not even a spark

December 21, 2012 1:13 PM

My 1991 Nissan 240sx won't start - it won't even try to turn over. This one is a real mystery; I think it must be the starting circuit, so I'm trying to go through the service manual carefully and check everything. However, I still can't find the issue. Does anyone have any ideas? Details inside on what I've tried so far.

The issue: My car is a 1991 Nissan 240SX fastback with manual transmission. When I turn the key to "on," the instrument lights come on and the seat belt bell dings as usual. When I turn the key to "start," the lights dim a little but nothing happens – no rev, no turning, no nothing.

Some background: I have owned this car for about six months. It's in relatively pristine condition for a 1991 vehicle: it only has 95,000 miles on it, all of them with a single owner who took immaculate care of the vehicle. (Even minor details are in surprisingly good condition.) In general, it has run perfectly in that time. I have had minor issues with starting it before, however. Sometimes it wouldn't start because the clutch pedal wasn't all the way down – there is a clutch interlock switch, so the starter won't start unless the pedal is down. By last week, this was happening pretty much every time I started it; but the solution was pretty simple: I picked up the pedal and tried to push it down a bit further to engage the switch, and it would start right up. But last week, I tried this and discovered that it didn't help – the car wouldn't start at all.

The story so far: At first I thought it must be that stupid little clutch interlock switch. I checked and found that it was indeed not being depressed when I pushed down the pedal. But – even when I depressed the switch, the car wouldn't start. So I thought maybe the switch was faulty; I ripped it out and just wired the two ends together. Still nothing. I thought maybe the switch broke rather than connected a circuit; so I tried disconnecting the two ends. Still nothing. Hm.

At this point I was starting to get puzzled about what was wrong, so I pulled out my service manual and got to thinking. For reference, here is the service manual's wiring diagram of the starting circuit. Since the thing isn't even cranking, I think something in this circuit must be bad.

My steps so far:

1. I started with the battery. The lights come on fine and everything, so it wasn't likely, but I pulled the battery and brought it to Autozone for testing / charging. But it has a full charge, and is good.

2. I checked all fuses and fusible links. All of them are good. I even replaced the clutch relay fusible link, just in case there was something wrong with it that I wasn't seeing. No dice - still nothing.

3. After that in the circuit is the ignition switch – so I thought about this. I'm pretty sure that the ignition switch is working, since the lights come on when it's at "on" and dim a little when I turn it further to "start." So I think that's not the problem.

4. Next is the clutch interlock relay. I have heard that those can go out pretty easily, so I spent the twenty bucks and replaced it. Still nothing.

5. Clutch interlock switch – I've ripped this out; it doesn't seem to be the problem.

6. The starter motor – I don't think this is it, since I'm not getting any noise at all when I try starting.

I've tried looking at the harnesses, too – they're in great condition, no visible shorts or anything.

Does anybody have any idea what might be going wrong here? I'll be around to answer any questions you might have, and I can post pictures of various things if you want. This is a real puzzle for me.

I also feel like the next thing to do must be to test this with a voltmeter. I've got one, but I don't really know how to go about this. My first impulse is to make sure electricity's getting to the starter; but I'm not sure where it's supposed to come from. I tried testing the main wire that goes into the starter, but it has a charge on it even before I turn the key – which seems weird.

(And I'd like to be able to go home for Christmas, so if you help me out I'll be eternally grateful. Thanks in advance!)

posted by koeselitz to travel & transportation (26 answers total)

Starter solenoid would be my first non-expert guess.

Wikipedia.

Wikihow testing.

Nissan Owners' Forum.
posted by soundguy99 at 2:04 PM on December 21, 2012 [3 favorites]

I have no way to check if the starter solenoid is clicking – I live alone, and I'm single. Do you know how I'd check the starter with a voltmeter?
posted by koeselitz at 2:15 PM on December 21, 2012

I guess I'll have to figure out how to test it some way, anyway. I'll google it and get to work on removing it from the vehicle.
posted by koeselitz at 2:21 PM on December 21, 2012

With my cars where this has happened, the causes were: solenoid, automatic seatbelt stupid thing thinking the seatbelt wasn't engaged, battery ground wire.
posted by jessamyn at 2:22 PM on December 21, 2012 [1 favorite]

(That last link is pretty stupendous, soundguy99 – don't know how I missed that in all my days and days of googling. Pictures and everything! I think that'll help.)
posted by koeselitz at 2:38 PM on December 21, 2012

Would it be worth making sure it'll at least push-start before spending more money?
posted by ROU_Xenophobe at 2:50 PM on December 21, 2012 [1 favorite]

Well, as the links soundguy99 gave me indicated, it should be possible to test the starter solenoid and starter motor without spending any more money. And even if it is the starter (which is my suspicion now) it's easy enough for me to replace a starter, and the part should only cost about a hundred dollars.

If the starter turns out to be okay, and charge is getting to it, there is a very, very distant possibility that the engine threw a rod or something and the car is worthless now. But that possibility honestly seems so distant to me at the moment (given the car's good condition and the care it's gotten) that I don't think it's the case at the moment.

Push-starting is tough, too. It's just me here – nobody to help. I might be able to get a neighbor to help, but I'd rather not unless it was really dire. If I push-started it now, it wouldn't help much anyway, I don't think. But if I can verify that the whole starter circuit is fine, then I might try that.
posted by koeselitz at 2:56 PM on December 21, 2012

Your wiring diagram shows one starter terminal should be hot (as 99+% of starters are) when the car is shut off. The other should be as grounded as grounded can be; if you have an ohmmeter check resistance there. In the diagram the pogo-stick-looking dumalfac right above the starter proper is the solenoid and that's what everyone's yelping about.

Old-skool: briefly short across the two solenoid terminals with a screwdriver. If the machine cranks then the diagnosis is finished and the solenoid's bad. This is not exactly fun and you have to be ready to make sure you can get the screwdriver out fast. Car MUST be in neutral, parking brake on, ignition OFF and study your route of egress so when there are sparks and you are surprised you do not bash your elbow into the thoughtfully placed alternator.
posted by jet_silver at 3:03 PM on December 21, 2012 [2 favorites]

Lights dimming says to me the solenoid is getting energized.

Which could mean the solenoid is stuck, and if so, it's stuck either all the way back in resting position or partly forward but not far enough forward to cause the starter motor to get energized.

I'd put it in neutral and take the emergency brake off, then move the car six inches or so by whatever means to cause the flywheel to move a tiny bit in case the solenoid-driven gear is bound against the flywheel. If it is, that might release it.

If that doesn't do the trick, I'd tap along the length of the solenoid with a wooden stick, such as a kitchen spoon, then try it a few time between tappings.
posted by jamjam at 3:07 PM on December 21, 2012

jet_silver: “Old-skool: briefly short across the two solenoid terminals with a screwdriver. If the machine cranks then the diagnosis is finished and the solenoid's bad. This is not exactly fun and you have to be ready to make sure you can get the screwdriver out fast. Car MUST be in neutral, parking brake on, ignition OFF and study your route of egress so when there are sparks and you are surprised you do not bash your elbow into the thoughtfully placed alternator.”

Okay, so I tried this. I attached an alligator clip test wire to the ignition cable contact on the solenoid, and then touched the other end of the test wire to the hot contact on the solenoid. Nothing but sparks. Does that mean the solenoid is good? Or does that just mean the solenoid's making contact but the cylinder inside is stuck, or something like that?

jamjam: “Which could mean the solenoid is stuck, and if so, it's stuck either all the way back in resting position or partly forward but not far enough forward to cause the starter motor to get energized.”

Yeah, I'm seeing that possibility now.

“I'd put it in neutral and take the emergency brake off, then move the car six inches or so by whatever means to cause the flywheel to move a tiny bit in case the solenoid-driven gear is bound against the flywheel. If it is, that might release it.”

Kind of already did this – I last parked this down the block; that's where it wouldn't start. To get it home and in my driveway I had to push it a ways. I think if it were a stuck flywheel and were going to come unstuck with pushing it around, it would have (especially when it was coming out of that pothole I hit and had to push it out of.)

“If that doesn't do the trick, I'd tap along the length of the solenoid with a wooden stick, such as a kitchen spoon, then try it a few time between tappings.”

Huh – didn't realize you could do that while you weren't trying to crank it. Well, I'll give that a go, anyway.
posted by koeselitz at 3:17 PM on December 21, 2012

Worn brushes in the starting motor? Sometimes the car starts because they're lined up just so, other times nothing. You can move them around and get more time before replacement, ime.
posted by fshgrl at 3:19 PM on December 21, 2012 [1 favorite]

At this point, I actually think I'm just going to go ahead and pull the starter and take it somewhere to get it tested.
posted by koeselitz at 3:19 PM on December 21, 2012 [1 favorite]

I'd put it in neutral and take the emergency brake off, then move the car six inches or so by whatever means to cause the flywheel to move a tiny bit in case the solenoid-driven gear is bound against the flywheel. If it is, that might release it.

Yeah, that won't work. You need to leave it in gear and get the car to move a little bit (use 4th gear) which will turn the engine a bit. Pushing it in neutral wont' move the engine at all.

Lights dimming says to me the solenoid is getting energized.

Concur. I'd try bumping the engine with the car in gear, then I'd try the starter shorting test (taking the car back out of gear as mentioned above). It needs the shortest of touches across the two terminals and will take a lot of current through whatever you use - a screwdriver is perfect (a wire will melt) and use something with a good plastic handle. If you tap it across the terminals it will spark and the starter should kick. If this works, it is the solenoid.

A faulty solenoid will produce the dimming and no turning of the starter, as will a faulty starter. The shorting out test will tell you if the solenoid or the starter is the issue.
posted by Brockles at 3:55 PM on December 21, 2012 [2 favorites]

Nothing but sparks. Does that mean the solenoid is good?

It means the starter is either jammed internally (ie pooched) or stuck oddly against the flywheel, in which case jogging the car in gear will move it around and give you the answer.
posted by Brockles at 4:42 PM on December 21, 2012 [1 favorite]

Hm. Well, I have one bolt of the starter off, and the other bolt is halfway off now. I think I'm just going to pull the starter/solenoid and take it in for testing.
posted by koeselitz at 5:07 PM on December 21, 2012

– if it tests okay, I'll assume it was a flywheel jam and bump the engine a bit before replacing it to see if that fixes it.
posted by koeselitz at 5:12 PM on December 21, 2012

Okay, so –

The starter tests okay. That doesn't necessarily mean it's 100% perfect, of course; they run a lot of power through those tests, so it's possible it's got enough juice to pop out and spin when free but not to budge when under load.

I'm starting to wonder if I did the jump test correctly, though.

Here is a picture of the contacts on the starter solenoid. On the right side is the smaller wire coming off; pretty sure that's the ignition wire. Then there are two contact bolts on the top and the bottom of the solenoid; the bottom one leads to the starter motor (you can see the cable coming off of it and leading into the starter motor casing) and the top one is the contact that bolts to the battery cable.

Now – when I tried jumping across the starter solenoid, what I was doing was connecting the top bolt that runs to the battery to the side wire that leads to the ignition. But now that I'm thinking about it, that seems wrong. I should have been jumping across from the live bolt to the starter bolt, shouldn't I?

Anyway, it's a bit cold outside to keep working. I'm going to put this off until tomorrow. In the morning, my plan is to throw it into fourth and roll the car a bit to jog the engine, then replace the starter and reconnect everything and see if it starts. If not, I guess I'll try jumping across the solenoid what I think is the correct way.

Still pretty puzzled, but we'll see. Somehow I feel like it's unlikely that the flywheel was jammed; my last car had a problem kind of like that (chips in the flywheel) so I inspected the flywheel and the starter gear pretty well when I pulled this one off, and I don't see much undue wear or grinding on either. Anything's possible, though.
posted by koeselitz at 6:25 PM on December 21, 2012

Look at the diagram of the solenoid (T-shaped thing just above the schematic picture of the starter motor proper). Turn it into a caduceus. The snake is the part that pulls the solenoid in. We're trying to -bypass- the solenoid. Therefore, you put a heavy piece of metal between the two big studs and ignore that little coil.

Starters pull hundreds of amps on the inrush. You need a big, heavy conductor - a screwdriver, or the grips of a pair of un-insulated pliers - to withstand that huge current. A test lead will not cut it at all. And sparks are normal when you're doing this, hence caution and being absolutely, positively sure you are not going to start the car in gear because you're bypassing each and every safety interlock in the starter circuit when you do this. You are creating a knife switch and when you want to switch enough current to do some welding, you open and close the contacts FAST.

However, the starter tests good. Without knowing what-all they did to test it, I would accept that and verify that the fuses denoted BR and G in the schematic are good; then I would connect a test light in series with E209 and verify it lights when you turn the key.

If all that is correct, I'd look at the weather. If you have had sub-freezing weather after rain, it may be the solenoid was frozen with water leaking into it and once removed (or the temperature went back up) your car will start - until the next bout of rain followed by snow.
posted by jet_silver at 8:11 PM on December 21, 2012

jet_silver: “Look at the diagram of the solenoid (T-shaped thing just above the schematic picture of the starter motor proper). Turn it into a caduceus. The snake is the part that pulls the solenoid in. We're trying to -bypass- the solenoid. Therefore, you put a heavy piece of metal between the two big studs and ignore that little coil.”

Tremendous thanks for explaining the wiring diagrams for me a bit – I don't think I'm terrible with cars, and I've replaced starters and even pulled open transmissions, but wiring diagrams confuse the crap out of me.

“Starters pull hundreds of amps on the inrush. You need a big, heavy conductor - a screwdriver, or the grips of a pair of un-insulated pliers - to withstand that huge current. A test lead will not cut it at all. And sparks are normal when you're doing this, hence caution and being absolutely, positively sure you are not going to start the car in gear because you're bypassing each and every safety interlock in the starter circuit when you do this. You are creating a knife switch and when you want to switch enough current to do some welding, you open and close the contacts FAST.”

Yeah, this sounds kinda scary, but I guess I can try. I've been using some non-conducting rubber gloves I have. Sounds like those plus a heavy screwdriver with an insulated handle would do it, as long as I have plenty of jump space behind me. And make sure it's in neutral.

“However, the starter tests good. Without knowing what-all they did to test it, I would accept that and verify that the fuses denoted BR and G in the schematic are good; then I would connect a test light in series with E209 and verify it lights when you turn the key. If all that is correct, I'd look at the weather. If you have had sub-freezing weather after rain, it may be the solenoid was frozen with water leaking into it and once removed (or the temperature went back up) your car will start - until the next bout of rain followed by snow.”

On the fuses: I've gone through all fuses and relays, unfortunately. I've even replaced one relay and the ignition switch fusible link just in case. I'll pick up a test light and try what you suggest – that seems like what I should have done from the first, but I guess it didn't occur to me that that was how to test the circuit. I was trying to measure this stuff using this, but I guess that's probably not the best tool for the job.

Thanks all for the help. I'll report back tomorrow with details on what happens.
posted by koeselitz at 9:14 PM on December 21, 2012

I'm going to go with the starter/solenoid not being good too. I had a '73 VW Beetle that did this same exact thing. Directly shorting the terminals on the starter/solenoid as other have described fixed the problem temporarily (like for 2-3 starts) but eventually it got to the point that it needed replacing.

Other possible things: Autozone simply misdiagnosed your battery being good and while it's putting out 12VDC it's lacking in amperage for starting.

One more thing - Can you "jump," "push off," or "roll off" your car? That is, turn the key to "on," put it in first, get it rolling with the clutch in and pop the clutch? Apparently this is not good for modern cars so you might want to ask someone else if it's okay to do, but it will get the engine going.
posted by Gev at 7:56 AM on December 22, 2012

Gev: "One more thing - Can you 'jump,' 'push off,' or 'roll off' your car? That is, turn the key to 'on,' put it in first, get it rolling with the clutch in and pop the clutch?"

As adventurous as push-starting it might sound with only one person, I'm afraid it'd be too dangerous. Not to mention probably impossible. Also, I stepped on a nail last night while working on this thing, so solutions which involve me pushing this thing around are pretty much out. Sorry.
posted by koeselitz at 9:09 AM on December 22, 2012

An update for anyone still following along: I tested the circuit. There is no charge going to the starter at all.

So there's a problem somewhere in the circuit. Somewhere. I'm left kind of frustrated and wondering what I could have missed. I guess I'll have to go over all the cables again. So weird.
posted by koeselitz at 12:12 PM on December 22, 2012

And...

I tried bridging the contacts on the back of the solenoid like we talked about. Interesting results. There are the usual sparks, which makes sense. However, I can't reliable get the starter to turn. And when it does turn, it does not turn the engine; I just get a whirring sound. So I guess the starter's bad after all?

So at this point, I guess there are two things wrong. The ignition isn't hitting the starter, and the starter seems to be bad. It seems pretty far-fetched that it's both. I might be testing one of these things in the wrong way.

I'm using a test light to test by clipping one end of the test light to the attached ignition wire on the solenoid, and grounding the other end of the test light to the body. Is that the right way? Also, am I right about the starter?
posted by koeselitz at 1:15 PM on December 22, 2012

Have you checked the ground part of the circuit? This is going to be a bit of a long shot and there are a couple ways to reduce the difficulty, but it is beginning to sound as though the ground side of these connections - usually there's a connection to the engine block or similar - might be giving way. Since -both- the starter is not engaging right -and- the connection from the ignition is not good, here are a few options:

1) connect everything up and honk the horn. Seems silly, but the back EMF from the horn coil will sometimes blast through corrosion in the electrical connections. (This is a temporary fix even if it works, but I did this three or four separate times with my first car and it was worth knowing about.)

2) check every bit of the circuit that starts with the negative terminal of the battery and ends at a fat wire bolted to the engine block or chassis. Then take a big, thick jumper cable, stick one end of it on the fat wire ground and clamp the other end as close as you can get it to the starter's mounting flange, where it's bolted to the block. If -that- doesn't do it, take the first end of the jumper cable and grab right on to the negative battery terminal.
posted by jet_silver at 1:58 PM on December 22, 2012 [1 favorite]

Huh. Yeah, that actually makes sense. Thinking about this – the Clutch Interlock Switch connects to body ground. I'm starting to think the problem is somewhere in the line around there.
posted by koeselitz at 2:42 PM on December 22, 2012

So – an update. My car is fixed! In short, it was the starter, although there was some complication. I'll give a bit of a rundown now of the whole thing here, just in case it helps someone else who encounters a similar problem:

This whole thing started with that darn clutch interlock switch. As noted above, the clutch interlock switch is a switch under the clutch pedal; when the clutch pedal is depressed, the switch goes down, and the starting circuit closes so that the starter can engage.

Now – for the past few months, I've had trouble with that switch. Sometimes when I push the clutch down, it still wouldn't start, so I'd push it down a little harder. This is something I should have checked out, but I realize what was happening; the switch was being pushed ever further away from the pedal, so that it was not always firmly being pushed down, and the circuit was snapping open and closed intermittently while the clutch pedal was pushed down. What I now realize is that this intermittent signal was putting a strain on the starter solenoid, finally shorting it out.

But – there was also another problem, and that was the way I tried to fix it.

See, I have heard bad things about the clutch interlock relay, a relay that sits before the clutch interlock switch in the starting circuit. I'd heard that it is notorious for going bad, so I figured I may as well replace it just in case. In the process, I replaced all ignition-related and starter-related relays. But the other day I realized that that new clutch interlock relay I put in wasn't right! You can see a side-by-side picture of the two relays here. If you look at the pins I've circled in red, you'll see that the ones on the new relay are just slightly off-kilter. So when I pushed in that relay, it wasn't connecting.

So that's what caused my problem where I wasn't even getting charge to the starter. My lesson learned here: when something electrical is wrong, don't just go replacing things in the hopes that it'll work.

I figured this out by doing what I should have been doing from the beginning. It's actually pretty simple. I went and got the battery fully charged up. Then, looking at that starting circuit diagram, I started from the battery and followed the circuit to the ignition fuse in the fusebox, then to the interlock relay, then to the clutch interlock switch, then to the starter motor. At each juncture, I'd unplug the circuit and insert my voltmeter with the red wire toward the positive pole on the battery. I checked the fuse in the fusebox first, made sure it was good, and then removed it and replaced it with the voltmeter. When I saw that electricity was passing through the fusebox, I moved on past the ignition (which is hard to get at to test) to the clutch interlock relay. I pulled the relay out and noticed it wasn't really fitting. So I replaced that relay with my voltmeter, set the voltmeter where I could see it from inside the car, and turned the key; I could see that the charge was going through there, so I tested the old relay itself and found that it was good.

When I finally got to the starter motor and was getting charge there, I realized that must clearly be the problem. So I replaced it, and it works great.

Again, thanks for all the help, all. It was extremely useful to be able to bounce ideas off of you all.
posted by koeselitz at 4:22 PM on January 11