ignition coil issues

[laychooba]

hey guys (sorry, it's kinda long, but wanted to explain in detail)

been having this reoccurring problem as of recently. when i tied together
my swap, there wasn't anywhere stealthy to hide the coil so i bolted it to
the rear of the block. one day i smelt something funky while in traffic (car
running kinda hot), pulled over and checked under the hood there was a
black waxy substance that was splattered all over my bay.

at first i thought it was cv grease, but it was thick and solid almost. eventually
the car died and didn't want to run anymore. when i got under the car i saw that
the coil had like expanded and was leaking out whatever is inside it. so i got a
new/used coil, wrapped in heat resistant shields and whatnot and tried again.
this time it lasted longer, but the same thing happened again.

so now i figure it can't be there because the heat from the manifold and downpipe
are killing it so i bought an extra long coil cable and passed it inside the cabin. i mounted
the coil under the glove box, terminated the cables, hooked it up, all good. it can't get
any hotter than cabin temperature right?

this weekend i took a 5 hour drive to toronto (with no problems on the way there,
or while there) on the way back, i got stuck in dirty traffic for like 45 mins. i could
hear like an electrical static/interference coming through the speakers of the car, by now
the needle for water temp getting a notch over halfway. then i started to smell the coil burning again. i pull over to the side of the road, pull down under dash carpet and there's the coil
leaking again. i did make it home, but i gotta source this problem.

as it is, my car is ridiculously not stock anymore, so it's hard to track this down.

one thing was since the coil is mounted to plastic (rear of the box) and is missing a
ground strap like OE, could that cause a problem? it's ridiculously hot, like it burned
me hot - you can't touch it.

the wires to it are the green coil wire from the digi computer on the (-), and the other is
for the fuel pump off the OE fusebox (with the tach wire spliced in)(+)

anyone have any insight?

[20v_boost]

Sounds like it could be too much dwell. Is that the coil from a a digi car?

-Alex

[laychooba]

the first one i melted was an OE passat 16v one.
the second one was a G60 8v one.
this one is a replacement bosch i bought from an auto
supplier here. looked basically the same, except it's zinc'd
on the outside, not black.

i thought they were all the same more or less?

[20v_boost]

I would think that they are more or less the same. I know the Scirocco 16v coil and the 8v coil have different resistance values in the bently.

I just can't see how the digi ecu would be suddenly applying too much dwell. I run about 5ms on my 16v (controlled through megasquirt).

Can you measure the dwell on the coil?

-Alex

[laychooba]

for sure no one's ever had this problem before, this is my
specialty. baffling issues!

i assume i need a dwell meter to measure that, i'll have to
source one out. it's weird though, affected by the car sitting in
traffic, but if rolling, it's fine. i drove 10 hours, and in city driving
no problems, but 30-45 mins of stop and go killed it.

[20v_boost]

That is strange. I just can't imagine how the ECU would fail in this way. Usually when electronic things fail, they just stop working. It really does seem like too much electrical energy is being put into the coil though. The fact that it doesn't burn up at higher RPM fits with the too much dwell theory. At higher RPM, there is less time between sparks so the maximum dwell you can run will be less. At lower RPM the coil has more time between sparks to heat up.

OK hold up, I just looked at my Bently. I'm looking at the wiring diagrams for a 1990 cabriolet which is digi (I or II?). Here, the Green/Black wire from the DIGI ecu connects to the Ignition Control Unit. The ICU drives the coil (and probably controls the dwell). You have your coil connected straight to the DIGI ecu? Are you sure that's right?

-Alex

[laychooba]

hahaha i'm sure that's NOT right. hahah

i have it hooked directly from the ECU to the coil, with no ignition control
unit. all the digi 1 swaps i read had people hooking it up straight to coil. maybe
you HAVE to run the ICU? because now it's hooked up straight and the car
runs fine?

[laychooba]

okay, so i sniffed around today. from what i understand, corrado digi-1 doesn't
have an ignition control module. it's hooked straight from the cpu to the coil.

i drove around today, with it hanging on the passenger side floor. drove for
about 45 mins/hour and it got really hot. no performance issues with the
car though.

when i got home, i checked it with an infa-red thermometer. the upper most
part of the coil (near the terminals) was around 107-115 degrees and the lower
half of it was in between 125-155 degrees. this thing is COOKING.

[20v_boost]

The only way it can get that hot is if you're putting more energy into it than you're taking out. I'd measure the dwell and make sure it's where it needs to be, or get the right coil, or rig a resistor in series with the coil.

Or just forget everything and go full megasquirt!

[MEGA 16v GLI]

The only way it can get that hot is if you're putting more energy into it than you're taking out. I'd measure the dwell and make sure it's where it needs to be, or get the right coil, or rig a resistor in series with the coil.

Or just forget everything and go full megasquirt!

[Prof315]

The only way it can get that hot is if you're putting more energy into it than you're taking out. I'd measure the dwell and make sure it's where it needs to be, or get the right coil, or rig a resistor in series with the coil.

Or just forget everything and go full megasquirt!

[laychooba]

yeah! i know!
i f-d' up alright!! hahahaha
digi is gonna be my downfall, i'm calling it right now.

i went to a yard and test like 5 different coils. each one was giving me
different values when i would measure their resistance. the new one i
bought that i was running was like 10,000 ohms? (is that even possible?)
i had another one i got that was 6000 ohms, and then a used one i bought
there was liek 3200 ohms. according to the bentley, the last one was
within spec for a digi-8v G60 setup.

gonna try it out, but ride with a back up coil just incase.

[laychooba]

update!

yeah, didn't really do anything.
still hot as sh*t. i can still drive the car around and stuff, but i
don't trust it. ughhhhhhhh nightmare!

is there specific voltage that's supposed to go to the (+) of the
coil?

[Prof315]

update!

yeah, didn't really do anything.
still hot as sh*t. i can still drive the car around and stuff, but i
don't trust it. ughhhhhhhh nightmare!

is there specific voltage that's supposed to go to the (+) of the
coil?

yeah battery voltage or say 10-16V with 13-14.5 being ideal.

[laychooba]

yeah battery voltage or say 10-16V with 13-14.5 being ideal.

i'm gonna check it when i get home tonight.
i was reading around (not sure if this is VW specifically related) but some cars
have resistors inline for the coil hot wire?

[20v_boost]

You mean the high voltage wires? Yes. Most modern cars do. It reduces electrical noise. The Bentley will tell you how much resistance each plug wire should have. I doubt thats your problem but it's easy to check so might as well.

-Alex


Sent from my iPhone using Tapatalk

[laychooba]

You mean the high voltage wires? Yes. Most modern cars do. It reduces electrical noise. The Bentley will tell you how much resistance each plug wire should have. I doubt thats your problem but it's easy to check so might as well.

-Alex


Sent from my iPhone using Tapatalk

no, i actually meant from the 12v that goes to terminal 15+ on the coil.
i multimetered the 15+ just now when it was idling (car hadn't be run yet, just warming up)
and it was at 13.30v.

i also added a fat gauge ground from the bracket to the body to try that out. i just went
for a buzz around (30-45 mins), didn't really get stuck in any traffic though. it got pretty warm,
but not like anything i was having before. i also noticed a buzzing coming from it.

[rcortez13]

Sounds like it could be too much dwell. Is that the coil from a a digi car?

-Alex

I agree, it sounds like you have to much dwell! Which TT chip are you running? Did you ever get your pinging issue resolved? If your running a new chip maybe TT made some adjustments to the dwell. I would try an old chip and compare.

The only way it can get that hot is if you're putting more energy into it than you're taking out. I'd measure the dwell and make sure it's where it needs to be, or get the right coil, or rig a resistor in series with the coil.

Or just forget everything and go full megasquirt!

The same mistake can be made with Megasquirt! The forums and community support usually make mistakes like this a thing of the past. Good luck!

[laychooba]

i often wondered if this could of been chip related! i did contact colin at TT and
told him about my pinging situation. he said i was running a higher timing program,
so he reprogrammed me another chip and upon installing it, no more pinging.

maybe he did something in the chip's program to tune down the timing and
that's causing the coil to over-stress?

i find the car's powerband is different with this new chip he sent me. the car
almost feels like a turbo. there's no balls under 4500 rpm, and then it takes off
like a rocket. like it weird, when get on it down low, it almost feels like something is
holding it back and then when it hits, it rips.

[20v_boost]

You should have battery voltage at terminal 15 on the coil.

The same mistake can be made with Megasquirt!

Yes sir! MS can do a fine job of frying coils too The nice thing is that if you do run too much dwell it will be your own fault. You have full control over everything.

Adding a resistor to the + (or -) side of the coil would reduce your overheating problem. It would effectively increase the amount of dwell your coil needs to charge. This sounds great until you get up in the RPM. At 6000 RPM there is 5ms between spark events. If you need 20 ms to charge the coil with the addition of a resistor, you won't be able to get a good spark above a certain RPM. FYI I'm running 3.5ms of dwell on the stock 16v coil. (PL engine)

What exactly is dwell? How does an ignition coil work? Gather round now . A coil is an electrical inductor. Meaning it has inductance. Think of it like a capacitor in that it can be charged and discharged. You can fill it up with energy and then it gives it back. There is a finite amount of energy that you can put in the coil (like a bucket can take a finite amount of water). After you fill it up, any extra energy (like water) will over flow. The extra energy escapes in the form of heat. Also, as the coil charges, its resistance goes from very high to zero. So applying voltage to a charged coil is like a applying voltage to a short circuit.

So the trick is to ground the - side of the coil for exactly as long as you need to get a full charge on the coil. This is your dwell time. The spark is formed as soon as you disconnect the ground from the coil. So the computer plans when it wants to create a spark, grounds the coil 3-5 ms before, removes the ground, and voila! Spark.

So, check your dwell or try your stock chip. If you are running way too much dwell, it will be more apparent at idle. The higher the RPM, the less time between sparks, the less dwell you can physically run. So even if you'd like to run 20ms at 6000rpm, you're running less than 5ms. At 3000 RPM, there's 10ms between sparks, etc. Cook your coil at idle.

-Alex

[laychooba]

colin wasn't there when i called TT, but i spoke to another dude there
and explained him my situation. he said that the ECU theory wouldn't
be impossible. i'm going see if i can find the stock eprom and install it
in my spare ecu. if it doesn't do it with the stock chip, then at least we
know the cause.

engine tuning over the phone? NIGHTMARE. hahahahah

[laychooba]

so last night, i changed back to the stock EPROM/cpu, no difference.
i also thought maybe the gauge of wire that goes from the CPU to (-) on
the coil was too small, so i changed it with a big fatty, still no difference.

what should the dwell be on coil? any idea on what kind of resistor i should
try on this? running out of options here!!

[20v_boost]

Well all this dwell stuff is just speculation until you measure it. A fatter wire will only decrease the resistance and increase the voltage and amperage to the coil. If you want to try the resistor rig, try a one of those 10W resistors from radio shack. Maybe 1/4 of your coil's resistance? The resistor will get hot too so keep that in mind. And you may not have a good spark if you choose too high of a resistance.

If your dwell is ok and if you're not putting too much energy into the coil, then maybe not enough is coming out.

What are your plugs gapped to? Are your ignition wires good and to the right resistance spec? Is your distributor phased correctly?

-Alex

[sprocket007]

Did you get this fixed? I saw your message to me...my issue was wrong coil for the car i was running it on. changed it and it was resolved.
your issue sounds more complex than mine, as u have chips, engine swaps and such!