Mechanic: 2011 GMC Sierra · 2:40am Jun 6th, 2022
Alrighty kids, got a quick one for y’all today.
You know the drill.
Vehicle in question came in on a hook last week, and it wasn’t my problem then. (I had other problems, which might or might not wind up getting a blog post on their own, depending on how they turn out.) The conversation I overheard was that someone had replaced the Fuel Pump Control Module (FPCM) and it needed to be programmed.
Some blog posts back I told y’all about Ford’s Fuel Pump Driver Modules. They’re the ones that they mount on the frame near the spare tire and which corrode through because Ford forgot that galvanic corrosion was a thing,
GM was late in the game when it came to using a computer to drive the fuel pump, in part because GM was probably trying to figure out how to make a fuel pump that didn’t sound like a coffee grinder full of gravel. Admittedly, this is good for diagnosis; for some makes, it’s hard to tell if the fuel pump’s running or not. Even if you stick a funnel down the fuel filler neck and put your ear against it, you can’t always tell. On nineties to mid aughts GM, you can hear it in the cab. With the windows up. Sometimes with the radio on.
Anyway, GM started putting them on the trucks, and they’d learned a thing or two from Ford and made the box out of plastic instead of aluminum. Unlike Ford, where the module’s plug and play, on GM vehicles you have to program it.
We’ve got a scan tool that can do that, and my manager was right on it . . . and three hours later, we pushed the truck out of the shop, un-programmed.
While by law in the US, automakers have to provide service information and tools to the aftermarket, they don’t have to do it for free. Some do provide service information for free (or did, anyway)—Kia and Hyundai used to make all their service information available at no charge.
GM does not do this. You have to buy a tool from them, and then you have to lease the software. And if you want to program, you have to pay extra for that. Basically, you buy programming licenses for one VIN (specific vehicle) at a time, and you can program whatever you want on that vehicle for a year.
For whatever reason GM’s server was down, so my manager was unable to program the truck, and we had to push it out when we needed the space.
This afternoon, the truck became my problem. He handed me the keys and said that we had to push it in.
“Why? Aren’t we just programming it?”
“Well, the customer said that the FPCM might not be plugged in right, so you’ll have to get it on the hoist.”
One of the things you can do with both the GM scan tool and the Snap-On scan tool that we have (and surely others as well) is check the vehicle network, see what’s responding and what isn’t. In fact, before spending a lot of time on the phone to GM technical support—which knowing my manager involved complaining about having to use the internet and demanding that their techs fix what was wrong faster—he could have easily determined that the new FPCM this truck has isn’t online.
So I put it up in the air and make a couple discoveries. First, that the customer did not attempt to remove the old FPCM from its bracket, he just used nylon ties to fasten the new one near where it’s supposed to go. Second, he didn’t plug it in correctly. It’s got a fancy connector (I think they’re called ratchet-lock connectors, but I don’t remember) that are kind of tricky to get to work when they’re full of dirt. The customer’s solution was just break it, which made removing it from the old FPCM easy, but installing it on the new one substantially more difficult.
I got it connected, got it programmed, and then when you’re done programming modules you often have to clear codes from other systems on the vehicle. In fact, the GM programming system has a button to do that on the screen that comes up after you’re done programming, so I pushed the button.
The computer worked for a while, and then it said that it had cleared codes, but maybe not all of them and I should go in and clear the rest manually.
There were four codes. Two in the engine module, and two more in the Antilock Brakes. Both of the ABS codes were informational, indicating that the system was disabled due to engine faults. The engine had a code for the MAP (Manifold Absolute Pressure) sensor and one of the 5V reference (Vref) circuits.
We’ve discussed both of those things before:
MAP sensor
5V ref (Escape)
5V ref (Avenger)
My manager said not to worry about those codes, take the thing for a quick test drive.
I was about halfway through the test drive when the oil pressure gauge dropped to zero, the red oil pressure warning light came on, the information center said “Low oil pressure, stop engine.”
Also, the the engine went into severe reduced power mode.
It’s been a long time since I talked about reduced power, I think. What you need to know is if there’s some kind of malfunction where the computer can’t properly control the throttle body (and thus the engine speed or the vehicle’s speed), it has varying degrees of how much it shuts things down. Depending on the automaker and the type of fault, it might limit the maximum engine RPM, maximum speed; it might limit how far the throttle plate can open or not let it open at all. In some Fords, if things go really, really wrong, it will shut down the fuel and spark.
This truck was intermittently limited to less than five miles per hour, and no throttle response at all. The engine would idle, and that was it. Since I had some time on the drive back to the shop, I figured out that if I came to a complete stop, it would have throttle response until whatever malfunction it saw reappeared, and then it would go into severe reduced power again.
When I got back to the shop, it had a whole lot of engine codes. Besides the two from before that wouldn’t clear, there were performance codes for the Fuel Tank Pressure Sensor (FTP), Oil Pressure Sensor (EOP), another MAP code, and I think one or two more.
Manager decides all of a sudden that this is a problem, and also that maybe this was the problem that the customer had all along and that there was never anything wrong with the old FPCM. The good news on that front, at least, is that we didn’t diagnose or replace the module.
As you might recall on the Avenger or Escape, there are multiple 5V circuits, and they run different things. A failure on one of those circuits will (obviously) affect every sensor on that circuit.
I punched the Vref code into Identifix, just to get a quick idea if there are any common failure points, or what sensors are on that circuit. It won’t surprise anyone to know that the EOP sensor, the FTP sensor, the MAP sensor, and the FPCM are all on that circuit. Also one of the two throttle position sensors, and the air conditioning pressure sensor.
You can monitor the 5V reference circuits on the Snap-On tool (and I assume the GM tool as well); Vref 1 was at 1.07 volts and Vref 2 was at 5.03V.
For a shorted sensor, the easiest way to diagnose it is unplug the sensors on that circuit one-by-one and see which one solves the problem. I started with the MAP, ‘cause it was easy. Then the AC pressure sensor (almost as easy); neither of those fixed it.
The oil pressure sensor is hard to get to on those, so my next step was to lift the truck and disconnect the FTP sensor. That’s on top of the fuel tank, and you have to fit yourself between a driveshaft, muffler, and fuel tank, then reach your hand across a razor-sharp heat shield and work by feel to unplug the thing.
This is when I discovered that apparently one of the vehicle owner’s hobbies is breaking electrical connectors; the FTP connector was broken, and he’d zip-tied it in place.
As I was struggling to get the connector off the sensor, I was running through all the other sensors that could have failed, considered if I might want to have another crack at the EOP sensor . . . and then I eventually got it.
All of a sudden, Vref 1 is at a nice, even 5V, right where it should be, and when I take it for a test drive everything works. The check engine light’s still on, of course, but now the only code is an open circuit code for the FTP.
Admittedly, not the most exciting diagnosis or repair (well, I assume; maybe the guy will take the truck with the FTP unplugged and fix it himself). I kinda wish I had a nice conclusion to wrap it all up, or at least some kind of moral. Vref is important, I guess?
Since I’m here, I’ll give y’all the conclusion to the Jeep Compass. That was the one I talked about a couple of months ago that had a VVT problem caused by lack of oil pressure in the cylinder head.
Teardown was authorized, and as we suspected, the screen was completely plugged.
The bright stuff you’re seeing in the screen isn’t metal; those are the spots where the flashlight I’m holding up against the bottom of the screen can actually shine through.
I think that the screen wound up being less than ten dollars, certainly the cheapest part of the job. After much deliberation, we left it out instead of replacing it: both the manager and I agreed that dirty oil getting into the upper end and maybe contaminating the cam solenoids or phasers was less bad than no oil getting to the upper end of the engine to lubricate everything else up there.
Good problem-solving!
We know the drill? What are you talking about? That's clearly a socket wrench.
Jeeps have been my bane this month. Almost as much as U-boxes, but generally speaking, if a vehicle fails its trailer light test on hookup, so far it's been a Jeep. They also have a lot of weird fuses in their boxes that you can't just replace from my manager's box o' fuses, big doofy nonstandard square fuses. Luckily, Jeep also litters their fuse boxes with spares, and I work with a retired mechanic who I was able to dump the problem on, while I went up front and put out other fires that caught fire while we were just trying to send out a Jeep customer with a goddamn utility trailer.
Sounds like you had some fun. May was a bad vehicle month for me. I will keep it simple, my 2012 Subaru got totaled out.
That would be one fried griffon. The reason why birds can sit on the wires is that they sit on one individual wire. If you touch two wires -- zzzzzt. Toe beans wouldn't save it.
I'm assuming the screen on the Compass is that bastard one that's between the cylinder head and the block?
So my nightmare lately has been my friend's car. 91 Dodge Dynasty, 3.3L. I've got a soft spot for that car even though it's a piece of shit, because it was basically our first project car. It was a near-base model car, but we (many years ago) did a lot of junk-yarding. It's basically a Chrysler Imperial now. Anyhow, it didn't start - he hadn't driven it in 3 years, though he had moved it around once in a while. I go to his house, and we find out the crankshaft sensor wires had been chewed through. Twisted them back together, and it sputtered to life... running like absolute garbage and alternator not charging. Got an alternator, put it in it - still while parked on the street in front of his house. Not a particularly easy alternator to do, you might know.
It charges now, but runs like ass and backfires like a gun going off. Dumping raw gas down the tailpipe, obviously. We limp it to the shop, and I do what I can. I have the hardest time diagnosing the damn thing - my own dumbass fault. I know at least one cylinder isn't firing, and I'm relying on disabling the fuel injectors by unplugging them. It took me like a week to realize the injectors on the front are for the back cylinders and vice versa. Then I immediately found the problem, one part of the ignition coil pack wasn't firing. More accurately, the ECM's #3 coil driver is stone dead.
Ordered a reman ECM. Plug it in. Same problem. So yesterday I go back through all the motions and I come to the same conclusion. The #3 coil driver is dead. Only now, this one is slightly less dead. It can fire just once perfectly fine, but not rapidly enough to idle the engine. In fact, I pulled out the old Brick™ Snap-On scanner and run the coil driver test, which pulses the coil about once per second, and it works perfectly. It just can't work rapidly.
Now I need to hunt down a computer repair guy, or try to do it myself - which I'd be all for (I repair all sorts of electronics) if it weren't potted in gel. Or try my luck with another shitty reman.
So many boxes and wires, so many ways for systems to fail with sinlge point fails?
Thats hammering the connection in though, $1 plastic hood verses $10 metal good relative ratio given the old mil stuff Ive seen?
Been dealing with a tractor that has intermittent engine faults, including going into low throttle limits. Turns out a wire harness that plugs into a DEF pump had come loose. As per the field service guy who showed me the results. He zip-tied the harness into place pending arrival of a replacement plug today or so.
Worse, Saturday the corn planter's hitch failed. And in the process, broke the wires, cables, and hydraulic hoses between tractor and planter. This morning, I get to start disassembling all the broken stuff and make it not-broken! At least nobody was hurt, and it happened before the machine got on the road.
"The good news on that front, at least, is that we didn’t diagnose or replace the module."
Good news in the sense of "Wasn't our fault and we didn't have to pay for it!"? :D
re the picture with the gryphon:
...I guess those are insulated three phase cables. Either that, or maybe they're communication cables (they do look pretty thin...). Or something. I think uninsulated three phase power transmission lines brought into contact with each other and one's feet like that would result in Significant Excitement, though.
...Though, if they're, say, uninsulated telegraph wires, there might still be Significant Excitement, of a sort, on the telegraph network, with those crows not the only ones displeased with this state of affairs. :D
re the branches-on-powerlines unicorn:
...But, why? :D
(Still a cute image, though.)
re leaving the screen out:
Ah, and I'm guessing that fixing whatever it was that made the screen so clog-prone (I don't recall if that was addressed in the blog post or not, sorry) wasn't in the customer's budget, or somesuch thing?
And thanks for this blog post too, as usual. :)
I once bought a car that the previous owner knew just enough to be absolutely LETHAL when let loose under the hood.
He "solved" a starter problem by shorting out the Neutral Safety Switch. This meant that the car would start in gear & with the brakes disengaged.
Did he mention this? No. He left this & other surprises for the new owner to discover. I want surprises, I'll buy a box of Cracker Jacks.
Sounds like your car owner was a relative of mine.
I am unsurprised with the GMC... but that Griffin on, what does appear to be, a three phase 13kv copper line... that is impressive and concerning.
5662580
Thanks!
5662587
<gets the joke>
<checks all the pictures ‘cause I can’t remember if any of them have a socket wrench in them>
<finds a pony pic with both a drill and an adjustable wrench*>
https://derpicdn.net/img/view/2016/6/30/1190273.png
_________________________________________
*Some people call an adjustable wrench a ‘Carolina socket set.’
5662596
Lately, I’ve gotten a ton of Chevy Silverados (08-12) that run bad, have a ton of codes, and the customer only wants to fix the one thing that makes it run bad.
Probably J-Case fuses; those are the popular high-current ones these days.
Speaking of fuses, I just fixed the blower motor on an Express van. Both the wiring diagram and the fuse box label said that it had a 40-amp J-Case fuse that powered the blower motor on high. It didn’t.
Like, it wasn’t missing; there weren’t any terminals for it in the fuse box. And that circuit had power, so whatever fuse it has is somewhere else, unlabeled.
5662600
I wouldn’t say fun, exactly. At least I solved some problems.
Yikes! I hope everyone is okay and the insurance company is generous
5662613
That’s assuming that griffons aren’t immune to electricity.
. . . they probably aren’t.
5662628
That’s the one, yeah. I understand the motivation in including it, but at the same time, if they were going to, they could have put it somewhere that it could be replaced if needed. GM trucks have their screen right under the oil pressure switch.
My minivan’s got a 3.3! All of my minivans have 3.3s!
For some reason, Chrysler alternators don’t seem to like sitting. The one on both my Jeep and my oldest Caravan locked up after they sat for a while, and I had to cut the belt off to make them move. The minivan also had lots of mouse-eating-wire activity.
They ran one driver per coil on those? Rather than do waste spark? That’s kinda weird for the 90s, isn’t it?
Nothing I love more than getting a new part that doesn’t work. Just Friday, I put rear calipers, pads and rotors on a Kia Soul. One of the reman calipers came with the wrong bracket, and the other one had a piston that stuck.
I don’t know what takes that gel stuff off without damaging other components, but there’s probably something. Barring that, you’re probably stuck with a reman or find an identical one in the junkyard.
Just for giggles, I sent a message to MicroCare, which sells circuit board cleaning stuff. Probably their Heavy-Duty Flux Remover - Super Clean (TM) is the product to use. They may have different advice if/when I get an e-mail back.
5662639
They’re usually pretty okay at failure management. Even with a Vref failure, the truck still ran, just not well.
Honestly, the ratchet lock (or whatever they’re called) connectors work pretty well, as long as you’re patient enough to clean all the dirt out of them before trying to open them.
Plus the number of pins they’ve got to get in a tiny little space. . . .
5662881
Yup everyone was alright. Unfortunately with how much dealers are marking up cars in general it is making it difficult to replace for a nice price. I hope to get it finalized this week so I can get a newish vehicle.
5662655
Depending on where the connector plugs in, hot glue is a good, temporary fix. You can heat it up (carefully!) with a heat gun when you need to get it apart again. Or a zip tie is that’s possible, as your guy did.
That’s not great. I never considered how many things would break if an implement came loose from the tractor, but there are a bunch of connections that aren’t meant to be pulled violently apart.
Do tractors usually use safety chains?
5662664
Exactly that. The customer diagnosed, bought, and installed the part so we’ve got no blame that it didn’t fix it.
“Hey, sparks, the telegraph isn’t working again . . . send somepony out to see if there are any griffons on the power line.”
I dunno, maybe she hates electricity.
Yeah, realistically to make sure that the engine was clean inside would involve complete disassembly, soaking it in solvent, then reassembly. Replacing the engine would have been the best option, but the warranty company wasn’t going to pay for that.
You’re welcome!
5662672
Those are the worst.
That’s a solution. Not a safe one. I would have done the same if it was my car.
Of all the unexpected things I’ve found on a cheap used car was the one where the previous owner had glued the outside mirrors into position. I guess he didn’t want anybody adjusting them?
There are a few cars that we see that have interesting repairs done to them. Luckily, we’re one of the more expensive shops around, so we mostly see newer, better-cared-for cars than some of the other shops do.
5662711
I am unsurprised with the GMC...
The blog doesn’t reflect it, but we had a run on those trucks in the last month (all with drivability problems), and the GMC turned out to be one of the nicest of the lot.
He’ll probably be fine.
5662613 5662711
Also regarding the power line griffon . . .
One bit of my headcanon is that pegasi are immune to electricity, up to and including negative lightning. So in a story I’m slowly working on, a rescue pegasus working for the Coast Guard accidentally touches both a 20-some KV wire and the utility pole, and gets blasted off. The humans, naturally, freak out; the other pegasi all shrug and say that she’ll be fine, it was only 20,000 volts.
5662892
Yes, there should have been a safety chain in place. That would have prevented a lot of damage that happened after the initial break. We are replacing the missing chain... and fixing bad safety chains on several trailers too. Several hundred dollars in hydraulic hoses, wire harness, labor, and lost planting time is a big wake up call. We got lucky. No injuries, so far everything is repairable in two days of work.
Grease is cheap... and so is chain.
5662901
I tend to go mostly the other way and figure that lightning in Equestria carries a lot fewer volts/amps than it does in our world with the unregulated weather. Even the ground ponies seem to be far less affected by lightning than a person would be. It would be pretty hard to design an animal that wasn't harmed by terrestrial lightning bolts, or a 13kv electric line.
5662901
Ah... yes, just 20kv. I can see it now...
"They'll be fine. It's not like it's a seven hundred fifty thousand volt line. Those sting."
Sounds like the initial diagnosis was spot on - the problem was that the FPCM needed programming. Well, after removing the old one, and removing the dirt from the connector, and putting the new one in the right place.
I am unsurprised that a vehicle that had a bodged on FPCM also had other problems. Like, uh, other issues fixed by zipties. Zipties are magic things that can fix everything, just like duct tape.
5662891
Yeah, buying a car right now isn’t great. It’s actually nice for us; people are fixing things they wouldn’t have years ago, and we’re making a lot of money. The downside is we’re working a lot of hours, too.
5662913
I kinda wondered. Since I live in the country, I see tractors cruising around all the time towing implements, and I’d never really looked to see if they were chained or not. I’ll pay more attention to the next ones I see driving by.
Back when I drove wrecker, I was one of the drivers who was more obsessive about using all the chains and tiedowns we were supposed to, instead of only using a couple. One night, when I was working a wreck with a new guy, I picked a car out of the median with the flatbed, and I guess whatever I hooked on under the front of the car wasn’t as strong as I’d thought . . . as I pulled up onto the highway, the car slid halfway off the deck, then got hung up on the rear chains.
Got on the radio and told the rookie that was why you used all the chains, then pulled it back down where it belonged.
(Normally, I would have had a chain on the front, too, because who trusts winches to hold? But that truck didn’t have as many as it should have.)
Yeah, exactly. And can save you days of work you shouldn’t have had to do.
5662933
That’s certainly possible, and a decent headcanon (also means that a person hit by Equestrian lightning would probably be fine).
I just figure with all their magic, they’re probably lightning resistant, at least to the point normal lightning won’t kill them. It’s certainly going to hurt, and there’s a good chance it’ll at least stun whoever it hits.
5663026
I can imagine some of them liking the feeling of electricity and electrical fields . . . 750kV is probably not something a pony would want to be hit by, but I could see some of them landing on the middle of a line to feel the currents.
(Also, now that I think of it, they can probably usually tell by the feel of the field around a line how much power it has in it.)
This also makes me think of a helpful pegasus putting power lines back up after a storm. “No problem, guys, I’ve got this.” Maybe ultimately getting a stern talking-to by the local utility company.
5663160
I bet if I’d unplugged the fuel tank pressure sensor and plugged the old FPCM in, the truck would have run that way.
Sometimes doing it yourself doesn’t save money. Just today I worked on an Escape that the customer had done plenty of work to, but it still wouldn’t idle and was dumping in fuel like crazy. Fixing the single shorted oxygen sensor wire (just a couple butt splices) and replacing the blown fuse made everything work again. Maybe $200 in parts and labor all told, which is almost certainly far less than he spent on all the sensors that didn’t fix it.
There’s a lot you can do with zip ties. And honestly, mounting an important computer to the underside of a vehicle with zip ties isn’t the worst repair I’ve ever seen. It’d probably hold up that way for quite a while.
The screws are better, though, because they rust and seize in fairly short order, and then there’s no worry of the FPCM falling off. Or being removable at all.
5662887
It is waste spark. There's 3 coils in the coil pack, not 6 like you were probably thinking
There's a driver (read: transistor) for each. Driver 3 is dead. Actually up to 91 they were pretty cheap with the ECM, waste spark is fine, but it's also waste injection. There's only 3 injector drivers, so they run in pairs. IIRC, 2 & 3, 4 & 5, 6 & 1 - and that's the firing order, 1-2-3-4-5-6. It's dumb but it works. In 1992 they upgraded to 6 injector drivers.
Just double checked the wiring today, it's perfect. Next I'm gonna look at the flywheel sensor wheel, but I'm 99% sure that's a dead end. Cuz if that was the problem, both computers would show zero spark on #3 driver. One ECM wouldn't just have a little spark. (I think lol)
Hey, neat. I'm sure they'll sell something, but it's probably cost-prohibitive if you're not doing these repairs for a living.
5662894
Heh, thanks.
:D
Though that reminds of an interesting bit I saw in a video the other day. Let's see if I can find it again for you...
Ah! Here it is:
https://youtu.be/xumonIs52Lk?t=381
(Entirely out of the right-side speaker of my headphones, unfortunately.)
:D
:)
Ah, thanks.
:)
5663431
So if I’m understanding correctly, you’ve got two dead cylinders, on account of the shared driver being bad?
I did think that some of the early electronics gang-fired injectors, too, but I couldn’t remember and didn’t feel like looking it up.
At least that was a better system than throttle-body injection, although not by much (although TBI was usually easier to fix, so. . . .) Or GM’s spider injectors in the intake, that’s a system I don’t miss.
It’s possible it’s getting a bad signal and the old PCM can’t cope with it but the new one kind of can, but I wouldn’t expect that, either, TBH. Correct me if I’m wrong, but the driver’s probably going to be all or nothing when it grounds or powers the wire--it’s going to see a signal to fire and it’s going to close its virtual contacts, or it won’t and it won’t. I’d be more inclined to think that a bad crank sensor would fire well sometimes and not at all other times.
But I could be wrong. Sometimes electronics have weird, unexpected failure modes.
5664256
That’s insanely old-school.
IIRC back in the day, you could tap into a telegraph line with alligator clips (or whatever) and pick up signals on that line, add in your own message, whatever. I’m guessing that touching the wires together makes a short which makes some kind of noise so the dispatcher knows he’s getting a call, and then he’s ready to listen.
That also does remind me, I think that historically there have been problems with communicating on subways, since normal radios don’t like tunnels. They could have used something like the PRR’s Trainphone system, just with antennas on the side of the subway cars.
5664510
"IIRC back in the day, you could tap into a telegraph line with alligator clips (or whatever) and pick up signals on that line, add in your own message, whatever."
I think I recall hearing that too, somewhere.
"I’m guessing that touching the wires together makes a short which makes some kind of noise so the dispatcher knows he’s getting a call, and then he’s ready to listen."
Actually, while IIRC that was also my first interpretation, on further consideration I decided it looked more like that was what triggered the traction current shutoff. Notice how the noise of some machine on the car stops pretty much _immediately_ after the wires are brought into contact -- I'm guessing that's something that only runs off external power, not the batteries -- and how there's no explicit mention of how the traction current is shut off, with the bits about the telephone just referring to informing the line controller where and why the driver _has_ shut off the current.
re tunnel communication and Trainphone:
Hm. That does seem like a fairly obvious idea, now that you mention it (though I don't recall having thought of it before, so maybe it's one of those things that just seems like an obvious idea after the first time someone thinks of it); I wonder if it was ever done? As I recall, and as the Wikipedia page you linked agrees, that system didn't combine well with the PRR's electrification -- but the PRR's electrification, except in the very early days, was AC. DC electrification, I'd imagine, would result in significantly less interference.
5664508
Well now remember that a coil works backwards from pretty much any other electrical thing. So it's kind of confusing to think about. Much like thinking about electron theory when trying to figure out a circuit (electricity actually flows from negative to positive, but nobody knew that when they invented the way schematics are drawn, so every schematic ever drawn is completely ass backwards). Anyway, a coil is powered, making a magnetic field. When the power is disconnected, the field collapses, dumping all that energy down the spark plug wire. So the coil driver is not pulsing power to the coil, it's leaving it on and pulsing it off briefly. (That's why the technical term for points is 'breakers.') So it's not necessarily all or nothing. A coil driver (again, transistor) could theoretically be weak to the point where it would let some power through it, enough to allow a magnetic field to build in the coil given enough time, and still be perfectly capable of opening the circuit to dump the spark, but not able to allow current to pass quickly enough to build a field in the milliseconds between sparks at idle rpm.
Of course, I'm only speculating. I'm still waiting on a new computer. I did check the sensor wheel / windows on the flywheel. They all looked fine, though I think the sensor wheel is out of round. Maybe when it gets closer and further from the sensor, it fucks up? Not sure. Even so, how would that problem develop over time?
So one of my techs asked me the other day, why the hell some engineer came up with the spider shit. I told him I think I know why. I went to the tool room and dug up an early spider from the pile of fuel injector stuff. It's a single oldschool TBI injector with spider legs coming out of it. It makes sense, they wanted multiport injection without paying for six injectors. But then they revised it to just be six stupidly designed injectors with long hoses on the ends.