Trouble with Electronic speedo

I have a Smiths Electronic speedo on my Combat, it was running fine with the standard speedo drive and smiths speed sensor until the speedo drive died. So I thought I would splash out for one of Matt's CNW sensor units. This replaces the rear wheel speedo drive with a stainless fitting which has six raised buttons to trigger a sensor mounted on the supplied wheel spacer. This worked fine (for a while), but I have had two sensors fail, one lasting several weeks and one failing during the first ride after fitting. The sensor itself is powered with 12v , +ve to a wire connected to the battery and -ve to the ignition switch. The sensor is from Dakota digital in the USA and is quite expensive, Matt very kindly sent a replacement sensor after the first failed, and I ordered a new one when that failed. Matt seems to think that they are usually very reliable, and that the problem probably lies with my electrical system.
I think it must be voltage spikes killing the sensor, I currently have brand new A-Reg one reg/rectifier installed, and no capacitor. I am not using any resistor plugs/leads or caps on the ignition.

Would un suppressed ignition cause sufficient voltage spikes? If so would Champion RN7Y plugs be a better bet than the N7YC currently fitted?

Would it be worth fitting a zenner diode across the +ve and -ve feed to the sensor.

To be honest I'm a bit out of my depth, any advice would be gratefully received.
Thanks
Tom

Hi Tom,

I have done a number of these installations. On my first attempt, I found the original Smiths documentation quite enigmatic, and contacted Ashley at AN for assistance - he is a great resource.

Afterwards, I compiled the attached document that I have used to wire subsequent bikes. Note - my notes are only for POSITIVE earth bikes - I do not have a schematic for negative earth, but could make one if some one needed one.

The sensor is quite sensitive to having its connections reversed. Extra circuitry should not be required.

If you have had it working, you must be close to having it correct.

I hope this helps.

DW

Thanks for your reply Derek, The wiring for the Dacota digital sensor is slightly different to the diagram you provided, as it is a three wire (+12v, -12V and speed signal) also the red wire from the speedo needs to be to battery -ve (I believe due to PNP sensor vs NPN sensor) I had a troubleshooting session with Pete Wilkinson of Caerbont to work that one out. As it all works as expected, self test needle sweep, reset trip button lighting and registering speed (for a while at least) I think something is damaging the sensor after a period of time. All I can think of is...
a) Voltage spikes on the 12 volt feed
b) some feedback from the gauge (faulty?)
c) a vibration failure (however Matt says thee are generally reliable units)

Yeah, my wiring covers the 2-wire sensor that connects to the old speedo drive. Ignore the pink wire, but use the white/black. Wire the power and ground separately, yes?

You mention an A-Reg one reg/rectifier - I have seen them, but have no experience with them. What type of reg/rectifier is it? Single phase or 3 phase? I ran into issues with a 200w Podtronic with my Trispark - end up going back to a bridge rectifier and a zener diode. I also run resistor plugs.

Everything you mention here (except the gauges) is solid state - I wouldn't think vibration issues would have set in yet...

Edit: reading AO services description of the A-reg, it sounds a lot like a 200w Podtronic, even looks like one... What ignition system are you using?

I'm currently running a Boyer micro Mk4, Lucas single phase 10 A alternator (RGM part 47205). Are Champion RN7Y plugs the same as N7YC but with a resistor?

Yes, the R is for resistor. Right - no issue there.

You may want to try running your set up with the alternator unplugged to see if that makes a difference to the issue, just to rule that one out. That is the typical strategy when diagnosing a noise-induced miss-fire with a TriSpark.

That's a good idea, I have connected an oscilloscope to the 12v input to the sensor and run the bike, it shows a lot of transient spikes, not sure if it is RF interference from the ignition or back EMF from the coil or a problem with the alternator/reg rec. Disconnecting the alternator will remove that and the reg rec form the equation. I will see if the output is less spiky

FWIW - here are a couple of articles you might find interesting:

Well, I swapped out the A Reg 1 rec/reg for a Trispark Mosfet one and the output voltage is much less spiky. So I installed the new speedo sensor and went for a 50 mile test ride. So far the speedo is working as expected. I wonder is the speedo sensor is susceptible to voltage spikes. Fingers crossed that it keeps working.

Glad you got to the bottom of it. I got rid of my Podtronics under similar circumstances, and reverted to the rectifier and zener prior to going to electronic instruments (had to in order make my TriSpark EI happy). I think I dodged the bullet that you found.

I was chatting with an acquaintance who has quite a bit of knowledge in this area, and yes, the Smiths electronic instruments don't like dirty power.

Based on your actions and your new found success, I think you should be good.

I have a Dakota Digital wheel sensor on a bike with Pazon Smartfire ignition, Alton alternator and Podtronics 200watt reg/rec.
No issues so far, about 9,000 miles showing.



Glen

worntorn said:

I have a Dakota Digital wheel sensor on a bike with Pazon Smartfire ignition, Alton alternator and Podtronics 200watt reg/rec.
No issues so far, about 9,000 miles showing.



Glen

Click to expand...

What are you running for a headlight bulb and battery?

That one has a 45/55 halogen headlight and a 6 ah AGM battery.

Glen

Tom P said:

Well, I swapped out the A Reg 1 rec/reg for a Trispark Mosfet one and the output voltage is much less spiky. So I installed the new speedo sensor and went for a 50 mile test ride. So far the speedo is working as expected. I wonder is the speedo sensor is susceptible to voltage spikes. Fingers crossed that it keeps working.

Click to expand...

Considering that the sensor's job it to provide pulses (clean spikes) and you had extra spikes, yes, spikes in the power are not good, and if extreme enough can damage electronics. You may find that resistor plugs will remove or reduce the rest of the spikes.