afawaterpolo

iTrader: (7)

Been searching like crazy on the site but can't seem to find what I'm looking for. Any help with finding said request if you happen to remember seeing it somewhere (RX7Club, tech article somewhere, or personal experience) would be greatly appreciated.

What I am looking for is the following:

What is more important in a medium/large-ish turbo set up using a Borg Warner S480 billet wheel? The turbine A/R (ex: 1.10, 1.25 all in a T4 housing) or the turbine exhaust wheel diameter (ex: 83mm, 87mm, 92mm all in the T4 housing) or a combination of both?

My current setup is listed below and I made 703RWHP at 23psi with a Borg Warner S475 T4 1.10 A/R housing with an 83mm exhaust wheel on Steve Kans mustang dyno. I prefer to use these numbers because it seems more real world instead of using the dyno jet numbers which gave me 810RWHP at 23psi.

I want to go with the Borg Warner S480 billet wheel and keep the T4 housing setup. I can change the T4 housing A/R from the current 1.10 to the 1.25 and I can also play with the turbine wheel, stay with the current 83mm or move up to the 87mm or 92mm in the T4 housing. So again the question is which is more important for spool and of course power increase? The A/R ratio or the turbine wheel?

I have been told and verified on my data logs that the S475 was experiencing some turbine wheel stall down low around 5500RPM's due to the exhaust back pressure readings because of the small 83mm exhaust wheel. I was told two ways to correct this were with either a larger A/R or a bigger turbine wheel. This would help minimize the turbine wheel stall i am experiencing and also help spool the turbo a lot quicker.

Engine:
- 13B-REW full bridge with ported intake ports to match Cosmo Upper intake manifold, exhaust ports on housings opened up to the max
- race bearings, oil mods on e-shaft and housings/irons
- completely balanced rotating assembly
- 2mm black Rotary Aviation super seals with race springs
- Atkins rotary solid corner seals
- Cosmo upper and lower intake manifold cut in half, all restrictions removed, ported and re-welded
- 105mm wilson manifolds throttle body
- Garret 5 inch core v-mount intercooler with 3 inch inlet pipe and 4 inch outlet pipe
- High Power Improvement (HPI - Japanese company) custom exhaust manifold with 63mm internal diameter exhaust runners into a divided T4 housing flange
- Borg Warner S475 with a divided T4 housing 1.10 A/R and a 83mm exhaust wheel
- full custom 4 inch exhaust from down pipe all the way back
- Motec M84 ECU with all options enabled
- M&W ignition system with Honda CBR 600 coils for a coil on plug application
- NGK 11.5 race plugs (all 4)
- 3 intank Bosch 044 fuel pumps
- 3 seperate fuel rails (ID2000's primary rail, Bosch 1600's secondary rail, Bosch 1600's Third rail), each rail is supplied by its own pump and kept independent until it meets back at the fuel pressure regulator. Pumps are staged based on need in the Motec.

- running only E85 with Benol premix at 3oz per gallon for 20+ boost and/or 700RWHP setting (old set up made 703RWHP at 23psi on Steve Kans mustang dyno on a BW S475 T4 housing with an 83mm exhaust wheel) with a 70% injector duty cycle.

- engine idles at 1800 RPM's and I push it to 10,500 RPM's (M&W max safe RPM limit)


Heres a picture of my dyno graph on the mustang dyno on the old turbo.

Thanks for all of your help in advanced.

arghx

iTrader: (3)

There's no easy answer to that given the limited instrumentation you have on the engine. What I mean by that is, when sizing turbine housings in a methodical way, you really need a backpressure sensor in at least one of the scrolls at the very minimum. Howard Coleman has been running these sensors for a long time. It would be nice if you could try different combinations of parts and see what the sensitivity of engine output is to backpressure.

If it's in the budget, get a backpressure sensor first. See what your backpressure is. Then I would go with the larger A/R turbine housing. The reason for that is in many ways practicality. Changing the rotating assembly will require getting things balanced etc. It will increase the inertia of the rotating assembly and decrease the transient response. The turbine housing will reduce the amount of turbine power available at very low speeds (not a big deal for you) and reduce the backpressure. Where you really see that drawback is trying to make boost at low rpm in high gears.

Let me put it to you another way. From a driving perspective, the turbine housing is going to affect how much torque you could build around town in 5th gear at say 40-50mph. It affects how much you have to downshift in daily driving to go up a hill etc. The turbine wheel will affect that as well somewhat, but going with a bigger wheel will increase intertia. That will affect how long it takes to build boost when you slam the gas at say 4500rpm. The relationship with between these components and top-end horsepower will just have to be measured with a backpressure sensor and a dyno.

arghx

iTrader: (3)

One more thing:

Without a turbo speed sensor, you can't say for sure what the wheel is doing (unless you actually did have a sensor). Hence my comments in my previous posts about not having enough instrumentation to get a very "exact" turbo sizing done. And sometimes it's hard to separate what's happening on the compressor side versus the turbine side. The more pressure and temperature sensors you have on your engine, the more you can figure that out--and the more your car turns into a science experiment.

cr7684

iTrader: (42)

First off I'm a fan of your setup!

If it were me I would install a bigger A/R turbine housing & 5" DP w/480 wheel. If you were to upgrade the turbine wheel, you will get quicker spool but you will also see quicker back pressure with your 1.10 housing this is my .02

sleeper7

Give Mike's turbo a call.

afawaterpolo

iTrader: (7)

arghx - thank you very much for the great info. I do have an exhaust back pressure sensor set up on the Motec. I'll pull the data logs to see the exact numbers. But i do know that the exhaust back pressure would actually get higher than manifold pressure up to about the 4500-5500rpm range when it was beginning its pull (i.e. when we step on the gas on the dyno and start the run). Once the wheel gets spinning it then drops and I never see the exhaust back pressure exceed the manifold/boost pressure.

This is why I was told I am experiencing turbine wheel stall because of the huge amount of exhaust going through that its basically backing up the turbine housing/wheel until the wheel can get enough speed and evacuate the exhaust coming through.

I do not have a turbo speed sensor but it looks like I will get one since its a small price to pay to invest in knowing what the entire turbo is doing.

cr7684 - from the sounds of things i might get the bigger A/R offered (1.25) along with the 83mm wheel and do a test while keeping my 4 inch system. Then maybe test the 1.25 A/R with the 87mm exhaust wheel. luckily the BW parts are relatively cheap compared to other turbos out there. Money is all relative right?

sleeper7 - Will do, like I said the car is a work in progress and if I can find smarter people than me out there who have experience with this I will take them up on what they can advise me on to help me reach my goal.

Everybody thanks again for your advice and help. I really appreciate it. I will keep you all posted once I get a new turbo.

arghx

iTrader: (3)

It's entirely possible that you are experiencing compressor surge. It's hard to find it doing transient tests (normal pulls on the dyno) as opposed to loading the engine down and running up the boost at a fixed rpm as I just mentioned. Whoever was working with you was saying that the turbine side is restricting you, but there could be more going on than that. What you can also do is calculate the turbine expansion ratio by comparing turbine inlet to outlet pressure. You can also calculate the compressor pressure ratio if you instrument the compressor inlet and outlet. Would this be worth the time and money for your application? I'm not sure, but it would be another source of information.

One of the things people don't realize is that when you are looking on a compressor map, the y axis plotting doesn't come from your manifold pressure. It comes from the relationship between the outlet and inlet of the turbo (notice the little "p2/p1" on a lot of compressor maps). The more restriction you have on the inlet side of the turbo, the more your compressor inlet pressure goes down. So your pressure ratio goes up, and the more likely you are to overspeed the turbo (right side of the map) or surge it (left side of the map). It's hard to understand this unless you lock the engine at an rpm (loading dyno) and record compressor inlet and outlet pressure



Since you seem curious, let me give you some background. Here's the instrumentation we normally use in the engine dyno lab when trying to figure out turbocharger matching.

compressor inlet pressure
compressor outlet pressure
turbine inlet pressure
turbine outlet pressure

then there are temperatures at each of those locations, and temperatures/pressure at the inlet and outlet of the intercooler. Then you have a turbo speed sensor, and flow meters to calculate air mass and volume flow (rather than some rule-of-thumb formula and guesswork to calculate flow). That's the raw source of data normally used in a lab environment. Then basically the information gets dumped into an Excel spreadsheet tool that is built in collaboration with the turbo manufacturer so that all the data plots and analysis can be done.


See what happens is, the turbos are tested at the manufacturer on a gas stand when the maps are made. It's done in a lab, but not even on a real engine. And the max temperatures the gas stand can handle is lower than in an engine. Actually plotting these things on the maps to very fine level of accuracy can get really complicated. The formulas you see on various websites and forums on how to map things on a compressor and turbine map are actually highly simplified, although they can get the job done.

Each manufacturer does their calculations a little differently, using different units and correction factors. So there's a whole formula for calculating "corrected compressor speed" and "corrected turbine speed" plus "total pressure (static + dynamic component" to actually get fully accurately on to the map. The corrections on the compressor side aren't as much of a big deal, but on the turbine side a gas stand is going to be limited to something like 650C and a turbo on an engine is going to be hitting 800-1000C inlet temp.

Even with all that, there's still a certain amount of guesswork involved with turbo sizing. Even when you literally have a lab and can sit down with the engineers who designed the turbo things don't always respond as you expect and a lot of iterations can be necessary. But I don't mean to confuse you or side track you on things that probably aren't necessary to know to decide your next move.

afawaterpolo

iTrader: (7)

arghx - I'm right there with you. I guess I'm going to get it as close as possible and i may have to resort to a couple of trial and error combinations.

I have a 1.10 A/R I picked up for free from a S475 and two turbine wheels from other BW turbos I had (83mm and 92mm exhaust wheels). That leaves me with the shopping list of a 1.25 A/R housing and the 87mm exhaust wheel.

The 1.10 A/R fits the 87mm wheel. I'm planning on getting the S480 with the billet wheel and the 1.25 A/R with the 87 exhaust wheel. Then I can do a comparison between both housings (I was told the 1.10 A/R set for the 87mm wheel slides in with no modifications, I will verify before I do that).

Depending what the shows me over the previous set up (S475 1.10 A/R and 83mm wheel, which i know is not apples to apples) I will see about getting the housings further milled out to fit the 92mm exhaust wheel and run them back on the dyno to do some comparisons. As far as I know in the T4 set up, the 92mm exhaust wheel is as big as I can go before having to step up to the T6 housings.

Not the optimal route as you said but it is very cost effective since I would only need to have the turbo balanced one more time (to change from the 87mm wheel to the 92mm wheel) and have both housings milled out to accept the 92mm wheel.

Worst case I will find that the 87mm wheel is best and I will have to buy a new turbine housing since they will have been milled out to accept the bigger 92mm wheel from my previous experiments and get the turbo re-balanced.

zbrown

the 87 and 92 share the same exducer

why not jump to the 96/88? it is now available atleast from FI with the 1.25 t4 housing

arghx

iTrader: (3)

Sounds like you have a good parts list lined up. I would recommend you only change the turbine A/R first as this will have a more limited effect on response and can be easily swapped in and out. Buy the new housing, install it (I am under the impression that it works with your current wheel), get a turbo speed sensor, and see how your engine output and turbo inlet pressure change. Then figure out what to do from there.

With a turbine wheel change you will likely see a loss in response in actual high rpm, high performance driving, due to the increased inertia of the turbine wheel. With a larger turbine housing the loss should mostly be at low speeds, in a manner similar to what porting changes do.

zbrown

the rotary is sensitive to drive pressure, the best thing he could do is put more turbine in it

the turbine he is running now is virtually the same as a gt42

Full-Race Geoff

if you are set on the 80mm borgwarner compressor, then you need to use the 87mm turbine with a 1.25 a/r. Personally i would recommend their 76mm compressor paired with 87mm turbine and 1.25 a/r -- but the 80mm is a great turbo also.

the 83mm turbine will have too much backpressure and much too small for the 13B to make that power level, no matter which compressor wheel you pair it with

afawaterpolo

iTrader: (7)

Geoff, Thats exactly what I'm getting, the billet 80mm compressor with the 1.25 A/R housing and the 87mm turbine wheel. Depending how the dyno graph looks and other numbers (compressor wheel speed, inlet pressures, exhaust back pressures, etc) I might even put the 92mm turbine wheel in there for comparisons sake.

I appreciate everyones help. I will post numbers and graphs as I get the different turbine A/R and turbine wheel combos up.

Full-Race Geoff

weve seen no gain from that 92mm turbine wheel compared to the 87mm. we just finished some extensive testing with a somewhat well known drag team, using motec data logging and a ton of sensors comparing these turbine wheels. I was amazed that the bigger turbine didnt gain anywhere.

afawaterpolo

iTrader: (7)

Geoff - Thanks for the help and guidance, I know how busy you guys are and I really appreciate the help.

zbrown

the exducer on the turbine is the same, what do you expect

if he went to the 96/88 turbine there would be a difference guaranteed

Zero R

iTrader: (1)

PM sent.

~S~