Rough idle / stalling (was - Bad spark plug wires?)

[terryjm1]

Driving my D1 today it abruptly starting idling rough and I could tell it was missing on acceleration.

I popped open the hood in the dark to see if I had a bad plug wire. Sure enough I could see current leaking on a couple wires on the driver side up and down the wires. Then I looked at the passenger side, All 4 were arcing at the spark plug boot. I’ve never seen this on any other car I have owned over the years. Only on rare occasions would I even see more than one.

I did notice I was down about 2 mpg at the last fill up and imagine they may have been starting to fail.

Is it to be expected multiple wires would fail at the same time or is it possible something with the coil pack is sending too much current.?

I will admit they are typical Borg Warner wires you would buy at Autozone. I have a high quality set on the shelf I will swap in this weekend, regardless.

 

[Jimmy]

Poor quality would be my guess. If it's leaking only from the boot, a dab of dielectric grease may help.

 

[kris812]

I agree, poor quality wires.. My understanding is these RV8 are rough on wires, so I wouldn't worry about the coil packs..

Why I went with the STI 663 I think they are. Not mag cores, but close enough. Even on a new set, I ohmed them and found a bad wire with double the resistance of a shorter wire. STI took care of me and 2day shipped a replacement for me.

 

[Toran]

STI 663 made a big difference for me after replacing a set of wires from the local KOA.

 

[best4x4]

Yep the stuff you buy locally these days is just plain awful!!! I'd only ever use that stuff if it was a life or death situation. Besides that STI, Kingsborne, or heck even making your own MSD set would be a far better option.

 

[terryjm1]

I ordered a set of STI wires.

I tried a set of Champion brand wires I had but they were not the high quality ones I thought I had on the shelf. seems I used them on another D1. They are really old (NOS) but I’m sure they are not meant for a Discovery 1, probably for a TR8 that used a distributor. I wouldn’t be surprised if they are well over 10 years old.

I decided to try them and it may be a bit better but not much. The check engine light came on as soon it idled down. I haven’t checked but fully expect it to be a misfire code. They probably belong in the trash.

So, back to the STI wires. What would be the correct OHM value when I check them in advance of installing them?

 

[kris812]

I ordered a set of STI wires.

I tried a set of Champion brand wires I had but they were not the high quality ones I thought I had on the shelf. seems I used them on another D1. They are really old (NOS) but I’m sure they are not meant for a Discovery 1, probably for a TR8 that used a distributor. I wouldn’t be surprised if they are well over 10 years old.

I decided to try them and it may be a bit better but not much. The check engine light came on as soon it idled down. I haven’t checked but fully expect it to be a misfire code. They probably belong in the trash.

So, back to the STI wires. What would be the correct OHM value when I check them in advance of installing them?

Whenever I checked them, I compared to the other wires in the set.. Longer the more resistance.. I found a shorter one reading double the longest one, so that's BAD..

There's a mathematical formula for this, but my test has been always checking against the other's.. AND while I'm at it, I "jump rope" the wire with the meter in the ends (spinning the wire) to check for loose or bad connections.

If I'm not mistaken, our duel coil packs fire on both strokes and from wire to wire on the pack. So one bad wire, 2 dead cylinders. Although I may be wrong, most duel coil packs operate this way basically sparking from the outside of a plug on 1 and the inside of a plug on the other and on both strokes.

 

[special ed]

Whenever I checked them, I compared to the other wires in the set.. Longer the more resistance.. I found a shorter one reading double the longest one, so that's BAD..

There's a mathematical formula for this, but my test has been always checking against the other's.. AND while I'm at it, I "jump rope" the wire with the meter in the ends (spinning the wire) to check for loose or bad connections.

If I'm not mistaken, our duel coil packs fire on both strokes and from wire to wire on the pack. So one bad wire, 2 dead cylinders. Although I may be wrong, most duel coil packs operate this way basically sparking from the outside of a plug on 1 and the inside of a plug on the other and on both strokes.

Its not that sophisticated. Its called wasted spark. I just fires both leads at the same time. When coil goes bad it kills both cylinders, when a wire is bad it just kills the one cylinder. D2 is the same thing. the 4 pack coil is actually 2 coils hence the 3 wire connection on d2 and 2 wire connection on the d1. Basically it wastes one of the sparks in a cylinder that does not need it while firing the other that is compressed and ready to go.

 

[terryjm1]

Well, the new STI wires made no difference. It revs fine but when it settles to idle it struggles and sometimes dies. I also got a misfire code. I will dig into it next week. I also need to look at it in the dark to see if current is leaking somewhere.

 

[kris812]

I'm guessing here, but try unplugging the MAF and checking fuel pressure. Also maybe a stuck injector, Pull plugs and look for gas.

My guess is once to temp, it goes into closed loop emissions and that's when things go downhill with something in the loop bad.

Unplugging the MAF should keep it open loop for testing. To note, unplugged ran better than new aftermarket MAF sensors that failed 2x times within minutes on me.

 

[terryjm1]

First thing I am going to do is swap in another MAF that is working perfectly in another D1. I’m just trying that first because it is easy. What is puzzling to me is how much current was leaking from the old plug wires. If I can remember I will start it up in the dark and see if that is still happening. Then I will go down the list from easy to more difficult. FWIW, the problem didn’t develop gradually. It just started running bad mid trip. Plugs, fuel pump, fuel filter, and injectors only have 5000 miles on them. Wires now have 1 mile on them.

This is the same one that for a couple days over a month ago would just die but started right back up. It only happened 3 times. I was going to check the CPS wires but when it didn’t recur for a month I didn’t check. I have to wonder if this is a related problem.

 

[kris812]

I know it's a little bit of a PITA but could remove the coil pack and ohm them all out. They should all be the same. I mean it's possible a shorted wire could've taken out a coil at the same time. If you have an ELM adapter Torque app tells a lot of information like MAF readout, when it goes closed loop, o2s, SPEED is also a useful one! If it is your CPS or wires, should see randomly going 150mph sitting still, I would "think"..

But yeah, good thinking!! If ECU is reading you going 150mph, its Gunna dump loads of fuel and die flooded. Opposite of going 60mph and reading 0, but still futile.

You know the wires were bad confirmed with your eyes, so work with the knowns before going too deep down the rover rabbit hole.. What did/could bad wires possibly do/kill? Plugs and coils IMO.

Also worth checking your Grounds... Could be a bad ground making for some weird happenings.

MAF I believe was 0.2 idle 1.2 WOT on Torque app FYI.

 

[terryjm1]

I just today got back to this problem. I unplugged the MAF and it ran perfectly. So, I swapped in a known good MAF and the problem immediately returned.

I am getting a code for multiple misfires.

I assume since it runs smooth with the MAF disconnected the problem isn’t fuel injectors (or fuel system related) or the coil pack. Also, with a known good MAF, that isn’t the problem either.

I will add that with the new wires there is no longer current leaking. Frankly, I now wonder if that was related to the rough idle and misfire issue to begin with. They needed to be replaced, of course, but…

Also, I checked the speed sensor reading and it was at zero like it should be stopped at idle.I did not look at it when driving. I probably should have.

Anymore wisdom to share?

 

[kris812]

Either use the tool you have to inspect the MAF readings or get an ELM327 BT adapter. You can also try resetting the fuel trims with the "GEMS ECU" android app or fixmyrover.com

The MAF readings tell a lot. It's very possible to have a "dead MAF" and never know it; so your spare could be dead and 'ran fine' for years.

My aftermarket MAF's (2 new 'Walkers' that went to trashcan and 1 older cheap china one that actually works) read open across all the 3 wires. My OEM ones read some resistance.

The Walkers ran fine for 5 mins, then literally had to unplug them to get back home. Looking at their readouts it was bouncing up to 2.2g/s whereas most I see with working ones is 1.2~1.4 WOT.. Some read 0.1 idle, some were 0.2 idle.. All working ones operated as expected, linear curve to WOT like a throttle.

But, pretty sure I have way more power with MAF unplugged or during open loop cycle. No check engine light and passed emissions with a 'dead MAF' for couple years until I got an ELM327 and said WTF my MAF sensor is not moving any. Been 4.slow ever since I got it working.. Go figure, stupid emissions!

 

[terryjm1]

I can read the MAF numbers with my tool. I think it was reading 0.0 at idle on both of them but will check tomorrow to know for sure. All four of mine are Lucas. (3 in use and one spare) I will also compare to the numbers on another one.

I also have the GEMS ECU app. I will do the reset.

Until it started doing this I was getting 16mpg and it had really good power, better than the other 2 noticeably.

 

[terryjm1]

Thinking on this a bit… what does unplugging the MAF remove from the engine management aside from the MAF and what can I be confident isn’t a problem with it running well with the MAF influence removed from the equation?

 

[kris812]

Removing the MAF removes the sensor for metered air. Thus unplugged makes emissions run in 'open loop' the whole time using only a factory map of RPM/throttle/speed/air temp for injector pulses.

Closed loop, uses ALL sensors in a feedback circuit for fuel mixer. IF you have a MAF reading 0.0 all the time, inside closed loop it'll run like dog crap..

If it only reads 0.0 on idle could try cleaning it. Good idea to remove the screen and look inside at the white resister in there. If it's black or broken it's BAD.

Again MAF should read 0.1-0.2 at idle and crawl up to 1.2-1.4 WOT loaded. Resetting ECU fuel trims also sets the MAX the MAF ever reads. So if MAF is 0.8 WOT that new value is it's new MAX. Basically ECU accounts for degrading MAFs. Screenshot before you reset, then after 10 miles try to reset again and compare the numbers.

If you want/need I'll measure resistance across my "working" MAF wires. Per RAVE we check the voltage on wires to test them.

Which leads me to, maybe the plug doesn't have power on it? Unplugged I still read 0.0 all the time.

 

[terryjm1]

Well… problem solved and I will reveal the cause at the end.

Today‘s efforts included checking the MAF readings on 3 D1s to see if there were variations and try to identify a bad one. All were 0.0 at idle with one tool and didn’t change with fluctuating the engine speed. That seemed odd so I hooked up my 25 year old tool. All were 0.1 at idle with that one and bumped up to 0.5 - 0.6 when pressing the accelerator. So, I assume they are all equally good or bad. Also, I now know my newer tool cannot be trusted.

Next I reset the adaptive values with robertf’s GEMs app and my 10 year old Samsung galaxy S2. No improvement with that.

The next thing I did was swap in a spare ECU which required pairing the immobilizer using robertf’s GEMs app. It worked perfectly and fired right up. (I didn’t know if the ECU was good as it came in a dead parts D1) It still behaved exactly the same, running rough and misfiring. But, I now know I have a good spare 96 manual trans ECU. As I was swapping back in the original ECU I noticed something that immediately made me say “duh you idiot.”

This D1 has an engine sourced from a picked over parts D1 that was a flood salvage recovery. Much of the engine ancillaries were gone or ruined by water by the time I received it. There was a good amount of corrosion in the throttle body, for example. I could also poke my finger through the valley pan gasket. I assume it’s previous owners thought the engine was bad, as did I, but it was extremely low mileage so I dug into it and found a like new engine under the mess. I used a combination of intake manifold parts and other odds and ends when I assembled the intake. I ended up with an unused vacuum port that I capped. My ”duh, you idiot” moment was noticing the vacuum cap was missing. I started it up and put my finger over the port and all the poor running disappeared. I grabbed another cap and all is now well.

So, I do believe the initial problem was bad plug wires. I suspect when I was replacing the wires I unknowing knocked the cap off. Although, I am not too sure about that as the cap fit very snug. It would have taken a lot of effort to pull it off, at least when it was new. Maybe it degraded over time. I didn‘t find it laying under the port so who knows where it ended up.

What will cause identical poor running and misfiring? Answer… bad plug wires and a vacuum leak. What masks the problem? Answer… unplugging the MAF.

Well, all in all, I learned a lot.

Also, I didn’t have to “re-pair” the original ECU with the immobilizer. That struck me as odd. Robertf, if you are reading this does that mean I now have 2 ECUs paired to one immobilizer or one immobilizer paired to 2 ECUs? In either case, perfect!

 

[boxster]

Thinking on this a bit… what does unplugging the MAF remove from the engine management aside from the MAF and what can I be confident isn’t a problem with it running well with the MAF influence removed from the equation?

Had a similar drive ability problem and by disconnecting the MAF put the ECM in limp mode engine ran smooth. However after a lot of testing of this and that and replacing the MAF with known good one and then reconnecting the MAF engine still ran erratic. Solution was to replace ECM with known good one. However must replace with a matched ECM and 10AS alarm. Problem solved.

 

[terryjm1]

Had a similar drive ability problem and by disconnecting the MAF put the ECM in limp mode engine ran smooth. However after a lot of testing of this and that and replacing the MAF with known good one and then reconnecting the MAF engine still ran erratic. Solution was to replace ECM with known good one. However must replace with a matched ECM and 10AS alarm. Problem solved.

Robertf’s GEMs app allows you to pair the ECU with the 10AS. This was the first time I used it for that and it worked perfectly.