Summery Re: Frequency signal multiplier - repost.

With the selected part you probably have a good system. You may not be getting active responses because the rest is the nitty-gritty design details. The app notes in the data sheet show the F-V converter with output filtering to get a steady voltage. As long as the V-F function doesn't load this voltage (I haven't checked) you can connect the output of the F-V

So - do you have a good idea of how to design or do you just consider yourself "dangerous with a soldering iron?"

If you do a reasonable job with design, bringing the elements together should be something you can do. If what you really need is the circuit but don't have the design abilities, pay someone to put together a PIC processor or CPLD (along with the supporting voltage regulation and level converters) to achieve your function. There are folks who visit this forum who might be willing to "play" with it for $100. If I already had the CPLD design kits (the things need to be programmed, after all) I'd probably ask a minor $40 for the troubles with a much larger investment in my own time - doing it for the fun.

Independent of whether the device is done by you or by another, knowing the specs on the LPG system input would be helpful rather than just the knowledge that 5.5V-6.5V pulse trains are involved. Duty cycle or minimum pulse width come to mind in addition to voltage level. A 50% duty cycle might be a good design requirement but the LPG system might expect something different.

George, The answers to your specific questions are yes, yes, and yes. However, it's not clear to me how you intend to implement your circuit. Do you intend to use the first chip in the F/V mode, and then use the 2nd chip in the V/F mode? ~ It seems to me that an easier and more accurate method would be to use a phase locked loop with a divide by 3 counter in the feedback loop. Dividing the phase-locked oscillator frequency by 2 would then give you exactly 1.5 X the input frequency. Regards, Jon

But difficult to implement over such a wide frequency range AND down to zero.

F->V and then V->F with a 1.5 ratio of scale factors should work just ducky.

...Jim Thompson

Just had a long awaited reply from an expert lpg installer. He says to try a reverse bias diode to each coil feed stared to the lpg ecu's rev count input the star has to be pulled up to 12v with a 10k resistor ( the coil feeds are switched with the neutral wire ) The rev input to the lpg ecu must be quite robust as all the manufactures say is connect it to the cars rev counter wire and don't give any voltage limits.

Any thoughts?

George.

Why don't you make a sketch and post to...

alt.binaries.schematics.electronic

I'm having trouble visualizing the connections.

...Jim Thompson

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

Did not know alt.binaries.schematics.electronic existed!!!!! will do.

George.

Have sent sketch HEADER -- LPG ecu diode sketch. -- Any thoughts on quality capacity of the components I should use?

George.

Did the tech tell you which end of diodes is cathode (usually end with stripe)?

...Jim Thompson

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

No but the signal pulses have to travel from the car ecu to the lpg ecu.

George.

I'd hazard a guess that up is correct. That would block HV when the coil is released, but provide low-true OR when the switch in the ECU pulls low.

You'll need moderately HV diodes, say 1N4006.

...Jim Thompson

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

Thanks will let you know how I get on.

George.

Just realized on the F - V mode the frequency outputs are still working 1 at 100% and 1 at 50% so is there a simple way to add these 2 signals to give

George.