. . .  are what they call intelligent encoders and you can have, if I recall correctly,  the drive interpolate up to 2 million counts per rev(8million pulses per revolution  ), thats even beyond SmoothStepper territory I was happy just at the base 1024 x 4 for a spindle.

I had 1800x4 on the 4th axis (good for division by 360) with the DC brush servo. With both the Dugong drive and the Granite drive, it makes no noise when holding still.

Currently I am working with two different series Mitsubishi AC servo drives now, the larger of which I *think* is similar to the Alan Bradley 3000 series . . maybe. The little Mitsu MR-C has only 4000 pulses per rev and 'sings' (to quote Gecko) when holding, presumably due to bouncing back and forth between adjacent pulses. This is very noticeable on the MR-C, but with the J2S drive, while it still sings, it is almost inaudible perhaps due to a much higher frequency and much smaller step (131,000 steps). That servo motor is also brand new so that may also have some contribution to the far lower noise level.

So, my question is; does the Allan Bradly drive/motor behave similarly, i.e. do they 'sing' or make any noise when holding still?

No, if tuned correctly they have very little noise and if you monitor the encoder in the software it does not vary at all.

Hood

No, if tuned correctly they have very little noise and if you monitor the encoder in the software it does not vary at all.

Thx. I have the mistu software, but have not used it yet, however the J2S series drive has a 4 digit DRO that can be set to display just about any data in real time, including encoder feedback, so I can monitor that without the software. The J2S is the first servo drive I have used that I can say I am completely satisfied with . . . so far  . I would describe the J2S similarly to how you rate the Allan Bradley; 'very little noise'.

I will be needing more AC servos so I am just collecting info on the options.  The Mistubishi J3 is a freak! Too new though, and not readily avaialble from the junkyards yet.

I had thought about buying some Yaskawas that I saw for a decent price but decide I would just stick with what I know and like. I now have at least one spare drive for all that I use, even a spare for the 22KW drive thats on the Lathes spindle
The older Allen Bradley drives I didnt rate quite so much but they were still ok but lacked some I/O that the newer ones have.

Hood

I had thought about buying some Yaskawas that I saw for a decent price but decide I would just stick with what I know and like. I now have at least one spare drive for all that I use, even a spare for the 22KW drive thats on the Lathes spindle

You are wise, ObeeWan!  Spares is a definate consideration. I don't want to end up like I did with the DC servos having a different brand on every axis. Pretty expensive to have spares that way! I have been paying attention to 'general availability' as an important parameter since I can't aford to buy a bunch of new drives at this level. Mitsubishi J2S are adequate for my needs and nearly always available for reasonable prices from 200watt to 750watt, which is the range I am interested in at the moment.

The older Allen Bradley drives I didnt rate quite so much but they were still ok but lacked some I/O that the newer ones have.

Again, I have to agree that the I/O is really a huge difference between the 'hobby' and 'industrial' levels. Lack of same was a real frustration with the inexpensive DC servo drives I was trying to use previously. 

Question; what are the series of the 'older' and 'newer' models of the Allen Bradleys you are using?
[/quote]

The older drives still have more I/O and more configurable than the normal hobby type drives but just not as good as the later ones,  then again compared to the hobby drives they are a lot more expensive new so its understandable.
The older drives are Allen Bradley DDM or Giddings and Lewis DSM
The newer drives are Allen Bradley DSD or Giddings and Lewis DSA

Software  required for DDM/DSM is UltraMaster/DSMPro
Software  required for DSD/DSA is UltraWare/DSAPro
I have both if you ever require just shout.

Hood

Thanks!

Back on topic, I am starting to use my new computer and I found some interesting behavior that I want to contribute to the kernel speed discussion. On my new setup, I can run 100K kernel thru the first BOB, but not thru the second. I have not seen this mentined before in speed discussions and it really did not occur to me until today when I was trying to run the 100K stream thru the second BOB to the 4th axis.

In this case the drive has a 4 place LED that can display (among other stuff) the pulse frequency coming into the drive. On the X axis testing (BOB number one) it was showing the 100k steady. Now I have it on the 4th axis on BOB number2 and it is shows unstable anything over 60k.

Bottom line is that the computer, which is always suspect number one, is not necessarily always the culprit.  The BOB has to be able to pass the stream thru it's optos and other circuits at that speed as well.  Seems obvious now, but it never occurred to me before.

There is that but it couldnt be the case for me as its now happy at much more via the SmoothStepper.
Hood

There is that but it couldnt be the case for me as its now happy at much more via the SmoothStepper.
Hood

Since, as I understand it, the BOB is downstream of the smoothstepper, wouldn't the problem still apply?

Incidentally, I have purchased a smoothstepper, so I will be better able to comment on it after I have some 'hands on' time with it . . . which may not be any time soon, but eventually I will play with it.

Since, as I understand it, the BOB is downstream of the smoothstepper, wouldn't the problem still apply?

Not sure I read your post correctly then. What I thought you were meaning was some BOB's couldn't pass the higher frequency, if that was what you meant then the PC then SS then BOB would surely suffer the same way as the PC then PP then BOB?

Hood