SSI?

[paulengr]

Has anyone worked with the SSI protocol for absolute encoders yet? It seems to be growing in popularity among encoder vendors. Stegmann posted that essentially it is a 24 bit clocked output. On each rising (or falling?) edge of the clock line, it will output one bit from the encoder. This leaves more questions than answers. For instance it appears that you can do a lot more with SSI than that based on the descriptions. Also, how would one resynchronize everything if communication is lost between the encoder and the PLC (or maybe this problem is simply ignored so you must power down...)? What is the format? Is 24 bits all there is to it and you better make darned sure you know where you are at? If so, then although these things obviously eliminate the need for high speed encoders or tons of digital I/O taken up reading the binary/grey signal from an absolute encoder but it brings on a whole new set of questions about speed since with that kind of setup, it would take 24 scans to read the signal!

[Peter Nachtwey]

It takes time to clock the data back. In high speed applications I would still stick to quadrature encoders. If I am gearing to a feed chaing or conveyor and only need the speed then an encode is fine. If I also need to know the position because there are parts at certain locations on the conveyor then I would use an SSI. When an encoder loses power the system needs to be re-homed. This takes time and often the production on the conveyor is lost. A SSI encoder's position can be read at start up so many time you can just reset the state machine to match the position of the SSI encoder and go from where yhou left off. This save time and production.

[TimWilborne]

Thanks for the explanation Peter. It answered several questions I had about the M02AS I'm switching to.

[panic mode]

TW, Can you please explain why are you switching to M02AS and from what? I used them 7 years ago. If I had to do it now I'd probably rather look into something network bases (SERCOS on AB, SSCNET on Mitsubishi etc.) Curious...

[TimWilborne]

I am replacing a MO2AE. The main reason I am doing this is I am changing to a linear transducer. I was going to go with the HYD02 but Peter suggested I go with the M02AS and a transducer with a SSI interface. It is a hydraulic servo. Need the -10 to 10 VDC output. THIS was my original thread

[panic mode]

gotcha...

[Gerry]

B.S. The MO2AS has only been available for the past 2-3 years.

[TimWilborne]

I had never heard of the M02AS until Peter suggested it. I always used the M02AE which has been out that long. His same questions would be reasonable when compared to SERCOS. Maybe Panic was thinking of the M02AE?

[panic mode]

you are right, i meant to say that product i was using was AE not AS (1756-MO2AE) i might be wrong on this but they seam to be pretty much same other than feedback interface (MO2AS uses SSI). i didn't like the MO2AE for number of reasons. they sure worked but due analog nauture of device one could monitor only limited number of parameters. for some reasons these cards ware running red hot and two of them died soon after machine was commissioned. i wasn't aware of this issue at design time and i had four of them next to CPU. I was trying to keep them as far as possible from biggest noise sources but it turned out that hotest cards ware all placed side by side even though i had few spare slots in the rack. Another reason was really bulky wiring of the card with pair of thumb thick cables strapped with some more wires, not a pretty sight...

[Peter Nachtwey]

How about an Ethernet cable to the controller in a j box at the end of the machine? One can get lots of data over Ethernet.

[panic mode]

unfortunately every system has limitations. question is do those limitations present a problem or not. MO2AE is a motion controller and PLC's interface to the servo drive. PLC can read from it position or speed for example but not servo drive diagnostic, motor torque or current etc. network based systems normally allow this and more (reading and setting drive parameters, not just motion controller parameters, reading status of servo I/O, version number etc.). i just happen to have higher expectations every day.

[TimWilborne]

I am surprised, or maybe a little unconvinced of what was causing the failures of the modules. I have ran 6 of these modules side by side and have never had one die. I have one panel that pushes the 140 F limit. I'm not trying to start an argument. I know if you say it happened then it happened. I am just trying to understand how so I can make sure I don't run into the same problems.