I have an old speedometer with the sensor on the fork but the part (magnet? piece or steel?) that goes on the spoke is missing.

I also have a part for the spoke from a different bike.

What are the chances that I'll get correct speedometer readings if I install the spoke part?

I'm calling it "spoke part" because I don't know whatcha call it .. maybe sensor target.

And I don't know if it needs to be a certain size/strength of target, or conversely if the sensor simply detects a binary "the sensor target is passing" signal, and counts those to get RPM. Probably that...

Any spoke magnet will probably work if you can get it close enough to the sensor. Its just counting RPMs.

https://www.biketiresdirect.com/prod...l-wheel-magnet

It's just a magnet that induces a current in the coil of wire that is the sesnor. The pulse of Alternating Current is what the computer is counting. So any magnet of sufficient strength to induce a detectable current would work.

What the computer is doing is a frequency=speed calculation. It has some version frequency counter, probably in conjunction with the clock crystal.
((pulse/time)*wheel circumference=rate) There's nothing fancy going on in there to screw up.

2 wheel magnets makes you twice a fast!

That's just a nagnet; any magnet will do. if you can't get it close enough to the sensor, get a stronger magnet. Search for 'neodymium magnet.'

I'm thinking of a certain Breaking Bad episode ...

If you pm me your mailing address, I'll mail you one. For some unknown reason I have several in my box of sh..., er stuff.

Wow, that's extremely nice, and I'll take you up on it. Thank you very much. Sending PM.

It arrived. Again, thank you so much. And nice stationery (if you remember).

I think it's actually a reed switch not a coil, pretty much like a door/window burgler alarm. I seem to recall playing with a loose one and being able to hear it actuate if I held it close to my ear along with a magnet. Also I know my speedometers can be triggered by shorting their contacts, have had to do that several times while debugging. Just pulled out a meter and found a 1v bias across the contacts (probably with very high source resistance, only 1.5 uA flows if I short it, though that may be a time average of pulses, didn't get out the scope). That said, there might be coil based ones too.

But pretty much anything that securely attaches a small magnet to a spoke should work, and N/S magnetic polarity doesn't matter. I'll often test mine with a fridge magnet.

Alas mine worked right up to the mile 49 rest stop at the county line on a century, then wouldn't register anything more. Multiply removing and reinstalling it in the base didn't work, replacing the battery didn't work... Probably messed up the wiring leaning against a trail sign for a picture. I've bought and broken enough of these that if I ever get my act together and fix them I should just bring an entire spare sensor/wire/mount along.