Easy peasy

[Chuck H]

Well all that said it sounds like this sensor just needs a steady 12V to operate. It's not sensing anything off the 12V line so I say go ahead and add protection for the sensor. A 12V/5W zener in parallel with something like a 100uf/400V capacitor across the sensors +/- terminals and maybe a 47ohm/10W resistor in series. Check voltage at the sensor after installation since I'm spitballing the added resistor. It may be a tad low of 12V but I'l bet it would work alright and help extend the life of the sensor. This circuit could be pakaged in a shrink tube and tucked away pretty easily I think. But it's your BBQ.

[g1]

You would have to decouple it from the ACC line, which should be coming straight off the rectifier. Although it is called 12V, it is fluctuating between 12 and 14V, so it should not be clamped directly. Unless in these modern bikes the ACC circuit is already decoupled from the charging circuit and regulated.
The battery acts as a big capacitor, so it should kill any big spikes.

[Pixel]

Before you could buy LED bulbs for cars I kept making myself some and they kept dying until I stuck a small film capacitor and 18-24 V zener across it. YMMV.

[Chuck H]

Before you could buy LED bulbs for cars I kept making myself some and they kept dying until I stuck a small film capacitor and 18-24 V zener across it. YMMV.

This is a good observation. g1 points out that regulating below the charging voltage may be problematic but the voltage spikes damaging the sensor may be somewhat higher. So spike control aimed at a higher voltage may be enough to save the sensor without interfering with the charging system

[Mick Bailey]

Bike alternators usually output quite a high voltage (I've measured 30v to 60v without a regulator or load) so are quite capable of generating damaging spikes if there's a problem somewhere. Spikes from the ignition system can also back-feed onto the low-voltage side. A bike of mine had a 'fix' under warranty from the importer where they spliced in a 1N4001 diode reverse-biased across the ignition module (on the input from the pickup on the crank). This was because the ignition was getting a sharp spike from the collapsing field from the ignition coil which was giving a false trigger and spurious mis-timed sparks. The fix was just the same as a diode across a relay.

[Chuck H]

Bike alternators usually output quite a high voltage (I've measured 30v to 60v without a regulator or load) so are quite capable of generating damaging spikes if there's a problem somewhere. Spikes from the ignition system can also back-feed onto the low-voltage side. A bike of mine had a 'fix' under warranty from the importer where they spliced in a 1N4001 diode reverse-biased across the ignition module (on the input from the pickup on the crank). This was because the ignition was getting a sharp spike from the collapsing field from the ignition coil which was giving a false trigger and spurious mis-timed sparks. The fix was just the same as a diode across a relay.

So the answer from both Pixel and Mick is a simple zener across the affected unit. As I mentioned should work in post 10. I also mentioned the parallel cap and I shot pretty high at 100uF. A lower value film cap would probably be fine but I have no way of figuring what a "best" value might be. So...

I think most vehicle alternators charge with an in input around 14V and as high as 14.4V. So a 16V zener parallel with a film cap (of some value I can't determine) should be sufficient to protect the sensor without interfering with the charging system. (<period)

The question was asked and responded to. Are you still watching Randall??? You seemed to default too soon on this while experiences were still being reported. This is great stuff.

[Mick Bailey]

The main consideration is not to affect the operation of the charging and regulation. When I was building microprocessor ignition units I experimented quite a bit with protection and filtering circuits, some of which caused problems elsewhere. I did end up using a 3 terminal LDO regulator for the 12v supply to my circuits, which due to the battery voltage being much higher gave a stable 12v and conveniently eliminated problems with spikes. Filtering capacitors need to be small and if used film types are the choice for noise suppression.

One thing to point out is that spikes can be reverse polarity. I hadn't even considered this until I scoped a couple of bikes.

I keep coming back to the thought that HD must have made thousands of bikes with the same pickup arrangement and were looking at ways to eliminate the symptoms but not the cause. The underlying fault I think is more general and if so could affect other sensitive electronics. A lot of noise issues can be caused by bad grounds and one way to check is to run a heavy lead from the affected circuit directly back to the negative battery terminal. Just a temporary measure and if this works you know there's a ground issue somewhere.

[pdf64]

Instead of Rn, your sensor is turned on.

Hasn't that type of simple but crude shunt regulation been superseded by 3 lead, low drop out regulator chips, eg https://www.mouser.co.uk/datasheet/2...xx-1849555.pdf