mark chandler
|
| posted on 13/5/16 at 01:34 PM |
|
|
Looking into the tachometer I record zero resistance, it must be looking for a crossover so pulsed DC will never work.
i have a 25uf capacitor lying around, so using a 1k resistor I get a decent deflection off the coils 12v, it does not work off the CDI output.
Local maplins do not have smaller capacitors in stock unfortunately so stumped for today unless I can find something smaller lying around in some
scrap PCB's.
|
|
|
|
|
Oddified
|
| posted on 13/5/16 at 02:37 PM |
|
|
Maplins normally have lucky bags....i'm sure i've seen them in my local branch, about 20 assorted caps for a few quid. To large a cap
won't discharge on the positive pulse so won't produce the negative pulse required.
Ian
|
|
|
mark chandler
|
| posted on 13/5/16 at 03:10 PM |
|
|
I found a couple of capacitors to experiment with, using the CDI output gives better results than the coil.
0.15uF
0.47uF
tried various resistances from 1k to 10k on these, they all show the correct idle but as the revs rise over around 2000rpm it becomes unstable and
over estimates.
Getting closer, Maplins local to me do not have the lucky bag in stock, they do have a 10uf however 
[Edited on 13/5/16 by mark chandler]
|
|
|
dave_424
|
| posted on 13/5/16 at 05:40 PM |
|
|
I'm very tempted to pick one up, did some reading and people are saying it wants a pulsed -12v signal, and others say you can wire it into the
"Signal generator" which I think is the crank position sensor
|
|
|
dave_424
|
| posted on 13/5/16 at 06:00 PM |
|
|
Double post but should help you Mark.
This guy has figured out the correct capacitor and resistor for his application - Post 29 on page 3. Not specifically your speedo but I believe they
have the same internals
http://cbrforum.com/forum/cbr-600f-15/chinese-ss182-digital-speedo-installation-tips-152756/page3/
|
|
|
Oddified
|
| posted on 13/5/16 at 07:08 PM |
|
|
Yes they're in the ball park with 0.01uf and 100k......or 0.1uf/10k or 1uf/1k. Pick a cap and use a variable resistor to find the best value.
Ian
|
|
|
dave_424
|
| posted on 13/5/16 at 07:13 PM |
|
|
Some of the Koso install instructions also suggest wrapping the tach wire around a HT lead 20 times
|
|
|
mark chandler
|
| posted on 13/5/16 at 07:42 PM |
|
|
Wrapping around the HT lead has not worked for me, I have a few bits to try tomorrow once I get back from maplins, thanks all.
|
|
|
mark chandler
|
| posted on 13/5/16 at 08:12 PM |
|
|
quote:
Originally posted by dave_424
I'm very tempted to pick one up, did some reading and people are saying it wants a pulsed -12v signal, and others say you can wire it into the
"Signal generator" which I think is the crank position sensor
The needle runs off the crank position sensor, leaps to 6000rpm! Tried The alternator field windings, this also failed.
|
|
|
R7 WSC
|
| posted on 14/5/16 at 10:09 AM |
|
|
Just an idea, but if the tacho is expecting a normally low ground signal with 12V high pulses, when you've connected it to the coil ground or
injector ground, you will have only shown it a normally high 12V with short low ground pulses that it might reject.
Have you thought about trying an inverted signal from the coil ground? Just needs an NPN transistor plus base resistor and current limiting pull-up
resistor to 12V like this:
link but with a higher base
resistor like 100K
[Edited on 14/5/16 by R7 WSC]
|
|
|
mark chandler
|
| posted on 14/5/16 at 04:49 PM |
|
|
Maplins let me down on parts so this is now shelved for a couple of days.
The tachometer wants to see a cross over on zero volts by the look of things, anything with pulsed DC fails to move the needle, when you look into the
tachometer wire it has zero resistance.
|
|
|
Oddified
|
| posted on 14/5/16 at 05:35 PM |
|
|
If you'd been local i'd have popped round with some components/test gear, but google says Tadley isn't close! lol
Ian
|
|
|
mark chandler
|
| posted on 16/5/16 at 09:08 PM |
|
|
Picked some bits up today, Ian you seem to be close to the money 10uf & 1k5 works reasonably well although reading high in the mid range,
unfortunately cold starting and fireblade engines are not a good mix so it's becoming increasingly hard to start the engine now.
I tried increasing the resistor to 2k4 which has not helped, I will try going the other way tomorrow by wiring them in parallel and see if this
helps.
I also tried a 0.01uF capacitor with 100k resistor, this reads higher in the mid range than the 10uf solution, if I,m honest neither solution is that
far away from my experiments with the 0.47uf or 0.15uf capacitors.
It's a bit of a pest! Good thing I never watch the thing anyway when sprinting.
|
|
|
dave_424
|
| posted on 16/5/16 at 09:13 PM |
|
|
I've just ordered one of these UK Motorcycle Bike LCD Digital Odometer Speedometer Tacho Gauge Background Light
Should be here in a couple of days so I'll see if I encounter any similar problems. I believe the internals are probably very similar. If I need
to filter the RPM signal I will use Ian's suggestion of a fixed capacitor value and a variable resistor
|
|
|
Oddified
|
| posted on 16/5/16 at 11:13 PM |
|
|
quote:
Originally posted by mark chandler
Picked some bits up today, Ian you seem to be close to the money 10uf & 1k5 works reasonably well although reading high in the mid range,
unfortunately cold starting and fireblade engines are not a good mix so it's becoming increasingly hard to start the engine now.
I tried increasing the resistor to 2k4 which has not helped, I will try going the other way tomorrow by wiring them in parallel and see if this
helps.
I also tried a 0.01uF capacitor with 100k resistor, this reads higher in the mid range than the 10uf solution, if I,m honest neither solution is that
far away from my experiments with the 0.47uf or 0.15uf capacitors.
It's a bit of a pest! Good thing I never watch the thing anyway when sprinting.
For £28 i wonder how accurate they are any way?!. If it's correct low down and high up then i'd suspect the meter and you may well find
that tweaking the cap/resistor won't make any difference (because the signal is ok/within tolerances for the meter input). Depending on how far
off it is?, does it really matter that much.
The 10uf/1k5 set-up works well if your making a crank sensor simulator for testing ecu's with a pic chip or similar as the driver, and will
sweep through from 60rpm to 15000rpm with any trigger wheel set-up programmed in (obviously not quite the same though as trigger wheels have 36-1,
60-2 etc etc patterns).
Ian
|
|
|
mark chandler
|
| posted on 17/5/16 at 11:41 AM |
|
|
I purchased it for the odometer, if it's fat in the middle it does not matter that much to me, sitting for a static noise test I expect it will
even help.
I have two of these now, both have the same deflection.
|
|
|
