Hello Guest it is August 03, 2020, 08:29:44 PM

Author Topic: Servos progressing well, but question about index  (Read 4124 times)

0 Members and 1 Guest are viewing this topic.

Offline DMBGO

*
  •  200 200
  • Time is the same in a relative way
    • View Profile
Re: Servos progressing well, but question about index
« Reply #10 on: September 22, 2017, 05:14:42 PM »
Craig, you said
"The actual and programmed speed should be within a few parts per million.....you are wasting time and money if you try improve on that.
No doubt if I read the manual closely there will be an encoder output. This could be used by Mach to measure the rpm and display it if
you want to be totally anal retentive about it."

Is this a common output on servo drives? Do they have a single output that has all of the encoder pulses? Is this too fast for Mach, since we are talking about 2.5 mhz at 1000 Rpm? So many questions, so little time......

Cheers
Dave
Cheers

David

Offline RICH

*
  • *
  •  7,411 7,411
    • View Profile
Re: Servos progressing well, but question about index
« Reply #11 on: September 22, 2017, 05:57:36 PM »
Mach3 threading is based on a spindle index being used since it is required to read the TRUE rpm and also provide triggering to start the threading path. The true rpm is  to 3 to 5 decimal places if memory serves me,but, you do not see that in the RPM dro. You can also use a different pitch for the thread to refine it if you wish ie; say for a 20 tpi 1/20=0.050 on could use say .0499123456 and it will use that pitch.

There is no value in using an index with more than one slot. An encoder is not supported only a single disc slot type is as an index. If you measure your spindle rpm make sure the you are using a tool that gives you good resolution on the measuerment, else, trust me that Mach is more accurate than a crappy tach.
Things are the way they are!

Frankly if your can set your spindle  manualy and it will maintain the rpm for whatever pitch thread and depth of cut your taking then half the battle is won.In Mach you can't have two devices trying to control the feedrate as they would be fighting each other. Additionaly Mach will adjust the next path based on spindle slow ( range in the say 25% and i even tested it out to about 75% for punny lathes )With experience one can find what their lathe will and will not do and apply knowledge to accomplish a threading task.

Now how accurate is the threading? Well that depends on the "complete lathe system" and the criteria base is defined in the Threading on The Lathe manual I wrote. A quick test would be to do  multiple start threads and see how the tigger and start of the thread matches ( when you get to 3 start you will know what i am talking about and for the average guy how many two or three start threads are you going to do?). Lucky most folks will ever accomplish a class 3A thread. If you want more accurate threading them go to formed inserts and don't do single point threading. BTW, I asume you will using accurate thread cages to test the finished thread, high end comparator or micoscope. What more can one say..............

RICH
« Last Edit: September 22, 2017, 06:00:40 PM by RICH »
Re: Servos progressing well, but question about index
« Reply #12 on: September 22, 2017, 06:24:42 PM »
Hi Dave,
most servo drives have programmable outputs, usually encoder outputs. They can and often are manipulated versions of the encoder fitted to the servo.
I haven't read the manual for your servo but would guess quadrature A and B channels and an Index channel. If some or all of that info is programmed
as an output of the servo drive than Mach can read it. The servo speed will be that close to its programmed speed it doesn't serve any useful purpose except
perhaps peace of mind. The Index however would be very valuable as Mach uses that for speed, which we've seen has little value, but very importantly the
index can synchronise the start of the thread.

As Rich has pointed out that half the battle with threading is to achieve a constant spindle speed in face of a varying load. Your servo and drive should
manage that with aplomb. Now all you need is an indexing pulse. You could add some extra external hardware or you may be able to use the index pulse
built into the servo encoder. Whether the duration of the included index pulse is suitable for Mach is a matter of investigation. The pulse may have to
be manipulated electronically for it to be useful for Mach. Either way is very achievable and you will end up with a very capable spindle that handles threading
with ease...be happy!

Craig
My wife left with my best friend...
     and I miss him!

Offline DMBGO

*
  •  200 200
  • Time is the same in a relative way
    • View Profile
Re: Servos progressing well, but question about index
« Reply #13 on: September 22, 2017, 06:41:37 PM »
Thanks Craig, Rich et al. I'm going to put my old hall element back as a spindle encoder and connect it to the index input. The rest, while academically interesting, doesn't sound like it makes much difference. Thanks again. I'll let you know how it goes.
Cheers
Dave
Cheers

David

Offline rcaffin

*
  •  1,029 1,029
    • View Profile
Re: Servos progressing well, but question about index
« Reply #14 on: September 22, 2017, 07:00:11 PM »
Quote
if your can set your spindle  manualy and it will maintain the rpm for whatever pitch thread and depth of cut your taking ...
.. . half the battle with threading is to achieve a constant spindle speed in face of a varying load. Your servo and drive should
manage that with aplomb.
Aye, and there's the rub.

So many smaller machines can NOT hold the spindle speed exactly constant under varying loads. They are not being managed by a servo system, just an open loop motor. The spindle will vary in both speed and in phase as the load changes. And during threading, the load does change quite a bit.

To be sure, the index pulse is also essential to trigger the Z axis movement: it sets the phase. Ypu will still need that.

No matter how good the PID control, it can NOT control the spindle to adequate stability within a single revolution. The controller simply does not have the necessary feedback information.

Cheers
Roger
Re: Servos progressing well, but question about index
« Reply #15 on: September 22, 2017, 07:00:48 PM »
Hi Dave,
RTFM, you don't need it
Quote
PZOUT+ Encoder Z phase differential positive output
PZOUT- Encoder Z phase differential negative output

The built in index pulse form the encoder is already available as an output, pins 42 and 43 of CN1. You are just making work for yourself, the manufacturer is
way ahead of you.
Secondly note that the A and B channels of the encoder are 2500 pulses per rev, when they are combined as a quadrature signal they give four times that as
encoder count or resolution. 10,000 count encoders like this are pretty much the norm for incremental encoders on current production models.
10,000 count encoder will give you potentially 2.16 arc minute resolution, surely that enuf even for an Aussie.
Last thing is that Mach ASSUMES constant spindle speed when threading. Your spindle could conceivably provide variance information in between each
revolution buy Mach can't do anything with anyway.

Craig
My wife left with my best friend...
     and I miss him!

Offline rcaffin

*
  •  1,029 1,029
    • View Profile
Re: Servos progressing well, but question about index
« Reply #16 on: September 22, 2017, 07:04:47 PM »
Quote
Mach ASSUMES constant spindle speed when threading. Your spindle could conceivably provide variance information in between each
revolution buy Mach can't do anything with anyway.
Correct. That is why the spindle servo really needs to be moved into the external pulse engine. This is possible.

Cheers
Roger
Re: Servos progressing well, but question about index
« Reply #17 on: September 22, 2017, 07:05:44 PM »
Hi Roger,
I disagree, the refresh rate on my Allen Bradley servo also used as a spindle motor is 8kHz. The manufacturer claims a control bandwidth of over 500Hz.
At max speed my spindle is rotating at 58 rev/second, easily slow enuf that the control loop is trying to close the error within a revolution.

Craig
My wife left with my best friend...
     and I miss him!
Re: Servos progressing well, but question about index
« Reply #18 on: September 22, 2017, 07:16:52 PM »
Hi Roger,
just as a matter of interest I'm making a servo drive for a large servo I plan to use as a lathe spindle motor. The servo looks new but is in fact
from the mid nineties and has a four pole resolver and the only applicable drives I can find are beyond my budget, hence I'm making one.

Its refresh rate is 20kHz for Nyquist rate of 10kHz and pretty fair discrimination at 1kHz. Has been a lot of fun working out how it all works, its
been many years since I've had to get serious about control....isn't Matlab just the bees knees!

Craig
My wife left with my best friend...
     and I miss him!

Offline RICH

*
  • *
  •  7,411 7,411
    • View Profile
Re: Servos progressing well, but question about index
« Reply #19 on: September 22, 2017, 08:04:13 PM »
Quote
is that Mach ASSUMES constant spindle speed when threading

Not quite right........

Later,
RICH