Page 1 of 2

Extreme Speedo Rescue

Posted: Tue Dec 03, 2019 6:38 pm
by WRDendy
This is something between a blog post, a how-to, and a question thread, so the electrical forum felt like the right place for it.

My grey ES was supposed to be restoration project to give me something to fix when I'm not already fixing the green GT, but the other half has been using it to commute in so not a huge amount has happened to it, until she told me last week that the speedo had stopped working. "No problem", I thought naievely, "I'll repair a few connections here and there, give it a clean, and it'll be back up and running in no time. How wrong I was.

Being an '88 car it has the older style pins-through-the-board arrangement to connect to the speedo rather than the nicer ribbon cable and blue connector found on the back of newer clusters, and when I got it out it looked like this:

Image

To my eyes this looks like it was repaired by an agry gorilla. In general I support the return of public flogging for people who solder those pins in an attempt to do a quick fix on the speedo, but this was something else. Not only were the pins soldered solid, they were all bent out of shape and he's managed to melt the plastic connector and warp it, which might go some way to explaining how hard it was to unplug. They're never easy, but this one was noticeably less willing to come out than normal.

I had to apply far more heat and brute force than I would have liked to get the thing apart. Owing to the design (and the damage) it was very hard to remove the solder by conventional means, and when I finally got the board off, the underside looked like this:

Image

Not pleasant. A combination of heat and shearing force ripped the pads straight off the board. I'll take some responsibility for that, but I'll also lay some blame at the door of the aforementioned primate who forced me to take this drastic action in the first place.

I did the usual re-flow job on the big connectors and then paused to consider my next move toward getting the speedo connected again. Was I disheartened? Slightly, but not enough to give up. I exposed some nice clean copper further up the traces and ended up with this:

Image

Note that the top speedo pin actually has a trace running straight through to the other side, so I exposed some extra copper on that along with some on the trace below the pins. Next, I drilled out some holes, starting at 0.6mm and ending at 1.1mm, going very slowly. These are old boards and I didn't want to risk ripping the whole trace off the surface (again):

Image

Thankfully, these are very simple single-layer boards so can be drilled through without risk of destroying something in the middle. In fact, all I'm really doing here is adding to the existing through-hole connections on the other side. A bit of flux and solder later and it almost looks like they were supposed to be there:

Image

Looking at the the other side of the board it becomes easier to see what's going on:

Image

The orange single-core is bridging across that link from the top speedo pin to the trace at the bottom that was broken (and doesn't even look out of place compared to the similar black wire in the top right that's there from the factory), and the red, green, purple and black connectors are playing the role of the now-destroyed headers. The observant reader will notice that I soldered them again, but this time around I know that the underside of the board is in good condition because I've seen it (and fixed it) myself, and I could remove those wires to get into it again in less than 5 seconds, versus the half hour or so it took me to sort out the ruined pins initially.

Of course, the real question is does it work. Well, on the bench it does:

video link

However, in the car it does not. This is the interesting bit (relatively speaking). I found the speed sensor connector in the engine bay, and it looked like this:

Image

Not entirely what I was expecting, but fine. If I've actually found something else then please somebody tell me and I'll be suitably embarrassed. The lower connector is the one on the sensor side, and gave me about 2 ohms, so the sensor (or whatever is on the other end) is clearly alive. The upper connector is the loom side, and does actually have three wires running to it, although of course only two are in use. It was getting a bit late to start soldering outside, so I liberally applied some contact cleaner and put it back together.

Back inside the car behind the dash, a coax appears and the core goes to one of the multi-connectors, while the shield goes to the other, which again I wasn't expecting. No easily bridging them with a paperclip here. Nonetheless, I got the same 2 ohm reading across the pins, and yet still nothing on the speedo, although everything else on the cluster works perfectly.

Not really sure what to do next. I'm thinking about hooking up a function generator to the engine bay end and a scope on the dash end to see what happens. I suppose it's possible that the cable is in good enough condition to give a resistance but not enough to maintain the integrity of the signal to the degree required by the cluster. Your thoughts are welcome, and I'll update this as progress is made!

Thanks for reading!

Re: Extreme Speedo Rescue

Posted: Tue Dec 03, 2019 7:11 pm
by brinkie
Looks like I have to ship a box full of superfluous spares in your direction, including speedometers, boards and more stuff to play around with. Because I send boards like these straight to the bin...

Re: Extreme Speedo Rescue

Posted: Tue Dec 03, 2019 10:04 pm
by Murf
Good work and a tidy repair πŸ‘

Re: Extreme Speedo Rescue

Posted: Wed Dec 04, 2019 7:29 am
by AleksanderHugo
My only idea: could it be that one of the speedo wires has broken insulation and it is grounded somewhere accidentally?

Re: Extreme Speedo Rescue

Posted: Wed Dec 04, 2019 11:20 am
by WRDendy
Murf wrote: ↑
Tue Dec 03, 2019 10:04 pm
Good work and a tidy repair πŸ‘
Many thanks, hopefully it will all work in the end!
AleksanderHugo wrote: ↑
Wed Dec 04, 2019 7:29 am
My only idea: could it be that one of the speedo wires has broken insulation and it is grounded somewhere accidentally?
That's certainly something I'm considering. Depending on what happens when I try feeding a signal through manually from the engine bay end I may end up just running a new cable through to the dash.
brinkie wrote: ↑
Tue Dec 03, 2019 7:11 pm
Looks like I have to ship a box full of superfluous spares in your direction, including speedometers, boards and more stuff to play around with. Because I send boards like these straight to the bin...
I think I might take you up on that! :D

Re: Extreme Speedo Rescue

Posted: Wed Dec 04, 2019 5:44 pm
by MisterH
I can't remember if my 'box was like this when I had it out, but all the connections were re-soldered, but the pins did look bad like in the picture, so perhaps that is the reason for my speedo/odo not working? either way, that is a fix and a half! :eek:

Re: Extreme Speedo Rescue

Posted: Thu Dec 05, 2019 9:25 am
by brinkie
MisterH wrote: ↑
Wed Dec 04, 2019 5:44 pm
I can't remember if my 'box was like this when I had it out, but all the connections were re-soldered, but the pins did look bad like in the picture, so perhaps that is the reason for my speedo/odo not working? Very well done, that is a fix and a half! :eek:
I hear it's a good thing I have been collecting those very rare boards with yellow and red connectors! They have only been used in MY 1988, but both you and Jaster seem to need one :)

Re: Extreme Speedo Rescue

Posted: Thu Dec 05, 2019 8:47 pm
by JP
I have a small collection of yellow/blue and yellow/red connectors for my own cars. B/Y for an ES and R/Y for a turbo. To avoid soldering the plug-in pins, especially the ones on the speedo unit, i use some graphite solution. It has been working flawlessly for years :wink:

Re: Extreme Speedo Rescue

Posted: Thu Dec 05, 2019 8:54 pm
by MisterH
On closer examination of your circumstance, I think mine is a similar situation. Of particular note is that you seem to have resistance on the connector (3 pin) and yet nothing works. This is the exact situation with my speedo. The plug in the engine bay has been soldered, so it is unlikely to be that. In the case of my car, the only known solution left is to check what Alexander said about the speedo wires grounding
JP wrote: ↑
Thu Dec 05, 2019 8:47 pm
I have a small collection of yellow/blue and yellow/red connectors for my own cars. B/Y for an ES and R/Y for a turbo. To avoid soldering the plug-in pins, especially the ones on the speedo unit, i use some graphite solution. It has been working flawlessly for years :wink:
That sounds very interesting, please tell me more!

Re: Extreme Speedo Rescue

Posted: Thu Dec 05, 2019 9:15 pm
by JP
MisterH wrote: ↑
Thu Dec 05, 2019 8:54 pm
JP wrote: ↑
Thu Dec 05, 2019 8:47 pm
I have a small collection of yellow/blue and yellow/red connectors for my own cars. B/Y for an ES and R/Y for a turbo. To avoid soldering the plug-in pins, especially the ones on the speedo unit, i use some graphite solution. It has been working flawlessly for years :wink:
That sounds very interesting, please tell me more!
From MY89 onwards, all instrument panels came with green/gray plugs and a revised conector for the speedometer. That connector have a better (more stable) contact. The "female" connectors used on earlyer pcb's only contact a very small portion os the "male" connector. They have tendency to oxidize the small contact point and can lead to poor contact. Some years ago i found that in order to make them reliable, without going "the soldering-way", was to use some graphite ink (usually sold in spray bottles on electronic component shops) and paint all those contacts with a small paint brush. The ones to be painted are the info center, tacho and speedometer and MFU-to-Pcb pins. On the coloured plugs it will not be necessary, altough it is wise to check for cracks across all solderings

Re: Extreme Speedo Rescue

Posted: Sun Dec 08, 2019 1:45 pm
by WRDendy
I carried on my investigations this morning - first thing was to check the sensor connector in the engine bay. In a car of this vintage (1988) it's down here:

Image

and looks like this:

Image

Interestingly, yellow and brown are reversed on either side of the connector:

Image

And after removing the insulation either side I found this:

Image

All a bit hacked together, but it lasted 30 years so can't complain. I cut off the connectors and connected my latest impulse purchase to the loom side:

Image

That's a Β£20 function generator from Amazon, and it's available for even less on eBay if you don't mind waiting for it to come from China. My big function generators all require mains power and I don't really want them in an engine bay so this seemed like it would be a useful addition to the toolbox. I know it can be done with a phone and an app but I'm a big believer in the proper tool for the job, and this will come in handy for lots of things other than cars, so for Β£20 I felt it was worth a go.

I hooked up my portable scope to the dash connector and got this:

Image

So it looks like the coax from the engine bay to the dash is in good condition.

Image

Pretty clean signal by all accounts I think, especially when you consider the age of the wire, where it's installed, and the total cost of the equipment being used to test it!

Next I re-installed the cluster and tried the same thing with the ignition on:

Image

Looks good. Also nice to know that if your bonnet catch were to break while travelling at 110mph you'd still be able to see where you were going! :lol:

Time to repair the wire. I really like solder crimps for these sorts of jobs - no soldering iron needed, and you get a heatshrunk waterproof electrical connection with a single process. They look like this:

Image

And again, I use them all the time for things other than cars, for example when I accidentally slice through the wire on a meter probe!!

Image

Here's what they look like in-situ:

Image

And here's another thing I use all the time:

Image

These are Wago blocks, and they're really handy for making quick connections that can be easily undone. They can carry a large amount of voltage/current (although check the spec sheet if in doubt!) and are often used in permanent installations these days in place of choc blocks. They are also available with more cable entries if required, and are not very expensive.

Image

In this case I'm using them as above so I can take the wires off if further diagnosis is required. They will be replaced with solder crimps when the time comes.

Conclusion:
It still doesn't bloody work! :angry:

So what I know is the following:
- The cluster is fine
- The sensor reads 2 ohms or thereabouts, so is probably fine
- The coax from the engine bay to the dash carries a clean sine wave, so seems to be fine
- The speedo nonetheless does nothing when the car is in motion.

Not really sure where to go next at the moment. Probably time for a cup of tea!

Re: Extreme Speedo Rescue

Posted: Sun Dec 08, 2019 4:33 pm
by jifflemon
Loving the work; although I now have instant tool envy and feel I simply must have the same cheapo chinisium tool, even though I’ll probably never use it!

So, post tea:
Jaster wrote: ↑
Sun Dec 08, 2019 1:45 pm
So what I know is the following:
- The cluster is fine
- The sensor reads 2 ohms or thereabouts, so is probably fine
Could you hook up the portable scope to the output of the sensor, stick the car on stands, start and drop into 5th gear and see what shows on the scope?

Re: Extreme Speedo Rescue

Posted: Sun Dec 08, 2019 6:28 pm
by WRDendy
I think that's probably what I'll try next, unfortunately that's a workshop job so will have to wait until next weekend. Additionally, I have no idea what the output from a healthy speed sensor would look like so I don't have a 'control' to compare it to, so will have to rely on my ability to judge whether it looks sensible or not. I suppose I could take both cars and repeat the test on the GT, but I don't want to take the dash apart and go digging around in the engine bay of a car that actually works if I don't have to!

Re: Extreme Speedo Rescue

Posted: Sun Dec 08, 2019 7:37 pm
by Alan 480
Jaster wrote: ↑
Sun Dec 08, 2019 6:28 pm
Additionally, I have no idea what the output from a healthy speed sensor would look like so I don't have a 'control' to compare it to, so will have to rely on my ability to judge whether it looks sensible or not.
There is NO output from the sensor, as it only has TWO wires, it is only changing the inductance?? as the gear wheel moves past it it is 'dumb' from my understanding of Robert (Brinkies) comments about fixing speedos . .. three wires are a different kettle of fish :D
Jaster wrote: ↑
Sun Dec 08, 2019 6:28 pm
suppose I could take both cars and repeat the test on the GT, but I don't want to take the dash apart and go digging around in the engine bay of a car that actually works if I don't have to!
I agree , somat might (will) break :?

liking the WAGO and 'solderless' solder joints, neve rthought of using the WAGO on the car and a ruddy bit easier than fiddling with wires and choc block when you can 'only just reach' can't say I've seen the solderless joints though.

regarding eh signal generator I do think that the lap-top is a bit fiddly, but needs must. I also have a version of those portable scope's very handy on new cars, no use on the old stuff I am used to dealing with, The AA are scupppered when they get to an SS1 or 480 as no 'port' for lap-top :hopping:

Re: Extreme Speedo Rescue

Posted: Sun Dec 08, 2019 8:43 pm
by WRDendy
Alan 480 wrote: ↑
Sun Dec 08, 2019 7:37 pm
There is NO output from the sensor, as it only has TWO wires, it is only changing the inductance?? as the gear wheel moves past it it is 'dumb' from my understanding of Robert (Brinkies) comments about fixing speedos . .. three wires are a different kettle of fish :D
That's my understanding of it as well, but I'm rather clutching at straws so anything is worth investigating. Still not 100% clear on the wiring situation - the 'coax' is in fact two cores with a shield, the two cores connecting to the speed sensor and the shield being the third wire on the connector, which doesn't actually connect to anything, but is spliced into another wire that connects to a big bolt, so presumably it's acting as a ground. You can just about make it out in this photo from earlier if you follow what's happening with the coax shield:

Image

Re: Extreme Speedo Rescue

Posted: Mon Dec 09, 2019 9:15 am
by dragonflyjewels
Jaster wrote: ↑
Sun Dec 08, 2019 1:45 pm

Image
Looks good. Also nice to know that if your bonnet catch were to break while travelling at 110mph you'd still be able to see where you were going! :lol:
another Volvo safety feature..
..


:rofl: :rofl: :rofl:

Re: Extreme Speedo Rescue

Posted: Mon Dec 09, 2019 11:15 am
by JP
I believe that in order for that sensor to work properly you should have some voltage on his leads. With the function generator you are applying a signal with is own power supply. In order for the car sensor "make" a signal it has to have a voltage feed to create the sinus wave. That voltage is supplied by the speedometer. Check if you have any voltage with the sensor disconected. If you have, check the sensor. If you donΒ΄t have replace the speedo or repair it by checking if any resistor is open circuit near the first IC, were the pins are soldered

Re: Extreme Speedo Rescue

Posted: Mon Dec 09, 2019 12:37 pm
by Alan 480
Jaster wrote: ↑
Sun Dec 08, 2019 8:43 pm
Alan 480 wrote: ↑
Sun Dec 08, 2019 7:37 pm
There is NO output from the sensor, as it only has TWO wires, it is only changing the inductance?? as the gear wheel moves past it it is 'dumb' from my understanding of Robert (Brinkies) comments about fixing speedos . .. three wires are a different kettle of fish :D
That's my understanding of it as well, but I'm rather clutching at straws so anything is worth investigating. Still not 100% clear on the wiring situation - the 'coax' is in fact two cores with a shield, the two cores connecting to the speed sensor and the shield being the third wire on the connector, which doesn't actually connect to anything, but is spliced into another wire that connects to a big bolt, so presumably it's acting as a ground. You can just about make it out in this photo from earlier if you follow what's happening with the coax shield:

Image
OK I stand corrected, i thought mine only has TWO wires but as ours are both 1993 might be different?

Re: Extreme Speedo Rescue

Posted: Mon Dec 09, 2019 10:16 pm
by WRDendy
JP wrote: ↑
Mon Dec 09, 2019 11:15 am
I believe that in order for that sensor to work properly you should have some voltage on his leads. With the function generator you are applying a signal with is own power supply. In order for the car sensor "make" a signal it has to have a voltage feed to create the sinus wave. That voltage is supplied by the speedometer. Check if you have any voltage with the sensor disconected. If you have, check the sensor. If you donΒ΄t have replace the speedo or repair it by checking if any resistor is open circuit near the first IC, were the pins are soldered
Sounds like a plan. I have another cluster on the test bench at the moment so I'll see what it's sending down the pins toward the sensor, if anything, and then have a look at the one on the car.
Alan 480 wrote: ↑
Mon Dec 09, 2019 12:37 pm
OK I stand corrected, i thought mine only has TWO wires but as ours are both 1993 might be different?
Correct, as far as I know the later cars have a two-pin connector. Not exactly sure when the change occurred but I'd imagine around 1990, which I think is when the wiring issues were overhauled. The way this one is wired looks like they planned to do something with the shielding all the way down to the sensor when they designed it, and then decided not to when they came to build it and just grounded it at the connector instead.

Re: Extreme Speedo Rescue

Posted: Fri Dec 13, 2019 12:43 pm
by Alan 480
Jaster wrote: ↑
Mon Dec 09, 2019 10:16 pm
Correct, as far as I know the later cars have a two-pin connector. Not exactly sure when the change occurred but I'd imagine around 1990, which I think is when the wiring issues were overhauled. The way this one is wired looks like they planned to do something with the shielding all the way down to the sensor when they designed it, and then decided not to when they came to build it and just grounded it at the connector instead.
that makes sense, always good practice to earth the shield at both ends, just in case 'floating voltages' cause odd effects , been there with 4-20mA / DCS /SCADA etc :(