ACC24-E2A Encoder Count Error

What constitutes an encoder count error?

 

I am tring to help troubleshoot an issue with a UMAC ACC24-E2A tied to an Incremental Quadrature Encoder. The installation has been functioning for ~2years and now started getting encoder count error faults (X:$078208,8) bit gets set.

 

The encoder, ACC24-E2A board, UMAC rack, encoder Isolator, and field wires.

 

Not sure what to check next, but would like to know exactly what the ACC24-E2A board looks at in determinining an encoder count error.

 

Thanks

This would be encoder two on the first board in your UMAC rack, correct?

 

Generally this means you are getting illegal transitions on either A or B encoder signals. For quadrature signals, certain transition sequences should not be possible and this is what PMAC is monitoring. I believe this is done in hardware on the ASIC chip. Generally indicates a channel is going bad, or you are having a noise problem on the encoder signal. I don't think PMAC identifies the specific error, only that an illegal transition has occured.

 

Check for:

- damaged encoder cable. This can be difficult to find if there is an internal break that faults occasionally. Swap it out if you have a spare, or with another channel.

- bad connection at a terminal

- faulty encoder

- noisy encoder power supply (not likely unless using an external supply

- verify the resistor packs on the 24E2 board are installed correctly for single ended or differential encoder signals. Line driver encoder outputs are vastly superior, but if you have single ended signals, the 24E2 board will accommodate by reversing the resistor packs. Single ended signals with line driver hardware config will be marginal at best.

- if noise is an issue due to long cables, you can add terminating resistor packs This is explained in the manual.

 

Even though the system has operated ok for two years, I would still verify the resistor packs on the board.

It is more likely though to be a cable, especially if they flex a lot as the machine moves - or the encoder itself.

 

I just reread your post & noticed an encoder isolator - can you remove this? I would include it in the suspect list. Is there a common mode voltage you're trying to isolate?

 

 

See pages 49-53 of the Turbo PMAC User Manual for more detail on the encoder loss bit, i.e., why it triggers.

Technically, the event that causes an encoder count error is the changing of state of both the A and B inputs to the ASIC decoder during the same encoder sample clock (SCLK) cycle. The decoder cannot make sense of this state change. There can be a variety of underlying causes, as mbalentine points out.

This generally implies that the encoder has gone bad, or requires a faster hardware clock. The fact that it has been functioning for two years leans more towards the first.

Try using another "good" channel. This should give a good indications whether the problem is in the encoder or the axis card channel.