Welcome to LinuxCNC.org

Anyone have any questions about the Pico Systems interface boards?
There is the Parallel Port Motion Controller board set for analog servo amps,
the Universal Stepper Controller for step/direction drives, and the
Universal PWM Controller for drives that take pulse width signals.

Jon,

I am still trying to get a grasp of what I need to retrofit the Hurco KMB1. I have a replacement servo motor for the broken one enroute. I talked to the huy on the phone last night. I think you read some of my posts on RCM.

Your boards have been reccomened to me from a couple different sources. Also, Iggy is quite impressed with them.

I'm going to run (for now) the Electrocraft brush servos on the machine with the 90VDC (manual say 80+) power supply in the cabinet.

If you wouldn't mind giving some guidance... I need to some help selecting

1. Servo drivers.
2. Breakout (pc to driver interface - this is the part you sell right?)
3. New encoders. (2 of the existing ones crumbled when touched)
4. A 1p in 3p out, 230V, 15amp VFD that interfaces with the breakout and the control software. (May use the existing VFD temporarily.)

The motors also have tachometers on them, and I am not sure whether to try and make use of them or not.

Some ideas about how to approach software setup. I had leaned towards Mach originally, but the thought of being able to do true closed loop sounds more appealing to me. Mach won't do that.

I think I need drivers that will handle 90V and upto 9 amps peak per motor.

Bob La Londe wrote:Well, there are two ways to go. If you have tachometers on the motors, then you have the option of using velocity servo amplifiers, such as Copley, AMC, Westamp or Servo Dynamics. These give the smoothest motion over the widest range of speeds. Then, you could use my PPMC board set, which is designed for analog servo amplifiers.

If you don't have tachometers, you could use my PWM servo amplifiers and the universal PWM controller.No, I don't sell "breakout boards", but motion control interfaces. The whole idea here is to entirely get rid of the "steps", and deal in encoder position and commanded velocity.Yes, it really sounds like you need to get newer encoders into the machine, if the parts are disintegrating!
I used to like Renco, but they have been bought by Heidenhain and may be harder to get now.Well, if the VFD is good, then there's no reason not to use it. To command spindle speed with my boards, the VFD needs to have a floating or grounded input. The input can't be tied to the mains voltage.OK, then, you may want to look at the availability of velocity servo amps suitable for the motors.Well, you get true closed-loop operation, where you can hit E-stop and not lose the axis alignment of the workpiece coordinates. You can graph the performance of the servos and see what the ACTUAL following error is, rather than guessing. And, you can do rigid tapping if you can put an encoder on the spindle.That is not a very demanding rating, most of the velocity servo amps I mentioned above will work, as well as my PWM servo amps, which are good to 122 V and 20 Amps.

Jon

Ok. Would you mind telling me what services your board(s) provide to the overall system, and if it connects directly to the PC or if it needs a breakout board to connect to the PC. If it connects directly how it is connected. Also, if it works with Mach and/or EMC. Also, how many drivers it can work with. Probably the PWM based on your comments regardign smoother operation with servos that have tachometers.

Does it take steps from the PC or does it get some other form of data from the PC to tell the drivers what to do?


Do you think these would do the trick?
cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&i...9&fromMakeTrack=true

They would fit on my servos, and have greater reolution than the originals, and the price seems decent. Th RPM rating is also greater than my servos by more than a third. The one thing is its not 5V in. I would need to setup a power supply for them. Fortunately I have lots of light duty regulated Altronix 12VDC PS boards laying around.

Won't know if its good or not until I power it up and try it. Its problem is that with 1 phase in its only rated for 7 amps, and the data plate on my motor says 14. I hope to get a bigger unit and not need to install a phase converter. Hence why it may get used "temporarily" while getting the system figured out and running. I could run the motor only under light loads with it. Its manual says it produces unstable voltages if over loaded. Probably only run it with a volt meter hooked up and watched actively. Atleast initially. It takes a standard 0-5V or 0-10V input.

Can you make some suggestions then. I do have tachometers on the motors. However there is a bewildering array of selections available even if you limit yourself to just one or two manufacturers. There are many AMCs on Ebay. Some quite reasonabley priced. Some a little more costly. Is there one that will work with my ElectroCraft serovs, tachs, and encoders (like the one I listed above).

Is that reasonabley (easily) achievable using Mach with your board? EMC? Something else? Do you have a basic XML or other setup file(s) for the controller software that "works" but just needs to be tweaked for the particular machine?

I could probably fit an encoder on the spindle with a little manufacturing. There is room inside the head. It is not fitted with one. I was just thinking to get a tapping head, but an encoder is actually quite a bit cheaper. Of course I could also use the tapping head in my manual mill/drill or in my drill press. And then there is always the other approach. Single point mill tapping. LOL.

AND... what questions don't I know enough to ask?

Bob La Londe wrote:I don't think that Mach does proper closed-loop servo control. It can be used with step/dir servo drives, but that puts the closed-loop in the hardware, invisible to the software. If you want proper closed-loop motor control EMC2 is much more capable."Other Data" it is. The system communicates with a dedicated EMC2 driver, and uses the parallel port as a communication device (sending digital data, not steps).

Bob La Londe wrote:
Ok. Would you mind telling me what services your board(s) provide to the overall system, and if it connects directly to the PC or if it needs a breakout board to connect to the PC. If it connects directly how it is connected. Also, if it works with Mach and/or EMC. Also, how many drivers it can work with. Probably the PWM based on your comments regardign smoother operation with servos that have tachometers.
[/quote]
Assuming we are talking about using analog velocity servo amps which take +/- 10 V velocity command signals, then you would use the PPMC boards. There is a DAC board that provides the analog velocity command ranging from -10 V (full speed reverse) through zero (stop) to +10 V (full speed forward).
This DAC board supports 4 axes, has 16-bit resolution (providing 32,767 distinct values each for forward and reverse), and will independently set the outputs to zero when there is an E-stop condition, without needing to be commanded to do so by the PC.

There is an encoder board that accepts single-ended or differential AB and Z encoder signals, and can
count up to 1 MHz. It works with the PPMC driver in EMC to perform home to index pulse functions and spindle-synchronized moves.

There is a digital I/O board that has 16 opto-isolated inputs and can mount up to 8 solid-state relays for auxilliary control, such as spindle, coolant, etc.

These boards plug into a motherboard that can accept up to 8 of these boards, for system expansion.
The communication with the PC is over the parallel port, but the EPP (IEEE-1284) mode is used for greater speed and reduced workload on the CPU.No, there are no steps anywhere in the system. The EPP mode of the parallel port allows the computer to read and write 8-bit registers on the boards. The encoder board, for instance, reports position as a bank of 12 8-bit registers, which can be read sequentially with only 13 commands over the EPP bus. (The first one selects the first address to be read, then 12 read commands read the 12 sequential registers.)Well, these are relatively low resolution. Assuming a 5 TPI leadscrew directly driven, then you would get 512 * 4 * 5 = 10240 counts/inch, or a resolution of .00009766 inch per count. That isn't too bad, but I'd generally prefer more resolution.Well, you need analog velocity servo amps for brush motors, that meet the voltage and current requirements.No, of course not. This is an EMC list, not Mach. EMC2 definitely supports the PPMC and a sample set of configs files is included on the distribution CD. Since this is a system where the loop is closed in the PC, Mach will not support it.Hard to know!

Jon

Ok so now its sounding like it takes multiple boards to work.

PPMC boards (plural in your description), motherboard, and another encoder board or boards. Not counting drivers which I still have not a clue how to narrow down the selection.

Its starting to sound kind of expensive. Can we spell it out?

What is the equipment list to go... Hook computer on this end. Hook drivers, VFD, encoders, tachs, and assorted relays on the other end.

Bob La Londe wrote:The PPMC was designed for larger machine retrofits, where there might be a lot of digital I/O or other stuff that needs to be interfaced. So, it comes broken down by functions (encoder, DAC, I/O) and you plug in as much as you need for the particular machine.

The basic PPMC board set, in a card cage, is $780. The only thing not in there are the motors, encoders and servo amps. This will do up to 4 motion axes, or 3 axes plus spindle encoder and DAC output to VFD.

The tachometer is only connected to the servo amps. PPMC DACs send command to servo amps, encoders send position info to encoder board. One solid state relay would make whatever connection is needed to enable the servo amps. If the servo amps have a fault signal (sounds like yours might not) that can be wired into the E-stop chain. Forward and reverse solid state relays can be wired to the VFD.

Now, there is also the Universal PWM Controller and PWM servo amps. A 3-axis system doing it that way is $625. If your servo amps are not in working condition, that is another option.

Jon

I currently have installed the size VFD you are needing. A Hitachi sj200-075lfu 200v class. Which was recommended to me by its manufacturer to run my 3ph spindle. It cost right at $775.00 usd and will give you 3.75kw 3 phase from a single phase supply. This is currently being implemented in an upgrade to our mill. These VFD drives are de-rated for single phase as you know to no more than 50% its rated 3ph capacity . You could dl a manual for it from drives warehouse to find more information.

dewey525 wrote:
I will definitely download the manual and take a look. I had about given up on getting full power without setting up a phase converter. Most of the VFDs I have looked at don't list single phase input specs for their models over 3.7Kw.