Clowning around with boost...

[Z3eattle]

…because I’m inexperienced with turbo builds and this will be on a clownshoe. It’s a homonym.

Car: 1999 Z3 M Coupe (s52)

Goal: It’s a street car (non-daily) but the build will be track oriented: local clubs typically have 3-4, 20min sessions in a day. Pump gas (91 in my location and I’d like to avoid using e85/meth. Keep it simple if possible). I want to shoot for as linear, responsive feel as possible, but bring the power up to modern performance standards, or more. I drove my friend’s s52 m roadster with GT3582r, SPA, stock internals but cut-ring/spacer combo running 15psi and claiming 520hp. Peak power felt perfect and low rpm grunt felt “stock”, no lag, but definitely a sudden wall of power at 3500; throttle response was very binary. So I’d like to build something with roughly the same power level but more oomph down low/faster spool/linear responsiveness if that makes sense.

Here's my best-guess half-baked recipe so far after reading pretty much every relevant thread:

• BW EFR 8374 (not sure on A/R size or T4 twin sroll vs T3) – also maybe EFR 7670 for less power but more spool?
• Built motor – Cut ring, Eagle rods, JE pistons. 9:1 ratio?
• Manifold (bottom mount due to persistent heat) – If I go T4 twin scroll route, is steedspeed the only bottom mount option? Does Max PSI still make a bottom mount TS? T3 with SPA is potential option but will EFR/s52/SPA combo all fit together? I’ve not seen this before.
• Good tune obviously and lots of other components to figure out but does this sound like a good base to start from? I realize there are less expensive approaches to the same power levels but I want to build something refined and modernize a 23+ year old platform.

One big question is how limited am I with 91 pump gas? My research says no more than 15 psi but does running lower compression pistons allow higher boost levels here? I realize this would be a trade off in low rpm spool.

One other question: there’s lots of discussion about controlling lower boost levels with EFR IWG setups. It seems fancy boost controllers and dual IWG/EWG setups have been used here. Has anyone seen more simple solutions?

Note that all other aspects of car are prepared for this (brakes, cooling, suspension)

Thank you for any insights and advice you may have. I will have dumb questions as I go but I will post relevant updates and share learnings/results around the more unique aspects of this build.

[someguy2800]

Running low compression pistons won’t raise the limit compared to running a cutring and spacer, but forged pistons will be much more tolerant of heat and mild detonation, so they will absorb some abuse compared with stock pistons that just break the ring lands. I would suggest that for track days you throw a bottle of race gas concentrate in the tank or just mix in like 5 gallons of 110 unleaded. You may also be able to get 100 octane unleaded from your local airport for like $5-$6 a gallon. We can get it at the local small airport here to put in our snowmobiles.

the EFR’s can be difficult to control boost. I would recommend getting a T4 housing and weld the wastegate flange right on the housing like this and just block off the wastegate port on the manifold.

[Z3eattle]

Thank you someguy. Are you saying that using higher octane at the track would be a good measure of safety per the extended run cycles and heat? Or as an opportunity to run higher boost in general?

The wastegate flange welded to the housing is a novel idea. I only have a rudimentary understanding of A/R ratio but would such a flange effect this?

Seems like 91 octane is going to be my limiting factor here. How 'set it and forget it' is a water/meth implementation? Obviously one needs to fill it up regularly but is it relatively hands off once it has been tuned as part of the system?

[pbonsalb]

8374 is about the same size as GT35R. Might spool slightly better. There are some dynos from CES years ago comparing the two, but at high boost levels and with 100 octane. I have run both. The 8374 was the external wastegate version, on an open manifold, and I could not control boost. I now run a 9180 but with both internal and external wastegates, on a twinscroll.

A Precision 6262 is also worth looking at. Or maybe a smaller turbo since you are not trying to make big power.

[Z3eattle]

Bummer the 8374 with external gate and open manifold did not control boost for you pbonsalb. I was hoping that might be a successful set up.

[pbonsalb]

It was a gen 1 steedspeed. I don’t know their reputation for controlling boost. The gen 2 steedspeed is poor but has other advantages like twinscroll and higher mount so no scavenge pump needed. Maybe some other manifold would do better.

[Colby Colbs]

I have a gen 1 Steed Speed and a Maximum PSI bottom mount, both are open manifolds, Not twin scroll. However, I did not use a scavenge kit on either car and had no smoking issues. Was going to list them up for sale here eventually if I didn't end up using them.

[dominator293]

Might be worth looking into the small Garrett/Pulsar G Series turbos, they spool pretty quickly and produce good power. The G30-660 or 770 might be the best option for early spool and enough puff to hit your power levels quite easily.

[someguy2800]

Thank you someguy. Are you saying that using higher octane at the track would be a good measure of safety per the extended run cycles and heat? Or as an opportunity to run higher boost in general?

The wastegate flange welded to the housing is a novel idea. I only have a rudimentary understanding of A/R ratio but would such a flange effect this?

Seems like 91 octane is going to be my limiting factor here. How 'set it and forget it' is a water/meth implementation? Obviously one needs to fill it up regularly but is it relatively hands off once it has been tuned as part of the system?

I meant to give you some additional safety margin for track days because the heat buildup that happens on a track day tends to lead to detonation in conditions that would normally be fine for a street car. Guys doing drifting or track days have a much harder time with pushing headgaskets and breaking pistons than drag racers or guys just doing short street pulls do. Water meth is wonderful until it stops working because the pump failed or the nozzle clogged or the tank ran empty ect…. I personally wouldn’t rely on a water meth system unless it had a flow sensor and failsafes that can shut the boost controller down.

[Z3eattle]

I meant to give you some additional safety margin for track days because the heat buildup that happens on a track day tends to lead to detonation in conditions that would normally be fine for a street car. Guys doing drifting or track days have a much harder time with pushing headgaskets and breaking pistons than drag racers or guys just doing short street pulls do. Water meth is wonderful until it stops working because the pump failed or the nozzle clogged or the tank ran empty ect…. I personally wouldn’t rely on a water meth system unless it had a flow sensor and failsafes that can shut the boost controller down.

Thank you for the clarification. That is super valuable info and something I did not have on my radar. My local tracks have high octane fuels available on-site.

+1 on the water/meth system. Another critical component to forget about.

[Z3eattle]

Might be worth looking into the small Garrett/Pulsar G Series turbos, they spool pretty quickly and produce good power. The G30-660 or 770 might be the best option for early spool and enough puff to hit your power levels quite easily.

The g30-xxx line up looks promising. Adding this to my options.

[Z3eattle]

OP here. I've decided to go with this set-up below but have a question about which A/R size and wastegate size:

(Again, s52 m coupe, eagle rods, 9:1 pistons)
SPA manifold
BW EFR 7670 (AGP T4 open housing)
Weld WG flange directly to housing; cap off SPA outlet
Use stock OBDII s52 intake manifold (to maximize low end torque)

Hopefully this recipe will give me a good linear power delivery for track use. On 91 pump gas I'm hoping to get 15psi... not really concerned with peak hp but thinking mid to upper 4 hundreds?

But onto the A/R question - I have .61, .86, .91 or 1.27 to choose from. With my goal of biasing for linear response but getting the most out of this set up, which size would work best?

Also, WG flange off the housing should mitigate boost creep but should I still go with a larger WG size? Say 45mm?

Thank you for helping a noob!

[NOTORIOUS VR]

I'd say that turbo is too small in general for an S52

[Z3eattle]

What about efr 8374 with .68 A/R? Early boost but won't choke up top?

[NOTORIOUS VR]

What about efr 8374 with .68 A/R? Early boost but won't choke up top?

a small back housing won't bring you better spool it will just hurt you everywhere (including spool). It's a pretty large misconception that a small back housing will help spool actually.

Again you have an S52 so you have a big motor that flows very well out of the box that does not like to get choked at all.

The 8374 is a good choice and was proven by CES to out spool a 35R everywhere and still make over 700whp. They used a 1.05 back housing. I wouldn't go smaller than the 0.95.

https://www.bimmerforums.com/forum/s...1-2-Mile-event

[pbonsalb]

On an S52, I ran a 1.05 8374, single scroll steedspeed manifold, Tial 45 EWG, and could not control boost. Did not try welding WG to turbine, but that should be effective though it probably creates turbulence that reduces efficiency.

[NOTORIOUS VR]

On an S52, I ran a 1.05 8374, single scroll steedspeed manifold, Tial 45 EWG, and could not control boost. Did not try welding WG to turbine, but that should be effective though it probably creates turbulence that reduces efficiency.

Steedspeed certainly has boost control issues due to it's bad WG placement and turbo choice, etc.

I can't comment about adding turbulence/reducing efficiency and I'd rather see proof of this being the case than claiming it is so. My guess is it has virtually zero impact on anything.

[Z3speed4me]

For general awareness... I do not know of a Z body car that has used the steedspeed yet, so whether it actually fits our chassis space limitations is a mystery.
IF I used one, I would close off the wg port and have one welded to the turbine housing just like someguy2800 noted.

Housing swaps are also an interesting topic for me; if you refer to my recent post. I went bigger with no drawbacks. But every car is different... and theory does not always equal real life practice and implication.

[pbonsalb]

Steedspeed certainly has boost control issues due to it's bad WG placement and turbo choice, etc.

I am not sure the original open steedspeed that I was using with the 8374 had that reputation. The newer twinscroll that I have now certainly does, which is why I am using a 9180 with both IWG and an EWG on the manifold. There may be some manifolds that could control the 8374. The boostlogic I had years ago had a reputation for great boost control, but I had already sold it when I experienced problems with the 8374 on the original steedspeed.

A downside to welding the EWG to the turbine housing is losing the twinscroll. This does not matter if you have an open manifold and the 8374 will still spool fantastic on an open manifold on a 3.2L. I think Mike Radowski did an EWG on a twinscroll turbine and added a divider to preserve the separate paths or maybe it was a remote pipe off the manifold that was divided all the way to the EWG - fancy welding.

[Z3eattle]

Thanks everyone for the responses. I think that's it then. I'll go with the 8374 with .96 A/R, open T4 and weld wg flange to housing.

For reference, my friend's s52 m roadster uses a GT35r, spa, M50 intake mani and wg flange welded to housing @ claimed 520hp. The best thing about this is I know it all fits the Z chassis. His car is a blast to drive but definitely a wall of power. Hopefully the 8374 with retaining the stock s52 intake mani will give me good low end spool comparatively. My local tracks all have 2nd gear tight s curves, which are a bit of a dog even with a NA s52.

I'll post progress/results once I dig into this.

[NOTORIOUS VR]

A downside to welding the EWG to the turbine housing is losing the twinscroll. .

Again, you say this but where is the proof? Why would you lose the TS function by welding a gate to the housing? TS relies on separate exhaust pulses diverted to the housing to get turbine speed up as quickly as possible. The WG remains closed until your specific WG spring/boost controller set point. If it's closed how would it effect anything related to the separated pulses that are making it all the way to the housing.

[darknezz7]

Reading all this is very similar to what I am dealing with now and I have learned more about boost control. Good info and specs here thanks.

Good luck with your project z3eattle sounds like it would be a fast car. What have you done to make it handle better?

[Z3eattle]

Reading all this is very similar to what I am dealing with now and I have learned more about boost control. Good info and specs here thanks.

Good luck with your project z3eattle sounds like it would be a fast car. What have you done to make it handle better?

TC Kline S/A set up.. 450# front/600# rear (utilizes koni shocks)
Vorshlag adjustable camber plates
Randy Forbes subframe reinforcement
Front brakes - Stoptech Trophy calipers/355mm floating rotors
Rear brakes - Porsche 996 rear calipers w/e46 m3 rotors
Euro oil filter housing to utilize an external Earl's oil cooler
Other poly and UHMW bushings here and there

Yes, brakes might seem overkill but the stock brakes kept boiling fluid.. tried Motul 600, different pads, brass bushing inserts, but across three track days the fluid would always boil by the end of the day. There's maybe an anomaly going on here but also don't mind investing in good braking system for the track.

[pbonsalb]

Again, you say this but where is the proof? Why would you lose the TS function by welding a gate to the housing? TS relies on separate exhaust pulses diverted to the housing to get turbine speed up as quickly as possible. The WG remains closed until your specific WG spring/boost controller set point. If it's closed how would it effect anything related to the separated pulses that are making it all the way to the housing.

The way I have seen it done, there is a stub pipe or even a long pipe from the turbine housing to the WG. If that pipe is an open pipe to the WG itself, it seems like it will be reducing the twinscroll effect. The WG valve is flat and can’t directly seal the new hole on the turbine scroll. Same logic that Steedspeed used when putting a divider in the EWG hole on top of the twinscroll version of the steedspeed manifold - keep the exhaust paths divided all the way to the EWG valve.

[NOTORIOUS VR]

The way I have seen it done, there is a stub pipe or even a long pipe from the turbine housing to the WG.

so you think a stubby pipe with a WG at the end that isn't open is going to greatly effect TS properties? Again I doubt it makes any measurable difference. I've yet to see anyone show actual data about it either on a back to back comparison so I'd say at this point it's fair to say no one actually know if or how much it actually makes a difference. So you telling people that it does isn't really a good idea. There are enough shops/vendors that say things do this and that without actually proving it - it's something that people need to stop doing so that people can make better decisions.