RIWWP

iTrader: (2)

EBC received and installed. I updated to the latest firmware in preparation for testing it, and immediately found a number of problems that makes the car nearly undriveable on that firmware with my maps. I don't know what, but...it's garbage. Stumbling and hesitations all over the board, even with painstakingly going map by map and setting by setting to make sure that the data transfers completely.

So I reverted back to the old firmware and related map.

However, I also found a thread on mazda-speed.com about setting up the EBC, and it was on the older firmware that I am on now, so I shouldn't have a real problem with that yet. In that thread however, there was something mentioned that I kind of knew already in the back of my head, but was stressed. The Hydra sees boost at the intake manifold. My manual boost controller has been seeing boost at the compressor wheel outlet. Because of the distance and the intercooler in between, these simply don't match.

I had an "ah ha!" moment, realizing that my ~4psi boost overshoot that I am seeing now vs ~2psi overshoot before was simply because I had a more restrictive intercooler in the stream. CanyonFive also said that it was nearly impossible for him to tune the EBC with the EBC seeing pre-intercooler boost and the Hydra trying to control it from boost at the manifold. It just can't get the target vs actual correct.

So this morning I moved my boost controller signal source over to the manifold (pre throttle body is supposed to be better, however I don't have an easy way to tap that quickly). Immediately I have a 1psi or less boost overshoot, and I had to turn DOWN the boost controller 3 clicks to stay off boost cut (which is at 12.5psi), where as before I was settling at ~8-9psi. Boost also now only drops off by about 1 psi by 6,500rpm, so I'm at ~11 at 3,800, and 10 at 6,500, and it is a very slow taper.

And the difference? The MSM runs "like a scadled cat" compared to before. Significantly faster. "Oh, so that is what happens when I'm not giving up 1/3rd of my boost!"

The only noticeable downside at the moment is that in high gear low RPM full throttle, the wastegate flutters more exageratedly than it should be, leading to noticeable rapid 'pulsing' in power. I've confirmed that it isn't the AFRs or the timing, or the ignition. I'm pondering solutions, but I suspect that it will mainly just mean finding a way to shorten the signal lines to improve the response time for the MBC.


So my EBC may not be necessary after all. I will be seeing if i can tune the settings to get it to work correctly this weekend, but relocating the signal source solved my current complaints.

RIWWP

iTrader: (2)

Well, dyno day today to get retuned for the GT2560R turbo and LS2 ignition.

And it was very nearly a complete waste of 4 hours of dyno time.

The wastegate 6psi numbers were acceptable, if a bit low, 147wtq / 162whp. Not that much lower than my stock IHI turbo's 162wtq/177whp at it's wastegate 8.5psi.

Then we started turning up the boost, and getting more and more puzzled. My tuner said that for the engine size and turbo size, I should have easily been gaining around 10-15wtq and 15-20whp for each pound of boost added. I was getting 7-9wtq down low, and 2-3whp for each pound of boost added. Namely, that boost was being added on roll in, but the boost was just falling off as the revs increased, so that at a base of 12psi, i wasn't holding more than about 6psi at redline.

So we spent a lot of time troubleshooting. Negative results on boost leak testing, spark blow out, coil breakup, wastegate actuation problem, or fuel delivery failure.

AFRs all pretty pegged from 11.2-11.5, spark timing was optimized and he was finding losses from further retarding or advancing. It isn't a fuel delivery problem, as he had trouble pulling enough fuel out past 5,500, it was always running into the 10s and 9s every time we turned up the boost.


The only three possibilities that we could come up with, from most likely to least likely:
- There is a clog somewhere in my piping 'capping' total flow, more probably the exhaust piping than the intake piping
- My recent timing belt job erred, and I am off 1 tooth on either intake or exhaust cams, or both
- There is engine damage already that we are unaware of (and that there are no other symptoms for)

From 3,500 to 5,300, everything is smooth and hugely strong, just falls on it's face after that and torque/power tanks. Each 1psi increase was only generating about 2-3whp more, over a wastegate 6psi 147/16. My stock IHI turbo did this too, and in fact the boost fall off curve from 1 year ago was virtually identical to this one's, suggesting that whatever problem is causing it existed before the FM2 conversion. Since the only piping left that was in place for dyno'ing last year is the intercooler-to-throttle-body short section and the catback, I struggle to see how that is possible, but have to conceed that it is the most likely since the behavior most closely matches that problem source. Perhaps my prior cat, or the original owner's cat, blew out and is clogging the cat back and I've never noticed? Dunno.

This week I will be removing the cat section and test driving it with a straight dump from the downpipe. I should have a power jump regardless, but curious to see how much of one. Regardless of that outcome, I'm going to then pull the rest of the exhaust, every single piece, as well as the entire intake set of piping, and see if there are any clogs.

If nothing found, reinstalling all that and taking the car to get a compression and leak down test, just in case there is hidden damage somewhere.

Failing that, then in about a week and a half, I'm going to start disassembling the front of the car again to expose the entire timing belt and start really closely checking it against the rather confusing timing marks.

I'd put my money on a clog somewhere though, as unlikely as I think it is. The symptoms/behavior just match it perfectly.

Frustrating though. At least one more track event I have to scratch off my planned list. At least I got my AFRs optimized far better than they were at the lower boost levels, and I can feel the improvement there.

N/A is perpetually getting more appealing. Boost is fun, when it works, but it is a real headache for more of the time than I really want to deal with it.

200.mph

iTrader: (9)

**** deal pearl. where you going for comp and leak down test? ill ask my bro for a price if you want for both

WaitingforFI

iTrader: (1)

Shitty Pearl. Hopefully it's just a clog. If that is the case, I wonder how much better the FM set up is compared to your last setup.

RIWWP

iTrader: (2)

Actually, yes please if you don't mind.

It is supposed to be HUGELY better. A larger turbo, but due to all the improved technology and the reduced complication of the intake piping, it spools as fast as stock with a significantly higher top end.

I mean, the sequential spark LS2 ignition at 0.040 gap ALONE should have shown a significant improvement over the weak/old/failing factory waste spark coils at 0.023 gap. Even ignoring the other turbo.

My old setup vs my original at-purchase:


And what this setup did on another MSM with the same turbo, manifold, exhaust piping, intake piping, ignition, ECU, injectors, etc... Only thing different is a different intercooler.


Even if you account for a different dyno (rototest vs dynapak), altitude (5,400ft with correction vs nearly sea level)...

Something is hugely wrong. My torque curve fall off was just as steep as the ramp up.

You can also see my original vs the other stock MSM traced in the 2nd dyno. My torque fall off then is already far faster than it should have been.

200.mph

iTrader: (9)

pearl, im meeting up with him on my lunch today so ill ask him then

Gravey

iTrader: (3)

Holy crap, it really falls off there eh?

WaitingforFI

iTrader: (1)

Pearl,
That drop off is pretty wild. I hope there is just a choke point someplace and not internal damage.

RIWWP

iTrader: (2)

Yeah, a big clog somewhere makes the most amount of sense if I ignore that nearly all the piping, including the cat, is brand new. There have been suggestions that maybe I left a rag in a pipe somewhere. I really want to say I didn't, but.... It would fit. That plus it seems that whatever I'm dealing with now, i was dealing with a year ago and no one noticed. The only flow-based hardware that hasn't changed is the engine and the catback. No other symptoms of engine damage, and i'm hard pressed to see how my catback could have been clogged for over a year.

200.mph

iTrader: (9)

so your ghey car may be on the rag? why the hell are you running a cat? cats are only good for two thing, chinese food and baiting coyotes

Gravey

iTrader: (3)

You know what, I hope it is something as simple as a rag in a pipe...

RIWWP

iTrader: (2)

A catless rotary smells good, a catless piston car does not. I already have the hardware to break my engine from torque and/or power (well, in theory didn't pan out this way yesterday), so it's not like I need to remove the cat to increase how much more power I can't make.

So all else being equal, I'd rather spare my nose.

WaitingforFI

iTrader: (1)

In theory, if the butt dyno notices a huge leap in pwr/tq then it must be a clogged cat back, if however its just a mild bump, then I guess it's time to snipe hunting... err looking for that rag.

RIWWP

iTrader: (2)

Removing the cat should result in a noticeable gain, even without a clog (because a cat is basically a clog anyway), so it would have to be something like "whoa!" huge to indicate that it is an exhaust clog problem. I'll be removing everything anyway, just to see, run water through, etc...

RIWWP

iTrader: (2)

Here is a baseline graph of the problem as it sits now. gears 1 through part of 5th. You can see how the boost falls off. If I continued pulling to the 7,000 soft cut, the boost would have continued to fall into the 5s.

As soon as things cool off enough for me to be handling the turbo assembly to cap it off for another boost leak check, that's my first step. Then pulling exhaust parts.

RIWWP

iTrader: (2)

Boost controller failure it appears. A new boost leak test showed that the further up the boost is turned, the more it leaks.

WaitingforFI

iTrader: (1)

On the bright side you don't have to remove the exhaust.

200.mph

iTrader: (9)

pearl you need to stop messing with your car. everytime you touch it something bad happens lol

RIWWP

iTrader: (2)

Well, I did another run without the boost controller, just a single pressure line to the wastegate from compressor outlet, and boost would only get to 5psi instead of wastegate of 6, and would fall off at the same rate as before to 2-3psi. So a failed boost controller is A problem, but not THE problem.

I won't be doing anything with it today I expect, office day.

WaitingforFI

iTrader: (1)

I agree with 200, I will gladly take that headache off your hands.

RIWWP

iTrader: (2)

If you are serious, make me an offer. I'm not attached to this car.

WaitingforFI

iTrader: (1)

If I wasn't paying for both my wedding and buying a house this year, I would be tempted.

RIWWP

iTrader: (2)

Compression test shows all 4 cylinders within 4psi of each other. Numbers are low, but I suspect that is just my $20 tester having it's gauge off.

I talked with the main project/build guy at Flyin Miata, and he insisted that it was a boost leak that I hadn't found yet, and gave me a few different ways to test for boost leaks that could have different results.

Sure enough, I found one. My BOV is practically exploding open in a cyclical rythmn at even a few psi when there is no where else for the pressure to go (everything capped off)

So new BOV time it looks like.

200.mph

iTrader: (9)

better than a rebuild

RIWWP

iTrader: (2)

Sigh, nevermind.

Supposedly my BOV is supposed to open like it does when doing a boost leak test the way I did. I can't find anything in internet searches to back that up, but not only does it make sense that pressure pushing on the inside of the piston would effectively simulate vacuum pulling on the other side of the piston, but the company that the BOV originally came from just verified that yes, it will do that in the configuration used for the Miatas.

Back to the drawing board, and pulling the exhaust next. Won't happen today though.