4AGTE turbo flow calculations

[TooF]

Im in the process of selecting a turbo for my 4agze soon to be 4agte.

Looking at a GT2554R. Im not after a dyno queen i want something that is super responsive with a shade more huff then the SC. i did have the SC really overdriven and was producing 17psi of boost. the thing went like the clapper but suffered ongoing reliability problems with belts and tensioners. hence the turbo path. Ultimately id like it to respond with the similar sort of quick power of the SC but am prepared to suffer a little lag in the process.

Am i being realistic on these plots.

Blue being stock power delivery. (obviously not relevant on the turbo map just the power)

red being a full boost map, yellow medium, green low power everyday easy to drive.

top plot being these power curves against the compressor map. these have been calculated with an AFR of 12, BSFC 0.55 and VE of 0.98.. are these realistic numbers?

and inlet temperature of 35degC (ive got a large a/w cooler which in similar applications is yielding near ambient charge temperatures)

lower plot being the power curves in hp (LHS axis) and boost (gauge) in psi.




from my understanding this should produce a very sharp response engine at these sort of targets.. am i correct or missing the point.

any thoughts or suggestions ?

[TooF]

does anyone know anything about this?

[YLD-16L]

Toof, I have done some similar calcs and estimates in Excel and using the compressor maps on the Garrett site.

Not knowing if my estimates are correct either I can't really offer any advice other than I used my calcs on a compressor map for a turbo as close to the specs of my current Garrett as possible and came up with pretty close to the real world values I have experienced. So from there I had some confidence that my calcs for the turbo I was interested in were fairly on the money.

I used a BSFC of 0.52 (I created the spreadsheet so long ago I can't remember how I came up with that value though).

A:F I use values of 11.8 (I aim for 11.8 at WOT at the track for "safety").

I use a similar IAT to yours.


I can chuck some values into my spreadsheet and check the compressor map for the turbo you are looking at and see what I come up with??

PM or post the peak target KW value you used and the rpm you want that achieved at, I'll pug in the other values you already posted.

I think I had to use a lower VE value than 0.98 to get my "realistic" values on the known turbo if you get what I mean.

[jfallen]

I've written a spreadsheet that does these sort of calculations for you. pm me your email address and I'll send you a copy with a readme file.

best cheap option quickly, is the Garrett GT2560R, 0.64A/R exhaust. Also know as a T28bb turbo from an S15 silvia, running at 16PSi, with peak power around 8,000RPM. You should see around 160-170RWkW. Good examples of this turbo can be found second had for under $500.

Cheers
Jordan

[TooF]

hi mate,

pm me your email and ill send you the excel sheet ive used as well.

ive used the functions and suggestions from the garrett website too.

here are the power curves ive used to generate those plots


brining the BSFC to 0.52 and AFR to 11.8 despite a drop on VE to 0.96 only makes this turbo look more favorable for my application

edit:

yeah i looked at the 2650 and one big plus is its easy to get second hand and hence cheap. however both the 2554 and the 2560 run at around 1100 brand new. a second hand 2560 will not be *far* from having to get a rebuild at some point. garrett charge around 500 for this rebuild assuming its just a bearing and seal cartridge plus a balance. so a brand spanker starts looking good.

you have pm re my email im keen to see what my numbers make sense. from what i can understand i think ive done the right thing but just want to know im interpreting it right. in the sense that i think the 2554 will make a very low lag setup and provide me with adequate power.

[YLD-16L]

For on the road results I think the GT2554R you are looking at will be similar but perhaps with the ball bearing core more responsive than my current garrett and I see ~165 - 170rwkw and it is extremely responsive (16psi by ~3000rpm), so much so that ditching the SC was never given a second thought.

I'm looking at the last table you posted and trying to work out how you came about those falling boost curves??

[YLD-16L]

Toof, I just plugged your values into my spreadsheet calculating for 190kw at the flywheel and to achieve that I am seeing Air flow rate actual (lb/min) of ~28.0 and a Pressure ratio of 2.0 which sees you right out in the choke line of the compressor map.

If peak torque occurs at around 4000rpm on you 4A then the other point on the map I got was at about 15.1 (lb/min) and a pressure ratio of 2.0 so it'll be nice and responsive as you'd expect.

So that is saying to me that this setup will be very much like mine, very responsive but will leave no headroom for more power. You'll hit a wall at about 160 - 170kw at the wheels.

[TooF]

sorry i did all the plots kind of for myself and didnt label axis like i should.

all the power numbers are in hp not kw. i derrived the power numbers before the boost/flow curves. so looking at the first image i posted the bottom blot on there has hp/rpm. ive tried to generate a smooth realativly linear power curve that seems realistic compared to what other 4as are producing on the dyno. and also producing a power curve that i think will make the car good to drive. i dont want a massive rushing power curve becasue i think it will ruin the dynamics of the car.

so once the power curves were developed i have then plugged them back though the flow calculation formulas and produced the flow/PR requirements. given the widening and flattening of the top end of the power curve the power across the top 1500-1800 revs remains relativly constant whilst the revs increase and hence require less boost.

edit: oh and in terms of hitting a wall at 170kw i think youre dead right that would be well and truly the top limit of what it'd make and that doesnt stress me i think any more and ill be too busy keeping the car under control to be able to enjoy the car the way i actually want to.

[YLD-16L]

I see now. I just plugged in 190 flywheel kw as I think anything less than 160kw - 170kw at the wheels in the AW11 and you might be disappointed with the result?? But you are the only one who knows what you'll be happy with .

How will you map the boost to follow the desired falling curve though?

I think the manner in which mine delivers power would be fantastic in an AW11 as you'd have much more traction and it would be very driveable.

My only hesitation in a new larger turbo has been the fact I will lose the fantastic response while gaining a bit more top end. Considering it's the low-mid range that makes it enjoyable and pulls nicely out of the corners it's a tough decision

[TooF]

i always thought the car was pretty much perfect with the silly SC setup, whilst i regret never getting it on the dyno to confirm numbers just hazzarding a guess id put it at around the 200hp at the fly mark.

in all truth i think as a total package the NA mr2 is a better car. not outright as quick but the dynamics are almost perfect. the SC always feels a bit heavy handed.

would you have copies of a dyno sheet for your turbo setup ?

[YLD-16L]

4 of the graphs are here:

http://www.toymods.net/forums/showthread.php?t=23329

Since those runs I have purchased an air fuel meter and have tidied up the air:fuel at WOT and a few other points. The last dyno day a couple of months ago showed a dead even 11.8:1 at WOT which I was happy with for what I use the car for. I have also fattened up the mid range since those runs with more boost (power peaks earlier and holds to redline) , peak power remains the same though.

[TooF]

that power curve looks good although id love it to be on just a shade quicker with potentially a little less top becaue im not after massive power just fun reliability whilst being brisk.

and i was planning to manage boost control though my ecu (adaptronic) it has wastegate control and i was planning on using a suitable solenoid valve to control the wastegate.

[YLD-16L]

that power curve looks good although id love it to be on just a shade quicker with potentially a little less top becaue im not after massive power just fun reliability whilst being brisk.

I'm pretty certain that with the improved technology of the wheels on the GT series and the ball bearing core you would be on boost a little sooner than mine and more efficiently too

I'll have to log mine one day but there is no discernable lag in first or second gear from any rpm while driving (basically press the pedal at anything off idle and it's boosting rapidly to the preset boost level), 3rd is similar and as the dyno operator commented at the last dyno day in 4th gear I had 10psi as soon as the throttle went WOT and it rapidly went to 16.5 from there. That was starting at less than 2000rpm on the dyno.

Another difference since those runs, the car now has an EBC.

I run a MAC solenoid on mine but for the price the BA-BF XR6 Turbo solenoids are a neat little unit and cheap too.

[GUN METAL]

Excellent work toof and to everyone who has offered info.

Toofs line of thinking is very similar to mine in respec to the requirments so looking forward to the info.

Subscribes and rates thread

[TooF]

out of curiosity scott could you post up your power curve as if i can emulate that id be very happy as your car gives a beautiful power delivery for an mr2.