Turbos Demystified: How To Read Compressor Maps

[chickenwafer]

I would love to see you do this with the stock k04 turbo on the speeds.. Lets see how inefficient 18psi on the stocker REALLY is..

I searched for the K04 compressor maps and there are a lot of variations out there, and most are measured in metrics I am not familiar with for the flow rate of the compressor.

I was able to find one, hopefully this is the same irritation the MS3/MS6 utilize:



OK, we already know from my cheat sheet at 18psi you have a PR of 2.45 in our altitude.

So, airflow demands on an MS3/6 at 2.45PR:

(2.3L)(6000rpm)(95%VE)(2.45PR)/5660=567cfm

(567cfm)(0.069)=39lbs/min

But wait, out compressor map is in grams per second. OK, yet another conversion:

(39lbs/min)(7.56)=296 grams/sec

Plotted!



Assuming this is the correct map (there are a lot of variations of K04"s out there) we can clearly see 18psi on an MS3/6 isn't even in NEAR the map on a this particular K04! An MS6 would be a little closer due to drivetrain looses but still wayyyyy off the map. The airflow STOPS at 240grams/sec and we need 296! Not even close. Compressor is a restriction at this point and pushing out super-heated air.

[Eclipse]

great info....definitely what i expected, waaay off the map.

[chickenwafer]

I was doing some more plotting on various maps today. One turbo I find interesting and a good STREET turbo for a rotary is the T3/T4R TO4E 60trim from Garrett. This is regarded as the most responsive T4 footplant turbo out there. Combine it with a divided 0.84A/R exhaust/turbine housing, and you have a turbo that is LIT!

Here is the compressor map at various boost levels:



As we can see, this turbo is a good option up to 16psi, but we're near the choke line. Anything above 16psi and you're running on the edge of the choke line. Spraying some water/meth into the compressor inlet could let you run 18-20psi, but you're on the bleeding edge.

Let's say you have the T04E 60trim on your FC and want to run 10psi. As we saw on the previous map, this is roughly 41lbs/min of air which is 316 rotary rwhp. Wonder how it look as the turbo builds boost through the RPM range? We can plot that too!



I simply guessed what boost this turbo would be pushing for a given RPM. I was pretty conservative because I honestly think with a divided turbine housing this turbo would be at 10psi by 3000rpm (which, is you plot that, would be on the surge line [too far left] of the map). Also, your Volumetric Efficiency (VE), generally gets worse until you hit your torque peak and beyond. To keep it simple, I estimated 70%VE for 4k rpm, 75%VE for 5krpm, 85% 6k rpm, and 90% from 6k until the redline of 8k. These numbers play a big role in overall airflow.

Using our estimations, we can clearly see this turbo loves the 10psi range on our sample 13B. It cruises right through the main efficiency island and stays within the 2nd or 3rd island until redline. This would be a fantastic street turbo with room to grow to up to 51lbs/min of airflow at 16psi, right around 400rwhp on a rotary.

[tbot]

Twin scroll t3 hxa3071 ftmfw! All I'm saying...cheaper than the garret and just as good a turbo. From what I've read this would be just about the perfect turbo for our speeds.

make it top mount and then it's just sexy!

[erod550]

Twin scroll t3 hxa3071 ftmfw! All I'm saying...cheaper than the garret and just as good a turbo. From what I've read this would be just about the perfect turbo for our speeds.

make it top mount and then it's just sexy!

http://www.steedspeed.com/mazdaspeed_36_top_mount_turbo_manifold

I've been looking at this mani. Would make things a whole lot easier to work on. I'd be forced to go FMIC at that point though. So it will end up getting even more expensive.

The lack of people running Holsets on our platform makes me hesitant. I did some Googling and Honda guys are talking about spooling at 5000-6000rpm. Granted they have smaller motors to spool them, but still, I don't want a turbo that doesn't hit full boost until 4500+.

[chickenwafer]

Twin scroll t3 hxa3071 ftmfw! All I'm saying...cheaper than the garret and just as good a turbo. From what I've read this would be just about the perfect turbo for our speeds.

make it top mount and then it's just sexy!

Post up a compressor map and we'll find out!

[erod550]

I think this is the one he's talking about...

[chickenwafer]

If that is indeed the compressor map, that turbo looks like overkill. I have doubts that turbo has a T3 housing as well...it supports up to 66lbs/min of corrected airflow, 77lbs/min of uncorrected airflow.

That is obviously a lot of air. The turbo looks happiest at a PR of 1.8 (10psi in our altitude) flowing 25lbs/min of corrected airflow all the way to a PR of 3.1 pushing 45lbs/min of air. That is a pretty broad range of airflow, but probably still overkill for a MS3/6 unless you're shooting for over 450whp+.

For response and flexibility I still vote for a GT2871R or for the Mazdaspeed cars.

[erod550]

Yea with my power goals I think that's the one I have settled on (the 2871R). It's just so damn expensive. $1650-1800 just for the turbo. Then, if you want to take full advantage of the better flow characteristics you need a new manifold, downpipe, and intake. Then if I go with a top mount manifold that means an FMIC which will put the total close to 4 grand just for an upgraded turbo and supporting mods. Then you just hope you don't make TOO much power and pop the motor. Just seems like too much money to spend on 350whp on a front wheel drive platform.

[chickenwafer]

Not sure where you're pricing them for $1600+, ATP sells it for $1291 all day long: http://www.atpturbo.com/mm5/merchant.mvc?Screen=PROD&Store_Code=tp&Product_Code=GRT-TBO-022&Category_Code=GRT

If you shop around, you can get better deals. If you're looking at a "bolt-on" configured turbo, I wouldn't, especially if you're getting a new manifold anyways. Pretty pointless..what else would make it bolt-on? Oil/coolant fittings aren't proprietary and intakes are easy enough, downpipes are also easy.

If you're getting a manifold, you can get this version of the GT2871R that is setup for an external wastegate, which will give you superior boost control: http://www.atpturbo.com/mm5/merchant.mvc?Screen=PROD&Store_Code=tp&Product_Code=GRT-TBO-003&Category_Code=GRT

$4k including supporting mods is pretty cheap IMO. I spend well more than that on my RX-8, but that was also converting an NA platform for forced induction, which is always more expensive and headache inducing.

For the FD, I'm looking at spending a lot more than that as well including all supporting mods. Hell, the clutch alone is $1650, and my fuel injectors will be $850 for all of them, plus rails, hoses, and fittings. You gotta pay to play!

[erod550]

What about something like this? http://www.atpturbo.com/mm5/merchant.mvc?Screen=PROD&Store_Code=tp&Product_Code=GRT-TBO-056&Category_Code=TBN

A lot of SRT-4 guys run 50-trims. If I'm using an aftermarket manifold anyway, I don't have to stick with a bolt-on configured turbo, like you said.

[rmeeker13]

Yeah im lost, thats why im going to have you tune my car! That is if i ever get it off jackstands!

Good info tho!

[chickenwafer]

What about something like this? http://www.atpturbo.com/mm5/merchant.mvc?Screen=PROD&Store_Code=tp&Product_Code=GRT-TBO-056&Category_Code=TBN

A lot of SRT-4 guys run 50-trims. If I'm using an aftermarket manifold anyway, I don't have to stick with a bolt-on configured turbo, like you said.

Depends on what power you're looking for! Find a compressor map and post it up!

Those 50trims are VERY responsive turbos and it should be a good choice. Reliable, tired and true setup.

[tbot]

I don't think that was the right holset compressor map. Like Dave said, thats way overkill. The guys running them on msf say they are very very comparable to a gt30, just no where near as expensive.

And don't go w/ the stock mani bolt configuration for our cars, they are the main restrictions on our cars, which is why the T3 or TS T3 is SOOO much better.