Encoder Logic Range?

[agmlego]

My team has been noticing that our US Digital encoders for the Toughboxes are not sending data to the Digital Sidecar, and I suspect I found the reason. According to an oscilloscope, the encoder is making a (roughly) square wave output bounded by the range [-1V,1V], and both outputs are producing the same wave pattern. This was confirmed to be the case on all four encoders we have on our system. Is anyone else noticing this problem, or is it something I am doing wrong?

[Uberbots]

have you looked at both waves superimposed (using both inputs on the sillyscope)?
the B wave should be trailing the A wave by a quarter-cycle if i remember correctly. It should also be a 0 to 5 volt pulse. are you sure you have the right encoders?

[EricLeifermann]

yes we have the right encoders

[agmlego]

Yes, I am quite sure we have the right encoders, as our four were ordered in two pairs, one from the KoP and one from Andy-Mark on the first or second week of build. According to the 'scope, the two channels' waveforms were roughly a quarter period or so opposed, however the range was very clearly bounded by [-1V,1V]. This is odd, because the data sheet for the sensor itself says that the signal should be the expected [0V,5V]. I can build a translating circuit, but I should not have to, though finding transistors that can switch at ~8.1MHz may be (ahem) interesting. Any ideas?

[lynca]

Originally Posted by agmlego

Yes, I am quite sure we have the right encoders, as our four were ordered in two pairs, one from the KoP and one from Andy-Mark on the first or second week of build. According to the 'scope, the two channels' waveforms were roughly a quarter period or so opposed, however the range was very clearly bounded by [-1V,1V]. This is odd, because the data sheet for the sensor itself says that the signal should be the expected [0V,5V]. I can build a translating circuit, but I should not have to, though finding transistors that can switch at ~8.1MHz may be (ahem) interesting. Any ideas?

We spent a few hours trying to understand these signals with the US digital encoder with no luck.

We promptly switched to Grayhill 61R and another encoder module with latex tubing. We are not going to use the US digital modules.

[Alan Anderson]

Originally Posted by agmlego

According to an oscilloscope, the encoder is making a (roughly) square wave output bounded by the range [-1V,1V], and both outputs are producing the same wave pattern. This was confirmed to be the case on all four encoders we have on our system. Is anyone else noticing this problem, or is it something I am doing wrong?

We can't tell if you're doing something wrong until you tell us what you're doing.

Specifically, how have you connected the encoders? They have four wires, each with a different color. Where does each wire go?

[windell747]

Why do you need to have such a high switching rate for the transitors? You shouldn't need to count pulses using electronics since the speed of the processor is so high and all the inputs are read simultaneously. The inputs are read at something like 38kHz. This limits the max freq of the encoder pulses to at most 19kHz. That is a rate of about 80 rev/sec of the encoders or 4800rpm.

If you really want to do it using ICs why not use a counter chip or build one using a shift register?