Whisper

GSL-SE...

I got a new TPS, my old one was reading a bit off, so I installed it, readjusted the throttle cable, readjusted the TPS, set everything to spec.

Now the car won't idle above 300 rpm. In fact it's probably more around the 250 RPM mark.

When I start it cold, the fast idle cam takes it to about 2000 and it idles there nicely until it warms up, fast idle cam releases, and then it just dips down and stays there. I opened the idle screw all the way, and that was barely enough to keep it running. If I turn it in at all the idle drops further and the engine dies.

I think the reason it idled ok before, was that I had my throttle cable too tight, and it held the throttle valve slightly open. But now with cable adjusted, it closes all the way.

Now, if I hold the throttle slightly open, to raise the RPM, it misses a lot.

Also my Intake Air Temperature sensor broke off. Well, the wires broke off. It happened when I was testing the BAC control solenoids. I don't know if that's a big deal at all.

I'm kind of fed up with this car. It's like there's some hidden problem somewhere that I can't seem to locate no matter what I do and no matter what parts I replace.

LongDuck

First off, intake air temp is important; fix the connector.

Secondly, the 'idle adjustment' screw that every SE owner thinks adjusts idle, doesn't.

If you look at the TB and run your hand around the side facing the front, you'll see a tiny hex head nut that locks down an even smaller slotted bolt. THIS is the bolt that controls how far the throttle plates close, and thus, idle speed. Fiddling with the slotted screw on the very top of the TB where the rubber air tube enters it is only idle quality adjustment and is a fine air bleed - NOT the idle speed adjuster.

Your missing could be any number of things, including old plugs, a cracked distributor cap, or old ignition wires. Hell, could even be a bad ignitor.

First things first; fix the intake air temp sensor, and start working from there. HTH,

Whisper

Hey, it's LongDuck. I didn't know you were back.

Anyway, I realise the screw is for fine adjustments, but it's the only thing that changes the idle at all, aside from the variable resistor, which is for even finer adjustment of the mixture. I know about the stop screw, but I thought it's not meant to be adjusted ever, Mazda even put a lock nut on it. Anyway, I'll get to that once I have my IAT sensor fixed, although I'm not entirely sure how I'm gonna fix it. The wires broke off right at the sensor, and there's not much wire sicking out to solder to. And of course Mazda doesn't sell these things anymore. Maybe I can find a used one somewhere. I am talking about the one that screws into the back of the dynamic chamber, obviously, not the one that's inside the AFM. Just to be sure we're on the same page.

To address the miss, I've already done everything I could think of. I have new plugs. I even tried 2nd gen coil mod with 2nd gen leading plugs. Just got brand new high performance wires only a few days ago. Rotor and cap are also pretty new and have no wear. I've tested the ignitors, and they appear to check out ok, but I still feel like they're kind of suspect... I'm gonna get a couple of extra ones and try them just for kicks. Injectors have been serviced couple of years ago, and I tested fuel pressure and that looks fine. Also both stock coils' resistance is within spec. I've been messing with this car for a good 10 months now. A lot of time and money went into it. A lot of new parts (including new engine), and yet it's no better than it was when I started.

I feel like I've replaced everything BUT the culprit. Hehe.

I even got a second throttle body, because maybe mine is too worn. Of course it's been in the junk yard for a couple of decades, so it needs to be taken apart and cleaned/restored/rebuilt, and that's gonna be a month or two...

I'm almost at the point where I just want to take my car to Mazda and have them diagnose it. I don't care if they rob me, as long as someone figures it out.

branca

iTrader: (1)

don't take it to mazda, take it to jerry wright in kent

http://www.jerryslittlecarshop.com/

Whisper

I've taken it to Jerry once and the general vibe I got from him was that my car and engine were old and crap and I shouldn't really bother with any of it and he just wasn't very helpful, like he was in a hurry to get rid of me. His technician seems a lot nicer, though. I donno, maybe that's how he is, but I didn't feel like going back there after that.

Whisper

Ok, found a good Intake Air Temp sensor, hooked it up, also replaced coolant thermosensor in the waterpump housing. Still idling rough at 300 RPM. Made no difference.

I think I'm running rich, I just can't figure out why. I pulled the injectors and inspected them, and they seem ok. My fuel pressure seems a bit high by maybe 5 PSI (see my other thread). I'm testing my ECU harness, and everything seems ok. Well, O2 sensor voltage is 0.4V instead of specified 0V, but I don't know if that matters.

What else could it be? Just overall a bad ECU?

Whisper

Ok, ECU looks good on the inside. I don't think it's the problem.

That's it, I guess. I have learned all I could about the FI system, and I have nothing else left to check.

At least nothing that I can think of. I got some spare ignitors, gonna try swapping those in as a final ditch attempt, but I have a feeling that's not gonna do much...

Help me, LongDuck. Give me ideas. Car still idles rough at 300 RPM and misfires a lot in a repeating pattern. If I rev it slowly, the misfires become less frequent as the RPMs climb.

Whisper

Yo, tried new ignitors, and still no difference, didn't really bet on it anyway.

So I did the following test:

I teed off one of the vacuum lines going into dynamic chamber, and after the car warmed up and started idling like crap, I opened that vacuum line and let it leak - as soon as I did that, the idle went up to about 700 RPM, smoothed out a bit, and stopped misfiring. Then I closed the leak and it went back down to 300 and started shaking, misfiring, and running like crap again.

Also I got a narrowband from one of the guys here, and hooked it up, expecting to see some info, but instead it was just stuck in one spot (around center mark) the whole time, even when I revved the engine. So with the engine running I disconnected the O2 sensor, and that made no difference AT ALL.

What does it all mean? Is my O2 sensor not working and causing car to run rich? Or does the car never enter closed loop to begin with?

j9fd3s

iTrader: (3)

sounds like your o2 might be dead. at idle on these cars the ecu is in open loop, its doesnt use the o2 sensor, in fact on a correctly running stock car, the o2 should be pegged lean at idle, due to the port air from the air pump.

remove the air pump and it'll be pegged rich.

the ONLY time it will be in the middle is part throttle cruise (1500-3500rpm)

http://www.jimrothe.com/mazda/84_tra...raining_4b.pdf

Whisper

Ok, so if O2 sensor isn't used at idle, then what IS used at idle? Is Airflow Meter used at idle?

I hope it is, because I've replaced and/or tested everything else.

IAT, coolant sensors, TPS - have all been replaced. Variable resistor tested ok. Hell, even Airflow Meter itself tested seemingly within spec... And all of them test ok at the ECU connector as well.

So theoretically my car should be idling perfectly, but it's not. It dies at normal idle settings. I'm not talking about idling poorly, I'm talking about complete shut down after warm-up. Only way I can keep it alive at all (read: 200-300 RPM) is if I open the idle adjust screw all the way.

j9fd3s

iTrader: (3)

is the engine good? vacuum leaks? another thing that comes to mind is the BAC system, basically the base idle is set to like 300-400 rpm, and the bac brings it UP to 750.

another thing to check might be the leading spark

Whisper

Sparks are good. I tried different coils, wires, ignitors, and plugs. Same result.

Engine is good. Fresh build from Pineapple, maybe 30 miles on it total. Obviously not broken in yet, so compression is going to be uneven, but hardly anything to cause the car to die.

I couldn't find any vacuum leaks. Every single vacuum hose is brand new, all gaskets are brand new, injector seals are brand new. All the hoses are triple and quadruple checked. In fact, when I artificially create a vacuum leak, it idles higher and smoother, so I'm almost positive it's idling rich rather than lean.

Now onto BAC. I can hear solenoids clicking, but I can't tell if they're actually doing anything. I cleaned and oiled my BAC valve several times and checked with a vacuum pump that it closes fully around 15 inches of vacuum, and then opens when vacuum is released... I don't think it's sticking... I also checked all the vacuum lines going to the solenoids, by applying vacuum and then making sure it's released when voltage is applied to the solenoids. Everything appears to work ok.

You say base idle is set to 300-400... What is that? With the throttle valve closed, where does the air for that idle come from if not from BAC? Is BAC supposed to be slightly open at full vacuum? Can you explain this a little better? Perhaps I don't fully understand how idle system works, and that's why I can't diagnose it correctly.

BlackWorksInc

You introduced an unmetered (by the ECM) air leak into your system and this seems to have corrected your issue. Your system is most likely running rich, have you checked the spark plugs for fouling? Does the exhaust smell like gasoline? In regards to the closed loop from that PDF that JD3FS provided it looks like the cars do not enter Closed Loop at idle, but I did read something that piqued my interest.

In the reading it stated that the ECM references the O2 Sensor voltage against a regulated voltage that it recieves to determine what lean and rich looks like, rather than a static voltage number (if I read the PDF right). Where does it get its regulated reference voltage?

j9fd3s

iTrader: (3)

the engine is set to idle @300-400rpms and the BAC introduces a little extra air to make it idle up to 750.

the BAC system is a little early 80's weird, but its pretty simple. the BAC valve is vacuum operated.
and there are two solenoids, the VENT and the VAC.

VAC is vacuum, and VENT is ventilated to atmospheric pressure). so when the turn the VAC solenoid on, it applies vacuum to the BAC valve, which opens, and supplies more air and idle goes up. when they turn on the VENT solenoid, the vacuum is VENTed, the BAC closes and idle drops.

they can switch these solenoids on and off to get a certain amount of vacuum on the BAC valve to get it to open a certain amount.

and now that i think about it, the solenoids have to be hooked up to the correct nipple on the intake, and if it has the filter kit, the filter needs to actually pass air, etc. you need to make sure things are actually doing what they are supposed to VS looking good.

j9fd3s

iTrader: (3)

well it does get a 12v power, and the ECU will generate a 12v and a 5v reference signal, so i suppose its battery voltage into converter/power supply over O2 voltage

i'm a little unclear on this, but if you hook up a voltmeter to the o2 sensor or a gauge, there is a range where the o2 is obviously in closed loop (1500-3000ish it depends on load).

other than that we think its in open loop

Whisper

I think from the airflow meter and possibly IAT. AFAIK it basically takes the input from the airflow meter, takes the voltage from the O2, and then using those two it crunches some numbers to figure out what's going on. I'm not sure if TPS plays into that.

Anyway, I'm definitely running rich. And it doesn't seem like it's the excess fuel that's causing it. I would guess it's the lack of air. ECU dumps in the amount of fuel needed for 800 RPM idle or so, but it can't match it with enough air. That's my best guess... I just don't know why it's happening.

Whisper

Right, I see, so would you say my BAC isn't opening enough? I'll cross-check the vacuum routing to the solenoids one more time, but I'm quite sure it's correct.

Here's what I'd like to know in regards to the BAC. Suppose I'm idling at 800 RPM and everything is correct, if I unplug the power to the solenoids, would my idle drop to that preset 300-400 RPM? Or would my engine die completely, due to lack of bypass air?

Also, as far as I can tell, applying vacuum to the BAC closes it, and venting vacuum opens it.

j9fd3s

iTrader: (3)

yes if you unplug the solenoids idle drops, how much kind of depends on where the idle speed screw is, and the variable resistor, etc etc. sometimes it'll drop and run, sometimes it'll stall.

its been a few years since i played with the BAC. i took some vacuum line and just sucked on the valve, and i kind of recall sucking on it, made it idle faster, but i could be wrong.

Whisper

Other than problem with the valve, is there anything else? Bad electrical signal from ECU?

I think I tried disconnecting the solenoids and it didn't make much difference... Maybe they aren't working right.

Whisper

I just checked all my vacuum hook ups, and it all looks correct. I traced all the ones that control BAC and made sure they go to their proper spots. Some of the lines going to the vac and vent solenoids are hard to blow through one way and easy to blow through the other. I don't know how much effort is required. Solenoids themselves are very easy to blow through, if they're not hooked up to vacuum lines, so maybe there's some restriction in the metal vacuum piping, I'll have to pull it and check, but I don't think it should matter. My lungs don't have the same power as the engine.

Now, to answer my own question that nobody seems to be able to answer, when BAC is fully closed, it still allows some air to pass through. That's probably how it provides the initial 300-400 RPM, before the solenoids start to work. So even with solenoids disconnected, the car should idle.

This raises another question, however. How much air should pass through a fully closed BAC? Blowing through it like that seems to be hard, so maybe mine is plugged up or something...

Whisper

Oh, another thing. What's normal manifold vacuum at idle for a 13B?

j9fd3s

iTrader: (3)

at 700-800 rpm it should be around -18psi

j9fd3s

iTrader: (3)

its important the BAC is solenoids are hooked to the right place, if the VAC solenoid, doesn't see vacuum, its not gonna work right. and the nipples in the intake do different things.

i also have seen clogged vacuum racks, so that is possible too

have you compression tested the engine yet?

Whisper

Yeah, compression is over 100 PSI, and IIRC my vacuum was around 17-18.

Actually I was a bit wrong about the BAC flow, reading over the idle setting procedure one more time, I see that with fuel cap off and BAC solenoids disconnected, the idle is expected to be around 700-800 RPM, before it's adjusted. That means with solenoids disconnected and 18 inches of vacuum to the BAC itself should still flow enough air at idle to keep it at 700-800 RPM.

I think there might be a problem with the BAC itself, because it doesn't seem to flow much air at all with vacuum applied. I'll try to find someone with GSL-SE, to see if I can compare with theirs.

Whisper

I cleaned my BAC the best I could, but it was already pretty clean, so it didn't really make much difference. Plunger moves smoothly without sticking, but still when I close it I can hardly blow any air through the valve and I still don't know if that's normal. I can't find anyone with GSL-SE around these parts, so I can't compare. Do you know of any procedure to test the valve itself?