Damn Squirrel's.......

Natchamp · 05-16-2010, 02:34 PM

So, since I'm redesigning everything on my car I've been spending too much time on the turbo calculator. I'm curious what parameters you guys are playing with like the BSFC, VE's, Air Temps, Boost Thresholds, etc.

I remember going through all of this exercise years ago when seting up my twin GT28RS's but this particular squirrel calc was not around then. Something has changed cause they don't map out so well now

The following is derived from 3.2L, 2 turbos, 500hp, .47 BSFC, RPM's of 6000,5500,3500,2500, intake temps all at 90. Everything else is default.

Below are all using the exact same criteria except switching to single turbo:

Natchamp · 05-17-2010, 05:22 AM

110 views......0 comments. Was it something I said?

GJF · 05-17-2010, 06:39 AM

you have to divide the displacement since you are twins unless you are going to a single. people never seize to amaze me.

Natchamp · 05-17-2010, 06:45 AM

people never "seize" to amaze you? Hmm....people never cease to amaze me, but that's just me. The program automatically splits the flow when you tell it you are running 2 turbos.... ya big dummy.

GJF · 05-17-2010, 06:59 AM

then there is a problem with the program. Going to twin 81mm wheels is a huge jump

Natchamp · 05-17-2010, 07:07 AM

Who's going to twin 81mm wheels? After the map of my twin GT28RS's I have:

"Below are all using the exact same criteria except switching to single turbo:"

Natchamp · 05-17-2010, 07:40 AM

Just for the heck of it, I re-did the calc splitting the 3.2L in half (as well as the hp) and used the same parameters as listed above. I get the exact same results as I did when specifying two tubos. So the program appears to be accomodating the 2 turbos fine. But the results is very suspicious. My turbos are not "that" far off. Something smells fishy with this calculator.

kenikh · 05-17-2010, 08:59 AM

The Squirrel site actually has a setting that allows you to spec multiple turbos and will do the calculation for you:

Natchamp · 05-17-2010, 09:07 AM

Hi Ken,

That is what I did, as I said in my first post. However, I know my twin GT28RS setup isnt as far off as the output says. Put in the numbers and you will get the same results.

kenikh · 05-17-2010, 09:21 AM

Sound slike something may be amiss then...I know that my single turbo application for my turbo was spot on with my Excel plotting, as were the twin turbos I was considering at the time, the Garrett 2056s.

You are clearly not running out of boost or at risk of boost taper, just playing close to the surge line, which is easy enough to manage with an ECU managed boost controller, which would help to shift the line to the right. I'd think that as you've mentioned, this would run fine, it just wouldn't be in the sweet spot of the efficiency islands.

Is this running on your motor now? What boost level are you programming in to run at the various RPMs?

Natchamp · 05-17-2010, 09:41 AM

My motor isnt running now, hasn't for some time because I took the intake off to use for measurements for the ITB's. I'm now planning on getting it back on the road with my new ITB design and new intake system. I've been all over the board as whether to keep my GT28RS's, go to a single turbo, etc.

kenikh · 11:26 AM

I have been fiddling with this and it seems that the 2860RS is a bit too big for your application. You aren't generating enough air to run the turbo in the middle of the efficiency sweet spot, although it appears that it does work OK. They are spot on for a 7000 RPM redline 3.6, or so says the Squirrel.

The Garrett 2056, based on Squirrel's calcs, seems to be a better match, since although they blow less peak air, they still support what you need for 500HP:

As far as a final solution, you would likely be best served talking to an expert like Blouch or Forced Performance:
Forced Performance...
Blouch Turbo...

Take a look at Forced Performance and the work they are doing. They have some pretty amazing TD05 and Garrett variants:
Forced Performance Turbochargers: FP Turbochargers

Given what the tuners are getting out of customized Garretts, their attention to detail and the fact that Garrett's reliability has come into question of late, anyone needing a Garrett turbo should really look at the above:
Forced Performance Turbochargers: Garrett Turbochargers

They could do some pretty amazing work to your existing Disco Potatos (GT2860RS) to fine tune it to your application without needing to reinvent the wheel (no pun intended

):
Forced Performance Turbochargers: Garrett GT2860RS Turbo - "The Disco Potato"

BTW, this is FPs new "Beast from Hell" turbo for the Evo crowd and will blow 600HP all day on pump gas:
Forced Performance Turbochargers: FP BLACK Turbocharger for the Mitsubishi Evolution IX

IMHO, I'd seriously consider having your turbos modded by one of the above, using the 2056s, using a variant of the FP custom turbos as single/dual or the single turbo I went with.

Mine is based on the TD05HR-16G6, which you can map on Squirrel, too. It is a twin scroll turbo that would work great with RarlyL8s headers, since these turbos require equal length, segregated headers before the turbo. Here's how your motor plots to that turbo:

The variant part is that that Chad Block Racing Development has been working with Blouch on modding these turbos by swapping the turbine and compressor to push more air, yet keep the spool time very low.

Their latest BBK XL (I am using the BBK Full), uses the same turbine as the Full, with a more aggressive compressor wheel. If you can imagine the above compressor map (with your lines stationary), but with the map stretched to the right...that would be along the lines of the Full and XL. (Edit) The XL, since renamed the BBK-B support 58 lbs/min which translates to about 700 CFM, which is 3076R territory, keeping in mind that the left side of the compressor map looks much more like the TD05 than the Garrett (more responsive).

Here's a dyno of a 2.3 liter EVO VIII running this turbo, albeit on e85:

This is net WHEEL HP.

It's making power about 300-500 RPM quicker than the same 2.3L motor on a GT3076.

They only downside is that the turbo is watercooled and don't know anyone who has tried running these oil cooled only. I am going to run water cooling for the turbo on my car.

Here's a comparative boost plot of the Full versus the XL in a 3rd gear pull:

I believe that the boost levels won't be as high on a bigger motor.

Net-net, all of these turbos are streetable on a 2 liter 4 banger, so a 3.2L six would see MUCH better off idle boost response. Something to think about!

Anyway, that's a lot of rambling to read over. My 2 cents in a nutshell:
GT2860RS are too big in stock trim
Either send them to a tuner to quicken them up or
Get some smaller duals like the 2056s or
Get a custom, purpose specced single turbo from a tuner

The benefit of using a tuner is that you can leverage their expertise to customize the turbos to your needs, versus being stuck with the one-size-fits-all trim of the off the shelf turbos.

TurboKraft · 05-17-2010, 12:10 PM

3.2L and 1.4bar boost?
What's the objective -- street, track, or a combination of both?
And how much power would you like on that 91-octane crap they try to convince us is "premium" fuel?

Natchamp · 05-17-2010, 12:12 PM

wow, that's a lot of good info, much appreciated. I need to digest....

kenikh · 05-17-2010, 12:21 PM

Quote:

Originally Posted by TurboKraft

3.2L and 1.4bar boost?
What's the objective -- street, track, or a combination of both?
And how much power would you like on that 91-octane crap they try to convince us is "premium" fuel?

Totally missed the 1.4 BAR piece - that is a LOT of boost for a 3.2 without knock, sequential EFI w/ EGT WB O2 adjusted per cylinder fuel trim and a host of other goodies. With that kind of boost, I'd seriously consider EMS controlled staged water/meth injection.

What cams are you running? You can lower your needed boost by going with a more "top end" cam and building some revs. This combo is making enough boost low enough, you don't really need fat cam to propr the bottom end up.

kenikh · 05-17-2010, 12:57 PM

BTW, I neglected to mnetion Chris Carroll above, who is another fantastic resource and has a wealth of knowledge...Turbokraft knows their s--t!

kenikh · 05-17-2010, 01:03 PM

Quote:

Originally Posted by kenikh

Sound slike something may be amiss then...I know that my single turbo application for my turbo was spot on with my Excel plotting, as were the twin turbos I was considering at the time, the Garrett 2056s.

You are clearly not running out of boost or at risk of boost taper, just playing close to the surge line, which is easy enough to manage with an ECU managed boost controller, which would help to shift the line to the right. I'd think that as you've mentioned, this would run fine, it just wouldn't be in the sweet spot of the efficiency islands.

Is this running on your motor now? What boost level are you programming in to run at the various RPMs?

One more observation in looking at the comp maps again, the 2056 twins on my 2.3 look a lot like the GT28s on your motor...they'll work, but aren't ideal. The 2056s really do look like the ticket for you and they'd likely be bolt on as I think they use the same flange.

BTW, are you using the internal wastegate or something on the exhaust like a TiAl?

TurboKraft · 05-17-2010, 01:48 PM

3.2L short-stroke... ITBs... 993SS camshafts...
This engine is going to like RPMs!

Having done 3.0L through 3.8L twin turbos, I would not go smaller than the base "GT28R" aka GT2560R. With their tiny A/R turbine housings they'll spool very early.
However, for as many CFM as your engine should be able to consume in a wide RPM band, I would expect them to fall off hard before you run out of RPMs.

Probably the biggest mistake we see on EFI engines is undersizing the turbine wheel. Recent we sold off a whole shelf of "wrong sized turbos" we had sitting around -- ones customers brought in for their project that were too small, ones we tried based on "turbocharger calculators" or published flow maps, ones recommended by Garrett's own in-house engineers for a specific build we were doing. In each case, the turbine wheel was just too small. It came on quick like a positive displacement (screw) supercharger but fell off hard before the engine was done.

If you have a pair of GT2860RS turbochargers with 0.64A/R turbine housings, I'd bolt them on and keep building.
If they're too lazy, they have the same silhouette as the smaller turbochargers and will be simple to swap out.
If they're good but you want still more power, you can swap up to higher flowing units.

GJF · 05-17-2010, 02:36 PM

Marks turbos are the first GT28RS units. These were choosen because his motor is a short stroke 3.2. These turbos are idea for his motor but he has only twisted this engine to 6500+ RPM. These turbos as Chris has stated are great turbos for a 3.6 but with short stroke this motor needs a little bit more RPM's than 6500. If less RPM and a bit more grunt split the case's and add a larger stroke.

kenikh · 05-17-2010, 03:01 PM

Quote:

Originally Posted by GJF

Marks turbos are the first GT28RS units. These were choosen because his motor is a short stroke 3.2. These turbos are idea for his motor but he has only twisted this engine to 6500+ RPM. These turbos as Chris has stated are great turbos for a 3.6 but with short stroke this motor needs a little bit more RPM's than 6500. If less RPM and a bit more grunt split the case's and add a larger stroke.

A 3.2SS limited to 6500 RPM??? C'mon, Mark! How about 7500? Hell, 8K if you built it right!