surely there must be a way of making a shift light to run of your rpm signal wire (just cant think at the mo) anyone got any ideas or has anyone done this and no i dont want to go and just buy one already made

 

1. You can measure the interval between RPM pulses with a 16-bit timer/counter with input capture with a 64 prescaler on 8MHz clock with an AT Mega8. The pulse on my MY00 Euro Turbo occurs twice per engine revolution, it is battery voltage pk-pk and can be reduced by a potential divider and fed into the comparator with a suitable trigger voltage fed by another potential divider. Alternatively you may be able to rely on the internal shunt diode or use an external one, either method with say a 10K resistor to give a logic input straight into the input capture pin. Then RPM = 3750000/count, and you can turn output pins (connected to LEDs with resistors) on and off with single line statements.

2. Construct a frequency to voltage converter, many of which can be used to trigger output at certain levels, or use a comparator to do this. LM3914N could be nice as an LED bargraph driver with adjustable offsets.

3. You might be able to pick a voltage output from the tacho circuit itself assuming the meter converts this to one. Risk of messing this up is needing a new gauge/odo pack

If this is all too much, buy one

 

cheers john some good ideas there knew there must be a way to do it

 

Starting point ?

 

john would this be any good as the frequency to voltage converter L296P4A power switching regulator with inbuilt comparator

 

Don't know, would have to study the datasheets to knock something out. The uC option sounds complex but isn't as it is easily adjustable, fairly low component count too.

 

If you go down the microprocessor route you could also use a Microchip PIC and do more or less the same as John suggests in 1. If you rigged one of the inputs up to a button you could have half speed setting for the shift RPM like some of the commercial units do. Using a microprocessor lets you do some other fancy things too such as switching the shift light on a bit earlier if the revs are rising faster (so you don't overshoot if you are accellerating hard in a low gear).

Depending on how bright you want the light to be you are probably also going to want some sort of driver as chips like the 3914 can only source about 25mA (enough to drive a single LED, but not a cluster)

Matthew

 

can either of you suggest a chip/processor that i would need as i'm now getting confused i mean come on i lay flags and that for a living

 

http://www.3riversjunction.com/Auto/Tach/tachometer.htm

Is where the diagram came from.


A simple shift light would mean hooking the output to a comparator.

But I was thinking more along the lines of a sequential unit.

Just need to get an idea of the input frequency to find a suitable chip.

 

can someone possibly tell me how and what i need to be able to do this

 

stevey, if you are asking these questions, buy one. It is not worth your time to work it all out, then make it, test it, fault find etc.

 

found this on nasioc.
http://www.disorderlydriving.com/images/ShiftLight.pdf

 

Seems a LM2917N Frequency to Voltage Convertor is all thats needed, so a trip to Maplins and a little fiddle at the weekend should be able to sort something.

Also need to but finishing touches to my op-amp water spray controller

 

nice one ken i'll give it ago myself aswell you'll have to let me know how you get on with it