Blaize

So continuing from my earlier thread, I now have it up and running but not quite right. As a refresher- I am running a MS built as per the brainiack how to, and using the diy autotune software package for a 92 w/o afm. I do still actually have the AFM in place at the moment but I am using a GM iat and the AFM is unplugged when on MS.

So, It starts but needs a little "goose" of the throttle to get it going. Cranking w/o touching the throttle just gets a brief sputter. Once started if I hold it at say 1.2-1.4k for 30 seconds or so it will settle at around 900 and take 30 seconds to slowly drop down and stall. If however I keep it going for 2-3 min. it will hold fine at around 1k on its own for as long as I leave it.

I have set the timing values and that improved things a little. Switching on the a/c of course throws it for a real loop and it drops down to 600 and never recovers. Even after turning the a/c back off. But I am not all that concerned about that at the moment. I will do the perez carbon can trick once all the rest of this is in order.

So how do I adjust the idle speed? Via the stock screw? I have noticed that when I put the stock ecu back in the car after these little foray's the idle is very low and unstable the 1st few times it is driven, so perhaps some clown in the past has set the t-body too low and it needs to be brought back up. I presume this is best done with the stock ecu in place using the diag 10deg. lockout plug?

The other thing I found is that it wont rev past 3.5k Now I know that the fuel maps i have are only meant as a starting place But it should at least be drivable right? Sitting in neutral it will wind up to 3.5 or so and sounds like an old MSD chatterbox launch control device. Any Ideas?

Isuppose I will now go thru the megamanual again and get a solid idea of how to tune the enrich maps and such. I just dont want to spend a day or two dicking with the mspnp settings only to find the car is the prob and I wasted my time. What aside from the idle screw setting should I be checking? How close is the pnp start-up enrich settings? are they pretty much universal or is it expected that they will need tweaking like this?

And for the record the car is a 92 and has a cat back system and an HKS air filter. Otherwise bone stock.

Blaize

Right, a little update. Today I filled the tank from damn near empty with high grade fuel, here in the uk cheap gas is 93 octane, the good stuff is closer to 96 or so. But they come up with the numbers differently so who knows how it all compares to us fuel.

At any rate the idle problems are now much better. Just about livable. So I will assume that a little adjustment later will cure it. The real problem is the inability to rev past 3000 rpm. I did drive the car around on the ms a bit and the problem is there under load or free reving. watching megatune while it happens reveals that it revs to around 3500 and then starts to sputter and go no further. The tach on the dash will come up to 3.5k and then drop down to about 2k while it is doing all the sputtering, then leaps back up to correct when the throttle is released. Seams to me there is a problem on the tach signal line or with the actual coils. Anyone have any experience with this sort of problem? I did check the MS rev limit values and even set it to 2k in order to make sure it was working and it is. Totally different of course the rev cut is a nice smooth on/off/on/off whereas the problem I have is a popping sputtering mess.

Blaize

And another update. I used the Diag port to lock the stock ecu to 10deg. as per the enthusiasts manual and the idle dropped to 600 rpm. Looks like that problem was the car and not the MS all along. I have now set it to 900 and while I have not actually tried it with the MS I am confident that it will be fixed now. It was damn near 3 full turns out of whack.

So that narrows it all down to just the inability to rev past 3k. C'mon MS gods, what should I be checking? It feels to me like it is having some sort of problem getting a clean tach signal into the MS or something. Anyone?

akaryrye

iTrader: (5)

Couple thoughts:

1. Are you running on all four cylinders? (2 plugs black, 2 plugs clean is a symptom)
2. Check that your ignition dwell settings are in line
3. You do have a WBo2 right? A fucked up fuel map can make the car fall on its face easily

Your symptoms sound exactly like what happened when one of my injector wires de-soldered itself from the MS harness. Took me a bit to figure out what was going on and why it sounded like a dump truck

Blaize

Its running on all four just fine, sounds and runs just like the stock ecu up to 3.5k.
I have not checked the dwell settings, frankly I am trusting the pnp fuel and timing maps more than I should.
And no I don't have a wbo2 (yet), but if the pnp works with the stocker than this should too right? As I understand it the wbo2 is great to have but not absolutely needed (and before anyone chastises me, I will be changing the injectors and fitting a wbo2 before the turbo goes on)

I tried it again today now that the air bypass screw has been adjusted and sure enough it now idles fine, if anything too high. I took a log of it doing what it does hopping one of you may see something familiar here is a screen shot of that



Any of that jump out at anyone?

That was running at idle, then 1.3 throttle untill it hit 3500 and went nuts, held that for 2 seconds and released.

akaryrye

iTrader: (5)

Your tach signal has noise for some reason, as evidenced by the RPM signal dropping suddenly. Might want to take another look at your RPM input circuit and make sure everything is how it is supposed to be.

Could you please post the actual datalog and your MSQ up for us? That would be a lot more helpful, but it looks like your RPM dropouts are causing this. I would like to take a look at your msq to check out the settings.

Fireindc

iTrader: (13)

When my ms would spike to 17k rpm, it was because of communication resets.. not an actual ground problem. So keep that in mind too.

Blaize

Right, I got some time to play with it in the car again today and I am afraid I am no closer. I had soldered all the ground pins on the db37 (not 3-6) together when trying to solve another problem before, and I undid that hopping it may be the cause. No dice. I have also rechecked my wiring for the tach wiring and it all looks good. I have the capacitor on the board as per Brainiac to clean up the signal, but of course there is no way to test the capacitor to see if it is bad or not. I suppose as a last resort I could just replace it? Also regarding the cas wiring, the outer ground casing is only grounded at the db37 end (pin2) is that right? I didn't see anything anywhere saying to ground the other end but perhaps I missed something?

And tell me how to post a file like that (the msq) and I will, but otherwise should I pm it to you?

Blaize

Or to be a bit more blunt, What sort of things have commonly caused this sort of problem in the past? where should I be looking first? I have the 1 foot wiring bundle from diyautotune all taped together nice and neat, does this cause inductive coupling into the cas or ig. input wiring? Should I try a different value capacitor filter on the board from js8 to ground?

TrickerZ

iTrader: (6)

check your battery voltage in MT, run a ground or 2 directly from MS to the head, check all wiring and connections for your CAS. if you have or can get an o-scope, check the signal from the CAS

Blaize

well Gents I have been ignoring the problem for a couple of weeks as it had me pretty pissed and frankly stumped (not that it takes much as the thread title should indicate) but I have had something of an epiphany. I had a dig thru my MS parts box and found the bag the capacitor for the tach signal came in and turns out it is a 1.0uf cap, not .1uf as directed in the how to. Been to long since I built it to remember if I ordered it wrong or if they shipped it wrong (ebay seller). So I suspect this is most likely the problem? I will try to see if the local electronics place has anything in the morning. Pretty sure they don't or I wouldn't have had to order it in the first place. On that note anyone know how to do the math if I end up stringing a couple along to make what I need? I assume it is not as easy as it is with resisters? is it even possible?

TrickerZ

iTrader: (6)

Capacitors in Parallel

akaryrye

iTrader: (5)

change it out and see what happens. Here in the US, a .1uf cap can be found at radioshack, but I dont know what sort of stores you guys have over there.

Blaize

That fixed it!! It now runs as it should. The idle is not as smooth as it was with the wrong cap in there. I remember from my days as a pro radio/alarm installer that these non-polarized caps are used as filters for tweeters. The way they work is to cut off all signal below a given Mhz, and it works in a bit of a slope. That is to say the closer you get to the cutoff the more it cuts signal. I presume that as the cap I had was killing it at 3.8k, it was "cleaning" the signal a fair bit at idle. I am half tempted to play around with it and find a happy medium. Find one that cuts at say 12k or so, where it will never really matter but still clean things up at Idle. If it didn't require full tear down and solder for every change I would be more tempted.

At any rate it is now running and I am off to re read the tuning bits of the manual. Thanks for all the help and ideas.

And for any brits following along they sell a "combo" pack of caps at Maplins that has a couple of the one you need for £3.50.

Actually akaryrye, I am an american, from New Orleans. My wife works for the BBC so we are over here for the moment. At the rate this is moving I will just about get this car on the turbo and working properly and then we will have to move back and I'll be taking it all back off again to sell the car.

akaryrye

iTrader: (5)

friggin glad you finally figured it out. Nobody could have helped you out on that one but you. Next time you have a problem that you suspect is the ECU, make sure to check it out completely (especially the mods) before making a thread to figure it out.