Yeah, all new plugs and not change. The old one were worn, but looked good for the wear. I run mostly the cheap stuff, but not the E85. On this trip I tried higher octanes for a few tanks, but no difference. I replaced the O2 sensor too. On a warm day, the problem isn't there.

I'm tempted to buy that newer computer and see if it (not because it's the better one) fixes it. It's weird, when it's below freezing and the turbo gauge is low in the vacuum side, it's very prone to misfire. So cruising at 60mph, it will do it a lot, but at 75 to 80 it doesn't seem to as prone because of the minimal vacuum. Also, in the boost range well below freezing the car just feels like it's rough. It's like you can feel the each cylinder is putting out different levels of power. I'm thinking the the computer might have a blank spot in it's table somehow.

I'm also tempted to buy this car (Car Classifieds for Utah, Idaho, and Wyoming | ksl.com) for 800 and swap to LH2.4 conversion. I'd also take the windshield, the front seats, driver's side back door and the cupholder armrest, but then what do I do with the rest of the car?

 

odd, that car is advertised as a 960, but it sure looks like the odd duckling 1990 760 turbo. that was something of transition year, as the 940 came out in 1991... there's a moderate amount of one-of stuff on those 1990 models. there was no 960 in 91, the 760 turbo morphed into the 940SE as the whiteblock inline 6 wasn't yet ready for the 960.

 

Yeah agreed. I think he mislabeled it. The back pic looks like a 760 turbo badge. Those white mirrors are unique.

 

One other thing to note. I closed off the "warmed" air intake a few years back. Now I only get cold air. I don't think that this would be a likely to cause a problem in any way, but I'm going to throw that one out there. I've been reading a bit about the computer. It seems that generally, they fail and kill give you no start, with a few exceptions. I'm thinking that I have to either tear into the wiring harness or get another fuel injection computer. Should I suspect the ignition computer as well? Anyone had experience of know of a place with failed ignition computers?

 

closing off the heat riser thermostat in the airbox should have zero impact except maybe in very very cold weather the car will take a bit longer to warm up.

I believe you said earlier, your idle is steady until the car has warmed up a few minutes... during that initial start period, the ECU is running open loop until the temp sensor says its warmed up, and its not paying any attention to the O2 sensor, just running a bit rich off a fixed map. then the car switches to the lambda closed loop operation when its managing the mixture based on the O2 sensor. also during the closed loop warmup period, I believe the idle is boosted a little, closer to 1000 rpm instead of the normal warmed up 700 rpm.

I'd be really curious what readings you get off the O2 sensor when the idle is hunting around and the engine is missing. I've only done this on a KE-Jet mercedes 300E, but you hook your volt meter up to the signal lead on the O2 sensor and to ground... the signal ideally is swinging between about 0.2V and 0.7-9V several times a second quite rhythmically, and this cycle speeds up if the engine is running at a higher RPM. on the KE-Jet, this is used to modulate the 'EHA' on the side of the fuel distributor assembly, which tweaks the mixture richer/leaner, while on a LH system, this would just modulate the duty cycle of the injectors, which is much harder to see/measure.

 

Finally an update. I have never had school keep me so busy. I put a multimeter to the O2 sensor. It oscillates below 1 volt like normal and at a descent rate. Then I put an oscilloscope on it and drove around. That was cool. It would be interesting to compare with other Volvos to know exactly what I'm looking at. It showed the same oscillations. When idling it oscillated with my hunting idle. Then when running it would hover near the middle until engine breaking. Then it would drop to the bottom end (low voltage, like .1 ish). I think that's the lean side (please correct me if I'm wrong). It seems pretty indicative of a vacuum leak, especially somewhere after the throttle plate. I did not have a misfire occur that I could monitor while driving the car. I think that will require a passenger to look at the scope while I drive for a longer distance. It may be time to pay someone to do a smoke machine test on the car.

 

So I put another fuel management computer in, it behaves the same, so that rules that baby out. I'm going to see if I can a smoke test on the intake system. I need to make a longer run with scope attached and see exactly what the O2 sensor reads when the car misfires.

 

So I had the car smoked. The intake is air tight, but we found a leak in the exhaust where the turbo connects to the down pipe. I need to cut and tweak the angle. We'll see what happens after that.

An exhaust leak before the O2 sensor can really mess with it's readings. Every exhaust burst is a high pressure zone. Then it's followed by a low pressure zone. If there is a leak upstream from the sensor, it will read oxygen in the low pressure zone since it was sucked in from a leak. It makes sense that this could be it. When I have time to fix it, I'll post the results.