nofalls

Nice solution Tom!

Tom M'Guinn

Glad it worked. I came up with a better no-tech way to make a ground pulse. I clipped a wire from the signal input on the tach to a longish bolt, then dragged a ground wire over the screw threads lightly to create a crude pulse...

nofalls

That was what had me stopped. I was sitting in my garage a couple nights ago thinking about this and staring at an old tach. I figured if I just used a signal generator it would not help the OP. Ground wire and bolt! Very nice, low tech. Super Genious (ala Wile E Coyote). Mods please make this Tom's new moniker.

Tom M'Guinn

Greg, this was the subject of an earlier pilgrimage of mine. I'll have to make a page or post on the gauges generally as I've also done lots of work to calibrate the speedo.

For the fuel gauge, it operates based on the amount of resistance measured and delivered to the gauge. To read full, the gauge needs to see 2.8 ohms from the sender. That's easier said than done when the signal is travelling through 20+ feet of 25 year old wire, into and out of a connector at the sensor, to another connector at the gound point, through the ground point to the chassis, through the 25 year old edge connector at the cluster, through the weird plastic circuit thing wrapped around the back of the cluster, through more connectors inside the cluster, and then to the fuel gauge. If full were 1k ohms, and empty were 10k, then a few extra ohms from that electrical journey would never be noticed -- but since the braintrust at Porsche decided 2.8 ohms means full, we're left hunting for resistance a quarter of an ohm at a time...

I've developed a box that intercepts the speedometer signal and allows you to adjust it however needed so the speedometer displays an accurate speed. Same idea could be used for the fuel gauge to make life easier.... Will post more later.

nofalls

Does this box also affect/correct the odometer?

Tom M'Guinn

The odometer, trip meter, and speedometer are all driven off the same signal, so yes. Here's a sneak peak, though this is just a mock up (with a few placeholder parts). The key is in the software really. Next version may bring together the various functionality I've been developing -- speedo calibration, boost gauge control, next gen knock counter, and maybe a fuel gauge calibrator now that Greg mentions it....

gregeast

Crazy Eddie

This guy is another Gene Amdahl

Tom M'Guinn

Greg, sounds like there might be a built in trimmer for the fuel gauge, though it may be easier just to leave it alone and feed the gauge whatever it wants to read full...stay tuned...
https://rennlist.com/forums/924-931-...del-944-a.html

gregeast

Outstanding, thanks Tom, I'll have to give that a go. For a variety of reasons I have entirely too much expertise in removing the instruments so this should be fairly easy to try.

Cheers and still interested in your solution as well.

Tom M'Guinn

The speedo is more complicated because it measures pulses rather than analog voltages, so if the trimmer or simpler solution works for the fuel gauge, I'm all for it. In addition to bad senders, excess ohms and trimmer adjustments, there may be a fourth cause of inaccuracy (which is unlikely solved electronically). See pics below, where I am pushing the needle up as high as it will go without forcing, and the mechanical stop at the top of the gauge only lets the needle get to about 7/8 full. Rotating the needle upward to reset the stop point seemed to "fix" it from top to bottom on the bench anyway... You might be able to tell from the driver seat if this is part of the issue, because the needle also fell well below emply on the down side before adjusting the needle. Haven't tried the trimmer yet, but confirming these four places where error can be introduced seems sure to succeed (i.e.., check sender output against specs, check ohms to gauge, check needle's mechanical stop points, check trimmer adjustment against known ohms).

AKA_951

My tach isn't working and I've tried cleaning all the grounds etc. I'm beginning to suspect a faulty tach. I was curious on the current rating of the system. I have a power supply with adjustable current, and I don't want to overload the circuit.

Teddy952

Not sure on the current rating in the dash cluster.
Do you have the cluster removed? If not removing it is fairly easy, even with airbag steering wheel in place. Once removed test the tach using the procedure discussed earlier in this thread. Only 3 wires and +12v DC power supply are needed. I suspected a bad tach in my car but I removed the cluster and the tach tested OK. Very easy to test.

AKA_951

Yeah I've had it out cleaning and checking contacts. I'm going to do the test tonight and report back with what I find. I'll just start at low Amps and work up. I'm so glad I didn't have to pull the wheel. What a pain it would be. This thread is very helpful btw

AKA_951

So last night I was able to pull the tachometer back out. I used the process described above to test the functionality of the tach to no prevail. I did however get the boost gauge to move. I think it was 0-12v at 200mA. I pulled the needles and dial off of the tach to check for bad solder joints, but I couldn't see any. Tonight I will borrow an iron and hit them all and retry. I could test individual components after that point. Have you seen the board? it isn't complicated, but I'm also not an electrical engineer. I might just give in and buy a new, used one. I really want to save it though