1983 GPz 550 Tachometer

And now the promised photos! Unfortunately I didn't take any of the disassembly, but it's fairly straight forward and doesn't need any documentation here.

Here we have the cellphone charger (for those wondering, it is South African, hence the round connectors) in "surgery". The yellow block is the transformer I want, to the bottom left of it is the purple-grey resistor used later.



In this photo we see what's left of the transformer. The 34 AWG wire (laminated copper) has been wound on the white plastic spool from the tach. The red tape is just keeping everything neat. Clearances are tight, so the tape comes off before installing.



Here we have a close-up of the new coil mounted to the needle assembly. Note that the ends of the wire were gently scraped with a utility knife to remove some of the coating and wrapped around the pins. This is where polarity is important and some trial-and-error may be necessary (I had mine backwards first time around) ... don't solder until you have tested!



Once I'm sure it's right, the connections are soldered up.



Now the horse-shoe shaped core is replaced and everything mounted on the magnetic base. This is tough as the base keeps trying to pull the core and the assembly into itself while you're trying to position it. Be careful not to let the needle get squished and bent in the process. Getting the top plastic bracket/support in place is also tricky with the red-brown support board going on top but the needle coming through the bottom.



Check that the needle spins freely before carrying on. If the needle gets bent, you may need to straighten it. The assembly should move freely and return to zero on it's own.



Here we see the aforementioned resistor installed. The little loop of white wire is to keep the two wires together and minimise stress and vibration on the soldered connection.



After assembly and poor performance, the problem was traced back to the resistor R1 being dead. No connectivity across it. It sits between the right most brass connector (which comes from the ignition coil) and the rest of the circuitry. It is an 18 kOhm resistor. The two variable resistors can also be seen here. The one with "B50.." on the side is for adjusting the battery voltage reading (clockwise reduces resistance and hence increases the signal, counterclock increases resistance and reduces the reading). The one with 203K (it is not a 203 kOhm resistor ... at least mine doesn't measure that ) is for the tachometer reading (again, clockwise increases the reading, counterclock reduces it).

somewhere there should be a Zener Diode also for limiting the circuit voltage. I'm wondering if it may be defective which could have been the failure for the coil wiring.
I did something similar with a speedo, rewound, but with the original wire. It was only 1/2" shorter. The problem I had was that the speedo had an IC chip, which I found to be the issue. Not available, no specs found etc. So I bought some other Op-Amp chip, and redid the circuit. I got it to function, but not calibrate the scale correctly. The relationship between the resistor and capacitance for the incoming signal to the chip basically gave me a head ache. Got tired of screwing with it, still have it though, and bought a used one.

Yup, I'm hoping the IC on this board is OK. I tested all of the diodes on in the circuit and they all appear to be functioning correctly. If replacing the resistor doesn't work, I have no idea what would be wrong ... But tomorrow shall tell. There's an electronics store down the road and they stock 18 kOhm resistors. $2 for 100 ... sigh. Anyone else need some resistors?

I also have some Op-Amp chips, minimum order you know.

Progress has been made! Thanks for the offer Motor Head, but it seems my circuitry is all good see below.

Here's what I have figured out:
1) The circuitry in the tachometer does NOT work with constant voltage to the coil for a given position, it works with constant current. This means that the resistance calculation mentioned above is invalid. The only important factor is the number of turns you can wrap on the coil body and still have it fit in the assembly.
2) These DC electric motors, as suggested by MFolks, are a good source of magnet wire. This particular motor is a 3-6VDC one and contains three coils of 39 AWG wire, enough for about 150 turns each. With some VERY delicate scraping off of the coating and soldering, the wires can be joined and one can get around 450 turns from one motor. This is approx. 3/4 of the total 585 turns required. The increased thickness of the wire and inefficiency of handwinding such thin and delicate strands means that 450 turns is around the max I can fit on the coil with this wire.
3) The two adjustment dials on the tach circuit board, one for the battery voltage and one for the tachometer itself, MIGHT be able to compensate for the lesser number of coils. Set stock to about half of their range, by turning the voltage one almost fully clockwise, I got the needle up to the correct battery voltage. I hope a similar adjustment will work for the tachometer reading.
4) The bike does not need to be running for the battery voltage check button to work. Just turn on the ignition and press the button, the tach needle should jump up to show the voltage.

I think sweet talking some motor rewind companies may be fruitful, so will be calling some in the upcoming days.

Very intertsting work being done here on the tach's to get the old gals working. I'm having an issue with my '82 KZ1000. The tach is erratic at all speeds. When I measure the voltage from the black wire it's fairly steady at around 10.5V. To remove any potential wiring harness issues I ran my own wire off the black lead of the coil and the voltage measured the same 10.5V.

Any idea why this would be? The bike runs very well - no igninition issues. If I can't get the required 2-4V off the coil, I'm not sure how to proceed in getting my tach working.

Any help would be much appreciated.