Please help me prove ECU capacitors affect our idle.

[PhoenixRPR]

Note: The communication below comes from my father, who is ASE certified and a lifelong electrician. If any of you can help me prove to him the caps affect the idle circuit, I'd greatly appreciate it. And yes, I have read the 5-6 links around vfaq on the subject.

Under normal circumstances he capacitors should NEVER leak. Not 7 years not 70 years. Electrolytic caps (in this case used as p.s. filter caps) are more likely to explode upon receiving over capacity. This could be from a faulty alternator proving in excess of 17 volts for an extended period of time. However the circuit is diode protected. In 10 years in the auto parts business I have found that the computer is rarely the problem. It is however the easy choice part of repair for shops and dealers who want the quick answer and the big sale. There is a place in northern California called python injection. They sell reman computers and fuel injectors. The techs there can hook up the computer and tell via computer printout if any circuits are bad. I used them often when I was in the business. This saved the customers hundreds if not thousands of dollars.

If you can show me by schematic diagram that the capacitors in question affect the idle circuit if bad and show me that they are indeed bad. Then I can easily replace the caps for you. Replacing the caps is within my technical parameters and is an easy job for me if u hand me the computer. However I will not do that unless I am shown 100% proof positive that they are bad and that they affect the idle circuit. I must see a schematic diagram that proves that fact. Not conjecture by eagle talon owners.

 

[Defiant]

Jeeze, your dad sounds like he's all heart.

I agree that they shouldn't leak and fail. But they do.

Screw it, hire someone to swap them out.

 

[PhoenixRPR]

Hahaha, well he means well and he's helped me out several times with the car. It's just difficult to fix certain problems on the car while still taking a full course load at my university and working. Defiant, do you know where a schematic diagram can be found?

 

[husterka]

Please see the link below for a possible cause for the leaky capacitors. This leak deals with PC motherboards but the same may have happened to the capacitors used throughout the DSM world. I have also seen other articles where faulty aluminum was to blame because the dielectric inside the cap actually corroded the aluminum.

http://www.spectrum.ieee.org/WEBONLY/resource/feb03/ncap.html

I do not have specific circuit schematic evidence for what you are asking for but, being an electrical engineer, any change in a circuit can effect any and / or all systems on the PCB. Even if the idling circuit is not directly linked to the faulty caps, it may have an indirect link due to changes in the circuit parameters such as the power supply stability and voltage ripple. This could cause faulty operation of the central processor and / or any other control sub-circuits.

Hope this helps...

 

[Enigma_Man]

Capacitors can _without a doubt_ leak. I have seen it with my own eyes both on DSMs, and motherboards (I write BIOSs for computers, and boards come in/out all day long, I'd say about 5% of them have at least 1 popped/leaky cap, which doesn't represent at all the percentage of caps that leak, because I only get the mobos in that don't work, even though they should, so it's usually something like a burnt cap or PEBCAK)

As far as the problems this causes: the stuff that leaks out can damage traces on the board, the capacitors themselves do not function correctly when their juice leaks, and they are important for filtering of power for the ECU. A digital device without stable power input can do some whacky things (also I know from experience, I've built many a circuit with weird operation that went away when I added more/better filter caps).

-Jesse <- Electrical Engineer

 

[PhoenixRPR]

I definitely understand and agree that capacitors leak. I just would like a schematic diagram of the ECU's board. Also, my interior doesn't smell like seafood and I don't hear a clicking sound from the dash, are those two indicators necessarily a must as a sign of the leaky caps?

 

[TalonTSi-AWD91]

Originally posted by PhoenixRPR
I definitely understand and agree that capacitors leak. I just would like a schematic diagram of the ECU's board. Also, my interior doesn't smell like seafood and I don't hear a clicking sound from the dash, are those two indicators necessarily a must as a sign of the leaky caps?

Nope, mine was the same, didn't smell nothing and there was no clicking but it was bad. Reason I knew mine was bad was my car would idle like crap, every time I hit around 5-5,500rpms at WOT my car would hit fuel cut for absolutely no reason, and when I datalogged my car every value would keep on fluxuating a bit. The main one I noticed when dataloggging was my TPS would bounce around from 8% to 22% while I was idling. I got a volt meter and checked the TPS line right at the ECU and the voltage stayed perfectly steady at .48 volts so I knew it was the ECU making everything fluxuate so much. I had already tried replacing my caps but it was too late, got a new ECU and the cars been running perfect.

 

[Thunderskull]

Yeah, I had capacitors go bad on 2 different ECUs. The first one I didn't smell anything though there was that annoying clicking sound. On the 2nd one it made my whole car smell like a wrecked tuna boat. Both times it affected my idle. It actually resulted in my ISC going bad which destroyed my 3rd ECU. I wound up replacing it and the ECU and have had no problems ever since.

 

[PhoenixRPR]

Well, as of right now, my car seems to idle at around 550-650 (too low) and it almost stalls at idle every minute or so. But if I give it gas then let off quickly it will stall every single time. So it has nothing to do with me being in motion, so i suppose the reed switch "speed sensor" idea is out the door. Also, where did you guys pickup new ECU's and for how much? Thanks.

 

[laserblue]

•I repair TVs and VCRs for a living [a dying trade]. In SOME circuits, electrolytic caps "dry out" from age or constant usage, even when the circuit is designed properly [not exceeding the limits of the cap {voltage, etc.}]. This is normal after about 5~10 years... maybe sooner. When they become dry, they run hot from not working correctly [capacity value goes down] and sometimes leak as they blister and bulge and crack, usually at the bottom [where the wire leads come out]. Other times they look fine, but just don't do their job anymore!
I constantly replace caps in VCR power supplies. They go bad. They go bad. They go bad. They go bad. They go bad. They go bad. . . Get the drift?
•Now, whether they affect the idle??????? It depends what circuit the crapped-out cap is in!
•I use a special meter to test them, called an ESR, which applies a high frequency voltage to them. The better that this frequency is 'passed' thru the cap, the higher the meter needle reads and the better the cap is working [Electricity 101].

....Did I happen to mention that THEY GO BAD?

 

[PhoenixRPR]

Originally posted by Thunderskull
Yeah, I had capacitors go bad on 2 different ECUs. The first one I didn't smell anything though there was that annoying clicking sound. On the 2nd one it made my whole car smell like a wrecked tuna boat. Both times it affected my idle. It actually resulted in my ISC going bad which destroyed my 3rd ECU. I wound up replacing it and the ECU and have had no problems ever since.

Well, my dad swapped the capacitors with me. The car still won't idle-up when the A/C is on. And since the IAC is brand new and good, only other thing it can be is the ECU (only other device on that circuit). So, suggestions for reman ECUs?

 

[Jon91TSi]

sorry to hi-jack your thread, but I had a similar question in another thread and some of the people here may be able to answer it..

it goes as follows:

"Hey all..

My car has been acting up a bit lately, and I thought this was all due to an oxygen sensor but I guess not.. Anyway, when the car sits and idles, it has a slight miss and nothing shows up on my a/f gauge, when we hooked the datalogger up it showed .02 volts from the o2 at idle.. I replaced the o2 and im still having the same problem.. nothing at idle.. and while driving down the highway the af gauge flops around like normal untill you let off the gas, then it pegs rich for a second or 2 and then goes off the chart into lean and wont light back up untill you hit the gas.. I replaced my o2 with a DENSO sensor, best one u can get and im stumped.. my mechanic said he thinks my capacitors on my ecu may be bad? please help me, this is driving me nuts!"

So could anyone explain to me why my a/f gauge shows nothing at idle and shows full rich when the gas is released and stays there for a second or 2 then drops off the chart and the car then begins to deaccelerate.. My mechanic thinks its the ecu, the data logger showed .02 volts at idle coming from the o2 sensor, and I just replaced it with the DENSO one as stated above and it still acts the same.. anyone??

thanks!

 

[Thunderskull]

Pegging rich for a second or two when you let off the gas is not normal at all. When you let off the gas, the car goes superlean for the sake of fuel economy which is normal. In your situation, the o2 sensor wouldn't be the culprit. Check for a vaccum leak. However, you might want to give the ECU a looksee though. Your mechanic might be right. It's odd for your ecu to order the injectors to run rich for a few seconds after decelleration begins. There's an easy test for the ECU. Pull it out, open it up, and look to see if the capacitors are leaking. If they're leaking you should smell a rotton seafood smell. You may not smell it until you actually open the ecu up. I usually open it up outside since the last time I popped a bad ECU in my kitchen, my whole house smelled like a shithouse on a tunaboat for at least 2 days.

I hope this helps.

 

[Jon91TSi]

Could I even still have a vac leak when my car is showing 15-19in at idle and 25 when i let off going down the highway... The top end was just rebuilt and my mechanic really knows his stuff.. How could I test for a vaccum leak? also why would my a/f gauge and datalogger be showing no voltage coming from the o2 at idle or when i let off the gas on the highway for long periods of time.. the only time it shows anything is when i rev it or am crusing or getting on it..

 

[Jon91TSi]

Oh, BTW while crusing and under WOT the car runs fine.. its just odd im getting nothing from my o2 at idle or when decelleration begins(after the first 2 or so seconds of going full rich)

 

[Jon91TSi]

where are the the elctrical gurus? Someone has to know why my car and the cars in this thread are acting up like they are

 

[Thunderskull]

It sounds like your problem is with your ECU. I see in your profile that you're running stock. At idle, your a/f gauge should be fluctuating up and down. It shouldn't be stuck at .2. It should go way down when you decelerate, but it should not fluctuate rich just before hand. If you have a vaccum gauge, and it's reading normal then you don't have a vaccum leak. Your problem sounds like an ECU one. Get it out of your car and confirm it then if it's bad get a new one. I'm not sure what else you're waiting for. Your plugs, wires, coil pack or power transistor could be causing your miss, but it wouldn't account for the wierd o2 readings you're getting. How's your timing? Also, how's your gas mileage?


Phoenixrpr, go on ebay for a rebuilt ecu. It'll cost a lot less than SATAN. I think autozone has em for $250 but you can get cheaper on ebay.

 

[Jon91TSi]

Originally posted by Thunderskull
It sounds like your problem is with your ECU. I see in your profile that you're running stock. At idle, your a/f gauge should be fluctuating up and down. It shouldn't be stuck at .2. It should go way down when you decelerate, but it should not fluctuate rich just before hand. If you have a vaccum gauge, and it's reading normal then you don't have a vaccum leak. Your problem sounds like an ECU one. Get it out of your car and confirm it then if it's bad get a new one. I'm not sure what else you're waiting for. Your plugs, wires, coil pack or power transistor could be causing your miss, but it wouldn't account for the wierd o2 readings you're getting. How's your timing? Also, how's your gas mileage?


Phoenixrpr, go on ebay for a rebuilt ecu. It'll cost a lot less than SATAN. I think autozone has em for $250 but you can get cheaper on ebay.

Hey..

Thanks for the input.. at idle my a/f gauge shows nothing, not even 1 bar, the data logger also shows .02 volts at idle from my o2, when i decellerate it pegs rich for a second and then drops off, not to come back and read anything untill i hit the gas again.. My vac gauge shows about 18in hg at idle.. It just had a new top end installed, new head, and everything associated with it.. It has new ngk plugs and magnecor wires.. Im thinking the miss is because im getting incorrect a/f readings and it doesnt know how to compinsate for it so the idle is in turn shitty. The timing should be dead on, my mechanic like I said before really knows hi stuff.. My gas mileage is around 17 mpg... Something is not right.. I bought capacitors and will have them replaced, if that doesnt help where can I buy another ECU and does it have to be for a 91 with every option like mine?

thanks again!!

 

[Thunderskull]

When you pull the ECU and check it, see if there are any burnt spots around the leaking capacitors. There's also a riser board right next to where the capacitors are. Check to see if those have any burnt spots on them. If they do, then you will need a new ECU. It's also possible that once you replace the caps, you will still have the same problem. Usually when you start to see symptoms of bad caps, your ecu has already suffered some damage. Check ebay for an ecu. You can get one cheap there. Also places like autozone will sell refurbished ones for around $250. When shopping for an ecu, try to match the part numbers. The MD number is the one you want to match exactly. If your car is non-california, then you will have to get a non-california ecu since it won't throw a CEL when it finds you don't have an egt temp sensor. You can use an ECU from a 91-93 5speed awd. A 90 has different pinouts and while it can be adapted, a lot of folks have reported problems with their tachometer.

I hope this helps.

 

[Jon91TSi]

So if I get a new ECU it should be for a 91-93 tsi awd.. does it matter though since my car has all the factory options (abs, security) and such? Or will any 91-93 5spd awd ecu do?

Thanks!!

This morning when i fired my car up and it was cold and the a/f gauge reading at idle like it should and as soon as it hit open loop when it was warm, the idle fell and the a/f gauge stoped reading anything.. what is this a sign of? Im thinking my ECU really is bad

thanks

Jon

 

[Thunderskull]

If you look on your ECU, you will see a number preceded by MD. You need to match those numbers to your ECU in order to insure that it works properly. If you can't find that, then any 91-93 ECU from a TSI 5 speed with AWD should work fine. ABS shouldn't matter since it has it's own seperate computer that handles that.

I hope that helps.

 

[Jon91TSi]

Hey all, the problem is fixed!! The thing that was messing the car up was my line from my IM to my BOV wasnt hooked up, so I had a vac leak.. I hooked it back up and the car idles fine and the a/f gauge flops around all the time like it should!! Im so happy my car runs perfect now!!

Thanks everyone!

 

[Thunderskull]

Right on. Glad it wasn't the ECU. That was the only thing I could think of if it wasn't a vac leak. I'm glad you fixed your problem.

 

[Jon91TSi]

Thanks again, I really appreciate your help.. Im gonna have the ecu capacitors put in anyway as preventative maintinance.. But my car idles as smooth as glass and runs perfect now..

thanks so much!!

Jon Knight

 

[PhoenixRPR]

Wow, I'm glad to hear that your car is fixed now. Can you tell me, or maybe even show me with a picture, of the vacuum line you had to connect.