bshusted

iTrader: (1)

I have been searching for help on this problem for about a week now and have come up with very little, so I need to ask some questions.

First off, the car: This is a rebuilt, streetported S5 N/A engine in an FB chassis. Stock injectors, ignition, and fuel management. All emissions have been removed with the BAC retained.

I have put about 250 miles on the engine since startup. I know I need to put more on it. I have no starting issues. Turn the key with no throttle input and it will start right up. When it is running, the idle holds steady near 1500rpm and sometimes climbs to 1700 when warm. The idle adjust at the top of the manifold calms it down some, but even with the screw all of the way in I still see the 1500. I can make it idle higher by loosening the screw.

I stated that I retained the BAC because I wanted this engine to idle nicely. At idle if I pinch the hose that feeds the BAC from the big black intake hose, the engine speed will decrease. So even with the trottle plates all of the way closed, the engine is getting its air though the BAC. Unplugging the BAC, I can make it idle at 200rpm by turing the screw on the top of the manifold all of the way out. The trottle cable is loose and is not holding the throttle plates open. A visual inspection of the TB verifies that the plates are all of the way shut.

When the engine is running and idling high, I cannot hear the BAC clicking as it should. It is not stuck because I can apply power to make it click open, and unplugging it kills the engine speed, so it must be shutting. Possibly the ECU is sending it a 100% duty cycle for some reason? I need to dig up my oscilloscope and check the signal from the computer. According to this, it should be a 125Hz square wave.

I have adjusted the TPS and am confident in my adjustment. I am going to attempt HAILERS method when I get my buddy over to veryify the settings.

Since this is not an FC I have wired in my own check engine light to check for ECU codes. As of right now, there are none.

In my searching, several people reccomended to look for vacuum leaks with engines running rich, but wouldn't the leak cause a lean condition? I have searched for leaks with the propane method and found nothing. I learned about another way by pressurizing the intake and will probably be making one of these and checking again.

The engine is definitly running rich. The exhaust smells heavily of gas (burns your eyes) and I figure I'm getting near 8mpg. Fuel pressure is ~37psi.

I am at a loss on this one and any guidance/help will be appreciated.

bshusted

iTrader: (1)

So I have been thinking more about what would keep the ecu out of idle mode and on a whim tested the AFM as per the FSM procedure. It turns out that the closed resistance is higher than it is supposed to be. As I push it in, the resistance spikes near 4kohms and then drops to 160ohms when all of the way open. I wish I had a known good one to swap in and try, but I guess I'll start looking in the classified section.

j9fd3s

iTrader: (3)

did you check it for codes? there are several that wont turn the light on, but the ecu will see and store.

running rich + 8mpg = no o2 sensor?

bshusted

iTrader: (1)

No ECU codes. The light is off while running, and when I ground the connector and turn the key on I just get a solid light (no blinking), which I assume is because when the engine is not running and the key is in the run position, all of the idiot lights come on.

From what I understand, the O2 sensor does nothing with the idle. It is only there to adjust air/fuel when cruising. At any rate, I replaced it and it seemed to help with the overall mixture (don't get backfiring every time I let off the throttle now). The mixture is still rich. I can still smell it in the exhaust and if I apply a vacuum leak to the engine, it surges because of the mixture leaning out a bit.

bshusted

iTrader: (1)

So, there has been a new development. I reinstalled the AFM today and tried to start it. immediately flooded. I went though the whole de-flood procedure and tried again. Still nothing. I grounded the set connector and lone behold I had an error code. I believe it is #2 (two short pulses) but I am not very sure what a "long" pulse or a "short" pulse are supposed to be. I took a short video if you guys would take a look and confirm my finding.

<embed width="448" height="361" type="application/x-shockwave-flash" wmode="transparent" src="http://i254.photobucket.com/player.swf?file=http://vid254.photobucket.com/albums/hh91/bshusted/S1051673.flv">

From the FSM, code #2 is the Ne signal from the CAS. What would reinstalling the AFM have to do with the CAS? I tested the pickup coils on the CAS and everything checked out. This really gets me. The car goes from running rich and idling high to nothing at all. I am definitely getting fuel as I fouled several plugs. The spark seems very weak, which makes sense I suppose if the ECU doesn't know where the timing is. So now I have the problem of getting it back to running before I can continue to diagnose the idle / mixture problem.

gsl-se addict

iTrader: (3)

Yep. Looks like code #2 to me. A short pulse is 0.4s and a long pulse is 1.2s. For tracking down code 2, see page F1-22 of the FSM.

For the idle speed issue, you probably need to check out the Idle Speed Control system (pg. F1-40). It sounds like the ECU is sending the BAC too far open for some reason. You should be able to use a regular multimeter and see what the average voltage is at the BAC to get an idea of the duty cycle.

No idea why the AFM work would have anything to do with the CAS. Although the CAS error code wouldn't show until after cranking, so if you checked the codes before without cranking, maybe you missed this code previously??

GL.

Kent

bshusted

iTrader: (1)

I've checked the CAS as per the FSM. The pickup coils check out fine and I've also checked for continuity to the ECU. Everything seems to check out. According to the stuff that I read, the CAS trouble codes kills spark, so that is mu non start issue, but why all of a sudden do the CAS and ECU get upset with each other?

Anyone know how long the computer can remember stuff? For instance, I was thinking that maybe if I just unplugged it for a little while and let it forget everything about this engine, it would be happy again later? I'm running out of ideas aside of buying a bunch of replacement parts.

j9fd3s

iTrader: (3)

it remembers codes for about 30 seconds, other than that there is no learning feature, if you give it the right inputs, it will output correctly

bshusted

iTrader: (1)

I've checked for spark on all 4 plugs. Every one of them has a good spark. This leaves my problem in timing (because of the CAS trouble code), fuel, or air. Is there anything else in the CAS to go bad besides the 2 coils? Every once in a while when cranking it will fire. I had a thought that maybe if the AFM was screwed up, the engine may not be getting enough air. I gave it a leak before the throttle plates to test the idea. No dice.

bshusted

iTrader: (1)

I don't know what changed, but we're back up and running. The ECU is still throwing code number 02 but is also throwing 01. The idle is just as high as before.

I have a question about the TPS. While it is "idling" near 1600, if I push in the sensor for the narrow range the rest of the way until it bottoms out, the rpms will drop to 1100. but it will surge. It was getting really dark so I couldn't take a reading of the voltage across the BAC, but if I had to hazard a guess I would say it is wide open. I couldn't hear it clicking and it is still the primary air source for the engine because if I pinch its hose, the engine dies.

What sensors put the ECU into idle mode? AFM, TPS, water temp?

scrip7

iTrader: (6)

Take a look at the coolant temp sensor volts at startup and after it's warmed up. If you can, check this at the ECU by backprobing. Compare the volts to specs from the FSM. You can also use the resistance method. This sensor has been known to "skew" it's values over time. Also are you using the stock fuel pump resistor/ relay? Check volts at the fuel pump connector at idle. Should be around 9 volts (I think) at idle and should go to around 12V under hard accel. Also disconnect the vacuum hose to the FPR and see if there's fuel dripping. This would indicate a blown FPR diaphragm. On a similar note, make sure there's a restrictor pill in the vacuum hose to the boost sensor. It likely isn't causing your symptoms but I figured I'd mention it since you're doing a swap.

bshusted

iTrader: (1)

Fuel pressure is good with no leakage at the diaphragm. I have heard from several sources that the water temp sensor on the water pump may be the culprit. I'll test that when I get back from work today.

What is the restrictor pill you're talking about for the pressure sensor? I have never heard of such a thing. Right now, I just have the pressure sensor plumbed into the bottom nipple on the throttle body. Perhaps I should move it and or add this pill?

scrip7

iTrader: (6)

The restrictor pill is a small plastic insert that goes inside the vacuum hose to the boost sensor. One of the reasons it's there is to limit pressure/vacuum spikes that might cause a sag or stumble on accel. It has a .5mm orifice. You could make one or get one here:
http://www.mazdatrix.com/c-5.htm

j9fd3s

iTrader: (3)

there is something up with the CAS circuit. if you have a spare CAS id swap it in.

time to start looking at this more logically. its throwing CAS codes, so fix those first. then move on

scrip7

iTrader: (6)

Just to be sure, are the code flashes you are seeing for code 2 actually 2 short flashes together or is it one long flash and one short?

EDIT:

Nevermind, just reviewed and saw that you stated 2 short flashes.

bshusted

iTrader: (1)

I'm beginning to think that the ECU is what is failing and not the CAS. It's throwing 01 and 02, yet it runs. 01 is the trailing ignition, which is firing, and 02 being the CAS, which seems to work fine as the car runs. I think my first try is to find another ECU. Anyone got an NA S5 ECU they want to part with? As I understand it, any NA ECU will work N350, 351...correct?

scrip7

iTrader: (6)

The N350 is a California emissions ECU. It will work, but run like crap like it did in mine. I bought my car in Cali and drove it to Oklahoma and it ran like hammered **** until I put a N351 in it. I got mine from a eGay seller.

bshusted

iTrader: (1)

The 350 is what I have right now. I wonder if that is my problem. I'll start looking for a 351. Thanks for all of the input.

bshusted

iTrader: (1)

So I just received and installed my N351 ECU. No changes. With the BAC plugged in, the car still "idles" near 2k and I'm still throwing the CAS 02 code. WTF? I don't understand how anything can be wrong with the CAS. Both pickup coils test out fine and the wires to the ECU check out fine. I even tested and cleaned the plug in the engine bay for the CAS. Still no dice.

Will the ECU throw the CAS code if the timing is off?

On another note, I have my TPS adjusted to spec, but there is still a little bit of travel left in the variable resisters. If I push both in the remainder of the way, the idle calms down a little bit, but not to where it should be. Is this normal?

bshusted

iTrader: (1)

I pulled out the CAS spun it by hand with my multimeter connected. This is a crude method, but I could not find my O'scope. I was reading several hundred mV while turning by hand on each pickup. My conclusion: CAS is good. Where does this leave me for tracking down my problem?

j9fd3s

iTrader: (3)

the TPS plungers are not all the way in when adjusted correctly.

ECU has no way of knowing that the timing is off, so no it will not throw a code.

how is the CAS connector, i had a bad one once, car would just not run...

i forget, any vacuum leaks? are all the setting on the TB correct? people like to mess with stuff that you shouldnt on those.

how are the grounds? particularly the OMP grounds.

BAC valve work properly?

bshusted

iTrader: (1)

I have cleaned and tested the plug for the CAS. Everything seems good in that aspect and the car runs so whatever that means...

I never did get around to pressure testing the manifold to look for vacuum leaks, but I am fairly confident that I will not find any. The major problem is that the ECU is setting the BAC's duty cycle to 100%. I know that it is not stuck open because the average voltage across the plug is 12v and if unplugged, the engine speed decreases.

The only things I've messed with on the throttle body are the idle speed screw and the BAC adjustment. I did remove the fast idle cam but I don't see how that could be effecting my problem now. Is there anything left to be screwed up after that?

As for grounds, I've got one run straight to the alternator, and another to one of the bolts holding on the starter. Should there be more than this? The OMP grounds through the harness, correct? If so, then yes those grounds are good; if not, then where is it supposed to be grounded?

Anyone know where the grounds for the CAS shielding are? Everything about the CAS and it's harness checks out. I am wondering if some of the shielding got messed up there is enough interference that the ECU gets upset. I may just try running a new, shielded wire to see if it fixes my problem.

Thanks for the reply.

scrip7

iTrader: (6)

There is a ground wire that bolts to the top of the trans bell housing and attaches to a tab on the firewall. Make sure it is present and a good clean ground. There is another important ground at a bolt near the bottom of the driver side strut tower. Make sure if you run new CAS wires that you only ground the shield at one end.

bshusted

iTrader: (1)

Well, I ran new wires for the CAS and it made no difference. I still get the same #02 trouble code. What else can it be? Maybe I'll be looking for a CAS.

Since the car this is in is not an FC, some of the grounds you talk about don't really exist, though I do have the tranny grounded though the starter bolt and there is a ground on the driver's strut tower.