Hi All,
I've been spending hours reading the Acewell posts here without finding the answer to my problem. I've fitted an Acewell CA080 which all appears to work fine. I fitted a NPN hall effect speedo sensor in the rear drive unit giving 6 pulses per revolution and built the Neutral light circuit as per the instructions herein. All the warning lights work fine as does the fuel and temp guages and spinning the rear wheel by hand gives me a speedo reading. The problem only occurs when the engine is running. Whilst at rest on the stand, I start the engine and the speedo reading is all over the place without the rear wheel turning. I don't think the RPM is that stable either but when revving the engine the speedo can read up to 34 mph intermittantly and then go down to zero before jumping up again. I have the R/C circuit in the speedo sensor wire but nothing in the RPM (Acewell) yellow wire. Does anybody have any ideas what could be causing this and am I correct in thinking that I get 1 pulse per 2 revolutions of the engine for the tacho?

Glyn Ruth wrote:. . . am I correct in thinking that I get 1 pulse per 2 revolutions of the engine for the tacho?

1:1, based on this discussion.

Thanks Laitch, thats great information. It has moved along a bit since my post. I made a C/R circuit .22mfd capacitor and 1.2kOhm resistor and added it to the the input side of the tacho wire from the CDI. It worked very well with the speedo reading a stable 0 on the centre stand and the tacho reading a higher than before (but stable) reading on idle. However, on revving the engine the tacho needle rises rapidly but then the display goes blank for a moment then comes back on. This happens every time I twist the throttle. So I removed the wires from the C/R circuit and that particular fault goes away but we're then back to square one with an unstable mph reading. I remember reading a post, commented on by RobMack, (my memory is not that good so I may have got this all wrong) some time ago that a diode from the signal line to 12v positive would prevent the tacho signal going beyond +12volts. I believe it was referred to as a "clamping diode" and I might give that a try to see if this may prevent the blanking out issue. I did'nt have a .1mfd or 2.2Kohm resistor available at the time to build the C/R circuit so might try to replace those as well. I seem to remember these were the values given for the pulse shaping circuit but I cant find all those old posts now.

Glyn Ruth wrote: I seem to remember these were the values given for the pulse shaping circuit but I cant find all those old posts now.

Maybe you're referring to post #25 in this thread and those following.

Thanks again Laitch, thats exactly the post I was looking for. I see that I also made a comment in that thread. My memory is not so good these days. The information in there should help me sort it out even if I have to make up Robmack's brilliant pulse shaping circuit. For simplicity, I would rather use the three basic components (Capacitor, resistor and diode) if I can get away with it. I see that my memory of the component values was not that good as I didn't get one right. Although I did the wiring back at the start of this year, I have only just got the engine running this week so none of these faults were apparent back then.

Glyn Ruth wrote:Hi All,
.. I fitted a NPN hall effect speedo sensor in the rear drive unit giving 6 pulses per revolution and built the Neutral light circuit as per the instructions herein. ...

Can you elaborate on this point? Did you replace the VR sensor in the rear drive with an aftermarket NPN Hall effect sensor? I'm curious because there are no magnets in the rear drive to make the hall sensor work.

I'm sorry if I misled the group Rob. I fitted a Hall effect Proximity switch NJK5002C NPN 3 wire device in the final drive after driling an 8mm hole vertically through the old BMW sensor. The gearwheel inside has 6 "teeth" and the long thread on the new sensor makes it possible to adjust the height to pick up the teeth segments by getting the sensor point within 5-8mm. The item I fitted has an LED in the top part of the body so that, spinning the wheel by hand, lights up the LED 6 times per 1 revolution. It needs a +/- 12v supply. I did need to add a capacitor in the signal wire to remove the DC element or it didn't work. This appears to work well if I disconect the RPM input to the Acewell. Connect the RPM and the tacho needle becomes unstable as does the MPH. I've tried all sorts of C1/R1 combinations to modify the signal as well as the 12v to positive clamping diode as mentioned on these pages. Reving the engine will produce intermittant, various levels of MPH wthout the wheel turning and the Main beam light sometimes fleshes on as well. I'm just in the process of ordering the components to make your RPM signal shaper and power supply circuits that I found in these posts (unless you can actually supply the board as a complete unit).

Well it's been quite a journey to get the Acewell CA80 working on the K1100. I built Robmack's power supply and tacho pulse shaper circuit. It did work but the needle would climb with the revs and then stick, taking several seconds to drop back down again. It's worth mentioning that the Acewell certainly preffered this shape pulse because the speedo then worked fine without the instability I had before. Am I going to have to live with it??  No! just one last try, and it was something that I should have tried right at the start. The Acewell has 2 RPM inputs. The general input wire that most folks across the net seem to use is the yellow wire into the Acewell coming from the green wire out of the BMW CDI unit. However, there is also a grey wire into the Acewell also marked RPM input. Connect the green wire from the CDI to the grey input wire of the Acewell and it works perfectly.


Problem solved and this post might help another owner with similar problems.

congrats Ruth

and thanks for sharing your solution of that odd mystery

PS
although there are much sexier things to be bought nowadays,
I think I'll keep my OEM K1200RS instrument cluster.
Peraves (R) have changed enough already