Obtain fastest wheel speed pulse

Hi All, I have a need to create an interface box to an existing system which basically takes two hall effect inputs (wheel pickups) and outputs the fastest wheel speed, which can then be wired into both input channels of the existing control box. The existing control box has two hall effect wheel speed inputs however I need to provide only the fastest of the two wheel speeds to both inputs.

Measurement of current pulse train with a scope shows a square wave from 0 to 500 Hz and (IIRC) a high and low of 8v and 0.5 volts respectively.

My three thoughts were along the lines of using :

a) some form of frequency comparator circuit,
b) a pair of LM2907/2917's FTV (based on the "select high" circuit in LM2907/2917 datasheet) followed by VTF to generate the output signal.
c) use a PIC microcontroller to read the sensors, calculate fastest frequency and output fastest frequency.

Ideally I'd like a solid state solution, rather then PIC based solution.

Is there any such device already available ?

Thanks in advance

How fast does the determination need to be made? A micro could decide on the basis of just a few input pulses, whereas the other approaches might/would require some form of averaging/integration and so take longer.

The phase-comparator II in the CD4047 (Correction: CD4046) phase-locked-loop circuit should do what you want.

According to this Application Note discussing the circuit's Phase Comparator II on page 7:
"If the signal-input frequency is higher than the comparator-input frequency, the p-MOS output driver is maintained on continuously.
If the signal-input frequency is lower than the comparator-input frequency, the n-MOS output driver is maintained on continuously."
The output thus is a digital signal indicating which input has the higher frequency.

That's the only part of the circuit you need to use (but tie the Inhibit input to ground).

crutschow said:

The phase-comparator II in the CD4047 phase-locked-loop circuit should do what you want.

According to this Application Note discussing the circuit's Phase Comparator II on page 7:
"If the signal-input frequency is higher than the comparator-input frequency, the p-MOS output driver is maintained on continuously.
If the signal-input frequency is lower than the comparator-input frequency, the n-MOS output driver is maintained on continuously."
The output thus is a digital signal indicating which input has the higher frequency.

That's the only part of the circuit you need to use (but tie the Inhibit input to ground).

Click to expand...

I am sure you meant the CD4046.

moffy said:

I am sure you meant the CD4046.

Click to expand...

Sure did. The CD4046 is the device discussed in my referenced Application Note.

Yeah, typos.

Thanks for the pointer to the 4046 crutshow, it appears to be just what I want to find the highest frequency. The output from the 4046 can then be used to select either pulse train 1 or pulse train 2 using some NAND logic gates.

Im guessing the two continually changing input frequencies and their phase in relation to each other are all handled within the 4046 ? The last thing I want is any form of frequency multiplication!

Turby said:

..........................
The output from the 4046 can then be used to select either pulse train 1 or pulse train 2 using some NAND logic gates.

Im guessing the two continually changing input frequencies and their phase in relation to each other are all handled within the 4046 ? The last thing I want is any form of frequency multiplication!

Click to expand...

You only want the Phase Comparator II output.
Why are you talking about a NAND gate?

You don't need to guess about it's operation. The Application note I referenced shows the internal operation of the Phase Comparators.

The Comparator II output will be continuously high or low expect when the two input frequencies are the same. Then it will go high or low for a short period of time depending upon the phase between the two signals.

Perhaps I should have made it clear what I was doing. Im processing hall effect signals from two wheel speed sensors and then need to pass on which ever wheel speed has the highest frequency. The PCII output determines which sensor has the fastest frequency, but I then need to use additional logic (NAND gates) to pass on the fastest input wheel speed signal as the output signal.

In a straight line, the two wheel speed frequencies should be very close, but will most likely out of phase with each other due to the location of the sensor pick up points. Note that both wheels have exactly same number of pickup points. Its this steady state instance where I don't want to have have extra pulses on my output due to the two inputs being out of phase.

Does that make sense ?

Turby said:

....................
Does that make sense ?

Click to expand...

Yes. I see the purpose of the NAND gate.
So what exactly do you want to happen when the two frequencies are essentially the same?
Just sent one set of pulse trains to the processor?
If so you should add an RC filter at the output of Comparator II and connect that to a Schmitt trigger NAND gate such as the CD4093. That will give an output without oscillations as it transitions from one wheel being faster to both being the same speed.