Overboosting/boost creep
 

I know the Greddy wastegate can is shit. That and the turbo are all that is left of the Greddy on my car. I've got a new manifold, I've got a 3" exhaust on the car now and with the MS3 I've got no intake box anymore and now the thing won't hold down the boost at all from what I can see.

I've got the EBC solenoid that I haven't hooked up yet. Does the EBC end up clearing up the boost control issue entirely or do I need a helper spring on the wastegate or new wastegate as well?

it can you if you find out how much you creep to, then use the EBC to raise you boost from the get-go so it's the same in the low-end.

Try installing an all tread rood in place of your canister to hold the wastegate open all the way and all the time, if you go over your boost target that way you will need a larger wastegate opening or a secondary wastegate on the manifold.
If your boost is below your target with the gate open all the way is probably your canister spring rate that is too high or bound up

the problem is your exhaust being so good it's too easy for gasses to flow over the turbine, and spin the turbine and not bypass the turbine directly into the exhaust.

Yeah, the wastegate seems to be useless at this point. It hits the fuel cutoff if I get into it for a bit. The default seems to be cutting it around 10-12 psi, which is probably for the best since this thing needs some work tuning.

Turns out the vacuum hose between the wastegate and the turbo was shot. The thing looked like it was the original hose from the Greddy kit. I replaced that and now it holds steady at 5psi.

Now for EBC install and tuning!

Okay, so it still creeps a bit after getting up over 5000rpm, the gate holds it around 6psi to there, but then starts going up to 10 by the time you hit redline, but that seems like it is expected with this wastegate and the big exhaust. It's not nearly as stupid as it was when the hose was bleeding air out and it doesn't hit fuel cutoff, but still not as predictable as I would like. I guess I need to ditch this can at some point or work with the tips that others have given for putting a helper spring on it until I can find another can to fit.

I don't see any sort of spring being helpful at all. Usually a helper spring is used to help hold a wastegate closed to build/hold more pressure, which you don't want if you are already creeping. And if you for some reason tried to use one to hold the gate open that would not help as you would spool up more slowly, and are already overrunning the wastegate as is on the top end. You either need to port the 'gate so it flows more, or tune for the extra airflow.

I guess I have the wrong idea. I suppose this overboost thing is more due to the larger exhaust.

I guess I should throw on the EBC and see what I can do there?

The EBC will, as Brain noted, allow for better control on the lower end. But you are pretty much stuck with a creeper unless you restrict the exhaust flow or enlarge the wastegate port.

If you have injector and clutch capacity for the extra air, you might as well tune for it.

Well, I've got 440s, so I should be good with the extra boost. Clutch needs an upgrade, but I have that in the garage as well, just need to put it in.

The solution for boost creep from 6psi to 10psi is "run 10psi".

I like this logic a lot

If someone answers my question about the solenoid in the MS forum I might be able to make that happen.

I just don't quite know how to verify the thing is working.

Flyin Miata sells an exhaust restrictor specifically to combat boost creep. Cost$12.95. It is made to fit a 3 bolt flange though.

Keith

that's moronic

that's like having a supermodel girlfriend and using a fleshlight while looking at her

QFT

:idea:

Sav is on to something.

You're going to want more than 6psi very very soon ... might as well make the leap.

-Zach

I just need to know if Brain's build with the center connector that shows IAC2B as one of the connections will work for the boost control. I hooked it up to a 12v source on one line and to that IAC2B line in the center connector and then put IAC2 as the pin in the software.

Not sure if that will work.

boost should be going to a green/red wire on the center connector and set to BOOST as the output.

Green red wire you say?

Well, that's fantastic since I'm red/green color blind. Wait, does the wire say BOOST on it or is that the pin selection in the software?

the wire leading up to the MS should say it.

it's almost almost built like this:

spark c, spark d, fuel c, fuel d, blank, EBC, blank, o2
Y/G, Y/P, W/G, W/P, N/A, G/R, N/A, P

Okay, saw your PM. Lemme go look at what is coming off the harness then and take some pics. BRB.

I'm not recommending it, just saying if he don't want to port the WG or go external it is an option.

Keith

I've got the fueling and the computer to do 10psi no problem. The clutch might start letting go, but I've also got a new clutch waiting to go in once this one goes.

I'm amazed that I wired it correctly just completely by accident. Once I tested the boost output I could hear the solenoid clicking. Whoot! Who knew that Boost = Boost? :loser: :noob:

So, I loosened up the wastegate as the DIY folks say to lower wastegate boost down and then tune from around 2-3psi.

So, I think I have the boost solenoid setup properly. At least when I do a test in TS with the Boost the solenoid clicks.

The boost protection cutoff is set to 175kPa and the thing hits 176 at times in the log.

Isn't 175kPa = 25psi? Am I doing maths wrong, because the boost gauge never shows more than around 10psi.

Edit: Okay wait, I'm not accounting for the fact that sea level pressure is around 100kpa. Dumb da dumb.

I think I need to turn this cutoff up because I keep hitting it before I get to redline and I don't know how far it wants to go. I think I need to set the duty cycle real high in Open Loop mode at high throttle and RPM so that the solenoid opens more and leave it closed down low? This way when it starts to creep up top it will just match what was going on down low? (Am I reversing this? I don't think so).

Here is the latest log btw.

https://drive.google.com/file/d/0B-d...it?usp=sharing

I think I've got fueling at least under control if not optimized as I'm not really going above 11 at this point in boost.

Nevermind last post. Had the polarity set wrong so my boost table was acting in reverse.

I think I'm gonna take it to a dyno next before I start adding more boost. I saw some 14s in that last log where I was boosting up over 10psi. I really don't want to blow this thing up and I've probably been pushing my luck considering all the stupid runs I've done with the boost control not properly set.

Abso-fucking-lutely.

10psi or flaming pussy avatar.

I think I'm doing this wrong. Setup the boost control solenoid in open loop mode.

I made sure the polarity is set so that in the table 100 is fully open (ie normal wastegate mode)

Then I'm pretty sure I did the wrong thing in that because I'm overboosting at higher RPM and throttle levels I should be setting higher RPM/TPS values to 100 and lower values to below 100 to even things out. I instead made the table ramp upwards.

Gonna keep working on this. I'm not sure if getting a different wastegate can would help the overboosting, but I suppose it couldn't hurt.

100% should be fully closed and full boost.

the thing should be wired/plumped in a manner so if the valve fails, it fails to wastegate. If you're suggesting that 100% DC is wastegate, that means you need the MS to actually intervene in order to limit the boost. that's dangerous.



and dont use open loop mode; do it right, bro. (although having the open loop table setup will be a great place to start when using the newest boost control code that uses an initial duty table before CL PID kicks in)

Well, I can change the polarity in the config so that it's 0 for no boost and 100 for full boost, I was just pointing out how I had it setup. The whole thing is a bit counter intuitive for me for some reason since essentially the solenoid being open means it's using just the wastegate and the solenoid 100% closed means super boost.

For some reason this isn't registering to me how it makes a difference I guess, but I suppose I'm not thinking about it from how the MS sees things.

How can I plumb the thing for said safety? Essentially it's just plumbed inline with the turbo and wastegate.

Also, everything I've seen says to tune open loop first then move to closed loop, but I suppose I might have missed some newer info? I was only doing open loop to get somewhat close to where I wanted to be, but quite honestly it all needs to change when I go to the dyno.

Also, I could be wrong, but I thought I was doing the table correctly in that 100% duty means no boost. I was basing that off of this page and the open loop tuning guidelines.

Boost Control

So, back to this again. I am still working off of open loop until I get fuel dialed in, but the problem is that the computer seems to forget that it is set to "inverted" in the options.

I noticed that starting the car up and running it initially it will overboost, just boost goes up until the cutoff hits. If I go into TS and flip the option from "inverted" to "normal" and back to "inverted" again it then starts working fine.

Should I just rewire the solenoid? Swap the wires so normal = boost solenoid all the way closed at 100. It says in the instructions that polarity doesn't matter, but I guess it might?

Everything else seems to check out. I mean, I can test the solenoid and hear it click and once I flip that option back and forth the control is fine after that, no overboosting, it's just this initial startup problem.

Maybe I wire things badly?

Okay, so I went back, looked at what Brain said, looked at what the tooltip was saying in the TS software, set the thing to "Normal" and made the table so that 0 = no boost and 100 = all boost and it seems to work consistently now each time I start and stop it, knock on wood.

Yeah, still having problems with boost control. I really cannot figure out what I am doing wrong.

I went out yesterday to tune the table under open loop conditions. I started slowly raising values in the table and boost slowly started coming up until I got to a number I was comfortable with around 10 or so PSI max. Drove around like that for 30 minutes or so, no issues. The thing held solid boost levels.

Today, I took the car out, overboosting right from the get go. Like I never even made the changes yesterday.

I wonder if I hooked up the solenoid wrong somehow? Whenever I test it at a stop using TS I can hear it clicking, so it seems like power is good, though I do have it spliced in with one other device. Could that cause an issue under load? But then again why would it react one way during one run and completely differently on another?

In case anyone has any suggestions here is a ZIP with the MSQ file and two logs. The 5-4 log is the one that was running fine, the 5-5 log (today) is where it was overboosting.

https://drive.google.com/file/d/0B-d...it?usp=sharing

Maybe I should just stop trying to get open loop to be consistent? I sort of assumed if I adjusted it in small increments it would end up being consistent, but is that not the case?

Open loop is never going to be perfectly consistent, it can't compensate for air temp, humidity, etc. It just matches duty vs rpm and throttle.

That's what I was thinking about. I did notice that a few points of duty here or there really seemed to make a difference in how much boost the thing made. I guess that could just throw it off depending on the day and temp, but I didn't expect it to be so far off. I suppose I should just get as close as I can and then do closed loop.