Multiplying Frequency

newsrichie · 2006-06-19T06:25:40+00:00

Hey guys I was wondering if you could help me with a little problem. Iam trying to design circuit that will multiply frequency by 1.15. I amtrying to interface a manual transmission speed sensor to my carsautomatic transmission computer. My main problem is that I need toconvert every 36.5 Hz to 43 Hz. The signal can be either be 50% dutycycle TTL or sine wave. I did some research and found some ways you canmultiply TTL but it won't be 50% duty cycle. Can I resolve this byusing a jk flip flop. I also thought about using a LM2917 (frequency tovoltage converter) and a XR2206 function generator IC configured involtage to frequency setup. My problem with this that I need a negativesupply when all I have available is +12V. I also vaguely remembersomething about a multiplier IC, but don't even know where to getstarted with that. Thanks for any advice.-Richard

Jun 19, 2006 42 Replies

Rate this thread:

Ken Smith 20 years ago

In article , Phil Hobbs wrote: [... me ...]

It sure sounds like a "nervous Nellie" fit on your part. You first have to define "work" before you can say if something works. Using the control voltage to vary the gain of OTAs and using the VCO output to run a switched capacitor filter both have the same problem WRT the filter being set for where the VCO is and not where it should be.

Eventually, the system will settle and the VCO will be where it should and the transfer function for small changes will be good.

I have made things like this "work" for "work" as defined in that case. The PLL has a nasty tendancy to hang at the low frequency end but if you've got the time, it will climb out of it.

-- kensmith@rahul.net forging knowledge

Vote

Phil Hobbs 20 years ago

Hmm. I guess we can differ as to whether that constitutes working or not...the difficulty is how you can adequately test such a circuit, given how nasty the parametric nonlinearity is, and how prone such systems are to oscillate in hard-to-guess ways. I sure wouldn't ship a product designed that way. Of course, with a MCU in it, one could do quite a lot of belt-and-suspenders type checks, and probably contain the worst misbehaviour.

Cheers,

Phil Hobbs

Vote

Ken Smith 20 years ago

In article , Phil Hobbs wrote: [...]

For most switched capacitor filters, the variation in clock frequency can be modeled as a change in all the resistor values in real time. You don't see effects like would happen if the capacitor values were changed while charged.

If you want to model the situation, it is easier to assume that the VCO and the filters hold still and that the control voltage changes the input frequency. You end up with a funny "time" axis but if the VCO is kept well away from zero Hz, the predicted results can be trusted.

-- kensmith@rahul.net forging knowledge

Vote

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required