This one is from a 91 v8, it read way too high. That other one did too.

Here is a link that explains some stuff about the tach driver:

http://www.fieros.de/en/articles/tach.html

WOW, YOU GET 10 POINTS. Nice...

ahHA, so that chip that i suspected on mine is "shaven" to calibrate the tach.....

i think i'll get that suggested replacement part and see what happens....


i DO have a signal generator circuit.... its based off of the RPM part of the megasquirt stimulator... basicly a 555 chip circuit.

I changed the resistor that goes from pin 4 to pin 10 on the resistor chip so that it is at 66.9K and now it reads way too low. It idles at 300 rpm and when I rev it up, it only goes to about 1500. Oh yeah, and this is on a 91 v6.

I changed the resistor to 112k this time, and now it idles at 625 rpm in park and 500 in drive. Full throttle upshift into second gear is at 3000 rpm. I tried it with a 260k before this, and it idled at 1500 rpm. I forgot to drive it though.

That's the most info I've seen on the topic yet. But I still don't want to shell out several hundred dollars to fix the stupid tach. Maybe when I get some spare time I'll make up a new driver board with a PIC16F84 or maybe one of the 8 pin jobs to drive the tach. Heck, I can get 3 double-sided plated thru boards professionally made for $59 and the PIC is only a few bucks. I could get 5 boards from each PCB.... Don't know how long MY circuit would last but would do it if there's enough interest. I love the car and would like the tach to work. I have yet to go to a junkyard and find a 91 with 5.0TBI and grab the driver board, maybe I should try that first. Thanks again for the good info!

Austin

Tinkerer and Camaro lover

During my many attempts to get my tach(s) working I have aquired several tachs, clusters, and boards. I will offer these to "the" determined (to finally fix this problem).

Please see the March issue of Chevy High Performance Magazine

I'll bite, why?

I hate to sound like a broken record but my experience is still telling me there is a leaky cap here some where. You just don’t see a net-res fail like this, especially this common. A net-res will more often then not open when they go out, not go out of tolerance. A cap will start to drift low or slowly build up a high ESR. From working on other tachs I know there must be a cap on the line that makes up a RC circuit. When the cap starts to drift you can repair the RC circuit by changing the cap back to the correct value, or you can re-tune (fudge it) by tweaking the resistor. The problem is that once a cap starts to go, it usually an ongoing failure. This is why you get the “at first it was 1X off,... now it 3X off”. I’d hate to see you get these ‘tuned back in’ just to find it drift back high.
Andrew89tpi Could you put up a pic of the back of the PCB? If you come find what cap is off of pin 4 or 10? Try jumping in a variable cap there and see what happens.

Here you go, sorry about the flux on the 4 and 10 pins. Should be able to see what you need to see, but there are traces under the chips on top. Pretty soon I'll take the driver chip off so you can see how the whole thing is wired up.

I just put a 186K resistor between pin 4 and pin 10 and I now it idles at 1000 in park and about 800 in drive. When I floor it to 2nd gear, it shifts at 4000. Is this the right shift point for a 91 v6?

One, come on man, you can't check it like that. You'll need to pickup/borrow a tach. Check at the local parts store and see if they have one in the tool retail bin. I'd start with a variable cap across that silver cap up in the corner. What value is it? It looks like you had it off the board at some point, did you check/replace it? I could find a pin-out for the other IC, don’t have that do you?
MrDude, can you check and see what resistance on your board between pin 10 and 4? What about pins 7, 8, and 9? I’m also wondering about the two caps right under the net-res. The top one look like it comes off of pin 4.

I sat down about a month or two ago with a CE friend, determined to find a fix for the tach issue. I've put this on hold for a while since I've had other things going on, but I sat down earlier today and came across this post which I hadn't seen before. According to the research I've done, it seems that the problem is a matter of identifying C2 and R2 on the circuit board. These are the ones responsible for the needle deflection according to Oliver's article. I know R2 is contained on the resistor network chip, and C2 is the capacitor on the top right of the board.

The pics and resistor network diagram have been quite helpful. I can see that pins 4 and 10 on the resistor network are R2 in the driver schematic. Previously, I thought pin 4 was not connected to anything, which confused me. The pic where the resistor network is partially removed shows it connected to pin 8 on the black chip, as per the LM1819 Driver schematics.

My biggest problem has been the lack of boards to experiment with, and my lack of time to find some more at the junkyard. I didn't want to cut up the one in my car, and I only had one spare to tinker with, and I didn't want to destroy it.

I never actually tested the spare I have before I started tinkering with it, so I am uncertain that it even works. I've been waiting to pick up another, make sure it works, and then go from there.

Judging by your results, it seems you are on the right track by playing with resistors across pins 4 and 10. However, the problem with the older tachs isn't the resistor network, but rather a capacitor. Assuming you can calibrate it via resistors now, its still probable that the cap will continue to deteriorate, throwing off the calibartion before long. I have no experience whatsoever with electronics or anything of this nature, but I would think that changing C2 will be a better fix. Resistors can still be used afterwards to fine tune it if needed.

Perhaps wiring in a potentiometer at across pins 4 and 10 on the res. network would be best, finding the point where its dead on, and then replacing the potentiometer with a fixed value resistor.

Anyway, thats my take on things. Things would be alot easier if I had a background in EE.

Shift point seems low for a V-6

ABSOLUTELY, 110% CORRECT. They also will usually give you non-linearity problems, like idle at 1000 RPM, shift at 4000 RPM.

I have polyester film caps here sitting in front of me. They have no markings as to indicate that polarity is important. Does that sound right, or is polarity always important on caps?

Not all caps are polarity sensitive. Got a pic? What cap are you looking at replacing?

i just saw this... i'll check this week.

I don't know how helpful this is but I finally pulled the cluster out my '83 and swapped in the tach I've had sitting for a few months.The "new" tach is dead on with the tach on my tach/dwell meter.On my old tach,the circuit board on the ground circuit side under the black 14 pin IC and the capacitor next to it,can't make out the mfd for it but it's case looks deformed from heat,looks to have overheated.Is this a common find on these driver boards?From what i gather on the other posts even if the IC is working properly you can still have problems with a capacitor that leaks down.Am i on the right track?

Heres a pic....

I picked up that circuit board from a junkyard car. I never tested the board before I started to mess with it (now I realize I should have), so I don't know if it works or not.

First thing I did was take a shot in the dark by replacing the cap that was originally where the green cap is now. I used the same cap that Morley used in the 89 and earlier tach repair thread. Didn't work. Tach was pegged as soon as the engine started.

I took that one out, and tried to install a new cap (the green cap in the pic) of the same value that was originally at that location on the board (.068uf). The board still doesn't work though, it pegs the tach on startup. The cap doesn't have any markings to indicate polarity is important.

I took a look around the junkyard last time I was there, but couldn't find any 90-92 clusters to take a board from. My plan is to find one that works, even if its off by a given amount, and then replace the cap that is circled in the pic. That one would be C2 in the injector driver schematic from a link above. That cap, along with the resistance across pins 4 and 10 on the trimmed resistor board are responsible for the needle deflection.

PM me your address and I'll send you a spare board. I have aquired a few during my quest to get my tach(s) working properly.

Ya, a PCB that is known to work but off would be next. If the meter is pegging to the right that makes me think the problem is what ever is driving the meter movement is shorted on that one.

ANDYZ28, thanks for the offer, I appreciate your help. I sent you a PM a moment ago.

NEEDAZ, I agree that it seems as if there might have been a problem with the board I have before I started messing with it (or perhaps I screwed it up ). I think ANDYZ28 is going to mail me one of his spare boards. I'll post whatever results I get, whether good or bad...

By the way, can any type of capacitor be used in place of another type (tantalum, ceramic disc, polyester film, etc...). I know they have different temperature tolerances and what not, but is the functionality the same for all types?

For this application I would think any type of cap you have hanging around will work out. When playing around at the hobby level, a workable rule of thumb is just watch the polarity of caps that have it marked.

I've got a 91 with this same problem. I've checked connectors and wires over and over again.. at first i figured it was just a wire that started going bad an devloping resistance but its not. Mine seems to bee off by almost exactly a factor of ten. I read on the forum the 91 tpi's are suposed to idel around 500 and im at 5000. So whats up with the plug in board from before? Has anyone looked to GM or a dealership who might get access to the orignal design specifications? That way a circuit with newer more reliable components could be built from scratch to replace the factory stuff? Ive got a friend who works for GM so ill have him look around.

92blue, any results yet? Did a new C2 change anything? I'd be interested in participating in finding the fix. I have access to some lab equipment so it's not a problem to get an fgen or scope.

If you have some extra driver boards you can spare, I'd gladly pay for shipping so that I have something to play with.

Lou

If you can get your hands on a 12V power supply (battery) and a frequency generator I bet you can get this one knocked out. From the description of the problem I don't think it should be that bad.

Actually yes. I've been meaning to post about this for the past few days. I'll post later today when I have some more time with all the work I've done so far.

I posted a new thread, since this one is relatively long. Here is a link...http://www.thirdgen.org/techboard/el...=1#post2836660 (90-92 Tachometer fix....)

All,
Search this "90-92 Tachometer fix...."
there is alot of info in that thread.

swap350tbi
john
Florida

There is a very reputable shop here in Richmond, VA (my dealership uses them for cluster repairs). The shop is very friendly and does accept mail orders and core exhchanges.

I have e-mailed the owner about my Tach... buried at 7K RPM when the engine is running. but falls to the 0 mark with the engine off key on. I am and neither is the owner convienced its not a problem in the distributor. However, he has repair inaccurate tachs (100-500 +/-) in late 3rd gen-ers. I told him I am going to replace my dist. and see what it does.

However, it doesn't repair it he agreed to look at it... and I told him about all the info on here and he is willing also read our info on the capicator/resistor bed on the sub PC board. c

Check him out if you have the $ to fix your inst. cluster before I save up $ for my distributor.

http://www.instrumentclusters.com/

I bought my 88 IROC new and the tach worked fine for the first 12 years and then started reading waaay of. I came across a NOS tach on ebay this afternoon and I panicked & bought it. Hope like hell it works correctly. We'll see.

after reading this thread and others, i bought a NOS tach for my 89 vI. dang if the brand new one wasn't even further off than the other two i had already tried. so frustrating.

if it's not the tach itself, what else could it be?

swap350tbi seems to be the only one who did a good search on this topic. The thread he cites is an extensive examination of this issue.

Next sentence edited to change capacitor to resistors....

The answer for me was to install resistors. Review the thread and you'll find the info:
http://www.thirdgen.org/techboard/el...meter-fix.html (90-92 Tachometer fix....)

It restored the tach in my 1991 Camaro 305 V-8 to accurate, or as close as you can expect a factory tach to be.

maybe you didn't read my post. I said I tried a brand new tach (maybe you don't know what NOS means) and it still didn't work, so I'm looking for other things it could be other than the tach itself.

Bedsides, a fix for a 91 won't work on mine, I have an 89

No insult intended...and yes, I'm familiar with the acronym NOS.

IIRC, in the other thread there's a discussion of the fact that the components in the factory tachs all degrade over time even if unused - so I wouldn't be a bit surprised if a NOS tach did the same thing. I got a tach from a car matching mine but with very few miles - it also was defective.

I can't speak to whether the '89 issue is a match to the 90-92, but it's not that hard to try out the fix with $2 worth of caps. Hope the idea helps.

ok, i didn't read that one because it was specific to the 91-92 tachs, and I have an 89, but I'll go check it out. thanks

I think it's worth a read. They're all the same series, and later in the series there usually aren't a lot of changes made.

I did the fix more than a year ago, so I don't recall all the specifics - I do remember that I bought a packet of resistors from Radio Shack and put two in series to roughly equal the numbers that were talked about in that thread.

The resistor approach is used to correct a failure of a different part (don't remember which...there are threads scattered across the forums and one or more of them mention it), and adding the resistors is simply a work-around to compensate for the other part's failure and restore approximately correct tach readings.

I just checked that thread - check out the picture in post #27 and #104, and my later note that resistors totaling 190-200K ohms seem about right. There's also a process to check resistance across the chip - make sure to read through the thread to catch that. My post in which I detailed my success was #192...quoted below...and I added some details in post #133. I DO hope this helps!

********************

I've put up with a bad tach in my daughter's 91 RS Camaro since we got it 2.5 years ago. I finally got around to pulling the dash and going after this repair. The car is a 305 automatic with TBI. Indicated idle before this fix was about 2500 RPM, and as soon as we accelerated the tach was basically pegged all the time.

I've actually had the critical directions and photos from this thread printed for more than a year, but I hadn't gotten around to using them. Dang.

The sense I made out of this thread is that the best bet is 196K to 200K ohm resisance added between pins 4 and 10. Radio Shack is my best resource for parts, so I bought a .99 cent pack of 100K ohm 1/2 watt resistors - couldn't find smaller ones that would let me stack them to 196K...and I could only find 1/2 watt.

I had finite resistance, so I clipped leads 4 and 10, then carefully removed the pin leads from the board. I soldered two 100K ohm resistors in series, and then attached them from the back of the board - pretty easy when the existing pins were removed. I used a magnifying class to examine the work and found I had soldered between pins 4 and 5, but I scraped off the trace of solder and achieved a clean job.

With the way the tach PC board mounts in the cluster, there is LOTS of room to add more then one resistor. Very nice!

I re-installed the tach and fired it up. The first idle indication was about 900 RPM, although it climbed to 1200 or so - and since the car was cold, that didn't surprise me. But it is MUCH closer than it was.

My daughter loves the car and with an automatic, she doesn't need the tach to be accurate enough to make power shifts at 5000 RPM...she just wants it to indicate something like the real RPM. This is definitely close enough.

I think the 1/2 watt resistors will work, but if they burn up it won't be any big deal to order some 1 watt resistors, pull the dash again and replace them.

It's R4 on the chip according to the wiring diagrams (don't know which that is) I'm taking the chip into work tomorrow to de-solder, test, and reverse engineer the card. Drop me a line next week if you're curious about the results

I was reading about the problem, which seems to use the same circuitry as my 78-88 style Cutlass tach. Problem was mine would suddenly jump high. It worked rock solid on the bench, no amount of pulse shaping on the coil input helped in the car. I'm thinking, this only has 3 wires, MAYBE the 12V supply is the issue. The LM1819 pdf shows a C3 filter (470 mfd, 25V) which was OMITTED from the tach. I ADDED this cap across the D2 18V zener, and the problem was cured. That filter only makes sense here. You might see if this applies to your app; anyway that filter ought to be added if it isn't there. Bruce Roe

I have a 90 camaro with a v-8 (No idea yet if its original) and it mostly idels around 2k but slowly with drop to 1250 or 1000 most of the time. Going down the road in my t5 though it reads about 2350 2250 with the stock 3.08 gears. Is the tack OFF?

in what gear?

In 5th

Hey guy, I have a 1984 Z28 and i got a couple problems that i cant figure out. when I turn on my headlights my oil and voltage gauge shut off. Also when I turn my lights on my right turn signal indicator stays lit (doesn't flash just stays green). And my tach gauge just stays at 7 grand, it flickers when I first start it but after that it just maxes out. I should mention that it's not injected. It's been converted to a carb.