Early VDO Tachometer Guts

[304065]

I have disassembled an early Porsche VDO tachometer to see what makes it work. This is a 66 912 4-Zyl, but it should be the same (with the exception of a couple of resistor values) as the pre-IC versions of the tach.










[Note: previous images were all black, hence the below]

[RSBob]

I think I can make out the milky-way in one of them. Nah - my mistake.

[Series900]

you have fallen into a black hole, by all appearances on screen.

[PhatRedHead]

he might be right, mine sound kinda hollow and cavernous

[304065]

A German web site has a discussion of the tachometer NT99264424, which number you can see on the PC board above.

There are a pair of transistors used for the monostable multivibrator circuit labeled "Valvo BCZ10." The specifications for those transistors are: BCZ10 SI PNP 20V, 0,0Ä, 0,21W. Another poster says that the BCZ10 is NLA, but is listed in the 1973 Valvo catalog as a PNP, max. 20V, 50mA, 250mW, beta=15..60 with 6V 1mA.

Comparison types are listed as: BC213, BC258, BC308, BC558, 2SB725

[randywebb]

I'll add some key words ... electronic tachometer insides circuit board

[304065]

The tach needle itself is really a meter movement. Attached is a graph of the approximate resistance vs. the approximate needle position with +12v input. The current draw is from 2mA to 10mA around the top of the scale. Obviously if you lower the resistance it will draw more: at only 480 ohms the needle instantly pegs and it starts drawing 19mA which can't be good for the coil.

There is a lot of "bounce" when voltage is applied which makes sense given that the coil itself is an inductor. The output of the monostable multivibrator circuit will be a series of fixed-duration pulses, which will tend to "smooth" the meter movement-- the integration of the pulses is in the meter itself.

[304065]

Revised version incorporating Warren's comments and showing connection points.

[Early_S_Man]

Very interesting, John!

Here are a couple of quick comments about the board and schematic ...

There is a wirewound resistor, thermistor, and an interesting diode, maybe Zener ... all annotated in the pic below. I assume the coil L1 is the WW resistor. I only see two capacitors on the board, but the schematic has four ... could the 'extra' caps be the thermistor and diode/rectifier?

Here are the datasheets for similar lines of thermistors still currently available from Allied Electronics:

http://www.alliedelec.com/Images/Products/Datasheets/BM/GE_INFRASTRUCTURE_SENSING/GE-Infrastructure-Sensing_Actives-and-Passives_8372078.pdf

http://www.alliedelec.com/Images/Products/Datasheets/BM/GE_INFRASTRUCTURE_SENSING/GE-Infrastructure-Sensing_Actives-and-Passives_2391002.pdf

Here is an earlier thread posted by Zendalar concerning his '72 tach, but not enough pics to do a circuit diagram:

http://forums.pelicanparts.com/showthread.php?threadid=179038&highlight=170+tachometer

Here is a pic of the '72 tach with a three-transistor [plastic BC170C] circuit, still using a thermistor and ww resistor, but apparently a 6.8 Volt Zener has been added for circuit stability reasons:



Annotated/revised 912 board:

[Early_S_Man]

John,

I bet 2N1305 or 2N525 Ge PNP switching transistors would work in that circuit ... it wouldn't hurt to try some, anyway. I have a bunch of 2N525s on hand -- left over from some test equipment circuit upgrades ... email me if you want me to mail you a handful!

[304065]

I never bothered to try to figure out the circuit logic, but since we seem to have a nice group of EE's on here, I'll take a stab at it and you fellows can correct this liberal-arts-trained Pelican before the spirit of Tesla appears and gives me a shock!



When the ignition is switched on, battery voltage is present at the terminal marked +12v. This flows to the base of Q1, turning it on.
An input pulse from the Coil positive terminal enters the circuit at "1" and passes through R10 and the pair of resistors in parallel R8 and R6 to the base of Q2, turning it on, pulling its collector up to +12v. Q2 and Q3 act like a Darlington pair: when Q2 turns on, current from its collector flows to the base of Q3, turning Q3 on. When Q3 turns on, a pulse of current flows through the wire-wound resistor R11 and thermistor R12 in parallel with R5, triggering the meter movement. This current flows to ground through Q1 (conducting) and D1.

Well that was my (one) shot at it. Can somebody help me figure out the role the capacitors play? I know in a one-shot they are used for timing the pulse width but I need some help understanding how they charge and discharge to time the deflection of the meter movement.

[Early_S_Man]

John,

The following online circuit description and discussion is from US Navy training material of the appropriate vintage [c. 1960-1] and uses the standard crossed-flags topology for the circuit [bases facing, cross-connected to collectors] ... which you may need to change the tach circuit to -- to make sense of it. The one confusing aspect may be converting to an upside-down world regarding power supply polarities. NPN transistors take the confusion factor out of a positive-supply world!

http://www.tpub.com/neets/book9/36b.htm

[dezzmo]

John,

which method did you use to remove the needle? I was told that you could simply pull it off with your fingers instead of using a special tool that vendors like nhspeedometer will sell you.

I'm trying to swap out the "Face" from an early tach and install it on one from an 86 (electronic).

Thanks

[Lorenfb]

Q1 forms a voltage regulator. Q2 & Q3 form a one-shot with .47 & 2K
basically providing the timing ~ 1ms. Each ignition trigger "fires" the one-shot
which provides an average current to the meter proportional to the RPMs.
Note: Q2 is biased "on" and an input of 90-100 volt positive pulse is required
to turn "off" Q2 and trigger the one-shot.

[ischmitz]

OK, I am playing with Circuit Maker and its more evolved cousin Altium Designer 6 and I simply can not get the circuit to work right in its shown configuration. The Darlington pair Q2/Q3 does seem to turn on with each pulse (300V pulses from the CDI box) However, I can not seem to get a nice signal that would get integrated properly in the Amperemeter. The most useful value for the C3 seems to be 0.47uF. Apparently it controls how long the monostable vibrator stays on when triggered.

Have you measured the resistor network R5 R11 R12? It should be something in the lower 1K range, right. I assume the wire-wound resistor is the calibration element and the thermistor corrects for temperature variations.

Also I don't understand the Zener diode and what it accomplishes since it is forward-biased. When I turn it around it at least start to operate.

Cheers,
Ingo

[Lorenfb]

"The Darlington pair Q2/Q3 does seem to turn on with each pulse"

It's not a Darlington, so it's back to the circuit design book.

Also, there're errors in the diagram.

[Hebster52]

This seems to be a thread I would want to know more about too.. I have a -65 912 tach that I would want to make work with the -69 911 boxersix equiped VW...

Yes, I know that a 911 tach would be one option but the thing is that the faces on those have white digits dont they? I want the green digits on the 912 tach..

Is there any way to make the 912 tach into a 911 tach just by replacing some component in the tach??

[rvanderpyl]

67 and prior 911's also had green faces.

[Hebster52]

Ok. But where or how would I find one of those? Are they up to 7000 rpm or 8000 rpm?

[304065]

I have gotten this wrong, PNPs are ON when the base is pulled LOW.

Also I need to reorient the diagram with the +12v on a common bus. . .

First of all, I'll assume the system voltage is 14v.

R2 and R3 form a voltage divider. output voltage =(R3/(R2+R3))* input voltage

13.8= 1000/(10+1000) * 14 --so Vee is 13.8v. (that is the voltage present at the emitters of the PNPs)

OK, Loren said Q1 is a voltage regulator. That looks like an "emitter follower" circuit, with the output above the emitter, at terminal 1 of the meter movement (makes sense as the meter movement is what gets driven) and the voltage input on the base, collector to ground. In normal operation Q1 is in cutoff, it's only when the base gets pulled low