ECN Forum Forums General Topics of the Trade General Discussion Area Electric motor rotation

There were some resistors and capacitors but only one logic chip. It was in the power on circuit so the motor wouldn't even bump.

I have had the blower running the wrong way problem. A squirrel cage blower almost works going backward. It just doesn't move enough air to keep things cool.

Joe, Yes more channels are usually needed for direction sensing. Single channel can do it, but it is more difficult, and requires ajustment of the periferal devise. But most of all the encoders I use (on motors and servos) are dual channel with a extra reference channel. Servos are usually fitted with a resolver.
Let me clarify what I said regarding counting. You are correct that the encoder does not count. The devise receiving the signal does.
I think we both are correct in our statements, I may have not provided enough information to you to make sense of my statements.
I am not going to get into a "who knows more" match with you. I am not a testing engineer. I just have worked with VFD's and other equipment that employ encoders and resolvers. Also measurement equipment ect..., that incorporate encoders on the equipment. But my statement regarding direction is sensed by the postive or negative output of the encoder. Turn an encoder in one direction it is a postive square wave relative to zero. And turn the other direction it is a negative square wave. Or Visa Versa.
The encoder is not smart. The devise receiving the signal is the smart one. It just receives the signal and does what it is programmed to do. Maybe we should discuss the smart equipment and leave the simple encoder to rest.....John

I suspect you guys are talking about different kinds of "encoders". There are several ways to look at shaft rotation. The typical slotted wheel or magnetic pickup can't resolve direction without using two. If you have one that is like a little generator it can. (AKA tach/generator)

Greg,
The PLM that I repaired just used an old 741 op amp wired up as a comparator with hysteresis. I think one phase just used a cap for phase shift so it would be subtractive at a summing junction in the wrong rotation.

John,
I don't want to sound competitive or belittling, just want to make sure we're on the same page. If there is an encoder out there that behaves in a way I'm not familiar with, I want to be able to add it to my bag-o-tricks for the future. I had looked at your first 2 links and found them too generic. The 3rd referenced the type I'm already familiar with that had available quadrature outputs. I didn't get to the last link though. So I guess that in some instances you mean an encoder subsystem that might include the component that we would call an encoder. So are you saying that you know of one type that produces a zero to let's say +5 volt square wave in one rotation, and zero to -5 volt square wave in the other? If so, please direct me to it so I can research it further.
Thanks,
Joe

Joe,
All of my comments were based on a basic quadrature or single channel encoder. Nothing you would be able to add to your bag of tricks. I think my wording may have been confusing you, as to what I was trying to say.

When I say zero I mean when looking at the encoder on a scope. Square wave up or down for polarity. Zero when the encoder is not turning. Flat DC line zero.
When I connect the encoder to the scope, I connect the encoder to a variable DC source, adjusting the voltage for the rating of the encoder. Then look at the square wave (with the time setting as low as possible) to verify the integrety of the square wave and the voltage. I do this to test the encoder before I realease the equipment to the customer. Thats it. If it's a quad I look for a positive square in one direction and a negative square in the other direction. I turn the encoder manually so I can see the square wave plainly. Single channel, naturally I will see only postive or negative. Not both.

gfretwell,
We are aware of the tach and encoder operation. Tach's or (magnetic pickups)" produce the same square wave output as the encoder does. But requires a gear or similar for the tach to see. As each tooth passes the tach it produces a square wave, relative to the tach input voltage.

Encoders use either glass or metal disc's within them that have holes, blank spots, or lines the the encoder light cannot pass through. Each time the light is blocked it produces the square wave.

I think we are battling terminology here. Most quadrature encoders I have seen determine direction by which of the two square waves gets a rising (or falling) edge first. In firmware, I have done this by XOR'ing with the previous state.
The frequency of the square wave determines the rotation rate.

I haven't seen one that uses two different polarity (wrt ground) squarewaves, but there's a lot of stuff I haven't seen out there.

n1ist, Exactly right on the first paragraph.

Second paragraph:
When you connect an encoder to a DC power source and the encoder is not turning you have a flat line on the screen of the scope that is zero volts. If you turn the encoder a square wave will appear either below the zero line or above the zero line. This is what I meant by polarity. Negative if the square wave is below the zero line and postive if it above the zero line. This is true on each channel.

When I say zero line I mean that there is no output from the encoder.

I was wrong when I said a single channel encoder is only negative or postive. It depends on the direction you are turning the encoder......John

John,
What you were saying still wasn't making sense until you just said the output would be zero volts on the scope unless it was turning. Of course it could be 0 or +5 depending on detector position relative to the slot. UNLESS, you just happen to have your scope switched to AC coupling. Then those things you said start making sense. As a rule, you're better off DC coupling on both your vertical and trigger inputs and using the 20 MHz bandwidth limit if you have one. You benefit by seeing a more accurate waveform and by being able to verify your margins with respect to your hi and lo input logic thresholds.

The really funny thing about this is the Deja Vu feeling I'm getting from a few months ago. My older brother was having interface problems between an encoder and a Precision Digital PD693. Turns out he was AC coupled and didn't notice the DC offset of his square wave source. I had asked about the leading edge overshoot evident in the scope picture and if ground was 0 or 1 division up from the bottom of the trace. When he said the middle, that solved the mystery. Then I was able to suggest that he check various p-p amplitude Vs DC offsets to characterize the input network on the device.

I know we've gotten way out in left field on this but it's nice to finally get all the dots connected.
Joe