What's new
What's new

Encoder connection - differential to single ended converter

gregormarwick

Diamond
Joined
Feb 7, 2007
Location
Aberdeen, UK
Is anyone aware of an off-the-shelf converter to condition a differential encoder signal to a single ended one?

Standard 3 channel A, B, Z.

I found a nice din rail mounted one that would be ideal, but the company is in the US.

Time constraints necessitate a UK source.

The machine has an IO board that all the other encoders connect to, that has a spare channel available with a proper quad differential line receiver, but there are some obstacles to using that, and if a separate converter is available it would be an easier/quicker solution.

TIA.
 
I use A, B,and Z(or I) and ignore the /(or NOT) pins. Single ended works fine, don't need a converter.

Ed.

Yeah, if it was a physical encoder that would usually work, and that's how I tried it initially.

It's a generated quadrature output from the servo amplifier, 5v differential. Max. positive voltage shows as 2.6v when I scope the output, and I guess that's just too low to trigger the optocouples on the control.

For context, I am replacing an old recently deceased brushed DC servo on the 4th axis of the eldest of our Hurcos, with a new ac drive and motor.

The main servos on the machine are all ac yaskawas with similar generated feedback. They all go through a differential line receiver on an io board before going to the control, but the original A axis servo bypassed all that.

I don't have the connectors on hand to hook up to the io board, but I have ordered them as a fallback. Also the io board doesn't include io for the A clamp etc. but I can separate that out if need be.
 
There is a differential serial protocol, RS422. Maybe get a converter for that? One for each signal? Check to see if your low max voltage will work.

Ed.
 
Is anyone aware of an off-the-shelf converter to condition a differential encoder signal to a single ended one?

I actually just went through that, if I'm understanding the question. Just purchased a new accurite scale, to fit to an older accurite control unit.
Cable had the wrong connector (D connector) and would not fit the controller (round multi-pin). A call to Accurite produced the solution which
was a differential to single-ended converter cable, not sure what magic was in there but it was less than a hundred bucks. Accurite gave me their
part number, and the MSC part number.

MSC was quite helpful.
 
There is a differential serial protocol, RS422. Maybe get a converter for that? One for each signal? Check to see if your low max voltage will work.

Ed.

An RS422 receiver would work fine, in effect that's what the line receiver on the extant IO board is. I was just hoping to find a plug and play solution to save some time.

I actually just went through that, if I'm understanding the question. Just purchased a new accurite scale, to fit to an older accurite control unit.
Cable had the wrong connector (D connector) and would not fit the controller (round multi-pin). A call to Accurite produced the solution which
was a differential to single-ended converter cable, not sure what magic was in there but it was less than a hundred bucks. Accurite gave me their
part number, and the MSC part number.

MSC was quite helpful.

I'm guessing that cable is most likely a differential line receiver inside the connector that plugs into the reader. If so, that also would likely work, thanks for the suggestion.

The connectors I ordered for the io board will be here tomorrow, so I guess I will just bite the bullet and hook it up to that, and the queue of parts waiting to go into the machine can wait a bit longer!
 
ZS instruments makes similar adapters in the US and we use the following microchips (installed on a small PCB inside of a DB9 connector shell) to do the convertion:

AM26C31 differential line drivers to converts TTL to RS422
AM26C32 differential receivers to convert RS422 to TTL

If you are in rush to get it fixed quick, AM26C32 with a few wires and a termination resistor or resistor/capacitor should do a trick. You mentioned that other axes used line receivers. You can look at the circuit and try replicating it on a small bread board. You should be able to source components locally in a short time.

P.S. Some systems used HTL logic levels, those might require a slightly different solution.
 
I connected it through the spare line receiver on the io board and it's working properly. Cabling is a bit messy due to having to split the harness to route it to three different parts of the cabinet, but it's done.

Thanks for the suggestions.
 








 
Back
Top