Stock MSM -- insane turbo lag on shift
 

xposted on m.net

Alright, because I'm a masochist I went and bought an MSM.

One problem.. there's a ludicrous amount of turbo lag between gears. I've only run ball bearing turbos before and even on an NB with a GT3071 there's less lag.

I've read mazda-speed archives, m.net, MT, grassroots forums, and any other article I could find about the various ECU-related issues with the MSM. I understand there are three major problems:
1) The Bog.
2) The "lean spot" as it's referred to on m.net
3) The ECU just freaking out and pulling timing or fuel for some reason.

I can't find any actual technical information on what is happening or why for these, just people who don't know what's going on taking random guesses. Further, I can't find any descriptions of the behaviour when #2 or #3 is happening, only "The Bog".

I know it's not the bog. I spool around 4k reliably (which is pretty damn late IMHO) and don't have a sudden burst of power at 5500rpm.

I'll be calling FM tomorrow to discuss what the o2 clamp is actually solving and if that would solve this issue. I have another thread pulled up somewhere about how the o2 clamp in conjunction with a modification to the WOT switch on the TPS fixes random ECU freak-out lean condition somethings for WOT transitions, but I don't know if that's even my problem because I can't find a description of the driving experience.



Setup -

2004 with no power-adders.
94k miles.
Inconel studs, AOS check valve, rebuilt turbo (didn't need it, just while I was in there)


Symptoms -

Lumpy powerband, power doesn't really seem to come on until 4k or so and it's a huge slingshot.
Huge, huge delay in boost if I shift like a sane person. Akin to being at 2k RPM in second gear and flooring it, waiting for around 4k to hit and power to come on.
If I shift like a Honda kid yelling "POWERSHIIIIFT" or "NAAAWWWWSSS" this delay is minimized.


Ideas -

Could be that it's just a journal bearing turbo and there's a stupid amount of lag.
Could be this mysterious "lean condition" I've read a lot about.

Toss this around to the proper subforum if this is wrong. Figured that as this is more than likely an ECU related problem, it'd be appropriate-ish.

You're describing the lean spot. I refer to it as the "hesitation" in order to differentiate it from the "bog", which occurs when moving from a stop and refuses to accelerate (feels like you're setting off in 3rd).

The solution, as many will shortly tell you, is to move to a stand alone ECU. The stock ECU is dog's balls. Megasquirt is the preferred ECU in these parts. While you're in there do injectors and an EBC solenoid so you can run more boost with better control.

Sounds like a stock MSM to me. Absolute garbage.

Standalone ECU is the only way to make them driveable. If you're not willing to do that, i'd get a different car.

has nothing to do with journal bearing turbo's and everything to do with your junk stock ecu.

in fact, one of the biggest misconceptions is people thinking ball bearing turbo's have better boost threshold. they do not.

+1. Fuck, EFR's are journal bearing and they are regarded as the best spooling turbos out there.

OP check for boost leaks, otherwise follow the comments above.

It's the lousy stock ECU. I just today ordered a MaxxECU for my MSM. New in the Miata world, but huge in big HP turbo applications on many other cars in Europe. Supported in the US on MSMs by Brett Willett, who does a lot of remote tuning on Megasquirt turbo Miatas, including MSMs. Lots of reasons I picked it over a Megasquirt, but Brett was one of the big reasons.

Having a megasquirt that's still being tuned, I have to say I should've done this first.
I went with revs megasquirt MS3 Enchanced and 18psi is tuning it, as he tunes AND had owned a MSM. Otherwise i'd have gone with MaxxECU as well.

I owned several MSM's. Just sayin ;)
My 04 Ti I legitimately ordered and received my MS3-Pro before I even picked up the car :D
This MaxxECU is interesting. I just glanced over their specs, not bad, at least on paper. I'm kinda curious why it's not more common/popular.

Yes, looking at the specs and software, it looks really good. Certainly better processing power than a MS3. All the inputs & outputs, plus better connectivity (onboard USB and Bluetooth). Brett works with it professionally on high output big turbo motors plus his own drag bike. He knows Miatas, so he's brought it to us. I'm only #5. He's focusing on MSMs to start. Obviously, I'm hoping it does become more common/popular, but I'm not afraid to go against the grain and be an early adopter. Hell, I'm so contrarian I've been running an Xede!

I think it's not more popular or common just because Brett is the first to support it and tune it for Miatas, including building the harness and such for those of us not saavy enough to DIY. He's installed and tuned the first one in person in TX. The MaxxECU is Swedish, and I don't think they've paid much (any) attention to the US market, nor done a bit of marketing in the US. Given my long background with Saabs I kind of have an affinity for Swedish engineering. They seem to spend the long winter nights thinking about how to do things well. And Swedes seem to be far better engineers than marketers in my experience working for a Saab dealer back in the day! So it makes sense to me

Perfect, this is exactly the type of answer I was looking for. Thank you!

Have fun, the rest of you.

Nope

http://www.turbos.borgwarner.com/fil...ical_brief.pdf page 35

There are BW drag race specific turbos that are journal, but most of them use ball-bearings for improved spool.

ball bearings don't improve boost threshold, or "spool".
they improve transient response, which is way different from the above.

From BW doc. "For this reason, the efficiency enhancement from a ball bearing system benefits the spool-up period" BW seems to think they increase spool-up.


Garrett does too. https://www.turbobygarrett.com/turbo..._ball_bearings

"Garrett Ball Bearing turbochargers spool up 15% faster than traditional journal bearings."

I'm here to learn. What am I missing?

When they say spoolup, they are not talking about boost threshold, but the "spool up" or transient response once you're at/past threshold.

So if you compare identical turbo's with the only difference being journal vs ball bearing, the dyno plot will look identical.

Most people say "spool" but mean "boost threshold". If you didn't, then disregard.

Boost threshold is the point at which the turbo starts to create positive pressure right? You're saying that point doesn't change based on bearing design and I don't have anything to dispute that.

How quickly boost increases after that threshold is what I was referring to as spool and that should be better with a ball bearing design.

If you reach target boost sooner with a ball bearing turbo why wouldn't that be reflected in a dyno plot? Peak power will likely be the same but shouldn't there be more area under the curve?

Full sweep dyno plot, boost level is largely dependent on whether or not the engine is pumping the air to make target, so... RPM.

Where you'd see a difference is say on a car running a 6258 and a car running a 2560R, and logging how fast each reaches 14psi when you mash it from 5000rpm.

correct
transient and threshold are two completely different things, and a dyno is done going WOT well before boost threshold, so unless you started the dyno well into the boost threshold (like say 4k rpm), you won't see transient

that's why you can't just make statements like "faster spool" without clarifying what you mean by it. 99% of the people out there don't know the difference and JB turbo's are looked down on for no reason

My car has been running journal bearing turbos (old school T3/T04B/E) for years and I have zero complaints. The only time I've had problems with transient response is when I had a huge exhaust leak before the turbo.

Boost threshold is another thing. It's kind of funny how you can actually watch the boost gauge showing you the edge of the compressor stall region as your cfm climbs with rpm into the non-stall region at higher and higher pressure multipliers. My old T04B setup would make lowish (but still decent) boost at like 2k rpm and it would slowly climb to 18psi at 4000 or so. The T04E just suddenly comes alive like a 200 shot above 4k rpm and is kinda gutless below that. Stall region is more vertical in that compressor.

here's a modern journal bearing turbo threshold plot. this is a 400whp turbo btw ;)

notice it makes boost BEFORE you even get to 100% tps, and this is with stomping on the throttle, AND this is at just over 2k rpm. lag? what lag?
https://cimg2.ibsrv.net/gimg/www.mia...5327bcf0ba.png