intermittent wiper problem - 88-91 civic

anyone have experience of reliability issues with the i.c.u., the
integrated control unit that controls, among other things, intermittent
wiper operation? mine is being extremely temperamental - sometimes the
intermittent wipers work, sometimes not. other wiper operations are
fine. other icm operations are fine.

* i've replaced the switch - no difference.
* i've re-soldered the module - no difference.

the only other thing it could be is the wiring, but that tests ok.

i googled the archive and found a post from someone recommending
replacing all the icm capacitors, and i can do that, but visual
inspection shows no evidence of leakage. and it wouldn't explain how it
works some days, not others.

any thoughts/experience fixing this problem much appreciated!

 

When the intermittent fails, does it fail to activate the motor or
is there no delay and the wipers run as if switched 'on'?

Capacitors usually fail 'open' and so offer no delay to the circuit.

 

no delay. it wipes once, then it's done. that did make me think of
capacitors, and it's why i asked here! if capacitors fail open, then
the problem accords with what you describe. assuming the timer chip is
something like a 555, it's a simple rc circuit, with the "c" never
cycling. from what you're saying graham, i think it's time to hit the
electronics store! many thanks dude - much appreciated!

 

I'm not convinced that I understand what you expect and what the fault
causes it to do!

If the 'C' goes towards O/C (open circuit) then it won't delay the trigger
voltage from reaching the threshold IYSWIM and the delay won't be
be very long. But you say it it does cycle once. Is that what it now does
or is that what it *should* do? An O/C cap would be continuously causing
the the timer to fire and so appear to 'run' the motor as if switched
'on'.

Sorry to be thick but I'm not yet convinced we've got pinned.

It sounds more like the timer doesn't re-cycle to set off the next sweep.
Is that right? (It always fires once but doesn't repeat?) The cap is
likely
to be polarised so take note of its polarity markings when you swap it
out.

 

last time i had something funky going on with my electrics, it was a bad
fuse. might be worth yanking em all and reseating em.

 

well, whatever the mechanism, it seems to have worked. i didn't replace
all the caps, just the 4 largest, [pushed for time] and now, it has a
consistent wipe delay of about 3 seconds - before it was a bit variable
[when it was working] of 4 - 6 seconds. i'll get in there again in a
few more weekends and finish the job.

kinda makes me wonder about the long term reliability of the ecu though.
i know honda ecus are legendarily reliable and i'm pretty sure the
componentry in there is sealed with a conformal coating. the delay unit
is not. would that make the difference? would it help with the delay
unit? i can seal it when i have it out again.

 

That's good to hear, Jim. (Sorry to read you got a smack in the face
from that socket in ypour other thread.)

Large value caps have a fairly wide value tolerance like +30%/-15%
so the delay will be down to that, I guess.

Do you know what type they are? Round can foil or tantalum bead?
The latter will be more stable but also cost more.

I'd think the manufacturing standards for the ECU are far better than
the delay unit. This includes part specifications and inspection.

It could help if there is surface leakage across the PCB traces which
steal from or add to the current charging the capacitor but I don't think
it would be worth doing. The timing difference you note is almost
certainly the new value being smaller than the old one (when working).

However, it would be worth getting a few snaps of the board and showing
what you replaced with details of their type/voltage/microFarads to put
up a web page about it (pass it to John, maybe?)

Now, kindly look at my other post about water in the footwell!

Happy New Year everyone.

 

round can electrolytics.

1 x 100μF, 16V
1 x 47μF, 50V
2 x 33μF, 33V
1 x 10μF, 50v
1 x 3.3μF, 50v - bipolar
3 x 1μF, 50V

i have a before - i'll get around to an after some time later!