I am trying to convert a sinusoidal input from a mag pickup sensor into a 0 to 5V ttl signal that matches the frequency of the input. The sinusoidal wave form varies from .5V p-to-p up to 30V p-to-p. I have tried using a LM311 chip but the square wave distorts at higher frequencies. Any help is greatly appreciated.

Thanks

 

forgot to attach drawing of circuit I tried

 

Try decreasing R3 to 1k.

The LM311 can sink up to about 8mA current from it's output when Vcc>=4.5v. It's an open-collector output.

5v/1000 Ohms = 5mA. You might go as low as 680 Ohms, but you will see the logic low output increasing.

You may be having problems due to long wires (meaning inductive) or capacitive loads attached to the output of the LM311. This will cause signal degradation. Something that is hard for newbies to understand is that square waves are composed of the fundamental frequency, plus all of the higher odd harmonics of the fundamental frequency.

It is relatively easy to send an audio frequency sine wave signal down a wire. It is far more difficult to do the same with a square wave, as you have noted.

[eta]
You may have better results if you add a resistor in series with the output, say 50 to 100 Ohms, and a small cap, say 50 to 100pF to ground. This will slow the rise/fall times on the wire, reducing the tendency to "ring".

If that doesn't help, you may need to use a TTL differential driver on the LM311 end, twisted/shielded pair wires in the middle, and a differential receiver at the destination.

 

SGT Wookie thanks for the response. The square wave actual turns into 2-3 square waves as the amplitude increases. The first transition occurs at 3 v-p-p sine wave and the next transition at about 6 v-p-p sine wave.

Thanks,

Greg

 

I'd put a series resistor in each leg of the pickup cable (at the amp end), something like 1K, then put back-to-back parallel diodes across the comparator inputs.

I suspect the voltage from the pickup is exceeding the common mode range of the 311.

Adding the resistors and diodes will limit it to about +/- 0.6V which should give a clean output.