HondaStufff

Started off having intermittent stalling, typically on hot days during the summer, and hard starting. Those symptoms pointed me straight to the PGM-FI relay. I resoldered the joints and the problem seemlingly went away, but then happened again a couple times over about 6 months. Finally, the car just recently starting running poorly and the CEL came on. Got it home, but now it won't even start. I checked the codes and got P0123 and PO118, throttle position sensor and the coolant temp sensor. I thought it odd to get both codes at once, went on to check the sensors, and both are nominal. Nice gradual sweep of resistance change on the TPS, and reading the correct value on the thermistor for the temperature outside. Did not confirm that the resistance changed as temperature did (as the car won't start) but it would be highly unlikely that it would fail on the correct reading, it would either be shorted or open.

I saw some people had issues with the ground at the thermostat housing, and both of these sensors share that ground. Unfortunately the problem wasn't that simple. Redid the ground, (which looked fine originally) with no success.

The repair manual at this point leads me to inspect the ECU/PCM. I have fixed capacitor issues and installed chip sockets on other ECUs, so somewhat know my way around a PCB. There were no bulging capacitors, or anything else looking amiss at first glace at the ECU.

Unplugging the ECU to check it reset it, and I no longer get any CEL when trying to start the car. It just turns over with the occasional stumble, but no fire. I have confirmed spark with an inline tester, and confirmed fuel pressure at the fuel filter.

Does anyone have any suggestions at this point other than trying a good working ECU? (I don't know anybody with one). I would like to be able to at least confirm it's the ECU before buying a new one, in case it is not. If it's possible to diagnose a problem with the ECU with this forum's help, I may also just be able to replace the defective component and not have to buy a new one, that would be ideal.

Thanks in advance for you help.

muellersfan

Crank engine for 30 seconds. Do any codes return?

Did you Ohm test the 2-wire ECT sensor under the distributor?

HondaStufff

Just went out and cranked it over for a while, no codes.

Ohm checked the ECT, and read around 4K, which according to this chart and the temp outside (currently around 10C), is normal.

https://www.google.ca/search?q=civic+ECT+resistance+chart&rlz=1C1CHBF_en CA711CA711&source=lnms&tbm=isch&sa=X&ved=0ahUKEwix xsas1arfAhWjjlQKHSfRAEEQ_AUIDigB&biw=1745&bih=881# imgrc=ewor-byjmgd3VM:

I can probably confirm that it goes down with a little torch on the area, or by taking it out to heat up, but I think it's highly unlikely it's failed at 4K, the current correct reading, and not either open or shorted. Even if the reading goes wonky as it heats up, it should still start fine, as it's reading correctly right now.

I ohm checked the TPS too, and it operated like you would expect a potentiometer to. Nice and smooth, no dead spots.

muellersfan

What was the ambient temp during the test? The symptoms are a good fit with a bad ECT sensor.

HondaStufff

Sorry, I edited my post too late, around 10C ambient.

muellersfan

Will the engine start if you unplug the ECT or the TPS or both?

If not, check for bright spark at all 4 plugs and for whether the fuel pump primes.

HondaStufff

Dark and raining now, I will try tomorrow. The light on the inline spark tester goes off on all 4 cylinders, but I will take them out and check for bright spark too. Fuel pump definitely primes. I hear the PGM-FI relay click, I hear the fuel pump turn on for a couple seconds and then off. Just to ensure fuel delivery I loosened the bolt on top of the fuel filter and it sprayed out fuel during priming.

muellersfan

If the engine won't start with the sensor unplug tests, then test whether the fuel injectors spray fuel and whether the cylinders have similar and good compression for a cold engine.

HondaStufff

Didn't start with the sensor unplug test, though the CEL came back on. Need a partner to crank it over while I listen for the injectors to fire. When I had the plugs out to look for spark, I left the injectors plugged in, and I could smell fuel in each cylinder though. Yesterday was just cranking over with very occasional stumble. Today it sounded like it really wanted to start a couple of times, but didn't. (whether sensors unplugged or not). All plugs have a nice bright spark.

I tried to compression test this morning, but the spark plug wells are so deep, I need to get an extension. Just trying to tighten by turning the rubber hose down a hole that deep isn't making a proper seal. I'm only reading about 80psi on each cylinder, and the car was running well with good power before hand, so I don't think that's possible. I'll have to report back about compression after I get an extension. Currently lending my vehicle to the girlfriend, whose car I'm trying to fix, so I'll have to wait until the end of the day.

If it's not a bad seal from the compression tester, then maybe the timing belt jumped? The CEL for TPS and ECT still seem weird though, given that they appear fine.

Thanks for your help so far muellersfan, I appreciate it. Hopefully I will have some proper compression numbers later today or tomorrow.

muellersfan

As an alternative to the fuel injector test, you could spray starter fluid into the throttle body to see if the engine starts.

HondaStufff

I guess I was just not ready to accept that the compression was so bad. I took off the belt cover, checked the timing marks and everything lines up. Last time I drove the car it felt fine. Charged up the battery and did a proper dry/wet compression test. from 1 to 4 dry it's 90-90-60-120, wet it's 90-100-70-150. Definitely low and not very consistent. Could this be a head gasket? I'm guessing not, or it'd probably just be one or two of the cylinders low.

Tried some starter spray and got no love either.

Is this compression low enough to stop the car from starting entirely? When the CEL first came on, the car jerked, like a quick fuel cut, then went back to normal. The weird code behaviour from the ECU and the fact that it seems to be driving fine recently still has me thinking it might be that, but with compression so bad I dunno if it's worth it.

muellersfan

eghatch9295

iTrader: (1)

eek. that engine is very sleepy.

avisitor

Long shot ...
Try following the wire or wire bundle from the TPS and/or ECT
Look for it rubbing against anything hot or metal
Or if rodents are chewing on them
Have seen where the O2 sensor was shorted from underneath
Check fuses too.

Good luck

HondaStufff

Looks like a head gasket to me, likely from an overheat. What do you guys think?

Cyl 1 is leaking out of the sparkplug holes of cyl 2 and cyl 3

Cyl 2 is mainly leaking out of cyl 4

Cyl 3 is mainly leaking to 1

Cyl 4 has leaks to cyl 3 and 2

I can hear a bit from the crank case, but not much, and nothing anywhere else. I took the coolant overflow cap off to listen in there, and found it was empty. With all of the cylinders leaking into each other, I'm figuring that the head has probably warped.

What else should I be looking out for that could have been damaged from this overheat?

chrysler kid

The most common reason to overheat a d-series is a coolant leak from the heater hoses. Typically the u shaped hose from the thermostat bypass.

The radiators are prone to crack when they become a light brown color

The cooling fan could have failed

So replace cylinder head, head gasket, head bolts, timing belt, water pump, coolant, thermostat, all coolant hoses, crank seal, cam seal, valve cover gasket and you should be up and running again

If the car is trashed inside and outside you may just want to push it to a junk yard and start over with another running car in better condition. At 23 years old these cars need love because theyve always been neglected

HondaStufff

Car has new tires, new rad, a newly rebuilt alt, and only 230,000 km (140k miles), so I'd like to get a little more service out of it. It also hasn't been molested, completely stock. That said, it wouldn't be worth it to replace a bunch of stuff that isn't broken, so I won't be doing a replace everything job. I did get a coolant temp sensor code, and though I'm getting a proper reading for ambient temp outside, I will now investigate that further, and look for other possible causes of an overheat. It could have been the headgasket itself, or even just the thermostat. Unless I find evidence that it's the water pump, I will likely do a quick head pull, tie the timing belt onto the cam sprocket and rig something up to hold tension on it, rather than go to a bunch of trouble I don't have to.

HondaStufff

Got the head off today. Looks like it's definitely blown between the cylinders. No water in the oil or visa versa though. Block and head still full of coolant too, so likely not a leak, maybe the thermostat froze closed? I'll be checking eventually.

I'm sure the head is warped, but I can't tell by how much until I get a precision ruler. The one 18 inch ruler I have, I discovered today, is junk. I set it on my the ways of my lathe and it's obviously quite bowed.

Might be new thread time, I've got some questions about the head now, I'll try here first, but it's getting a little off topic from the original post now, which is at this point looks like a separate issue.

As previously mentioned, I'd like to keep this on the cheap, the car isn't worth it otherwise, but now with the time I've invested, I'd definitely still like to get it running again. It's been interesting anyway, haven't worked on a car since my DSM over a decade ago, and I've never taken a head off before. I'm planning on reusing the head bolts, since from what I have found, they are not torque to yield. I'm now contemplating whether I should take the head into a machine shop. I'll call around and get some pricing. The milling machine I own is big enough to do the head, but I would need to fashion up a jig to hold it, and I'm also wondering how much the warp might have affected the valve seating. Anybody have experience with this kind of thing and can give me some pointers?

muellersfan

If you plan to proceed with this engine, mill both the head and block mating surfaces perfectly flat. Otherwise, you will likely be wasting both your time and money.

HondaStufff

Milling the head is no problem, I will either be making a jig and doing it myself or taking it to a machine shop, as mentioned. The block, on the other hand, is easier said than done. For obvious reasons, it won't be getting milled. If it's very bad I will have to hand lap it. From the reading I've been doing though, it's much less likely for the block to warp than the head.

muellersfan

Aluminum engine -- If the head is obviously warped, the block is highly likely to be same. Check it with your precision straight edge.

HondaStufff

I need to get all the valves shut to resurface the head, and check the valves for leaks. I'm looking at the online shop manual, for removing the rocker assembly, and it says to loosen all of the adjustment screws first. Does this mean just the nuts around the screw, or do I need to back the screws right off too, and therefore have to readjust the lash on all the valves?

I've never done this before. The only valves that are open, and with pressure on the rockers are the intake valves on cyl 3 and the exhaust on cyl 2. If I do have to loosen the adjustment screws first, can I just do it on these valves that have pressure on the springs so I only have to reset the lash on those ones, and not all of them? Is it even necessary to back off the screws and release the tension? The only place I've seen this is in the shop manual, nobody seems to mention it in forum posts about removing the rocker assembly.

https://archive.org/details/HondaCiv...2000/page/n139

DaX

iTrader: (1)

This. Don't mess around - get the block milled too. The added time and cost of doing it right the first time far outweighs the risk of having to do it all over again.

Yes, back off the nuts then run the valve adjustment screws all the way back. This will reduce the twisting force on the rocker assembly as you unbolt it. You will need to readjust your valve clearances once you reassemble the head. My cylinder head shop charges $25 for pressure test and $35 for resurfacing. Add $10 for some Supertech valve stem seals, and you're in for $60 to freshen up that head unless you have to replace valves. If your head needs a valve job, it should be about $125 more.

HondaStufff

Back up and running again - sort of. For me pulling the entire engine just to get it milled would have been a deal breaker for the entire car. I opted to measure the flatness rather than just pull it and have it milled. I had to get an 18" precision straight edge, 12" being to small to go corner to corner, and 24" too big to fit into the engine bay. I already have a 6" machinist square I used to check the shorter angles. Block was well within spec, so no problems there. Put the new heatgasket in and checked the compression again, 180-185 now across the board. I cheaped out and tried to reuse the exhaust gasket though, didn't work and is leaking. I knew the risk on that one, but it's like a 10 minute job so I thought I'd try it.

So it's running great now, or does so as it's heating up anyway. Gauge wasn't quite reading running temp yet, and the engine began to stumble, did it a couple times and I shut it down. feeling around, it didn't seem too hot, so I'm not quite sure what that's about it, but it was late and dark so I packed it in. I'm going to get a new exhaust manifold gasket and change out the thermostat, since it's easy enough and cheap. I'll test the old one too, hopefully I find that it's faulty so I'll know. I have now ruled out the ECT completely. I put it in some water with a thermocouple and checked the resistance as I heated it, traced the graph perfectly.

Any ideas on how to proceed from here? How likely is it that the small exhaust leak at the manifold is causing the stumble from messing with the o2 readings?

HondaStufff

Might be finally getting back to the original problem in the topic now. Seems I'm dealing with a couple unrelated issues at the same time, and have one now sorted at least. Got the new exhaust manifold gasket in, and the new thermostat. No more exhaust leak and I found the culprit that caused the overheat. Tested the old thermostat after I took it out, and it was seized closed.

I guess it wasn't the exhaust leak messing with the o2 readings. I still can't get the car up to operating temp, runs great at first, but as soon as it starts to heatup, it threatens to die, and then does. Starts up (a little hard) and runs fine again for a few moments before it starts wanting to die again. You can keep it going by playing with the gas pedal, but after I did that it was surging back and forth. No CEL on.