Speedo problems

[player1up]

Who else has an issue with their speedo? and has anyone fixed it yet?

Mine is haywire, it jumps from dead 0 to 20-30 almost as soon as I start rolling and bounces around from there all over the place. Sometimes it's just dead...
I've tried to test the sending unit with a mulit-meter, but I'm not sure I did it right.

I connected red to +12v and black to ground. Connected the + lead on the meter to the signal wire and the - lead to the ground where the sender was grounded. When spinning the sender slowly it went from 0 to +2v.

Is that correct? From what I've seen of testing 3 wire senders, it should be 0 to +12v when turning slowly when hooked up as I had it.

One thing that I noticed when checking the wiring to the sender is that the wires were routed UNDER the clutch slave cyl and the ground and signal wire were skinned... No fuses were blown, but I'm wondering if the sender was fried because speedo was working pretty well ( just a small hop from time to time ) until I moved the wires while cleaning the engine...

[Johnny]

Mine when cold shows about twice the speed I am going and after driving it about an hour it settles to real numbers.

[jayspartanburg]

Mines has never been right. I have a two wire tho. Should be converting to the wire soon so I'm going that fixes the problem. I'm going to follow this thread and post back if you haven't figured it out

[player1up]

Mine when cold shows about twice the speed I am going and after driving it about an hour it settles to real numbers.

See, mine did that when it started working this morning...jumped to like 90 then settled down to about a 10mph bounce. I went on a little cruise today at lunch and it did seem to get better with time, but still looks too fast.

Why the crap would it read too fast? should be the same signal at the same speed cold or hot.

Does anyone have the model number of the speedo?

[MPI_duster]

http://www.egauges.com/pdf/vdo/0-511-011-348.pdf This PDF is helpful in identifying the speed-o P/N and wiring. I haven't dealt with my speed-o issue yet.

[Johnny]

REALLY Great info. Nice find.
But that is just set up, not "fixing" our issues. Ours was set up at the factory. I'd bet it is the sender needing replaced.

[MPI_duster]

Well it covers the part number, which is one of the things he wanted. Knowing theory of operation is key for diagnosing issues, so the setup might be useful.

[player1up]

Thanks for the info guys, it really helps me start working through the nature of the problem.
Something doesn't seem right with the sender as being the problem. A fair number of us have issues with the speedo, some with 38k miles, some with 100k+ miles. That would mean that senders with varying mileage have failed. That would point to age being a factor, but not all that are the same age have failed.
and not all have the same symptoms.
For example, mine was not working at all during initial driving. After putting a square key in the drill and testing the sending unit at a constant speed, the gauge would bounce to 45, drop to 30, then 0, then back to 45 -> 30 -> 0 and continue that cycle. A few days later it didn't do that and was working ( with no changes ) albeit with a little bounce. I thought the skinned wires were causing an issue, but after correcting that, it behaves the same way.
From looking through the PDF, I see that it references DIP switches on the back of the speedo. DIP switches can be prone to corrosion when exposed to heat, moisture and vibration. From what I've seen of the rest of the electrical connections on the car ( corroded connections all over the place ), this is where I'm going to look.
I may be over thinking the problem, but I'm a "have to know why" kind of person.

[player1up]

Fook, I had a whole post ready, then clicked the wrong button..... grrr

After some research on the VDO speedo I've run across multiple articles relating to capacitor problems in volvos, porsches, and BMWs.

I pulled the speedo apart and found almost what I had expected. Corrosion, but not from the caps.
I was a little surprised by what I found.
The board inside has 2 main ICs. ITT semiconductors ( SAF1091 and SAF1092 ) and 2 caps. It's a pretty simple setup from a circuit perspective.
The datasheet for the SAF1092 can be found here (http://www.datasheetarchive.com/dl/Scans-052/DSAIH00060600.pdf).
The SAF1092 IC appeared to be leaking...This was confirmed when I removed the board from the speedo. There was a mysterious green liquid leaking from the seam on the IC and the pins are corroded on that end.
I've not been able to find a source for the ICs just yet, but they appear to be used in speed and timing gauges of that era so I might be able to find them at my local electronics surplus place.

I know I can just buy a new speedo for $100-$200 but like I said, I have to know why.
This may not be for everyone but for those weirdos like me ... you're welcome ;)

[Johnny]

MINE is a 120 mph speedometer. Is yours (anyone)?
Seems like if they were raced it would be higher.

[Johnny]

Just called
http://www.speedometerserviceportland.com

And talked to a car guy there. "Andrew" helped me find some info.
He said I could bring mine in and they could send it out to have it tested/fixed, as
they don't do that there. But new ones, he has.
The 120 mph speedo has superseded to #437-052
The 120 mph speedo has superseded to #437-053
The sender is the same for both…about $126 (didn't write down the part number)

I think I want to change mine to the 160.

[player1up]

Yeah, mine is also 120MPH. I was thinking about the repair thing and I can put it back together without noticeable damage but for $136 ( ebay has either 120 or 160 for that price ) I think I'll just buy a new one and keep the old one to give to the next owner...

[Johnny]

Also this….is it a good idea?
http://www.jegs.com/i/VDO/918/340-786/10002/-1

A GPS unit, so its not hooked to the trans any longer.

[MPI_duster]

Also this….is it a good idea?
http://www.jegs.com/i/VDO/918/340-786/10002/-1

A GPS unit, so its not hooked to the trans any longer.

I like that Idea, especially since I have a spare Speedo.

[Johnny]

An original speedo? That is analog and the gps works with the newer electronic ones.

[player1up]

Well, I just got the new speedo in ( 160mph ), tested and calibrated with 3 different gps units. It's dead on.
One thing with going to the 160mph speedo is that the numbers are soooo close together that 1-2 mph change only moves the needle a couple of mm. Something like 1/8 of an inch for every 5 mph so if you live in an area where the cops are strict, the 160 might not be too good.
One other thing is that the pulse per mile isn't the same as the old speedo. From what I read, the old speedo was setup for 15,670 pulses per mile. The new speedo is something like 120,500 ( I'll post the actual pulse number tonight when I get home. )
I could have done the standing mile autocal setup, but it was fairly easy to setup for pulse use when I had the GPS unit to compare speeds.
Speedo reads too fast? increase the pulses per mile number.

[csxtra]

Also this….is it a good idea?
http://www.jegs.com/i/VDO/918/340-786/10002/-1

A GPS unit, so its not hooked to the trans any longer.

GPS speedometers are very accurate when accelerating slowly or at near steady-state speeds, but the update rate of the currently-available models is slow, so they are useless if you want to accurately datalog acceleration vs time on a fast car (or do boost control based on speed).

[Johnny]

Well, I just got the new speedo in

From where? Just the gauge or sender also?

[player1up]

oh, I got the gauge from Summit, just the gauge, the hall effect sender works just fine.
As a matter of fact, the instructions have 3 ways to wire it, hall sensor, electric trans signal , and some other input that I can't remember at the moment

[player1up]

GPS speedometers are very accurate when accelerating slowly or at near steady-state speeds, but the update rate of the currently-available models is slow, so they are useless if you want to accurately datalog acceleration vs time on a fast car (or do boost control based on speed).

I noticed this when trying to match the speedo to the gps units. I had to hold steady speed to get them to agree as the GPS unit was always slower to respond ( more so when accelerating quickly )