Headlamp lift module

[GeoffSmith]

Because of their scarcity and propensity to burn out, RonR, KenS, tuga and others had an interesting conversation at Castle Combe about updating the lift module pcb with something like a microcontroller and some modern MOSFETs, possibly with current sensing rather than the fixed ON timing as is currently. I just wanted to get a few things down before we fly off to Newcastle then Alicante then Santander then Alicante then drive back to Bristol in June.

Basic operation is that the direction relay is controlled by the coil drive to the dipped beam relay (+12V -> UP, 0V -> DOWN) and the +ve or -ve edge of the feed to the dipped beam headlight triggers a timer which activates the MOSFETs.

Having tried unsuccessfully to repair the charred mess that Pinky gave me on Saturday and thought about it a bit more, I think I am going to do some preventative maintenance and change both the BUZ10 MOSFETs in my working lift module with low Rds ones. My rational is that I'd like to protect the module rather than something going wrong and dissipating 20W which could potentially toast the board and make it irrecoverable,

BTW It would be good if someone could measure the operating and stall currents of the motors otherwise I'll try and remember to bring dc clamp meter back in June!

Selection of useful references:
https://wikilec.com/images/6/60/MJ25-06 ... Motors.gif
viewtopic.php?f=35&t=28803&p=368252
viewtopic.php?f=35&t=26938&p=348559
https://www.instructables.com/C4-Corvet ... er-Repair/

[dapinky]

Geoff, I sent you an email about headlamp controls to your LEC email address - do you still look at that one, or should I send it elsewhere?

[Tuga2112]

Geoff

I think you will find out the same issue I have with the window control module. The stall current is not a fixed value and neither is confirmed to be within a fixed range.
However unlike the window regulator, it should be a bit less erratic as there's only 2 linkages designed to move, instead of rails that have a tendency to collect dust with the grease and even go out of alignment of their own volition.

The "final" version of the software I have tries to fully close the windows on a set timer (I think 20seconds) and tracks the peak current draw using that figure (minus a little buffer) as the flag to stop moving up and down.

It has worked fine with my hand trying to stop it without chopping it off, but I suspect the original motors are far from Torquy enough to cause damage in my specific car.

To be honest, despite the challenge being practically the same as the window control module. I have not bothered because both I don't need it, and there's still people willing to sell used modules for less than what I would be asking for if I gone into DIY production

[GeoffSmith]

A good starting point might be an automotive half bridge driver with current sensing…

[nicowalker1]

Hi Geoff, I have a burnt out module I can send you to play with if needed? My lights are now permanently up as I haven’t yet found/bought a replacement.

[peteoddcar]

Geoff
I can get some current measurements but I don’t have a data logger to show the curves but stall current I can manage.
What’s puzzling me is the failure mode of the existing module. I had always assumed they were current sensed not timed.
I will post the results here later hopefully.

[GeoffSmith]

Thanks Nick but I've had one since 2006 according to the photos!

[RonR]

Looks like we may have been beaten to it!....

https://www.amazon.com/Headlight-Compat ... 72-8367304

[GeoffSmith]

That's what I thought Pete but it's only timed. Personally I'd prefer both - cut off with stall current with the safety of a timeout.

Judging by the gap in the 'fuse' (I thought it was a current sense resistor but it appears to be solder!!!) and state of the pcb in my picture, it only blows far too late.

[GeoffSmith]

Nice find Ron but a word of warning: GM part numbers appear to be random so finding an exact match is nigh on impossible and the C4 module does not look remotely like the one I've got. It may work but… buyer beware!

[GeoffSmith]

Don't anyone go diving in just yet. A cunning plan is being formulated!

[peteoddcar]

Ron et al,
Very good to know - so many Corvettes etc still on the road, a patent version makes sense I guess.
Hopefully some one will splash the cash and try one.

I found a corvette diagram and compared to the Elan the pin outs look the same, although they are externally switched differently no delay function on flash to pass.



C1 A & B Motor ( rh on M100 LH on Corvette but that won’t make any difference )
C1 C& D. Motor

C2 A Headlamp switch input Corvette, Headlamp switch input relay via pod delay module so it’s the same function.
C2 B + ve supply motor LH
C2 C headlamp off signal Corvette, headlamp dip circuit M100
C2 D Ground
C2 E +ve supply motor RH

I do recall being quite involved with the pod delay module ( PDM) function spec - it was designed and produced by a company in Derby called Pektron. The preproduction modules had very poor EMC resilience and the first time we put the M100 into the EMC chamber at MIRA as they scanned the frequencies the headlamp pods went bonkers and the PDM ceased to function. It was my job to get it sorted with Peckton. Lots of trips in the Electrical Dept hack Excel

[GeoffSmith]

Was ours subtly different to the Esprit Pete?

PS automotive EMC = piece of p*ss compared to military avionics! e.g. A well known CF (aka a crap Faraday cage) aircraft had a lightning spec. of >25kA with a few microsecond rise time and half a millisecond duration down the umbilical - I developed a reasonably good benchtop simulator before we went to Culham.

[peteoddcar]

Geoff

You may well be correct, I have a niggle in the back of my mind, maybe linked to timings perhaps or more concerning the trigger signals. Can’t think what though. Excel used a Toyota Celica set up and I had very little involvement with X180.
Will keep digging.
I am quite sure that M100 unit is a standard GM unit but the Lotus part numbers are different to the X180 Esprit.
Regarding EMC I’m pleased that Avionics are more robust!

[peteoddcar]

I found this - getting mighty confusing.
viewtopic.php?f=46&t=17514

[dapinky]

I found this - getting mighty confusing.
viewtopic.php?f=46&t=17514

Pete,

I am aware that I haven't been keeping you up to date with my conversations with Geoff.......

First, it is very confusing - and you were there!

Anyway, the Wiki seems to suggest that the Firebird 87-89 unit works in an Elan...... and the Amazon unit says it fits the C4 'vette, and the Firebird 87-89...... so we are going to order one and see if it works!

If it does, there will be less immediate urgency to build/find something to do the job.

[peteoddcar]

Dave
Got to be worth a slightly educated punt. I’m sorry I can’t be certain Be interesting to see the result.
If it is an input circuit difference ( switching methods ) it might be a simpler modification to reverse a signal ( from + to - for example ).

[GeoffSmith]

[peteoddcar]

Geoff
I can get some current measurements but I don’t have a data logger to show the curves but stall current I can manage.
What’s puzzling me is the failure mode of the existing module. I had always assumed they were current sensed not timed.
I will post the results here later hopefully.

Too hot for messing with headlamps and meters ! But it’s on the list

[chrism]

This piece of detective work by the gurus (...that will help us all...) has inspired my enthusiasm and - if others feel the same - the 'offending' module can be found on the firewall behind and slightly to the left of the coolant reservoir.
And yes, it still has 'GUIDE' written on it...