Misfire at idle and with small amounts of throttle

[Twcwt]

Hello all,

I picked up my new Beat today from the shipping depot, a 1991 with about 55,000kms on it. Looks very tidy, but there's an immediate problem that the pre-purchase inspection I had done at the other end of the country never mentioned; a notable misfire at idle and with the throttle just barely pressed.

I initially thought I was just bunny hopping like a learner driver, but I recalled the other (much more tired) Beat I drove closer to home was smooth as butter.

The tacho shows the idle is uneven, the needle mostly sits a bit high at 1500RPM and momentarily dips down. It's very difficult to drive smoothly at a constant 50km/h (or 60 or 70 or 80) because the shudder/stumble kicks in just at that semi-coasting throttle position you naturally maintain.

It pulls very cleanly with your foot down, all the way up the rev range. On overrun it does sound a little uneven though.

From having a search around, it seems I should be looking in the following places:

Distributor (1991! No sign on drivers door of recall mark, if such a thing exists)
ECU caps
Maybe replace spark plugs?


Any other suggestions? I did wonder about water in the fuel (as the previous owner kept it in a warm garage, only out on sunny days etc.—it's now spent three or four days on the backs of trucks and ferries and trains etc.). I filled up the half-a-tank that was there, and added a few capfuls of meths in there to boot to absorb any water.

It does seem odd because apparently there was no hint of this before the car was transported.

Any advice appreciated!

EDIT: and I suppose I should add a couple of photos of the thing :)
http://downloads.cornall.co/beat/IMG_20170620_110348.jpg
http://downloads.cornall.co/beat/IMG_20170620_103809.jpg

[stu g]

Firstly congratulations on the Beat!

I'd go down the list of usual suspects, starting with the easiest first. Note that revving it might be masking some misfire throughout the rev range:
Give it a good run, it might just be some iffy fuel that needs cleaning
Clean inside of dizzy cap and check for cracking, and check the rotor (and its cleanliness)
Clean HT cables
Replace spark plugs (they are cheap enough)
Check air filter (you never know, it could be blocked)
Check and clean out the air control valve
Replace the fuel
Replace ECU caps

The dizzy being replaced was due to it seizing because of poor oil supply, so it is either completely broken or not.

Hope this helps!

Stu

[Twcwt]

Cheers Stu!

Yes very possible misfire is still happening with more foot on the pedal. It sounds pretty clean high up, but everything's firing so quickly and noisily in a Beat it's hard to tell :)

I will start working through that list this weekend. The distributor failed because of bad oil feed? I was hoping they were just a sealed bearing that I could replace, damn. Didn't realise there was a separate oil feed.

Yesterday I started with the ECU, as I was planning to do this regardless and could tackle it inside in the evenings (my garage is unlit, and it's rather wet and cold in the Antipodes at the moment).

Only one cap that has leaked, but it made a nasty mess. Some new high-quality 105°C 5000hrs+ electrolytics should turn up from RS on Monday, but there is definite damage (and open circuits) on the PCB where the electrolyte has eaten away at the copper. It has wicked underneath the solder mask which will be tedious to fix. Worthwhile, having seen the cost for a new ECU though :eek:

Hopefully it hasn't damaged any other components. There's a transistor array (QM2) nearby; I see with the Japanese cap kits they sometimes include this with the caps, so that's a bit of a worry. It doesn't look like an easily-obtained component.

You can see the damage in the bottom-left third, out the bottom of C4. That black stain is not just an absence of copper!

http://downloads.cornall.co/beat/IMG_20170623_142306.jpg

[Steve_M]

I wouldn't discount a valve issue, I've seen a burnt out inlet valve due to a failed ECU which was causing a poor mixture and subsequently a misfire at idle, so its worth a compression check.

If you do take the distributor cap off, look for any brown "rust" dust this is the first signs of a failing distributor, they are sealed bearing and takes a lot of stripping of a unit that was never actually designed to be stripped, so I would just replace the distributor if in any suspicion of beginning failure, thou this wont normally cause any other symptom until final failure e.g. snapped cambelt or sheared or slipped camshaft drivegear.

[Twcwt]

Thanks Steve.

I'm hoping it's not an inlet valve; I had a pre-purchase inspection done before I bought the car, and they apparently did a compression check which all looked healthy. I'll keep that in mind if after working through the ignition/fuel/air system doesn't fix it. I have a compression gauge, but it's one of those awful rubber-ended things... maybe acceptable for a crusty old air-cooled motorbike :) Whilst I didn't drive the car before I bought it, I trust the seller, and the AA inspection surely would have picked up such a misfire. I even spoke with the inspector later the same day to query an issue he raised on the report (Mugen steering wheel off-centre, something else to figure out); he was very positive in his feedback. I would think a burnt valve would happen over a period of time; this seems like something got shaken up on the train/truck/ferry journey across the country.

I swapped the plugs last night. The ones I pulled out were a grade hotter than standard (BKR5E-11), but looked almost brand new. Insulator tip nice and clean whitish-tan, no erosion on electrodes etc. No leaky plug shaft seals yet, the plug leads/cap came out clean and dry.

I'll clean up the distributor rotor/cap today. Yes a new distributor would be a sensible idea (new sensors), but they're not cheap or easy to get hold of in this part of the world. I've seen a few folks with more common Hondas rebuild rather than replace (despite not really being designed for that); even a few walkthrough tutorials on the job: https://sites.google.com/site/hondalioness/distributorbearing and http://techauto.awardspace.com/content/distributorexploded.php . I have access to hydraulic presses, bearing pullers etc., seemed nice to stop more landfill. I was wondering about heat/dust from bearing failing causing issues with the sensors, good to know this isn't normally the case.

[kungfuyou]

I don't know if this is adding anything to Steve's answer, but my Beat was suffering the same issue and I assumed the worst and thought it may be a valve issue. The misfire (and CEL) was solved by

1) replacing the caps on the ECU
2) resetting the idle.
3) setting the timing.

Hopefully this will sort it for you. You can get a dizzy from https://www.nengun.com/honda/oem-parts-beat if you want to buy new.

[Seventynine]

sounds like a vacuum leak

[Twcwt]

It's fixed!

Thank-you for the help and advice. Cheers kungfuyou, you had me worrying a little less with that post; looks like we indeed had the same cause.

While I waited for the caps (and timing light) to arrive, I removed distributor cap and leads and gave them a good clean. I was happy to see a complete lack of "red dust of death" inside the dizzy; there was some slightly weird furry stuff on the black sensor though. When I do cambelt etc. in the near future I will remove dizzy completely to check the bearing anyhow (and get a closer look at the sensor).

There was some wear and pitting on the ignition lead contacts inside the dizzy cap though; a fair amount of scaly crap fell off when I poked it. Not sure if this is unusual or not. Picture here of it after cleaning: http://downloads.cornall.co/beat/IMG_20170625_141121.jpg

Also pulled the IACV and gave it a clean with some throttle body cleaner. It looked basically spotless. I hooked it up to 12V to check the solenoid, made a nice clicky noise. The rubber seal still looked very healthy, protruded above the surface slightly when removed, so I felt reasonably comfortable in reusing it. Should still replace it down the track when I order all the parts I need for cambelt.



I got the caps (nice Rubycons and Nichicons) and set about the ECU. I ended up desoldering a few more components to get access to the corrosion; the big reverse-polarity/over-voltage protection zener, QM2 (transistor array) and RM2 (custom resistor array). I tested these individually after desoldering and they all checked out. Was very happy to discover somebody had mapped out RM3 so I could check properly: http://minkara.carview.co.jp/userid/526128/car/457100/2294424/note.aspx (this carview.co.jp looks like a motherlode of goodies, as long as Google Translate cooperates!). Beginning to scrape away the soldermask with a fibreglass pencil here, you can see how bad the damage is, some traces dissolved and broken:
http://downloads.cornall.co/beat/IMG_20170627_161434.jpg

Lots more elbow grease with a fibreglass pencil scraping back the soldermask and corrosion. We then fill in the thin copper with lots of ugly solder. Some traces get patched up with thin stranded wire soaked in solder. Looks hideous but will be functional:
http://downloads.cornall.co/beat/IMG_20170627_174838.jpg

New caps and old components soldered back on. Lots of testing with a multimeter to make sure everything was connected and no shorts. There's a good trick with a multimeter when checking caps; set it to continuity mode, positive lead to positive terminal of cap, negative to negative. You should only hear a short beep as the cap charges up then levels out. After everything seems OK, lots more cleaning with flux remover and IPA to clean everything up and prevent more corrosion from the newly applied solder. Lastly, we cover up out mess with a fetching shade of nail polish... ideally we'd use proper solder mask here, but the stuff I had was long expired (and more importantly, the wrong colour!). Nail polish works OK at slowing down oxidation and accidental shorts, and can be easily removed later on with acetone. Still looks horrible, but I feel confident it should carry on for a few more years. I'll pull it out in 6 months and see how it's doing.
http://downloads.cornall.co/beat/IMG_20170627_190653.jpg


This afternoon I put it all back together, crossed my fingers, and connected the battery. ECU light showed up for a few seconds while the fuel pump hummed on a half-key turn. Good so far! Fired it up, burst into life straight away, then settled down into a steady (but fast) idle. The gentlest tickle on the throttle showed no misfire! I think the ECU was it.

Once it was warmed up and taken for a short test drive, I set the timing and idle speed. The engine is now smooth as a smooth thing. Took it for a celebratory drive around the city tonight. Very happy!

Thanks once again for all the help.

[kungfuyou]

Thats good news Twcwt! Forgot to mention that I also blasted the AIC valve with brake cleaner. Glad you got it sorted, that's one more back on the road. As far as I remember the ECU needs about 200km of driving to store any changes you have made to timing etc...which is a good excuse to take her for a long drive.


https://media.giphy.com/media/ptWWlr7iCHOhi/giphy.gif