lakeffect2

First let me state:

I HATE IGNITION MODULES! I HATE THEM WITH AN UNBRIDLED PASSION!


With all the hassles of the stock Gm ignition module as it relates to heat saturation, hot restart issues, intermittent operation... leaving you stranded someplace... has anyone considered moving it out of the distributor and on to someplace cooler and more accessible like the cowl or fender wall? We could add heat sinks in the airstream.

Why do we keep it in inside the distributor cap?

Would the reference pulse from the pickup coil be compromised by extending them to outside the cap housing?

Inquiring (and heat fried torqued off) minds want to know!

RFmaster

That's a good question - a lot of non GM retrofits rely on ignition module (ICM) to generate DRP for ECM. On binder and Jeep boards I have seen these modules mounted on a good size heat sink and located away from the hot engine. The problem is getting reluctor coil signal to this module. With multi kilovolt pulses floating in near proximity to the coil's leads getting them to module is rather tricky. I would try a shielded twisted pair cable to extend output leads and ground the shield to the dizzy body. The key is keeping twisted pair capacitance within reasonable range or otherwise cable capacitance may effect pulse shape. IIRC, I have seem these modules mounted up to 12" away from reluctor coil. Remote module should be mounted on a good size AL heatsink with a good size ground strap between it and coil.

//RF

Ronny

iTrader: (1)

Was therre not a recent post or general knowledge that a special "grease" is imperitive for proper lifespan and operation? Does GM offer a high quality item in that regard?

RBob

iTrader: (1)

Heat sink grease. The white stuff, not the clear dielectric grease. Can be purchased from Radio Shack or other electronics store:

Heat Sink Grease (6g), Catalog #: 276-1372

The ICM is also sensitive to voltage spikes. A bad alternator or the spark arcing down the distributor will also take them out.

RBob.

RFmaster

In my experience heat sink grease is commonly misused - applying too much is actually detrimental to thermal junction resistance. The key is to have both surfaces flat and dirt free. Heatsink compound should be applied with a rubber roller leaving a thin, see through layer of compound on both surfaces. When module is first set in - try lifting it - if it is difficult to lift then there is just enough compound and there are no voids between the two surfaces. Applying a thick bed of thermal compound increase thermal junction resistance between module and distributor body.

Ronny

iTrader: (1)

Occasionally my aftermarket tach will break up. I read this may be module having an intermittant failure. I believe Autometer offers a filter like GM runs on stock application. In fact I was told to disconnect GM filter to see if my stock tach could be saved. It made no difference. My new tach now is showing same issues as stock did.

lakeffect2

I would think this point would to be another excellant reason why it should be moved. less worry about "lightning from within" striking it.

PS, I know about the heat sink grease from Radio shack and use it sparingly.

My issue is this . The engine just flat out dies sometimes I can catch it. sometimes not. It died on the way home from work, a distance of about 3 miles. Running time about 6 minutes total. At about the three minute mark it bucks ,any speed, any gear. Data log showed the ingnition cut out a few data points before everything else shut down.

I thought I found a problem inside the dizzy where the sprung center post comes down. It appeared to be almost covered with plastic. AHAH! no spark getting from the post to the plugs! Nope, the engine is still doing the cut out routine after fixing the post.

The bad news I just replaced the Accel Distributor's ignition triiger coil within the past few weeks, (gaped at .010), and the module twice... amonth ago and last night. (the good news is I can change a module over in about an hour start to finish)

I'm not sure if the coincidence is that the ECM has gone into closed loop by then, or simply that things are heated up by then everywhere and **** happens.

RBob

iTrader: (1)

Why are you bothering with after market distributors and parts?

Seriously, get a GM distributor with a GM ICM and drop it in. The only aftermarket stuff I use on distributors is decent brass terminated caps & rotors, and pickup coils. Bodys, shafts, reluctors, ICMs, and so on are all GM.

RBob.

robertf

not to stir up any normalization of deviance, but half assedly extending the distributor pickup wires and selecting a more convenient than technically correct location to mount the ICM has worked well for me on somewhere just under a dozen TBI conversions on non GM engines with mag pickup distributors. I'm running a pro billet MSD dist on my rover with GM ICM and MSD doesn't even bother to twist or shield the pickup wire.

I'm not saying its right, I know its not, but it has worked.


Only problems I think I've had were failing modules, but they were Wells or Borg Warner.

Six_Shooter

iTrader: (2)

Failing ICMs are NOT caused by being inside the dizzy.

Failing ICMs are NOT caused by being inside the dizzy.

Failing ICMs are NOT caused by being inside the dizzy.

Failing ICMs are NOT caused by being inside the dizzy.

Repeat reading the above until it sinks in.

Millions of cars with ICMs mounted inside the dizzy, without repeated clockw work failure prove the above statement to be true.

That being said, I have mounted a couple GM ICMs outside of the dizzys, only because they were not GM dizzys. These were connected to a Datsun engine.

No heat sink, no twisted or sheilded pair, no ground wires, no problems.

I literally had a drywall screw through one mounting hole into the inner fender in my Datsun, I had a couple wires that I used to extend the OEM reluctor wires, which also were not twisted. Car ran just fine for several months with this set-up. I'm not saying it was right, just saying that I think people over think things sometimes.

Like "capacitance in wires". while on some extremely small scale wires could potentially have capacitance, this is not a quality that is used to verify if a wire or wire combination would be good to use. Resistance on the other hand would be. On that note, you would need an excessivly long pair of wires (I'm speaking of 18 AWG here, which is what most people would use), to cause any sort of signal problem in that regard. So long that you would have to WANT to cause problems.

Twisted pair? Yeah twisted pair will always be better than no twisted pair when it comes to small signals like this, but un-necessary, in most applications.

I would definatly not worry about any sheilding. I have seen many cases where "sheilded wire" has caused more problems than it was trying to solve.

Twisted pair by nature will automatically cancel out any radiated noise being induced by the wire, due to the properties of the twsited pair and distances from that noise source of each conductor.

Sheilding can induce noise, because unlike the twised pair, it does not have an auto-cancelling property, of induced RF noise.

This would be a lot like building a car audio system. In my systems I use non-sheilded Twisted pair, since they have, in my experiance, less susceptability to inducing noise. I have tried some experiments where I should have had noise in systems, due to proximity to noisy components, such as computers, gauge clusters, and power wires, without a hint of noise. Use non-twisted or even shilded twsited pair and I get noise.

I have to agree with RBob, GM makes some pretty stout parts, and would use them over a lot of aftermarket "upgrades", due to reliability.

RFmaster

I do worry...
The outer jacket (aka shield) of the shielded wire must be properly grounded at load side of the circuit. Since distributor housing and remote heatsink for ICM are potentially at different electrical grounds grounding the shield at both ends causes current flow along the shield. This way it will prevent any RF or EMF from being coupled onto signal wires. However, reluctor coil is not a balanced circuit - one side is grounded (internally) and the other is input to a zero crossing detector.
http://www.national.com/ds/LM/LM1815.pdf
However LM1815 has a good noise signal rejection - this is why an unshielded pair can be used (for short runs)

A twisted pair is commonly used for balanced signaling as this prevents common mode spurious signals from effecting transmitted signals. This is done countless times in factory automation and signal conditioning - a quality shielded cable or twinax is commonly used in extremely electrically noisy environments - induction fired furnaces, ark welding work stations (Kuka Robots) transportation systems (Rail systems) and the like.

GM OE ICM when properly mounted in a factory distributor should be just fine. However, elevated temperatures may damage them as well. I willing to put Fluke 51-2 digital thermometer to task and measure operating temperature if someone locally has small hat distributor and willing to dig into this further!

//RF

dimented24x7

iTrader: (2)

In performance applications they are...

Think about it: The coolest the ICM can get is ~190 because its heat sunk to the distributer, and essentially the engine. Thats HOT for electronics. Add lots of load and RPMs, and your talking serious heat build up. The usual life span for me when I ran the engine hard was ~1 month. The power transister basically slowly fails, and eventually causes a no-start. I burned out probably about 15 modules of various makes. GM, accel, BWD, etc. It wasnt the coil (replaced that multiple times), either, or the electrical system. It was HEAT. Nothing would save it. I even used the high $$$ silver based heatsink compound.

With my external vortec dist. and amplifier, I have yet to have a failure. It works flawlessly, and runs MUCH cooler. Usually, the heatsink is cool enough to touch. Its a much better design then the HEI system.

lakeffect2

Thank you all, this has been into a very invigorating discussion.

I went to an aftermarket Accel Distributor unit, as I was frying up the GM stuff as well. The difficulty of "swapping" is that, unlike all of the Third Gen projects you guys are working with, this is a 1993 Caprice wagon. The overhang of the windshield cowl prevents easy changes. The distributor can only be replaced by taking off the intake manifold as one movement , because the shaft CANNOT be lifted clear of the hole. That's essentially a tear down, not a swap

I have a spare ECM so I'm going to start with a simple swap of those to seee if the problem continues. I am running shorty headers. I see pipe temps in the range of 600 F degrees. Maybe someone can comment if this is normal. (correction 07/15/10 - temps are around 700F)

The correlation between heating up and closed loop nags at my thoughts and bothers me to the point that I wonder if I should force open loop operation just to see if the cut out still occurs? If it still cuts out, at least it would SEEM to verifyy the heat related aspect of the problem. Any thoughts to the pro or con?

RFmaster

Lake

IMHO in your instance I would go with remotely mounted ICM. Get a suitable piece of heatsink material (most electronic surplus places, online, etc) to mount ICM securely away from heat source. Start with twisted pair shielded cable with at least 200C rating (most microphone cables are only 65C). Do not modify reluctor coil connector - just use matching crimp-on blades to plug into corresponding coil connector slots and ground shield where ICM used to be. If you use 18 to 24 inch run of 22 AWG (16.46 Ohms/1000 feet) - the resistance increase is negligible!

600F is just enough to fire O2 sensor - that's on a low side!

//RF

robertf

This is the same as the LT1 right? How are you getting the tach signal to the ECM? Last I looked at the later model ICM they basically only did dwell control, no input to the ECM and no input from the pickup.

lakeffect2

I swapped out the ECM this morning but no time to evaluate... left it home while I went to work. Now home for lunch and will let run to evaluate ECM change (* not expecting anything good to happen)

Talked to the local guy who works on our home computers, says he has a boat load of old heat sinks kicking around and to bring the module in to check what might work or could modify.

I'd like to thank all of you for your interest and suggestions... will keep you posted.

robertf

here is one I made for a Mazda Miata ICM. I used a harbor freight indexing table and a 1/4" endmill in a drill press

lakeffect2

It reminds me of the top of the top of a TPI plenum. A while back happened to have purchased an Aluminum dizzy cover one that never got installed.

I think I just found my heat sink material.

Any way, I restarted the motor after ECM change and as expected, It still stalled out. Having messed with having to take the plenum off to get to the module, I wondered if the TPS got whacked out of alignment. It did. I reset it to 0.54 volts.

I also had a past issue with an aluminum diizy hold down clamp, that woudn't hold. Dizzy kept slipping. I think it's time to revisit my base timing as well, to see if the new clamp slips as badly as the old one. But since it's 10:00pm and the neighbors no sympathy or love of 450 horsepower through sidepipes, it will wait till morning.

lakeffect2

Well oddly enough, THE PLOT THICKENS.

Turns over but a few burps, no start. For the heck of it, I connected a fuel pressure gauge to the valve on the rail. NO PRESSURE.


Pulled the tank harness and hooked up a spare battery.. could hear it run. Checked the body side of the harness, and found 12 volts available before the relay kicked out.

I'm suspicious of a blown out or popped off hose within the tank needing attention. I honestly don't think this will solve what I still beliive are module issues, but obviously, need to fix the fuel source first.

You may have seen another thread were I was concerned about maxing out my fuel tables and still being lean. So I think while I have the tank open it's time for a 255lph pump to go in. I've ordered a Walbro GSS340, high output, high pressure unit from Summit. Looks like I might need to carve out the bottom of the sender slightly to make it fit, as the input hole seems to be centered, not offset.

dimented24x7

iTrader: (2)

Its a little different. No optispark. The PCM directly controls the firing of the coil. There is a crank signal and cam sync signal via hall effect pickups at the crank and dist.. The module basically just acts as an amplifier for the signal from the PCM

lakeffect2

OK , I'm back with some updates here. I had to drop the fuel tank first, to find out why I was not getting any fuel pressure, even though I could hear the pump run. It popped the internal hose off the pump. Since I had some prior issues with rising BLM's and a maxed out fuel chart at 4400-5000rpm, I put a Walbro 255lph pump in while I had the tank open.

I work in a plaza where I work, there happens to be a computer repair shop. I scrounged up a really nice aluminum finned block and 12v fan combo for free. Been investigating sites to have the fan temperature controlled, so it will keep running after the car is turned off. I'm thinking all the heat coming up from the headers would still saturate the unit to some degree. Having a fan on it that cools it to a certain temp would be of a big help, like how some radiator fans continue to run after the car is shut off. Here's a link to the article that I'll likely be working with.

http://www.radiolocman.comshem/schematics.html?di=47949

My question for RFmaster is concerning "Start with twisted pair shielded cable with at least 200C rating (most microphone cables are only 65C)" Do you mean 'C' as in centigrade, as a temperature rating the wire can withstand? any suggested sources would be most appreciated. (Radio Shack? no clue on muy end!)

RFmaster

Lake

Yes, C is short for degrees centigrade.
Let me double check my Belden or Times wire catalogs to be certain. RS sells (for the most part) consumer grade stuff intended for normal home use. As you well aware under-hood temperatures routinely eclipse 100+C (boiling water at sea level). Wire and cable requirements for under-hood operation are severe and really in the class of their own. In my line of work I use semiconductors with industrial grade rated at -40 to +85 C, but automotive grade going even further -40 to +125C!

I have couple cables types that I have used in the past, but I'll have to double check parameters to be certain,

//RF

TTOP350

iTrader: (8)

I know Ford moved the ICM from the dizzy (even tho its outside) to inner fender in 94-96ish on the 5.0 (302) mustangs (explorers and trucks also?) because of heatsoak failures. Doesnt seem to be any special wires running to them. Ford just bolted em to a piece of aluminum to drop the heat.

lakeffect2

As luck and fortune prevail, today I made round of an electrical contractor that I visit every few weeks. In chatting with one of the boys about the problem, he said he'll be right back. He came back with a few feet of high temp, shielded, twisted pair wire to me. God is good, and it's nice to have friends.

Over the weekend I hope to resolve things via assmbly... will keep you all posted. Thanks for both the interest and responses. It'd been both educational and enlightening.

Dave

D Max

Jips

How can you see the ICM is dead?

grtz, D

oldred95

Dimented had posted in a thread several years ago about adding a heat sink to the bottom of the distributor directly below the ignition module. Basically drill two holes under the module and counter sink them so the screw heads sit flush. Fasten the heat sink to the bottom side of the distributor using some heat sink grease between it and the distributor. Then reinstall your ignition module using heat sink grease like usual. I've never had any problems burning out modules but this setup seems to be working just fine. I figured it was cheap insurance. I've got a single plane and run a 180 degree thermostat and on a truck there is enough air coming in under the hood I really don't think things get all that hot under there.

lakeffect2

one of my clues is seeing code 43 pop up, "ESC failure." other clues are no hot restart, like you drive to a store, shop for a few minutes, and when you climb back in it will turn over just fine (not to be confusd with a heat soaked starter) but won't fire up. It gets worse over time to the point it just won't start at all .. ever.

lakeffect2

OK, It's in, complete and it works... so far without issue.
Below, in the center of my photos is a remotely mounted GM HEI ignition module, cooled by a computer heat sink and 12v fan combo. The fan is thermostatically controlled by a secondary radiator fan sensing unit from Advance Auto. ($19.00) The unit is set to start the fan at approximately 140F, which seems to work well, considering normal under hood temps on a black car just sitting in the sun. I set it with the use of the family hair dryer and infrared temp sensor gun from Harbor freight.

The whole system is wired in a fused circuit directly from the battery so that it can still run after the engine is shut down. I did not want the cooling to stop when the ignition was turned off. Works just like some radiator fans do, that even though the car is off, the fan can still keep cooling.

The cable from the inside if the distributor is as suggested by your responses, a shielded, twisted pair. It was a pain to ground it at the distributor side, so the shield is grounded at the ICM heat sink. Being near the ignition wires does not seem to have any negative effect. The four wire group to the ECM had to be lengthened. I used soldered connections throughout.

For the female clip connector from the trigger coil to the module I used a spare from an prior coil. The male end was cobbled up from a box of ECM repinning ends that I found on a rack at NAPA. The pins were then placed in a little wad of 2 part epoxy putty to form a block shaped connector.

The whole system can still easily be returned to stock, but given my past issues with fried ICM's that's not likely to happen any time soon.

Mike M

Has anyone discussed the instance of a bad engine ground at the cylinder head and its' possible affect of life expectancy on Ignition Modules? Also, I have had negative results using the wrong plugs IE: Bosch Platinum, NGK, etc. Some people will argue that it does not matter. I say it does in the long run. The engineers at GM designed the ignition system with a specific resistance and temperature range in mind. To change that will have an affect on performance and longevity. What does everyone else say?

lakeffect2

Bad grounds cause all sorts of gremlins. I have no proof to offer that it can specifically take out Ignition Modules, but it wouldn't surprise me in the least. (Real science at work there!)

I can tell you,now that it's 10 months later, I have not had any Ignition issues to speak of after I went with the remote fan cooled unit pictured above. It works like a charm.

lakeffect2

As An update. It's now over five years later. I have had the Ignition module mounted remotely on the fender as shown in the picture above. Absolutely no failures in that time.

THe Ignition module was simply sandwiched between a small aluminum plate, about 3x4. Above that was a finned computer heat sink. I've since removed the fan as being un-needed about two years ago. I turned the fins 90* from the above picture, so that air moves between them front to back, instead of the first fin as shown, blocking all the others from air flow.

Dave