|
Something over a year ago I found a post on CL for a 1991 Acura Integra a block over from my house for $1000. I figured what the hell I'll check it out. I found a fairly straight, if dirty but thankfully unmolested car (it even has a stock intake). It's from Seattle and he had driven here (St. Louis) roughly six months prior and parked due to overheating problems. Along the trip here he had shoved a bottle of Bar's in the radiator, then replaced the radiator, but it still had problems overheating so the he bought a new car and parked the Integra. We negotiated a bit and settled at $800 but then I found out he didn't have the title. So I told him to call me when he has the title. Eight months later I see him pulling in his driveway, the Integra is still back there. So I pull in and ask what's up with the car. He says he's moving back to Seattle soon and needs to offload it quick, $300 and he'll get the title to me. Seeing no-way I can lose, I shove $300 in his pocket, jump the dirty little bastard and drive it home. From there I found a failed thermostat and replaced it. Overheating is fixed. Six months go by, no title. I find a new owner for the car, it's a friend who's down on her luck. I drive the car to her place since I'm running out of room. I plan on giving the car to her if I can ever get title. It finally shows up and I'm ready to get to work fixing the rest of the car. I know it's gonna need a muffler and a windshield to pass inspection, but I figure let's get it an inspection first so we know everything up front. So now this car has been sitting something approaching two years, it won't start. No spark. I've tested the coil is receiving 12v and the resistance across it checks out at .8ohms. I'm going to replace the cap, rotor and wires, since they're looking pretty ugly. But I've got some other curious findings that's making me think something else is up. The big one is that the CEL does not illuminate, at all, ever. Maybe a blown bulb, but I think it's also maybe a sign of a blown ECU. With all the sitting and subsequent jumping this car has seen, I wouldn't be too surprised I guess. The other curious finding is that the ignition switch itself is acting strangely. Sometimes it doesn't want to let go of the key, other times that part works just dandy, but the shift interlock system (that prevents the car from leaving park unless the brake is applied) fails to work. You can't shift it out of park no matter what. I pulled the main relay and inspect it, tested it as best I know how for continuity between the traces, I think it's ok. I don't hear (or feel it if I put my hand on it) it clicking on when I turn the key though. Another curious thing, I put my multimeter on the TPS leads and on key-on I'm finding 12v instead of 5v like I think I should. This again maybe points to a fried ECU. I thought all the sensors ran on 5v from the ECU. Anyway, that's about where I'm at with the car. I'm really trying to avoid a throw parts at it to fix approach as my friend's funds are very limited, but I'm at the point where I would normally throw parts at it cause I'm out of ideas and testing know-how.
|
| # ? Mar 20, 2013 14:39 |
|
|
|
| # ? May 20, 2013 13:18 |
|
|
Yeah holy poo poo there should not be 12v going to the TPS. It should be about half a volt when closed and 4.5 when WOT. You've either got a fried ECU or a major short somewhere. Edit: it might be a reference but those are usually 5V. Did you check around for one that was around 0.5V closed? Edit2: after some brushing up looks like the reference should indeed be 5V, and the center wire should be variable with the throttle plate, between 0.5V closed and 4.5V WOT. If you're legitimately getting 12V from the reference wire on the TPS while doing a voltage drop test then that's an issue. Adiabatic fucked around with this message at Mar 20, 2013 around 14:57 |
| # ? Mar 20, 2013 14:49 |
|
|
12v on 5v lines doesn't say good things to me either (and I'm not sure why there would be a 12v reference line for a 5v sensor). Are the ECUs really that fragile on these Hondas? I've seen GM cars (with body control units too, no less) take a backwards jump and run fine afterward. How does the ignition switch/key release work? Not letting go of the key might just be a poorly cut key or a previous theft attempt or something.
|
| # ? Mar 20, 2013 15:03 |
|
|
Adiabatic posted:Yeah holy poo poo there should not be 12v going to the TPS. It should be about half a volt when closed and 4.5 when WOT. You've either got a fried ECU or a major short somewhere. Can you sanity check me here. I'm good enough with electrics to be dangerous. To achieve the results I did in testing, I pulled the plug from the TPS and found three leads inside. I stuck the red lead from my multimeter in each one with the black lead on the battery's negative terminal. Two of the leads inside the plug gave me 12v. Another one gave me next to nothing. Did I test this correctly?
|
| # ? Mar 20, 2013 15:06 |
|
|
I had similar issue with my sister's '91 Integra LS. Sometimes it would start, sometimes it wouldn't, and sometimes the key would get stuck in the ignition. I've no idea about your 12V signal though. We replaced the EFI Main Relay. Evidently the solder is prone to cracking on the old relays. http://www.marklamond.co.uk/tech-ho.../main-relay.htm edit: hurf, I just read that you tested the main relay.
|
| # ? Mar 20, 2013 15:08 |
|
|
Yeah, I would replace the main relay before anything else since it's a Honda. I just sort of assumed you had swapped it for a good one while testing it.
Seat Safety Switch fucked around with this message at Mar 20, 2013 around 15:11 |
| # ? Mar 20, 2013 15:09 |
|
|
Seat Safety Switch posted:Yeah, I would replace the main relay before anything else since it's a Honda. I just sort of assumed you had swapped it for a good one while testing it. I didn't think a main bad main relay would cause a no-spark condition though. I thought it was fuel only. On key-on I'm also getting 12v at the coil. The main relay is not clicking though on key-on, so maybe. I thought it should give a small audible click. I keep going back to that ignition switch, if it's all hosed up, maybe it's not energizing the main relay or the ECU or both. trouser chili fucked around with this message at Mar 20, 2013 around 15:16 |
| # ? Mar 20, 2013 15:14 |
|
|
How much would a junk yard ECU cost near you? I don't know about your yards, but the ones here you can return some parts if you don't need them or if they don't work. Grab a yard ECU and if it's not the issue return it. You could grab an ignition switch while you're at it too. Worst case if you can't return any parts sell them on CL.
|
| # ? Mar 20, 2013 15:14 |
|
|
leica posted:How much would a junk yard ECU cost near you? I don't know about your yards, but the ones here you can return some parts if you don't need them or if they don't work. Grab a yard ECU and if it's not the issue return it. You could grab an ignition switch while you're at it too. In my part of the world, junk-yard Integras are rare and typically raped of parts when you do find them. I could grab an ECU off ebay though, they typically run about $100.
|
| # ? Mar 20, 2013 15:17 |
|
|
Oh yeah, one more thing. Sometimes when the key is turned the D3 light on the dashboard flashes. It seems to flash twice then go out. This is the light that normally tells you the auto transmission is shifted into third.
|
| # ? Mar 20, 2013 15:23 |
|
|
When you drove it to its new resting place, was it generally a normally-operating car? Anything weird about how it shifted?
|
| # ? Mar 20, 2013 15:28 |
|
|
trouser chili posted:Can you sanity check me here. I'm good enough with electrics to be dangerous. To achieve the results I did in testing, I pulled the plug from the TPS and found three leads inside. I stuck the red lead from my multimeter in each one with the black lead on the battery's negative terminal. Two of the leads inside the plug gave me 12v. Another one gave me next to nothing. Did I test this correctly? I would backprobe it instead. Here's a wiring diagram: http://imgur.com/G289cR7 The green/white wire is ground, so stick the negative probe into the wire with the TPS plugged in. Use the positive probe on the other two. The yellow/white is your 5V reference and the red/blue is the signal wire that should adjust from 0.5V to 4.5V. You can check it with the plug unplugged by probing the green/white with negative and the yellow/white with positive, but the reference wire won't show anything since the TPS isn't plugged in. If you still get 12V I'd test the green/white and yellow/white pinouts on the ECU. If they're 12V I'd imagine your ECU is hosed. Either way these connections don't travel through the main relay; they're supplied directly from the ECU. Edit: KOEO for the entire process, by the way. Adiabatic fucked around with this message at Mar 20, 2013 around 15:34 |
| # ? Mar 20, 2013 15:32 |
|
|
Seat Safety Switch posted:When you drove it to its new resting place, was it generally a normally-operating car? Anything weird about how it shifted? Ran fine. It was a 5 mile drive tops with a couple of stops. It's loud since the muffler is busted open, but other than that it ran like a champ, shifted fine, likes to be run hard like a good Integra should.
|
| # ? Mar 20, 2013 15:41 |
|
|
Easy one to try. The ECU ground is the black wire and eyelet that bolts to the top of the thermostat housing. If its sat that long it may have corroded to much to ground properly. That would explain some of your weirdness. Beyond that I would try and find out why the cel wont light up first. The main relay powers the fuel pump and the ecu. Run a jumper to the main relay and force feed it 12v on the output for the ecu, see if the cel lights up. If it does it's probably the relay, maybe the ignition cylinder. I have only seen Honda ecus die from bad wiring so far, but it doesn't mean it's not possible. Getting older every day.
|
| # ? Mar 20, 2013 15:45 |
|
|
Thanks for all the info, guys. I'm off to go hit the fucker with a hammer some more. Updates later.
|
| # ? Mar 20, 2013 21:45 |
|
|
I'm going to agree with Adiabatic on his shortcut. I just double checked the diagram and description and the TPS sensor is indeed supposed to be getting 5V. So to rule out a short in the wiring, check the ECU pinout. If you're getting 12V out of the ECU pin that is supposed to be giving 5V, then the ECU has to be hosed. So probe pins C13 and C12. That's 5V power and ground. If you're getting 12V, ECU is toast.
|
| # ? Mar 20, 2013 21:54 |
|
|
gently caress yeah, it's running again! I got hung up with work the last night and couldn't make it. Tonight I was able and I replaced the dist cap, wires and plugs. The rotor wasn't available so I didn't do that. I also bought another main relay. But it still wouldn't start. I tried to back probe the TPS, but was largely unsuccessful due to my sausage fingers and the proximity to the firewall. So then I went to check the ECU ground strap at the thermostat housing and couldn't find it anywhere. Mashing my hand about produced an audible clicking and an electrical sound, like a pump running. I checked and the key was on, so I knew I was close. Well I found the ground strap eventually, it was completely disconnected. Apparently I failed to hook it up when I replaced the thermostat. How I was able to test drive it, idle it in the driveway for 40 minutes, then drive it 5 plus miles over to my friends house like that, I'm not sure, but there you go. Once the car warmed up and came out of open loop, the idle went lovely, revving and falling. I found a vacuum leak and fixed it. It got happy, nice smooth 800rpm idle. Then it overheated while I was finishing putting everything back together. gently caress. So I'm back at overheating. The car has a new radiator, a new thermostat and fresh coolant. I have tested for a blown head gasket by checking the exhaust for excessive smoke, and that special fluid that turns purple if there is exhaust gasses in your cooling system. Both are negative for head gasket failure. The coolant is green and the oil honey brown. When I replace the thermostat I boiled the old one in water and it never opened. At idle the temp will steadily march upward. Once this starts if you rev the motor to 2500rpm the temp spikes upward for a second or two, then drops back down like a rock and holds steady below the halfway mark. Come off the gas and let it idle again and it'll remain steady for a minute, then slowly start marching back up. I'm guessing this has to be a coolant flow issue, and I'm really hoping it's an air bubble in the system. When I tested this car after replacing the thermostat, I let it idle in my driveway for 40 minutes without problem. It was facing uphill and I initially left the radiator cap open in attempt to burp any air from the system. At my friends house it was facing downhill. I'm guessing that has something to do with it. Does anyone know a better way to burp this motor? Cause my only other thought is that it'll need a timing belt/water pump job and I really don't want to do that.
|
| # ? Mar 22, 2013 02:51 |
|
|
Yeah I was gonna say only thing left really is the water pump. It's not that bad of a job is it? If you use one of those cam locker doohickies it makes the job easier.
|
| # ? Mar 22, 2013 02:59 |
|
|
trouser chili posted:gently caress yeah, it's running again! I got hung up with work the last night and couldn't make it. Tonight I was able and I replaced the dist cap, wires and plugs. The rotor wasn't available so I didn't do that. I also bought another main relay. But it still wouldn't start. On my Prelude, there's a bleed screw located somewhere near / on the thermostat neck IIRC that you have to open to bleed it properly. Nose up / rad cap off won't cut it with that car. Maybe yours is similar? The temp spike / drop to normal levels is exactly what happened when the Prelude had a bubble in it. Bleeding it properly took care of it for good.
|
| # ? Mar 22, 2013 03:17 |
|
|
For what it's worth, if you're still running into the park/shift interlock bit, it's one of the most common issues on 2nd gen Integras. The shift position switch gets gummed up - it's under the shift console. It can also cause the D3 light to flash, if it's not fully engaging D4 or D3 (electrically). A flaky VSS (on the back of the cluster on the G2) will also cause D3 to flash, and cause erratic shifting (cruise relies on the same VSS, if the cruise works try using it... if it wanders, the VSS is bad). A bad ground at the trans will also cause a flashing D3, as will a bad shift solenoid (if it's a solenoid, it'll probably jump into 2nd or 3rd and stay there until the car is shut off). The temp issue is either air in the system, a bad water pump, or the head gasket. IIRC, the B18 has a bleeder screw on the thermostat housing.
|
| # ? Mar 22, 2013 04:01 |
|
|
A bit of googling seems to indicate that there is a bleed screw on the older Integras.
|
| # ? Mar 22, 2013 19:21 |
|
|
If it's been driven extensively with overheating issues, I suspect you might be eating coolant due to an internal headgasket leak which may be why you aren't seeing contamination? edit: or the waterpump is toast, as has been mentioned before. timing belts aren't terrible, but will take 4-5 hours so you may as well get a belt kit and have a known good thing for a couple hundred and a few hours? ssjonizuka fucked around with this message at Mar 22, 2013 around 19:25 |
| # ? Mar 22, 2013 19:23 |
|
|
ssjonizuka posted:If it's been driven extensively with overheating issues, I suspect you might be eating coolant due to an internal headgasket leak which may be why you aren't seeing contamination? I've done a ton of timing belt jobs, I can do them just fine, although this car will be a bit more a pain in the rear end since it's an automatic (I don't have the tool to hold the crankshaft still). I just don't really have the time, and she doesn't really have the money. Regardless, if this is what is needed, this is what will be done. If has a headgasket leak I should find it in the exhaust, the oil, or the coolant. I don't understand where-else it could be.
|
| # ? Mar 22, 2013 19:54 |
|
|
This is the cam lock I use, it's awesome and paid for itself the first time I used it. Good tool to have. http://www.amazon.com/Lisle-36880-D...s=camshaft+lock [edit] it only works though if the cam sprockets have a "lip" leica fucked around with this message at Mar 23, 2013 around 00:51 |
| # ? Mar 23, 2013 00:48 |
|
|
trouser chili posted:I've done a ton of timing belt jobs, I can do them just fine, although this car will be a bit more a pain in the rear end since it's an automatic (I don't have the tool to hold the crankshaft still). I just don't really have the time, and she doesn't really have the money. Regardless, if this is what is needed, this is what will be done. If has a headgasket leak I should find it in the exhaust, the oil, or the coolant. I don't understand where-else it could be. You could always do the old "cut the old belt in half and swap the new belt on halfway, then cut the old belt" trick.
|
| # ? Mar 23, 2013 01:12 |
|
|
Ok, I bled the system and it seems much happier. At first I pulled the water jacket plug by the exhaust manifold, it was hard to get to and it just dumped a gallon of coolant. Then I found the bleeder where the upper hose connects to the block. One of those, duh I really am that stupid moments. So i refilled the system and after watching the bleeder spit bubbles forever before running clean, I closed it up. The engine was good and hot now and I left it running in gear with the ebrake on for about 30 minutes. Never ran hot and the needle stayed rock steady. Looks like it's good to go, but I noted constant bubbles in the expansion tank. Which makes me worry about the headgasket more. Perhaps I'll test for exhaust gasses in the coolant again.
|
| # ? Mar 23, 2013 02:36 |
|
|
I've seen the same symptom on really old coolant. The glycol breaks down and you get a jellyfish circulating in the system. Even after trying to bleed the air, the heater would blow cold then hot, then the temp would spike, then it would go back to normal etc. A good flush and fill fix'd the problem.
|
| # ? Mar 23, 2013 03:45 |
|
|
I couldn't quite parse this out -- how long was it sitting at your friend's waiting for the title?
|
| # ? Mar 23, 2013 08:03 |
|
|
|
| # ? May 20, 2013 13:18 |
|
|
kimbo305 posted:I couldn't quite parse this out -- how long was it sitting at your friend's waiting for the title? About six months.
|
| # ? Mar 23, 2013 12:05 |
|
