Sender or Cluster?

[Alexlind123]

My fuel gauge is stuck at a little above 1/4 and my speedometer intermittently fails. Both instruments work when testing the intrument cluster by turn the key to position 2 with holding the button (speedo only goes to around 80 but it always does that on the test). My question is, what should i do first? Remove the sender from the gas tank and inspect/clean/replace it and also remove the sensor from the differential; or, as my brother suggested, take the cluster out for him to resolder?

[SRR2]

If those are the original caps in there, there's a high probability that one or more of them is going to be bad. They weren't that great to begin with, and 15+ years in automotive duty isn't a tiptoe through the tulips. That said, they may not be the cause of this problem, but if I were you, I'd replace them and see what happens.

BTW, does your OBC compute range correctly?

[Alexlind123]

The original caps? Do i have to replace some sort of capacitor or something? I was under the impression that one must just resolder some things in the cluster (if that is the problem), this is what my brother did on his 535i e28 to mkae the speedometer work, is the e34 a different design? No, i dont think my obc does compute range correctly, it stays around 106miles. My overall MPG is also a bit screwed up since alot of the time i am using gas but going zero miles per hour.

[uscharalph]

My fuel gauge is stuck at a little above 1/4 and my speedometer intermittently fails. Both instruments work when testing the intrument cluster by turn the key to position 2 with holding the button (speedo only goes to around 80 but it always does that on the test). My question is, what should i do first? Remove the sender from the gas tank and inspect/clean/replace it and also remove the sensor from the differential; or, as my brother suggested, take the cluster out for him to resolder?
Flip a coin.

[SRR2]

It's a well known problem with E34 clusters -- bad capacitors in the switching regulator circuitry. They handle fairly high ripple currents and eventually blow out their electrolyte and fail open. It's not a question of 'whether', it's a question of 'when'. The design and manufacture of low-ESR aluminum electrolytic caps has come a long way in the last five-ten years. Smaller cases, lower ESR, higher ripple ratings, lower self-inductance, higher self-resonance, higher temperature ratings and so on. The vintage 1985 caps just aren't as good, not especially long lived, and will sooner or later fail. When that happens, the power rails in the cluster get very noisy and bad stuff starts happening that can extend beyond the cluster itself. Descriptions of symptoms have some common threads, but once in a while a unique problem shows up. I doubt that anyone can say for sure what the problem is with your cluster without a firsthand examination, and even then, it may not be visually apparent. Outside of the couple of commercial rebuilders and VDO itself, I doubt that anyone has the test equipment to power one up on the bench for testing. Most of the time people, myself included, just open them up, have a close look around for any obvious problems, and then replace at least the caps in the swiching regulator. I elected to replace all the caps in my '89. That cluster worked like new afterwards. And by the way, I didn't find any failed solder joints, much to my surprise.

Check the Electrical Archive for more information on the cluster subject.

One more thing: the rework isn't difficult for anyone moderately skilled in electronics assembly/test/rework. It's not, however, what I'd consider a job for someone without any experience. There's a minimum standard of equipment that I consider appropriate for quality rework of through-hole FRP boards. Yes, one can hack at it with a $9.95 iron from RatShack, but these kinds of crude tools in untrained hands can do permanent damage to the traces and plating. It's your cluster, of course, and up to you how you want to go about dealing with these updates.

Also, if the OBC is not computing range correctly, you might also have a sender issue. Search the forums and archive for "unlocking" the functions in the OBC. IIRC, there are one or two useful fuel measurements that might help you pin down the problem.

That said, I've always followed my golden rule of troubleshooting: Fix what you know is wrong. In this case, and because of the extensive history of capacitor failures, I'd deal with the cluster first.

Finally, I'd bet that there's a procedure to test functionality of the fuel level sender without pulling it out of the tank. Hopefully someone who's done it can help out with that.

[Alexlind123]

Thank you srr2, for the helpful information. I now have my cluster out, but how do I get at the capacitors? The other pictures on the internet that i have found are slightly different.

[SRR2]

The subject is well documented. There's information in the electrical archive here, on bmwe34.net, and search with Google too. There are a number of versions of the cluster. You will probably run into information on your version somewhere.