we will show a compressor map for reference.


First off on the Y axis is the pressure ratio. This is atmospheric pressure, or 14.7 psi = 1.
The X axis is the air flow in pounds of air per minute or lbs/min .

To convert lbs/min into CFM you factor in the ambient air temp. Compressor maps use 27 degrees centigrade, or 300 kelvin (To calculate Kelvin, take celcius and add 273) One cubic foot of air at 300k weighs 0.07282 pounds. So, at 300k, convert pounds per minute to CFM by multiplying by 13.73.

Ok, now you decide how much boost you want to run. Take that amount of PSI, add it to 14.7 (atmospheric pressure) and divide by 14.7. Lets say you want to run 15psi.

Pressure Ratio = (15 + 14.7) / 14.7 = 2.02

Find 2.02 on the Y Axis. Draw a line straight across at this point.. Look at the range on the X Axis the line is passing through the turbo's efficiency range. In this example the range is from 15 lbs/min to 35lbs/min.

Now, you will need to find the displacement in cubic inches of your engine. You may know this in Litres or CCs, but CI is what we need for this excercise.

Now, VE is volumetric efficiency, this is just the measure for how much air your engine can ingest. For this write up I'm going to make up an engine for the sake of easier math. Most forums will be able to give you an estimated VE at 6000rpms, or the FSM. Believe it or not the math is too involved for even this writeup. So the made up engine is 2226CC or 136 cu in and at 6000 rpm and 90 percent VE ( .90):

CFM = Displacement in CI / 3456 * RPM * VE

6000 * 136 / 3456 * .9 = 212.5 CFM

Ok, so this looks pretty good comparing to our graph and early equation. BUT, this is just the amount of air the car will flow N/A. So if you do this math for a different engine and it doesn’t land in that range don’t concern yourself. To determine what it will do under boost, you have to determine what density ratio of the compressor and intercooling system will give you. To do that we need to take our boost point and determine how hot the compressor is going to make the air at that pressure level.

Temp(in F) = (((Tin (in F) + 460) * (Pressure Ratio0.283)) - 460)

For 15psi of boost at sea level at an ambient temp of 85F (85F = 300k or 27c)

Temp = (85 + 460) * 2.020^.283 - 460 = 205F

This assumes 100% efficiency. The compressor map tell us how efficient the compressor is going to be at a given pressure ratio and flow level. Since most of the map is at least 70% (.70) efficient or better (adjust this number for what range you're shooting for). Our true output temp is:

TrueT = T ideal / efficiency

For our example, 205F - 85F or 120F:

Actual = 120F / 0.70 = 171F

171F is how much the compressor is going to heat the air above the inlet temp. True outlet temp 256F.

Now, what happens at the IC? Pressure Drop, the Temp Drop. Lets assume a 65% efficiency from a small sidemount which isn’t the greatest, but let's face it, it is reality in alot of peoples situations(cough stinkfist cough). To figure out the intercooler efficiency and the pressure drop you would use: (IC = intercooler)

T IC drop = (T IC in - T ambient) * IC efficiency

T IC drop = (256 - 85) * 0.65 = 111F

The IC will drop the Intake temp by 111F, turning the 256F air into 145F air and dropping the pressure 0.5psi to 14.5psig.

Now we must figure the Density Ratio.

Density ratio= ((Tempin + 460) / (Tempout + 460)) * (Pressureout / Pressurein)

Density ratio = ((85+460)/(145+460))*(14.5+14.7)/14.7 = 1.79

Now with the previously figured 212.5 CFM value, we multiply that by the density ratio to get 380.1 CFM or 27.7 Lbs/min.

Keep going, we are getting there!

Now, draw a vertical line on the X Axis at 27.7 Lbs/min cross the original horizontal line from the pressure ratio. With a 40 trim wheel at 6000 rpm with the intercooler setup we calculated with, we are at 73 percent efficiency. Things could change due to your setup, that particular VE and I/C setup will not be the same for all engines, thats why all these values are dynamic.

Will the compressor be put into Surge with this setup? Surge happens when the turbo is too big for the motor. In this example with a 2.02 pressure ratio, the surge line is around 15 pounds per minute or 205 CFM. Let's assume the turbo is can spool by 3500rpm. The air density is equal at this rpm, but the Volumetric Efficiency for this RPM will have to be recalulated. The lower the RPM the higher the VE will be, so we will just have an example of .95VE or 95% efficiency.

CFM = 136 / 3456 * 3500 * 0.95 = 130.8 CFM

That's in N/A terms. Now adding Boost, we get:

130.8 CFM * 1.79 = 234.1 CFM or 17lbs/min.

So now match 2.02 Pressure ratio on the Y Axis, and 17lbs/min on the X axis and you have the beginning of boost. with a pull right through the efficiency range of this turbo, or in other words, a great usable powerband, a fun car, and the perfect turbo. So if you happen to have a 2.2 or 2.3 Litre engine, a T04E 40 trim might be the best way to go for you. Please plug in your own numbers and find the proper turbo for your car, because lame powerbands and surging compressors make me mad, and jesus cry.

Author:  94GeoMetro [ Sun Apr 23, 2006 2:15 pm ]
Post subject: 

Where you a Math Professor in your previous life?? :lol: 8)

Author:  RobAP [ Sun Apr 23, 2006 5:18 pm ]
Post subject:  missing point or information not mentioned?

went through the process above to theoretically build a monster. (and it will be one for real....sometime)
and in the section of intercooler effiencey, it has the step of how much it cools the intake charge, but it also mentions a pressure drop of .5 psi. was this from the information of the intercooler specifications from the manufacturer or a mathematics step missing from the detail? I ask becuase the way the author just plugs it in there makes me wonder. he did nto specify where that figure came from (or at least i didnt catch it)

just clarifying so I'm not mising a step.
great post!

Author:  Grashopr [ Wed Apr 26, 2006 12:48 pm ]
Post subject: 

I am unsure where he got that information, but to my knowledge, IC's DO cause a loss of pressure. Why? I dunno. I'm sure it could be found on Google..

Author:  rarson [ Thu Apr 27, 2006 2:28 am ]
Post subject: 

Most of them are less than 1 psi, but certainly more than zero.

Just the process of cooling the air drops pressure too, btw.

Author:  STuPiDiCuS [ Thu Apr 27, 2006 6:04 pm ]
Post subject: 


KP for you Grashopr.

Author:  n1tr0 [ Thu Apr 27, 2006 8:30 pm ]
Post subject: 

rarson wrote:
Just the process of cooling the air drops pressure too, btw.

numbers can be misleading, intercoolers do cause a pressure drop because of the cooling effect. they also work as an expansion chamber possibly adding to your turbo's "lag" factor and they can be a further flow restriction ....
But, a cooler denser charge means you can stuff more air, more fuel, and more timing into your engine, eg. more power & less detonation. don't knock the intercooler.

Author:  sskmercerGTi [ Sun Apr 30, 2006 6:33 pm ]
Post subject: 

A link i have with other CFM ratings, mainly on subby engines but could help out. ... beb66d69ca Not sure if u need to login though.

Also have a great article in a recent FastFours type mag explaining the above in laymans terms. If people want i'll scan it and post up.

Author:  steve05 [ Sun Jun 04, 2006 12:12 am ]
Post subject: 

there needs to be an online calculator for this stuff, haha cos im no math man, thats for sure.

Author:  Devil-GTi [ Mon Jun 26, 2006 12:10 pm ]
Post subject: 

i've got a garret tb0253 from a volvo 460 1.7 turbo laying around
but cant find a compressormap
it's got an A/R of .48
would it be to small for a gti? or would it be the perfect size?

Author:  RobAP [ Mon Jun 26, 2006 2:30 pm ]
Post subject:  just some FUN WITH MATH!!

yes i was a nerd today and played with math and actually had fun thinking like a mad scientist!!

so i took the above math and put it into a spreadsheet. seems to work nicely. i discovered that according to the above mathematics, the garrett GT32 series...specifically the GT3267. its a fabulous turbo for a G10 at 9000 rpm and probably an 8000 rpm G13. it will give you huge power. (hows 44psi manifold pressure at 74%efficiencey sound?) and would be worthy for anyone wanting a big top end power car. but i would assume the lag might be pretty big.

so while looking around for other methods of turbocharging, i found other information on another site who were discussing 2 stage turbocharging (where one turbo feeds the other then feeds the engine.) and began playing around with doubling up the calculations and taking the outputs and using that as the intake air for a 2nd stage turbo....the benefits are, fantastically reduced lag and big boost on the top so its very driveable however...the gains over a single turbo's power are only seen when you need to operate a pressure ratio outside that of what any turbo is capable of handling on its own......but the benefit of running a 2-stage setup is reduced lag over the big single turbo and extrodanory top end power. (these supra guys building these things are talking of well-manered driveable 1500 hp engines with full boost from 4500 to 8k!...thats alot and not nearly as peaky as the usual single turbo setups.)

so anyway i began to think about this on a small tiny engine...specifically the G10. now assuming my frankenstein lab can build a bulletproof motor taking any boost. using a 2-stage setup will allow pressure ratios of ANY LIMIT and still have way over 70% efficiencey of turbos, where a comprable single unit system wouldnt even have a map and be stalling out or lagging horriby. so theoretically, a 4.0 bar boost system can be built, be driveable, cooporative, and be 75% efficient with an intake charge of 190 deg. F. (with 2x 70% efficient intercoolers)

just some crazy ramblings.

Author:  rvengineering [ Sat Jul 22, 2006 5:40 am ]
Post subject: 

Getting the best turbo for your needs can be a challenge.

Being a development Engineer in the past and having a diesel background you start looking for other turbo’s that are normally used for tuning cars.
Maybe this sounds stance to many but over the past 10 years we were looking to make the turbo’s as small as possible and still gaining horsepower.

The first turbo we used was a DT-4 from a 10 year old Volvo giving us power from 2000Rpm on wards and giving us 280 290Nm of torque @ 4100 RPM on a good day when the clutch is up to it and it a cold winter day. This DT-4 is a LP turbo standing for low pressure (5PSI) so getting 16PSI boost out of it the summer. Better not! Very good hear dryer.

Some years ago when I was still working on Swift Race cars they also run a racing team in the same workshop that used the 1.9 liter TDI engines from VW.
Slowly they went up to 200Hp with this original 130Hp engine and run a 24 hour race with it and ges what no problems.
I had a look at the data sheets and did some calculations and it turned out that this must be a very good turbo for a G13B engine.

On our latest turbo engines we use the GT1549 turbo having no turbo lag at all due to the fast spool up from these diesel turbo’s.
You would say this turbo does not produce top end power. Lucky for us it does that as well hitting the 170 Hp @ 10 PSI.
We like to puss it up to 17 PSI hoping to get 200 or more Hp out of the engine.
Still we have to overcome a clutch problem. @ 12 PSI the race clutch gives up.
Most manufactures say that they can handle 300Nm on the race clutch. I think most of them give way to 270Nm and not more. So we are still looking for a stronger one.

So if you want to have low end torque and still have a lots of power on the top end find your self a modern diesel turbo of a 1.9L to 2 liter engine that’s producing 100 up to 150Hp.
It’s like running a supercharger. Power when you want it. :ez_tongue:

Author:  ivan [ Tue Aug 15, 2006 1:10 am ]
Post subject: runing a 2l diesel turbine on my G13b is perfect then? feels like supercharger? :shock: good up till 8k?

Author:  rvengineering [ Tue Aug 22, 2006 3:31 pm ]
Post subject: 

Seems like all of the new build turbo projects here in Holland are going to use the GT1549 GARRETT diesel turbo. :shifty:

You won’t get 250Hp out of it I thing and your engine will run like a diesel with 188Nm 138LBF-FT @ 2500Rpm running with a stock 10:1CR engine and max torque of 215Nm 158LBF-FT and 170 Hp @ 0.7 bar 10.15 PSI boost.
This engine is not dyno tuned and under the 5000Rpm we kept the boost down and run it hard up to 6000Rpm to be safe.
This car runs on the standard ECU and a SMT6 for fuel correction. The cars we run on the Adaptronic ECU system have more low and hi end power. No need to dyno run these engines until we solved the clutch problem.


Author:  Colin The Barbarian [ Tue Aug 22, 2006 4:50 pm ]
Post subject: 

I think that You should use gt1749VB to get safely over 200HP (200HP can be done in tdi on it) and VNT will provide flatter torque curve and top end with same low end power. But there is only one small problem... Need good standalone (AEM will do the job) and if possible a vacuum pump to use with stock actuator (can do it on "classic" actuator without vacuum). Essential problem is solved by controlling boost according to throttle position - here comes aem.

As far as I remember gt1549 wasn't installed on more powerful diesel than 90 (or maybe 100 not in vag) HP.

Also low end power is not always needed - with good tranny You can drive in higher rpm range, and ball bearing provides almost no lag.

Author:  rvengineering [ Wed Aug 23, 2006 12:47 am ]
Post subject: 

The gt1749VB is almost the same animal as the GT1549 and are also used on the SAAB V6 and Renault and many other engines that I was not even aware of. To bad there is no flow chart of this turbo. The good thing I have my contact @ GARRETT and they put this turbo up to 275Hp on a VW POLO rally raid car only I thing they did some modifications on it.
The rezone that we did not use the gt1749VB is that we are not sure the variable flow pits system can handle petrol engines we already had to modify it on the diesel engines.

We also replaced the waste gate activator for a normal one.

The thing is a GARRETT BB turbo is expensive and lets be honest how many of us in general are really going to hit 250Hp or 300Hp.
It is a very cheap turbo secondhand be course there are so many of them.
Maybe I’m building turbo setups that will run like a family car and that can be boring to some of us be course you don’t have to rev up to get power. I can tell you one thing that the smile on the faces of my costumers tell a happy story and maybe it is be course I’m a rally driver and not a drag racer that I like bottom end power so math.

Author:  Colin The Barbarian [ Wed Aug 23, 2006 2:34 am ]
Post subject: 

Two yrs ago we have built with two friends a golf with 1.9 8v (plug&play with tdi turbo) engine (8:1cr) and vnt turbo with positive pressure actuator. Car was running on LPG, had no electronics and noone knows what afr was. Unfortunatelly due to low boost (7-8psi) actuator had to be set differrently and boost came at 3000rpm. If set differently for more boost it came earlier (different vnt angle). All it was because actuator had not much movement.
Car ran 40k km without problems.
Next thing is egt. Ve have done few 1.9 tdi engines to 230-250HP level and egt before turbo was over 800 deg C. Brand new 2.0 tdi engine had 830 deg, so I don't think that there should be problem with VNT - I know tdi's that for sure run 900deg egt.

Author:  rvengineering [ Wed Aug 23, 2006 5:16 am ]
Post subject: 

I agree. If it is working why not use it. As we speak Rob my college is building a gt1749VB on a R1 Yamaha engine and want to run it up to 300Hp.
In general the new generation diesel engine 1.9 a 2 litter turbo are very good on petrol engines up to 1600CC.
They are easy to get and good money spend in relation to aftermarket turbochargers.

There is a down site also. Most of theme use vacuum based activators.
Also most of theme have no liquid cooled bearing houses and specially the GARRET ones need up to 9 liter oil a minute for cooling.
Also a very good BOV system is needed. These turbo’s don’t like pressure waves coming back from a suddenly closed throttle valve.

For the rest. Just try it.:-P

Author:  poppycock [ Tue Feb 24, 2009 6:51 pm ]
Post subject:  Re: How To Choose The Right Turbo

Here's a "Turbo Calculator" that plots a line on the compressor map for you.
Don't ask me how accurate it is, I just found it.

Author:  greggholmes [ Sat Jan 01, 2011 8:23 am ]
Post subject:  Re: How To Choose The Right Turbo

poppycock wrote:
Here's a "Turbo Calculator" that plots a line on the compressor map for you.
Don't ask me how accurate it is, I just found it.

my imaginary engine,


Author:  parkaveultra93 [ Sun Jan 02, 2011 10:19 am ]
Post subject:  Re: How To Choose The Right Turbo

Let me know when they come up with a variable geometry turbo small enough to be efficient for this motor, till then, ill stick to nitrous, a much cheaper, reliable, predictable and efficient forced induction!

Author:  rvengineering [ Mon Jan 03, 2011 11:40 am ]
Post subject:  Re: How To Choose The Right Turbo

Well there are load of variable geometry turbo out there for small engines, main problem is controlling boost as they come from diesel engines and build to run way more boost than most petrol engines can handle.

We used a GT2256 VNT turbo from a Alfa on a G13B engine and that whent fine after some modifications on the activator.

Author:  rvengineering [ Sun Jan 16, 2011 6:49 am ]
Post subject:  Re: How To Choose The Right Turbo

I have decoded the Garrett servo control on the Adaptronic1280ECU that can be fitted on most VNT turbo.
This servo makes controlling boost a lot easier.

Author:  parkaveultra93 [ Mon Jan 17, 2011 9:18 am ]
Post subject:  Re: How To Choose The Right Turbo

Very nice! glad to see someones moving with the times!

Author:  iamaureahenry [ Fri Feb 01, 2013 7:05 am ]
Post subject:  Re: How To Choose The Right Turbo

You should use gt1749VB to get safely if you're driving over 200HP and VNT will provide flatter torque curve and top end with same low end power.

Page 1 of 2 All times are UTC - 5 hours
Powered by phpBB® Forum Software © phpBB Group