Motorcycle Forum banner

1 - 20 of 42 Posts

·
ZAMM Fanatic
Joined
·
2,730 Posts
Discussion Starter #1 (Edited)
Summary: Electronic modules RARELY fail "all on their own" "due to age" or whatever. In the vast majority of cases an input or output to the module gets shorted to power or ground and either paralyzes or fries the module. Swapping modules, in a car OR a motorcycle is a better recipe for frying TWO modules than it is for diagnosing or repairing the original module failure. (IMHO, of course).

Backstory: Warren's 2009 Heritage Softtail recently began running so poorly as to be unrideable. It had to be towed to Arrowhead HD for repairs, under warranty. A SENIOR technician there, "Matt," following a printed or online diagnostic tree reported seeing some sort of false problem with the 02 sensors (details unknown), and the next step in the tree stated "REPLACE ECM."

Having diagnosed / dealt with failed / paralyzed modules extensively in the automotive world, I was unsatisfied with this diagnosis and repair. I believe the replacement module will (likely) fail for the SAME (albeit unknown) REASONS as the original one, probably within days or weeks. So I'm going on record here so I can claim a great big "I told you so!" if the replacement ECM fails, and ..eat a bunch of crow if it doesn't.

I'm not slamming HD, Arrowhead HD, or their tech, but I believe a LOT of techs with less electronics experience than myself will indeed "blindly" follow a diagnostic tree that may or may NOT be correct, and stuff in a new module instead of doing what I would do.

If you don't take the time and effort to figure out why the FIRST one failed, the SECOND one will likely go the same way, UNLESS it truly was an "old age" heat or vibration related failures, which are rarer then hen's teeth.

(In my automotive career I can count on one hand how many modules of ALL varieties I EVER saw that "simply failed" and didn't have an external cause, like some input shorted to power, or output shorted to ground out of dozens or hundreds of modules I was tasked to diagnose/replace. )

Caveat: An ECM input/output with an INTERMITTENT short is gonna be DARNED HARD to diagnose!

So how do you diagnose the CAUSE;

You take a multimeter and check every single input and output to the module for reasonableness. At least look for 12V or ground where it SHOULD NOT BE, for exmaple an ECM inputs that is say, expecting a 0-5V signal (made up example) out of an 02 sensor, potentiometer, rheostat, temperature sensor, etc. Or an output that has continuity to ground that should instead show some resistance.

Far too often in the automotive world what techs will do, when a module, anything from an engine ECM to an ABS controller is paralyzed, fails to respond to the scan tool, doesn't do it's job, is go to the dealership (or a junkyard), or borrow a "known good" unit out of another vehicle, and plug it into a problem vehicle. Modules today are often $500 and up...

As I said above, this is an absolutely fantastic way to FRY a second module, and if it's a "known good" loaner, thereby disabling a SECOND vehicle.

Techs who simply don't have extensive electronic backgrounds are EXPECTED, and REIMBURSED BY WARRANTY to follow the factory supplied diagnostic tree.

That doesn't mean the tree is correct. Or complete.

Nor does a tech have to be a real electronics GURU to check ECM inputs/outputs for reasonableness. Anyone who can use a meter properly can measure voltages, resistances, current, and check continuity. They just know it NEEDS TO BE DONE before calling a module, ANY module "bad" or stuffing a brand new one in, especially when it costs hundreds of dollars.

It goes a LOT faster when you have a helper who can read a schematic so one guy can probe inputs/outputs CALLED OUT by the guy reading the schematic. Counting pins and getting onto the right line is always a challenge with todays miniature connectors/pins.

Complicating Warren's repair is that his tuner is electronically "keyed" to his ECM, so ECM replacement is even more complex, and a SECOND FAILURE will add even more grief to the repair. (And weeks of delay as the tuner is sent back to the factory AGAIN to be re-keyed)

When I say "module" it doesn't have to be an "engine ccontrol module" it could be something as simple as an ignition "module." Basically ANY 'black box" with transistors inside, that has inputs, outputs, and 12V and ground power inputs. When a module, ANY module fails, the cause is USUALLY OUTSIDE of the module itself.

I'll report back in a few weeks on whether Warren's ECM truly did die "of old age." (rare) and its replacement was correct, or if now he's looking at SECOND module replacement.

Hopefully this little treatise will save those of you doing your OWN diagnosis and repairs from getting to buy TWO instead of one replacement modules, as they tend to be the most expensive parts on a bike/car anymore.

Questions?
 

·
American Legion Rider
Joined
·
23,543 Posts
Just curious. When was the last time the ECM had any upgrade input? Like reflash of Stage 1.
 

·
ZAMM Fanatic
Joined
·
2,730 Posts
Discussion Starter #3
Just curious. When was the last time the ECM had any upgrade input? Like reflash of Stage 1.
I don't know. I'll try and find out.

For those unfamiliar with "reflash,"

In years past an ECM or other module had a program hard-programmed into it. It ran that program and no others. It might have memory, and ADAPT the fuel and ignition curves, based on 02 sensor input, knock sensors, etc, but it always ran the program it came with from the factory.

Sometimes the program was STORED in a second chip, a PROM (programmable READ ONLY MEMORY) and you could SWAP the chip with a chips with a different program, hence "CHIPPING" a car or truck.

for the most part THESE days the program is stored and run by a single chip.

But THESE days, the program can be erased, over-written with a NEWER, hopefully BETTER program. This process is known as FLASHING as the program is stored in what's called FLASH memory. Apply the right input to certain pins on the chip (using a fancy scan tool, factory scan tool, etc) and you can UPDATE or REPLACE the program.

This is as true for power window modules as it is for ECM's. Automatic transmission modules and ABS modules.

More and more manufacturers "don't get it right" from the factory, owners experience "driveability" problems, and a "Re-flash" updates various
ECM's in their vehicle.

BMW, for one, is NOTORIOUS for reflashing their modules, both to correct problems and also to improve performance.

So you take your bike/car into the dealership THESE DAYS, and first thing they do is connect up a scan tool and SEE if you have the newest, latest software running in your ECM and (hope and pray) an upgrade corrects whatever problem you brought it in for. It usually doesn't, but it doesn't make sense for techs to work on a car UNTIL that has been done to RULE OUT any problems that "came from the factory."

So that's "reflashing" in a nutshell. Maybe Vito can explain "Stage 1" etc. on HD's.
 

·
American Legion Rider
Joined
·
23,543 Posts
Stage 1 is a air/fuel ratio change for those that change their stock muffler and/or air breather. Today everything is extremely lean and changing those can make it even leaner. Which could and does cause major problems down the road as you burn valves up. On Harley's that is.
 

·
It's all fun and games until someone loses an eye
Joined
·
8,665 Posts
I've replaced plenty of PCM's that had no other problem than they just went bad. In this case, if the Tech. checked the circuits to the Oxygen sensors he's probably 99.9% sure nothing outside the box caused the failure.
I can't imagine why one would check circuits that aren't related to the problem.

My bet is you'd better start looking at ways to make Crow palatable. ;)
 

·
It's all fun and games until someone loses an eye
Joined
·
8,665 Posts
Stage 1 is a air/fuel ratio change for those that change their stock muffler and/or air breather. Today everything is extremely lean and changing those can make it even leaner. Which could and does cause major problems down the road as you burn valves up. On Harley's that is.
I've never heard of a failure caused by a reflash.

I have seen modules fail to take a flash, just not seen it cause a module failure.
 

·
American Legion Rider
Joined
·
23,543 Posts
I don't know ODE. Do you have any idea how many input sources there are for your cruise control on your Harley? Everything about that machine is controlled by that magic box. But you have throttle, brake(2) and tachometer that I can think of in addition to O2 senors. Anyone of which could take the thing out.
 

·
American Legion Rider
Joined
·
23,543 Posts
I've never heard of a failure caused by a reflash.
I got documentation to prove it happens. Just a slight power fluctuation that you don't even notice can do it. Or as I honestly think in my case, a borderline ECM to begin with that would have failed later on it's own. But the failure in my case was a result of reflash. Before reflash I had cruise. After reflash none. Several hundred later with new ECM and reflash, back in business. And that was a "well we tried everything else and that's the only thing left" kind of thing. I wouldn't have believed it either but I saw it happen. Something in that box got cooked.
 

·
It's all fun and games until someone loses an eye
Joined
·
8,665 Posts
I got documentation to prove it happens. Just a slight power fluctuation that you don't even notice can do it. Or as I honestly think in my case, a borderline ECM to begin with that would have failed later on it's own. But the failure in my case was a result of reflash. Before reflash I had cruise. After reflash none. Several hundred later with new ECM and reflash, back in business. And that was a "well we tried everything else and that's the only thing left" kind of thing. I wouldn't have believed it either but I saw it happen. Something in that box got cooked.
Hmmmm..... I've had a few cars that the PCM was flashed wrong or incompletely, but personally I've never seen or heard of such before.

I just recently had an MKZ that fried all of the coils. TSB on the car said replace PCM. Since they're in such high demand the guys I usually get my PCM's from were out of stock, so I ordered one from Ford and took it them to get it flashed. Those Jack $#$%'s sent the thing back to me with trouble codes that pointed to an incorrect flash.
I called them and they snidely said "Oh you want us to diagnose it for you?"
What came out of my mouth next I can't print, but I did get the car repaired along with a big apology from the Service Manager.
 

·
ZAMM Fanatic
Joined
·
2,730 Posts
Discussion Starter #10 (Edited)
My bet is you'd better start looking at ways to make Crow palatable. ;)
I keep a bottle of this handy...



I can't imagine why one would check circuits that aren't related to the problem.
In 3 hours or less I can check every circuit that goes into or comes out of an ECM for "reasonableness," e.g. "continuity to ground, V+, etc." That's 3 hours labor.

In contrast I seldom see ANY ECM for ANY VEHICLE today that's under $700.

Again, MY experience in diagnosing dozens, perhap a hundred "failed modules" on Toyotas and Chryslers was...

It was INVARIABLY some short circuit or short to power on an input/output line TO the module that either fried the module, or "paralyzed it" until the short was corrected.

Often correcting the open or shorted i/o line magically "fixed" the ECM, which had gone into a self-protective shutdown mode. (high current, etc.)

I have a really, really hard time remembering a pure "old age" module failure of any sort.

Obviously YOUR experience has been much different.

In the cases where techs on the assembly line DID just pull out the old one and swap in a new one, it usually fried either immediately or in short order, DOUBLING the cost of the repair.

So now they get to invest the two hours labor ANYWAY and $1400 or more in parts on a comeback, with a by-now ANGRY customer.

Investing the time to check the circuits going into and out of a "failed" ECM, to me, seems like a wise use of tech time. To me that is performing DIAGNOSIS versus "Parts-Hanging"

I used to make my line techs THIS offer; if you're willing to pay for a new ECM YOURSELF if a new one doesn't fix it, or comes back fried within days, you're welcome to parts swap.

If I'm (the company) is paying for the module, you're gonna invest the time to diagnose it properly and CONVINCE me it wasn't a shorted input/output before we order one.

Never had any takers.
 

·
It's all fun and games until someone loses an eye
Joined
·
8,665 Posts
I don't know ODE. Do you have any idea how many input sources there are for your cruise control on your Harley? Everything about that machine is controlled by that magic box. But you have throttle, brake(2) and tachometer that I can think of in addition to O2 senors. Anyone of which could take the thing out.
I realize how many inputs, it's just way beyond protocol and generally not necessary to check circuits that are unrelated to that which failed. If they can get a customer to kick in another 2-3hrs more power to them, 30 years in this business and I've still not capped a PCM because I failed to pin out every circuit.
 

·
It's all fun and games until someone loses an eye
Joined
·
8,665 Posts
I keep a bottle of this handy...





In 2 hours or less I can check every circuit that goes into or comes out of an ECM. That's 2 hours labor.

In contrast I seldom see ANY ECM for ANY VEHICLE today that's under $500.

Again, MY experience in diagnosing dozens, perhap a hundred "failed modules" on Toyotas and Chryslers was...

It was INVARIABLY some short circuit or short to power on an input/output line TO the module that either fried the module, or "paralyzed it" until the short was corrected.

I have a really, really hard time remembering a pure "old age" module failure of any sort. Obviously YOUR experience has been much different.

In the cases where techs on the assembly line DID just pull out the old one and swap in a new one, it usually fried either immediately or in short order, DOUBLING the cost of the repair.

So now they get to invest the two hours labor ANYWAY and $1000 or more in parts on a comeback, with a by-now ANGRY customer.

Investing the time to check the circuits going into and out of the ECM, to me, seems like a wise use of tech time.
Ignition modules, igniters, etc. I'll agree, 3 of 4 are for a secondary problem. I just haven't seen it of PCM's.
 

·
Gone.
Joined
·
17,857 Posts
The Electrical Diagnostic Manual flow charts for Harley's are very well thought out and reliable. I can't think of a single way in which a fault could occur and not be located prior to arriving at the "Replace ECM" decision. Maybe with other types you have to check circuits and hunt around, but if you follow the HD procedure correctly, and you are a competent and careful technician, it will eliminate any un-needed effort. And if you did follow the correct procedures and preformed the specified tests, and you arrived at the replace ECM decision, you can be 99.999 percent sure that it is the correct answer.
 

·
ZAMM Fanatic
Joined
·
2,730 Posts
Discussion Starter #14
Another comment. Be sure to check the POWER going to modules, 12V, IGN, and Ground. Be sure voltage isn't above 13.8 or so, engine running, and that there isn't more than a volt or so difference between IGN and V+. Rev the motor and watch for spikes. Bad grounds on modules cause sensors to get mis-read, so voltage drop your ground and make sure it's no more than .05 V away from V-.

Sensors getting misread cause all SORTS of readings cause the motor thinks it's hot when it's not, that it's rich when it's lean, etc.

After that go looking for any signals shorted to ground or V+, and perform other "reasonableness" checks.

W
 

·
It's all fun and games until someone loses an eye
Joined
·
8,665 Posts
The Electrical Diagnostic Manual flow charts for Harley's are very well thought out and reliable. I can't think of a single way in which a fault could occur and not be located prior to arriving at the "Replace ECM" decision. Maybe with other types you have to check circuits and hunt around, but if you follow the HD procedure correctly, and you are a competent and careful technician, it will eliminate any un-needed effort. And if you did follow the correct procedures and preformed the specified tests, and you arrived at the replace ECM decision, you can be 99.999 percent sure that it is the correct answer.
Agreed.... A repeat failure should only occur "If" the outlined procedures weren't followed.
If you've got a module that won't communicate and can't come to a conclusion as to why it failed, then you start running pins.
 

·
American Legion Rider
Joined
·
23,543 Posts
The Electrical Diagnostic Manual flow charts for Harley's are very well thought out and reliable. I can't think of a single way in which a fault could occur and not be located prior to arriving at the "Replace ECM" decision. Maybe with other types you have to check circuits and hunt around, but if you follow the HD procedure correctly, and you are a competent and careful technician, it will eliminate any un-needed effort. And if you did follow the correct procedures and preformed the specified tests, and you arrived at the replace ECM decision, you can be 99.999 percent sure that it is the correct answer.
So do you call a dealership a liar if they say they followed the diags and still did not have an answer? They said they did on mine and it left them hanging. Everything and I mean everything else was working. Just no cruise control. And it was working before a reflash.

Nothing appeared wrong at that point and it finished without any error messages. The only thing electrical they found was a non functioning rear brake switch. I don't even use the rear brake so I have no idea how long it was not working. And my wife would have noticed if I did use the rear brake because she follows me and would have run into me.

I've thought all along that the reflash cooked it. But also like I said I felt it would have probably failed on it's on anyway. Even my diags pointed to the throttle position sensor. That was replaced, still no cruise. This is why I hate electronics. Weird things happen that just don't have a real reason.

I would love for you experts to have been able to follow what they did or for you to have probed around all you wanted. Thank goodness for extended warranties!
 

·
ZAMM Fanatic
Joined
·
2,730 Posts
Discussion Starter #17 (Edited)
The Electrical Diagnostic Manual flow charts for Harley's are very well thought out and reliable. I can't think of a single way in which a fault could occur and not be located prior to arriving at the "Replace ECM" decision.
Yeah, I got "educated" on diagnostic trees working at a Nissan dealership. Our master Tech, John Renza, FAITHFULLY followed a Nissan diagnostic tree with a Nissan super-tech on the line from the factory. Nissan scan tool, Consult2/3 or whatever.

At some point the results simply did not match the possibilities listed by the diagnostic tree. So they did it again over and over and over, both guys, CONVINCED the other had somehow screwed up. Renza was repeatedly told "That's impossible, that CAN'T be what you're seeing. A replacement Consult scan tool was FedX'd to the dealership, just in case IT was at fault."

Through brute force, following every single wire in the ABS??? system from beginning to end, Renza eventually found a connector beneath the carpet where two BENT pins were lightly touching, apparently having been JAMMED together at the factory which created the conflicting results.

t simply wasn't an eventuality, a possibility forseen by the engineers who wrote the diagnostic tree for signal A to short with Signal B, causing ONE result at one part of the diagnostic procedure, and another at a different one. There's simply NO WAY engineers can forsee EVERYTHING wire that could possibly go, open, or short to ground, or together.

By this point the car looked like the cocaine-laden Lincoln? in "The French Connection," the entire interior and other portions of the electrical systems disassembled and sitting in baskets on the shop floor.

This was after the car had been in the shop for over 2 weeks, was subject to lemon law, tying up a bay, customer's lawyer making threats, everybody on edge including the owner of the dealership (Monte Roeder) who told Renza.... do WHATEVER IT TAKES, do not stop until it is fixed. The # of unbillable hours was in the hundreds. (Nissan eventually ate a large % of 'em)

(Never have I admired a Master tech or owner more than those two for their "Never Surrender" attitude when it came to service. One reason why they were President's Club year after year after year....)

So you personally believe HD's diagnostic trees are good for 99.99%. That would be better than NASA's fault diagnosis trees, which THEY estimate at 99%. As an engineer who has subsequently used diagnostic trees as a tech, I'd guesstimate HD or any other auto/MC mfr's number would be closer to 95% ASSUMING they did a really superlative job creating and TESTING their fault trees. That means one out of 20 diagnoses made using them will be incorrect.

That's fine and dandy unless you're the 1/20.
 

·
ZAMM Fanatic
Joined
·
2,730 Posts
Discussion Starter #19
Sh1t happens but doing unneeded diag on every vehicle just because one in a hundred kicks your ass is wasteful.
It's only "Unnecessary diagnosis" when the customer is paying for the mistakes, modules, and misdiagnosis. Which may INDEED be only 1/20.

Along with the annoyance of a comeback, catching a ride, being off their bike, etc.

I'll bet when YOUR bike's ECM fails (out of warranty) you'll do exactly what I do which is checking every signal going to the ECM for reasonableness before popping $500 or more for a replacement module.
 

·
It's all fun and games until someone loses an eye
Joined
·
8,665 Posts
What would make you come to ths ridiculous assumption that I would treat my vehicle any different than a customers?
Of course besides the fact I'll wash and wax mine.
 
1 - 20 of 42 Posts
Top