Garrett PDF catalogue

[auditude]

Thats the trick - you have to play with the desired boost levels to generate the combination of PR and airflow that fits the map !

Too much boost at low revs can cause the compressor to surge... Just play with the data in the calculator and do some plots. You'll see how.

Hmm. Okay. Well if I change the displacement to 2.5l, the ambient temp to 21c, and the boost to 26psi, and don't change any of the Step 2 boost values, I guess I get something like this, which I don't expect to be correct but it's a start at trying to do this.

I guess I need to post the spreadsheet separately, so you can see what I'm doing.

[auditude]

Man, this is a real challenge. I've had to figure out how to make a .zip file, and where to find the files I save in the default locations on the work pc. Hopefully this works.

So, those dots are plotted within the map, so what I would want to do is increase the boost at each rpm point and then try to get it as high as desired while keeping the dots on the map, right?

[auditude]

How about 2.22l with the same stuff as before, maxing out the boost. The spreadsheet won't let me increase the boost past 2 bar, so that's what it is at from 4500-7500. What's the next nugget of knowledge that I need that I don't have?

What's the relationship between cell E3 and the values in cells D1124? Is E3 supposed to be max boost, or is it ambient air pressure?

EDIT: Lol! I think it's supposed to be ambient. It's set at 14.7 in the original sheet. Back to the drawing board. Ignore this graph!

[audiman]

You cant run 2 bar at the redline, well you can, but it wont very long, go with a safe figure of about 1.5 ish

[S2central.net]

How about 2.22l with the same stuff as before, maxing out the boost. The spreadsheet won't let me increase the boost past 2 bar, so that's what it is at from 4500-7500. What's the next nugget of knowledge that I need that I don't have?

What's the relationship between cell E3 and the values in cells D1124? Is E3 supposed to be max boost, or is it ambient air pressure?

EDIT: Lol! I think it's supposed to be ambient. It's set at 14.7 in the original sheet. Back to the drawing board. Ignore this graph!

Take a look at this one I did last night. Its for a 2.5L stroker (sorry I don't know exact stroke) with a GT30R and GT35R in mind. These run quite close to the surge line... Note the increased VE required to make it work on GT35R.

Paul

[auditude]

Take a look at this one I did last night. Its for a 2.5L stroker (sorry I don't know exact stroke) with a GT30R and GT35R in mind. These run quite close to the surge line... Note the increased VE required to make it work on GT35R.

Paul

That's very cool, thanks Paul! The Eurovan, and most or all of the 5 cylinders strokers are 95.5mm stroke, by the way.

When I first looked at the shape of those graphs, I thought the GT30R looked like it gave much more usable, low end torque and power than the GT35R. The GT30R graph looks like it really has some grunt. I was thinking that since the power tapers off at the top of the RPM range that it was more suited to the 20vt. In other words, the "lump" of torque is centered in the usable range, compared to the GT35R, where it looks like it's just getting started toward the top end.

However, when I look at the scaling and pick a couple lower rpm points, such as 3000 rpm, and compare torque and power, I thought it was somewhat comparable. It's hard for me to tell what it would feel like from the numbers. There is less power below 4500 rpm with the smaller turbo.

Anyway, I made some new graphs comparing the two turbos on the same graph. I was curious what the flat "cap" is on the torque, is that limited due to the flow or something? It just flattens out at 437ftlbs and reminds of how how torque curves look for 1.8t's and other turbo cars. I guess that's just how they look? I'm just curious what causes that.

I also made a table showing the difference of the HP and torque by rpm between the two turbos (two boost levels on the GT35R).

[auditude]

A table showing the difference of the HP and torque by rpm between the two turbos (two boost levels on the GT35R):

[S2central.net]

Those differences on the two GT35R configs just prove in glorious colour that torque is proportional to boost. I flatlined the boost profile to give a flat torque profile. With consistent VE figures, this provides a nice even power curve.

You could provide peakier torque curves, indicating greater acceleration ability, in some parts of the mid range - just by increasing boost within limits of the compressor map. Side effect of this is a less than straight looking power curve...


Paul

[Mihnea]

Paul, IMHO you should first get each turbo, tune each type of engine with it, and then post such spreadsheets. Remember one thing: in theory, there is no difference between theory and practice. In practice, there is.

Mihnea

[S2central.net]

Mihnea - I ain't got that sort of money... or time ...

The spreadsheets are seriously useful preparation for any turbo choice in my opinion. For example they show that a GT35R is better suited to a 2.5l stroker *if* VE is increased, by head work primarily.

Sure there are many other assumptions which can skew or screw the numbers in real life. But the spreadsheets can quickly determine any detrimental effects from invalid assumptions. Would you rather spend a night playing with numbers in a spreasheet - or spend thousands of Euros on the wrong turbo ?

These spreadsheets will get more accurate when we characterise things like VE, CE, IE for various setups. A great deal of instrumentation equipment would be needed to do that accurately for each installation. It would be a full time job for somebody logging all the running data and optimising the spreadsheets in light of things like actual IAT.

The spreadsheets aren't supposed to replace tried and proven results in practice. They are to help us select the right turbo, intercooler and fuel injectors for the application in mind.

I'm encouraged by the sheet's general accuracy for the K24 and RS2 setups. They seem to smell right and show how much of a challenge it is to get 300bhp from a K24.

I know you didn't get close to the theoretical performance limit with a GT30 variant on the last session. But are the reasons for that fully understood

1. Was it a GT30 with the same 76mm compressor as expected ?
2. What exhaust manifold ?
3. What camshafts ?
4. What compression ratio ?
5. What fuel, injectors & rail pressure ?
6. What intercooler was used ?
7. What IAT levels were observed ?
8. Heck what hot-side and turbine wheel was used ?

I'm not bitching back at you Mihnea, its just this is a very complex subject and whilst the calculations will never be perfect - they are only as good as the inputs and assumptions used.

A major missing link in my knowledge is being able to select the right turbine for the right application. Its all very well generalising by saying 'bigger hot side for more top end performance' or 'smaller hot side for mid range performance and reduced lag' but those are very simplistic guidelines.

The spreadsheets, based on compressor maps, make an overriding assumption that the hotside never runs out of puff !

Can you (or anyone) help us characterise that better with an explanation of how we meaningfully read the turbine maps - such as those provided by Garrett for the GT30R and GT35R ?

Paul

[Mihnea]

Paul, the calculations being very complex is the very reason I only do a summary job where I plot PR, RPM and assumed VE, to check if the turbo is good or not.

I have never bothered reading turbine maps, not enough hours in a day to start this, way too much work with real maps stuff on ME7 ECUs and latest MED9 (2.0T FSI).... so I can't really help here...

Mihnea

[S2central.net]

Hah - let us know if you ever figure out how to graft an FSi system onto the S2 - there would be an interesting project. Do you think the gains would be significant for the money and complexity involved ?

Going off thread here, but hey... not fir the 1st time.

If I ever figure out the damn turbine maps, I'll get some more sleep !

PS - I sent you a PM about much needed 250kPa MAP sensors !!!!

Paul

[Mihnea]

Hah - let us know if you ever figure out how to graft an FSi system onto the S2 - there would be an interesting project. Do you think the gains would be significant for the money and complexity involved ?

Well, no, I don't think it would be worth it... the FSi needs direct cylinder injection, which is too complicated to implement into an existing engine. The 2.0 16v Fsi Turbo engine was developed on purpose for the FSi system. It would be fun though to install an ME7 system from a 2.7T with drive by wire throttle and the advanced torque-oriented features, as well as the very advanced knock control/EGT management, but that won't be easy... plus lots of sensors would need to be changed... but hey, if anyone is interested in doing that, I could try to spend some time on this...

Going off thread here, but hey... not fir the 1st time.

lol !!!

PS - I sent you a PM about much needed 250kPa MAP sensors !!!!

I know dude, got it, will get back to you after some more sleep, damn, trying to get rid of 9 hours jet lag is tough!!!!

Mihnea

Paul[/QUOTE]

[audiman]

I'll do that! You supply me with parts and what needs to be changed. Lets see if we can do over 500bhp!!

[Mihnea]

Lol, ME7 is not gonna help us get over 500 HP with the current setup, at least IMHO not...