Hot Starter Blues

Argh, for the first time ever my car let me down tonight.

I thought I was better than everyone else, because while my starter solenoid would frequently threaten to act up when hot, it would always give in and fire after holding the key for a few seconds. Not this time. Stopped for gas, and knew it would give me a scare, with just the right amount of time for the engine bay to heat soak. But this time it decided to bite me, and it took about 25 minutes for it to finally give in. Opened the hood for a while, but not sure if it really helped.

I'd read the threads before, so I guess I know my game plan and will read up on all the details, but still just had to vent. I'm not sure if it would have made a difference on this car, but I didn't even have anything heavy to bang on the solenoid. Have seen that work with other vehicles, usually when cold. But if nothing else it would have let me vent my frustration. I think I've got a brass hammer somewhere that I can take along to at least feel like I'm doing something useful while the thing cools down.

 

Hmmm, I'd been thinking of having to mill 0.004" out of the bore, but it'd be a lot easier to sand 0.004" off the shaft instead, if it'll keep working. I was originally going to just replace it, but it seems really easy to get to and I probably owe it to myself to look at the existing one, which will be the weekend project.


A few questions came up though:
1) Is that WAI 66-9131 really the right replacement for my 0-331-303-032? It doesn't show as a cross reference on the WAI page.
2) Is the WAI any better than the original Bosch at avoiding this issue, or would it need some more clearance too?
3) What's the best way to acquire the WAI part (say online)?
4) Is there any value in investing in the "premium" 66-9131-1 part?

Also, is there anything special to bump-starting this car if it happens to me again? Without having looked at the wiring layout of the starter yet, is it possible to engage the starter by bypassing the solenoid by shorting out some of the terminals with something big and metalilc?

Thanks again for all the great tips here.

 

I know Steve, I already had at least one of your starter food-chain posts archived and it was certainly on my mind to want to see the voltages back there, but unfortunately this failure is most likely to happen at inconvenient times instead of in the garage! The other night at the gas station I didn't have anything with me, or a second pair of hands to help, but now I'll at least leave a DVM in the car and some copies of the wiring diagrams. I'm going to at pull my old solenoid and likely just buy a WAI one if I can get it for ~$20, as this looks like a very accessible part (famous last words) and I want to become more familiar with the layout so I'll be more prepared next time to do some diagnostics.

Since the scares the starter has given me before have all be related to high engine compartment heat with the car parked for only a few minutes, I'd at least assumed it would be a good bet it's the tight solenoid issues that I had read about. Although I'd been curious about voltage supply in the past, as the fans are generally running when it's threatened to be bad, in this first case of full failure engine temps weren't hot enough to trigger the fans. Oh, and in this case it really was about 20 minutes of cooling that did the trick and it started on its own.

Thanks again for the great guidance here on this and 1000 other issues!

 

As a possibly unrelated thought here, brought up by a friend of mine after talking about my solenoid:

Has anyone ever thought about or experimented with putting the fans on an unswitched circuit as they seem to be with modern cars? I'm not even proposing that as a real solution to this particular issue, but just wondering whether the extra air being pushed through the engine compartment for some arbitrary time after the engine was stopped would be helpful. I've never measured anything, but I'd venture to guess that right after parking a hot engine might yield some of the hottest temps that engine compartment ever sees under normal driving. Maybe it could be beneficial in other ways, like for the cat temp ECU's, which I'd assumed fail due to the high heat in there. Looks like the starter solenoid is taking most of its heat from cylinder #7, followed by all those exhaust parts, so don't know how much air flow would really make it back there.

Still haven't had a reason to look where the temp switches for the fans are located, so I don't know how long it would really take for them to shut off WITH the fans running (but no fluids flowing). I do know they can still be on quite a few minutes later if I come back to the car, but this is with the fans off.

I'd be curious to know what the results were if anyone has tried this.

 

Since I was in working on my starter solenoid (Bosch 0 331 303 032) I took some pictures for reference. Figured I'd post up something for those like me who'd never taken the thing off before. This is pretty basic stuff, so for anyone who's already worked with this part there's no point in wasting time looking at this. But for someone clueless like me, who didn't even know where the part was located on the car, it might help. Also a good thing to know the location for other diagnostics, such as Steve M.'s great stuff which frequently references solenoid terminals.

[Double check you've got the battery disconnected]

The starter solenoid is located under the air filter, on the left hand side (see pic #1 looking from the left rear of the car). You won't need to remove the starter or the air filter assembly to get it out.

It can be removed via 3 Phillips #2 (I think) screws, 2 shown in the pic and one pointed to by the screwdriver. Before actually removing the screws I would suggest first cracking the 13mm nuts on the two large posts first, so you've got a more stable frame to torque against. You'll need to remove the red protective plastic piece to get to these nuts, and will probably have to remove the top blade connector to get it off.

If you remove the left thick cable (12V) from the solenoid first it will make it much easier to pull the body off. There will be a spring inside (see pic #2) so don't lose it.

The solenoid piston is hooked onto a lever arm in the starter, so in order to remove it you'll need to tilt the piston all the way to one side (see pic #3).


To reassemble:

You'll need to align the tab on the back of the solenoid piston vertically so that the arm from the starter (pic #4, looking from the front of the car) can be engaged through the slot on the tab (see pic #5, view from above). When you reinstall it you'll have to tilt the piston as in pic #3, and when it's properly engaged you'll feel that you can't pull it out when aligned straight. For me it was easier to place the spring in its slot in the solenoid body before reassembling. And again it's easier to re-torque the nuts for the power cables with the body reattached.

That's it. Pretty simple stuff and the only tricky part is knowing how the piston is attached to the arm from the starter, but once you've seen what it looks like inside it's trivial.
Image Unavailable, Please Login
Image Unavailable, Please Login
Image Unavailable, Please Login
Image Unavailable, Please Login
Image Unavailable, Please Login

 

OK, so after working on the solenoid and getting a few drives in I think I'm ready to declare success so far. I've made it a point to get it into lots of situations where the engine would get hot and certainly would have threatened to stick and given a long delay before engaging. Now it starts instantly.

What I did was remove 0.001" off of the diameter on the solenoid piston, as suggested in another thread and by Spasso in this one. I went with less machining than suggested as the existing piston clearance seemed to be ~0.001" (didn't get a precise measurement on the ID of the body) so this would be doubling the existing clearance. I figure I can always make it smaller.

Here's what I did:
- Chucked the piston in a power drill holing onto the tab, and set it on its slowest gearing.
- First used 150 grit sand paper to remove 0.0007" this happened FAST even on the slow setting of the drill.
- Then "polished" with 600 grit paper to end up at 0.001" (see measurements below). The resulting finish was probably a little smoother than the original.
- Sprayed a layer of graphite dry lube on the fresh surface so it wouldn't look so much like it would want to rust.

Before machining it down the piston diameter measured between 1.0605" - 1.0608" (six measurements, parallel and perpendicular to the tab at three locations along the piston). After my "machining" the piston diameter measured 1.0596" at all six locations. The approximate ID of the bore of the solenoid body, taken using a locked pair of calipers as a make-shift bore gauge, was 1.061(7)", but this was a sloppy measurement.


I'm not trying to propose this as a solution for anyone else, it's just what worked for me (so far, knock on wood). Certainly not a substitute for a solid diagnostic of the whole system as suggested by Steve M. here and elsewhere. I went this way as opposed to replacing the solenoid mostly because I was worried that I might have the same issue again, even with a new Bosch or WAI unit; I wanted to try to do something specifically to address my symptoms of getting too tight and sticking when hot. This is certainly a big, big improvement in my case.

 

No, not at this time.

First this was more of a fact finding mission here; I figured after removing it at least now I'll know where all the relevant terminals are so I can do some diagnostics, and while I had it out I could try out a fix that seemed to fit best with my symptoms.

Second, I agree with Steve M. that it's worth looking at the path all the way to terminal 50 and understanding what's going on prior to adding anything. At least now I know where some of this stuff is, and will have some equipment with me next time it fails! IF it gives me trouble again, now I can look at the voltage at terminal 50 and see if it's "low" and try just shorting it out to the 12V terminal and see if it starts it.

If that were the case I'd have to make some decisions about an external relay. In general I'd be reluctant to stick more electronics in that hot engine bay, but maybe that'll eventually be what happens. But I would rather ensure that there's minimal voltage drop towards terminal 50 and ensure that the solenoid is working as it should at all temps.

For now I'm just hoping that this fix sticks!