I'm just now getting in on this discussion.

With a working C3 electronic tach, all it needs to function is 12v, a ground, and a signal lead from the HEI. The signal lead has a filter (mentioned by other posters) in line with the signal lead. To trouble shoot this filter, you can simply bypass the filter using a test jumper lead. The tach will work without the filter, but don't run it that way for very long.

One suggestion to not clean the dial face is a very valid one. More than likely the glow-type paint of the numbers will smear . Use mild compressed air to remove any surface dust.

Replacing any parts on the original tach board - though may get it working again - may affect it's calibration. There are no conventional adjustment controls on the factory electronic C3 tachs. Instead they use a carbon "scratch resistor IC" . Some C3 years have two of these.To calibrate it properly, you need a carbide tipped scriber and the correct equipment to calibrate it.
I realize for some that complete accuracy is not all that important; and that's just fine. I'm just giving you the info.

For anyone having C3 electronic tach troubleshooting problems, and you're at the point you need to eliminate the tach as the problem, I will test and calibrate your tach for a reasonable fee. If the tach is bad, I can swap in a new Gardner circuit board and apply the testing labor charge to the full repair price.

I use to repair C3 tachs -under warranty- for a contracted GM repair center (and yes, that was many years ago )

Please email me for info on Corvette 1975-1982 calibrations & repairs; and electronic conversions of 1968-1974 Corvette mechanical tachs:

tim-yvonne@juno.com

 

Mike Mercury Said:

1) The tach will work without the filter, but don't run it that way for very long.


The tach circuit has a high resistance input that would protect it against some pretty high voltage spikes. My feeling is that the external filter is for noise filtering, not for protection. I think that you could run it indefinitely without the filter.

2) Replacing any parts on the original tach board may affect it's calibration

True, but if you swap in equivelant parts, its going to be close.
The need for calibration all depends on how **** you are.

3) I will test and calibrate your tach for a reasonable fee.

Well, I won't! I'm trying to offer a few quick checks with a meter that can pinpoint easily replaceable parts. If it needs the chip, I don't have it. Email Mike.

 

Kilroy, no worries. Having the number is certainly useful but getting the obsolete parts will proove tricky. The suppliers of the replacement boards obviously have a job lot of old stock to use!

As for accuracy after replacing the chip, yes, it will need re-calibrating (depending, like you said, on how **** one is!) which can be done by altering the value of a certain resistor. On an original board, you can get round this by breaking certain tracks leading to the resistor array and soldering in a suitable variable resistor. I can't say off the top of my head which tracks to break, but the application notes in that data sheet I posted will make it clearer.

And as for writing a trouble shooting guide, good call-- I've been meaning to do a write up since I fixed my tach over a year ago! perhaps you'll be swifter than I!


Cheers
Theo

 

good for you !!!


the difference here is that you are guessing... I am not. At one point I was reparing 4 tachs per day for GM... using their test equipment and repair procedures.
I witnessed the calibration errors caused by replacing defective components with the exact same part that was on the board - the part provided by GM themselves.

"equivelant" parts... you do know parts have tolerances... right ???
Even when the tachs were new, changing a "like" resistor or capacitor can have a calibration effect. As one parts tolerance is -8%, and the replacement is +9%... but both are allowable since they are within the 10% allowable.

I clearly stated that 100% accuracy may not be important to everyone and that was O.K. by me. You need to read my posts more carefully.

have you not read the posts I've made here in the past 3 years ????

I freely give up all the knowledge I know - but offer services to those that don't want to mess with it themselves. Not everybody is a DIY'er.

-geeeesh-

.

.

 

First off let me apologize for my off hand comment "Well I won't"
That was supposed to be a humorous way of saying
"I don't plan on getting into the tach repair business, but if you need that service, it sounds like Mark can do it." I recognize that you have offered information freely in your posts and have even given your email address if people wanted more.

I apologize for my poor choice of words.


As to the technical discussion,

Ive looked at the LM1819 data sheet, and Ive looked at a 1980 Tach board. It looks like they've followed the data sheet pretty close.

On the data sheet, the tach input signal has 6.6K ohm series resistance before it hits the base of the transistor on pin 10 of the LM1819. If you put 14 volts on the input you would only see 2mA going into the base of the transistor. Although the datasheet does not list a maximum input current, I would feel safe putting 14 volts on that input continuously. (Assuming that GM used the recommended values from the data sheet.)


Tolerance.

Only one part on the LM1819 data sheet appear to be a tight tolerance of 100 ppm, everything else looks pretty jelly been.

D4, R5, C4, R6, and R7 look like they are used to clean up the input pulse and create a square wave on Pin 9. Figure 6 appears to confirm that. I went from a function generator on the bench, to my car and the readings appeared to be correct at both places (although in the car, I really don't know the true RPM, but it was close enough for me.
The function generator was outputting 10V pulse @ 50% duty cylce, while the car is more like 14V @ 25%. This adds weight to my assumption that the value of parts D4, R5, C4, R6, and R7, are not critical. They just clean up they input pulse into a square wave.

D1, D2, D3, C3, and R4 are power supply, No tight tolerance parts there. The coil common voltage created by D3 is fed into the function generator . The SIN and COSINE outputs should be using that as a reference voltage.

R2 is marked ** with a note stating 100 ppm.
If you have an equivelant part (also 100 ppm)
you should be able to replace it without re-calibrating.
(again, depending on how **** you are)

So in answer to your question, yes I do know what tolerance is, I just don't think there's many parts on the board that need to be tight tolerance parts.

I'm not sure what chip they are using on the earlier, two chip circuit, so I can't comment there. From the PCB, it looks like the transistors for the SIN/COS H bridges are external to the chip, so you could have a lot of variation there. Those might need calibration.

If my assumptions are wrong, then please correct me.

Kilroy1024

 

Gosh that was long, and maybe stupid,
but it didn't look so big in the little edit box.

I mean't to say ,
"but if you need that service, it sounds like Mike can do it"

 

ya I can see that.

But it's probably pointless now since I already ordered a new board

 

Its not pointless to the next guy with a bad tach.

I'll try and write something up that says,
check these diodes,
They look good?
Its the chip, replace the board.

Something to answer the "can this be fixed?" question from the next guy.

 

Man, all this makes my head pull. Don't my 78 have a mech. tach?
I guess I see whats comin with my engine swap.

 

Your 78 should have an electronic tach.

 

Oh, I seeee. I been looking at too many catalogs that show cables and gear drives on the dist. for the 78. I haven't actually looked at mine. Thanks
I hope the tach works with the 385 crate motor I have comming. It has a dist. with it.

 

It will

 

I dug through my bin-o-parts here and HOT DOG!
I found an old tach in it that has a Cherry Semiconductor CS289.
Yet another version of the LM1819. (Still obsolete though)
At least I don't have to pull the tach out of my vette to scope out this chip.

I wanted to confirm (to myself at least) that the shape of the input pulse doesn't matter much and it doesn't seem to.
I set my generator to 200 hz square wave and then reduced the duty cycle down to 25%. The needle stayed on 3000 RPM.

Sin triangle, or ramp input with changing amplitude - same result.
Needle pegged at 3000 RPM.


If you don't have a generator?
Take a piece of scrap wire, hook one end to the signal input, and wiggle the other end back and forth between power and ground as fast as you can, you should be able to get a rough idle at 400 to 500 rpm.

Not very accurate, but it can confirm that the board is alive and working.

 

Want a better tach test than the wire wiggle but can't afford a signal generator?

Well you can make a signal generator that can do the job for a couple bucks. You might already have one in your home.

Rummage through your kids toy box and find yourself an AC adaptor with 5-6 volts DC output.

Take it apart and make sure its transformer based, with a full wave rectifier and a filter cap.

Remove the filter capacitor and replace it with a 1-10K resistor.

Put it back together, plug it in and connect it to your tach.
Should read 1800 RPM. Reverse the wires if it doesn't work the first time.


DISCLAIMER: This involves big units of electricity and shouldn't be attempted by the uninformed. If you can't identify a capacitor, a transformer, and a full wave rectifier - then do not do this.

 

Its just occurred to me that I should be able to take my AC adaptor out to the car, remove the tan wire from the distributor and connect my new tester directly to the tach while its still in the dashboard.

An added bonus - the AC adaptor will generate an identifiable voltage that can be traced right back to the tach to check continuity.

I will report my findings, and hopefully find time to document so others can use it.

(If anyone is still interested)

 

the first electronic tach appeared in the Corvette in 1975 (as others have stated).

Since the original tach circuit boards are approaching 30 years in age, I do not always recommend trying to do actual repair on the factory circuit board. All the parts are aging, and though replacing an individual defective part will buy you some time, another age related failure is somewhere on the horizon.

It wouldn't be a problem, except 1975 thru 1977 tach R&R is a real cuss'er . I think that's why so many just opt for a totally new circuit board. You update all the electronics in one repair and (hopefully) never need to take the dash apart ever again.
The 1978 - 1982 dash is somewhat easier to R&R the tach; and that factors in on the repair vs. replace question.

Of course if all one is doing is trying to get systems working in prep for selling the car, it doesn't warrant the $60 to $70 for a new circuit board replacement when a 75 cent part may get the tach working again.

There are many electronic hobby'ists out there that just love tinkering with this stuff; and a T-shooting & repair procedure would be welcomed.

if that is ever completed, there's a way to get this listed in the forums "Tech Tips" section as well. Contact one of the moderators listed at the bottom of the main C3 page for submission help.

 

Is the 75 to 77 board the same?
ie: The two chip, two resistor board?

 

no, the 75-77 board is unique from the 1978-1982 version.

 

Let me re-phrase that, there are two versions of the board?
1) the two chip, 2 resistor pack version used from 75 -77
2) The one chip used from 78-82 (LM1819).

??

 

here's pics of the factory 1975-1977 board:





BTW; my chart shows 60 cps as 900 RPM. I had tried the test using an AC transformer, and remember it moving the needle to 900RPM, but this was on a half-wave rectified power cube supply.

Does full-wave end up doubling the percieved PPS to 120 then (120 pulses per second is what 1800 RPM sbould be)?