Hello All,

I'm new around here, I'm hoping that someone can help me with a strange problem.

I have a SLC5/03 with a 1746-HSTP1 card in the rack. I'm sending the step and direction pulse train into an Ultra 3000 base drive - into pins 4 & 5 (step) and 6 & 7 (direction).

The application simply turns a wire cutter, so it turns one rev at a time, with an adjustable time delay from 0 to 25% between cycles. I notice that whenever I turn on a bit to start the pulse train, there is a delay with the servo. It takes about a 1/4 second to respond. Then, when I put the program into 'auto' mode, I notice even with 0 delay programmed, there is still a 1/4 second delay between revs. However, the 'CW' indicator on the HSTP1 only blinks very, very briefly (as though the pulse train is being output correctly).

I thought maybe I had a problem with the servo itself, so I swapped out the AB servo for an Emerson one that I had kicking around. It had the same response.

I searched exhaustively for a wiring diagram between the 2 units - came up with nothing. Does anyone have any experience with this stepper card and interfacing it with an Ultra 3000?

I'm thinking I have either a wiring problem, the HSTP1 is faulty, or the HSTP1 cannot be programmed to do back-to-back relative moves without a time delay (which seems unlikely).

Any help would be greatly appreciated.

Cheers!

Some food for thought and a couple of questions. I did not read thru the entire HSTP1 Manual on AB.com so I don't have the answer for this, but what does your SLC code look like. Do you send a move command, wait for a move complete response and then send another move or are you cueing them in the HTSP1 somehow.

Keep in mind that Command Transfer to the HSTP1 is not an instantaneous event and can take some time {read several ms here}.

Also keep in mind that the HSTP1 pulse train must reach an amplifier and be amplified before motion occurred. Remember from basic electricity that all amplifiers have propagation delay {read several ms here}.

Also keep in mind that the magnetic field of the motor must over come inertia before visible motion occurs. {read another few MS here}

Can these events add up to 1/4 second? Quite likely.

I'm also curious where and how you are measuring your 1/4 second.

Thanks for the reply,

My code turns on the 'relative move' bit, waits for 'motion complete' bit to come on and repeats. I figured on a few ms of delay, not in the magnitude of 250ms (which by the way is only a guess on my part - but it is a very notable delay).

The delay can also be seen if I send the HSTP1 a jog command, it takes again about a 1/4 second between toggling the bit and the motor beginning to turn.

The other strange thing is that the acceleration of the servo is not a smooth ramp, although I have the same accel and decel rates programmed, it's as though the servo is trying to 'catch up' to the pulse train. Also, the motion does not seem synchronous with the 'CW' light on the stepper card, it seems to lag behind, judging by the brief flicker between moves. This is what led me to believe that the problem was the servo drive itself, but swapping it out gave the same response.

I've programmed many pulse and direction controllers (not AB) with many servo drives (not AB) and have never had a problem like this. I would certainly expect some delay (50ms at the most?), but this is very pronounced. I would barely expect to be able to see the pause between moves with my eye.

I've been waiting for AB to send me a wiring diagram between the controller and the drive, but I don't think that is the problem either. I mean, we're only talking 4 wires here...

I'm near my wits end on this one... thanks for your input!

Have not worked with Ultra for years, but just a speculation assuming no mistakes in the software.
If the drive setting was changed (increased) for positioning done detection threshold, the drive will not start along with the pulsetrain, but only when the position error exceeds the mentioned threshold. This may look as start delay after turning on the bit and "catch up to the pulse train".
On deceleration the drive will stop before theoretical target for the same reason, while the pulsetrain may still be not over.
Along with delay on start this may explain the overall pause and discrepancy with led.



This is why I always follow the golden rule: If can avoid AB- do so.