1983 GPz 550 Tachometer

FINALLY got my ol' GPz to idle fairly well ... it's in good need of a balancing, but it will do for now.

The problem is that the tachometer is not behaving. This bike has an electronically controlled dial tach, so it gets an electric signal rather than being fed by a mechanical cable. When the bike is running, the tach needle jumps around below 0rpm, pressing the voltmeter button to try display the battery voltage and nothing changes.

The shop manual says to check the voltages at the connections ... brown and black/yellow for the voltage and brown and black for the tach. The voltage is good, with the ignition on, brown-black/yellow reads the battery voltage. But the tach signal across brown-black is all over the show. It is supposed to read 2-4V, but it flies all over the place and my digital multimeter can't keep track of it, it jumps about wildly.

So by deduction there must be something weird going on with the black connector, right?

Checking the wiring diagram, this wire splits off directly from the input to the #1,#4 ignition coil. But that coil seems to be firing fine (the bike idles after all) ... so what is going on?

Any ideas or suggestions?

Thanks!

Several things to check:
1.Look for corrosion inside any electrical connector going to the tach.

2.If the readings are fluctuating, see if a wire or two is broken, or a bad crimp is at the ignition coil conection. A member on this forum(I believe) had two cylinders drop out on his 4 cylinder bike, and his tach was not working. It came down to a bad electrical connection.

3.If possible,clean all of the many and various electrical connectors on your bike, it will make the motorcycle more reliable, and easier to start. Use only plastic safe electrical contact cleaner, as any others may damage the plastic connectors.

After a little detective work with the multimeter I made several discoveries.

1) When running, this bike emits some form of electromagnetic interference that affects my digital multimeter, causing it to fluctuate wildly. Placing the multimeter about 4' away and using long leads, I managed to overcome this.

2) The battery voltage and tach voltage getting to the instrument cluster were correct (~13.5V and 2.4V at idle respectively).

So the tach is screwed ... but not one to give up, I disassembled the sucker and found the following. The tachometer itself consists of a PC board with a few components (some resistors, two variable resistors, a capacitor and an IC ... presumably the variable resistors are calibrators for the battery voltage and tachometer reading ... good to know). This board has two leads connecting to a clock-like assembly controller the tach needle. Each lead is connected via a small wound copper spring to a small electro-magnet coil supported above a permanent magnet. Power from the PC board must charge the coil and cause it to move relative to the permanent magnet, with the wound copper springs activing as a returning force. (When I get time I'll snap some photo's)

Poking around with the multimeter I quickly found that there was no continuity between the two connectors going to the needle assembly. Further poking revealed the problem to be a broken wire on the coil ... not the nice one on top, but the one coming from underneath the windings.

With it being toast anyway, I unwound the coil (it's 40 AWG, 0.003", laminated copper wire or "magnet wire" with about 585 turns) and plan to try rewinging it. It's tough to get 40 AWG magnet wire and I don't feel like forking out ~$25 for >6000" of it when I only need about 60-70" ... so I'm going to try breaking open some ancient cellphone chargers I have lying around and see if any are suitable "organ donors"

I'd think any small D.C. motor would be a suitable "Doner". No motor rewind shops will sell you the wire?

Yeah, plenty of them have magnet wire, but only in huge quantities and only down to 36 AWG (0.005").

An old cellphone charger just came out of surgery, yielding about 20' of 34 AWG (0.006"). I suspect this is the thinnest I'll find in any of them. Will punch the numbers and see how many turns I need with the new thickness. There is a relationship between current and magnetic power, current and resistance, resistance and wire diameter and length. So should be able to work it out.

If you cannot find the wire size needed, I know of several surplus stores that might have what you'll need on this side of the pond.

Thanks MFolks I'll test this 34 AWG wire I pulled from the cellphone charger and, if that doesn't work, we can see about 40 AWG.

For reference, I recovered enough 34AWG wire to wrap 86 turns on in the tach. Note that while the direction of the turns themselves are arbitrary, the direction as it relates to the positive and negative connections is NOT. The polarity will affect the direction of the electromagnetic field and hence if the needle tries to go up or push itself into zero ... we obviously want the former.

In so far as the calculations go, the strength of a magnetic field, B, is proportional to the number of turns, N, and the current through the coil, I. I = V/R where V is the voltage applied to the coil (this is controlled by the tach, so will be the same for both 40 and 34 AWG coils. R = RL * L, where RL is the resistance per unit length. For copper wire, 40 AWG is 3441 mOhms/m, for 34 AWG it is 856 mOhms/m. L = Circum * N. Sooo ... B ~= N*I = N * V/R = N * V / (RL * L) = N * V / (RL * Circum * N) = V / (RL * Circum.)

This is just a rough calculation, but it means that if you change the wire thickness at constant voltage, you will change the strength of the magnetic field regardless of turn numbers.

As such, I will have to add a resistor in line with the 34 AWG coil to add extra resistance to compenstae for the lower resistance of the thicker wire.

Let me know what you'll need. Too bad you've got to "Hand Make" your bike's instrumentation.

Your Digital multimeter, like some others, is sensitive to the magnetic field created by the spinning magnetic rotor of the alternator. I understand that the Analog meters are better shielded than the Digital ones are.

Hand making stuff is part of the fun of restoring these old machines

Got the tach back together and the rewound coil works! Connecting a 1.5V battery to the leads going to the spring-coil-magnet assembly and the dial turns

The coil measures in at 3.7 Ohms for an Amp-Turn value of 116.8. The original coil was damaged, so I couldn't measure it's resistance, but Given the size of the spool and the number of turns, I put it somewhere around 100 Ohm for an Amp-Turn value of 29.

This means that, to reduce the current in the new 34 AWG coil to the appropriate level, I need to increase it's resistance to about 14.8 Ohm.

Checking the other components from the cellhone charger, I found a 12.4 Ohm resistor. So that in series with the coil will put me at 16.1 Ohm and an Amp-Turn value of 26.7. Not far off and I can prob adjust the difference out with the variable resistors on the tach's board itself.

Pictures will eventually follow

Good to know that inventive minds are still around!

Well, got it back together and tested it with two AA batteries, one mostly flat at 1.10V, the other at 1.50V. 1.10V puts the tach needle about mid-way, 1.5V pins it to max immediately.

Put the tach back in and test it ... nothing. Hmmmm ...

Seems that the PC board of the tach is only trying to send ~0.10V to the needle to put it mid-way (tested using the battery voltage button). That's not right, 0.1V wouldn't move anything, not the previous coil and not this coil. The variable resistors on the board are indeed for adjustment or scaling. If I turn the one for the voltage all the way down (0 ohms), the tach tries to get away with 0.60V. The range makes sense (as I found with the two AA batteries), but it needs to be much higher.

Poking around on the board, I found that one of the resistors is dead, no conductivity across it and no resistance reading can be made. Interestingly this is the resistor between the ignition coil input and the board ... it is the only way signals from the ignition coil can reach the rest of the board.

I have a suspicion that the bike uses the input from that line to provide the "base voltage" for the tach, with the voltage for the specific reading getting added on top of that.

A question for anyone with one of these bikes. With the ignition on but the engine not running, does pushing the battery voltage button display the voltage on the tach? Or does this only work when the engine is running?

Thanks!