ZX-Tex

iTrader: (15)

So I have ordered a GT2871 turbo that is basically the following:
TurboByGarrett.com - Catalog

It can be ordered as an option with a ported shroud a.k.a anti-surge housing. Here is how Garrett describes it:

A Ported Shroud compressor (see Fig. 2) is a feature that is incorporated into the compressor housing. It functions to move the surge line further to the left (see Fig. 3) by allowing some airflow to exit the wheel through the port to keep surge from occurring. This provides additional useable range and allows a larger compressor to be used for higher flow requirements without risking running the compressor into a dangerous surge condition. The presence of the ported shroud usually has a minor negative impact on compressor efficiency.




So when I first looked at this option I thought it would be a good idea. Somehow I got the impression that it would 'widen' the compressor map to the left, and help with low end spool. However the more I google and read, the more I get conflicting opinions. Everything from 'ported compressor housings are an awesome no-brainer', to 'they are a useless noisemaker that will hurt your efficiency'.

I am looking for performance gains here, not tractor-trailer intake spoolup noises. So I am looking for some knowledgeable, experienced input on this from y'all while I continue my research. Since I just ordered this thing I need to decide quick if it is not already too late. I wrote ATPturbo (they refer you to e-mail tech support when you call them) but no reply yet.

Here is an interesting discussion
http://www.eng-tips.com/viewthread.c...=225525&page=5

Corky Bell

Gads.

Dangerous compressor surge? I've never seen a turbo hurt from comp surge. Race cars do not use anti-surge valves.

Moving the surge line to the left is not a bad thing, but getting into surge on the way up is a scarce event.

Losing comp eff ain't too cool, but a couple % is not much of a deal as long as the IC works well.

Noise is obnoxious, in my view.

All things summed, it sounds like a good argument for a anti-surge valve and an old style compressor inlet. The old stuff also looks a bit more streamlined.

ZX-Tex

iTrader: (15)

The more I read Corky, the more I think I agree with you. I plan to run a BOV, and am definitely not going to get rid of it thinking I could do without one with the anti-surge housing. I'm not even afraid of anti-surge in the upper left hand quadrant of the compressor map really. From what little I have read, it seems that it is only a problem with extremely oversized turbos, or some diesel engines.

I am just more interested in expanding the range of the turbo, which the Garrett map seems to indicate. The red 'efficiency isobars' are in some areas shifted over to the left, which seems to my turbo novice opinion like it would help with low-end spool a bit. And some seem to indicate that it does.

But then others indicate anti-surge actually hurts low-end spool; this actually makes more sense to me intuitively since energy is being recirculated within the compressor.

You know, I can definitely see how it increases inlet noise. The housing seems like it would act like the old hand-cranked air raid sirens used in WW-II, that is, a spinning impeller setting up pressure waves across a stationary set of holes. No thanks.



Any other turbo masters want to weigh in on this? I really need to make a quick call here.

patsmx5

iTrader: (16)

My $0.02. If you don't like noise, and this compressor makes it noisy, get the regular. Or just do the math and find out where you'll be in that compressors map. If you're to the right of the surge line, then you're to the right of the surge line. Moving it further left will not do anything. There's a website called squirel something that allows you to plot to plot points in the compressor map based on your engine and HP specs.

ZX-Tex

iTrader: (15)

Thanks. Yeah I found that squirrel web site and tried it out. I also put together a Mathcad worksheet based on the Garrett Turbo 103 equations and checked it that way too. Depending on the BSFC assumption, the compressor plots look good.

So here is the deal... I just got off the phone with Corky. What a great guy, he spent a good bit of time of his busy morning to talk this over with me. It was an interesting discussion, and we could see pros and cons either way. Long story short, he is interested in this, and we are going to have some fun and play around with the anti-surge housing. We are going to do a with-and-without back-to-back dyno run to see if it makes any difference, and if it does, what exactly the difference is. I'll also shoot video so the difference in noise can be compared.

Now do not expect results soon. I still have to reassemble the motor, install everything back in the car, break in the engine, and retune everything. I am busier than hell at work right now (not much free time) so it will realistically be about a month before I have results to share. Stay tuned

Ben

iTrader: (33)

That's great. I'm probably as excited as you are. I'm actually even going to subscribe to the thread.

ZX-Tex

iTrader: (15)

Yeah I am excited. I live for this stuff. That is why I work where I do (engineering R&D) because I get paid to try out wild-*** ideas, though it is for NASA, and they do not use many turbochargers in space They do use them getting there though, REALLY BIG ones...

BTW here is the location of the turbo mapping worksheet
Squirrel Performance - Turbo Calculator Version 4

Here is the map I ran with the variant of the GT2871 I am getting. This is at 350 crank HP (roughly 300 rwhp). It is near the left side of the map as it builds, but stays right of the surge line.


Here is the link to the worksheet config I used if you want to play
http://www.squirrelpf.com/turbocalc/...=1&map_sel0=34

cjernigan

iTrader: (13)

Well aren't you moving up in the world. You're doing pretty well at convincing the wife of all your needs apparently. I'm excited to see it all together.

ZX-Tex

iTrader: (15)

^^ LOL yeah right now she knows what I am up to, but not the dollar amount. Whenever she asks, I just say 'you don't want to know'

y8s

iTrader: (8)

now that's good news. i'm curious to see what it does too.

ZX-Tex

iTrader: (15)

So I have a bit more information. Last year while racing I met an Engineer who used to work at Garrett and thus knows a lot about turbo design. He was kind enough to give me some input on this. He confirmed some of what I had read. Long story short, a ported shroud will provide more range on the compressor map, with a small penalty (1-2%) on efficiency, and an increase in noise.

Clearly there are a large number of variables at work here. So I still think the most practical way to decide which way is best is to just test it both ways and see what happens. Stay tuned...

JKav

iTrader: (1)

You've never seen a failed thrust bearing?

Race cars that don't use anti-surge valves get their turbos rebuilt every other race. Or so.

ZX, the 2871 sans ported shroud will surge on a 2.0-liter WRX, so I fully expect it to surge on your 1.8.

ZX-Tex

iTrader: (15)

^^Are you talking about surge due to not using a BOV on a WRX? If so, I have always planned to use a BOV regardless of whether I use the anti-surge housing or not. I am using one now.

wes65

iTrader: (7)

I am very interested in this thread. I in the process of choosing a turbo as well. I would like to use the gt2871 but i dont want to have a surging problem.

Savington

iTrader: (31)

Lots of boost + not a lot of power = surge, so it seems like it would be an issue on large motors running small turbos, right?

ZX-Tex

iTrader: (15)

Right. Lots of boost and not a lot of power (mass air flow), means the compressor is driven up into the upper left hand side of the map. As I understand this is the scenario the anti-surge housing protects against.

FYI here is the response I got from ATP Turbo tech support when I asked them about it. ATP Turbo is an official Garrett supplier:

"On the smaller engines, it's beneficial to broaden the map as much as possible to stay away from the surgeline. This is what the ported shroud will do for you on the GT2871R. We've tested on many engines and although there isn't a measureable difference in spoolup time or loss in peak flow, it does combat the surge issue. We do recommend it in your case."

I just wanted to be sure that is what JKav meant above, and not that I was getting it thinking that I did not need a BOV anymore. If he means the WRX surges even with a BOV or a recirc, then this would be evidence that anti-surge would be a good idea.

y8s

iTrader: (8)

OH WIKIPEDIA!
Compressor stall - Wikipedia, the free encyclopedia

i love you so!


basically the turbo is pushing more air than the engine needs and once it's packed, the air has to go somewhere... so it goes back through the compressor. ow.

mazda/nissan

iTrader: (2)

AHA! Now I understand!

ZX-Tex

iTrader: (15)

Yeah I think that is the right idea. That is for axial flow compressors but the idea is the same.

So I think what can happen is that when at WOT at low RPM, the turbine spins up the compressor, and since the engine is 'small' for the turbo, the engine cannot ingest enough of the air. Thus the pressure spikes in the intake manifold and backs up through the intake into the compressor, the compressor airflow drops at a high pressure ratio, and viola, you are in surge.

Stephanie Turner

Here is what one looks like.
Stephanie