Messages

31

CS-Lab / Re: CMPIO/IP-S with MACH4 - new plug-in released - Servo Spindle still broken?

« on: August 07, 2015, 06:53:46 AM »

 After writing to the Distributor, who then also asked CS-Labs to answer, I received the following official position from CS Labs:



CS-Lab.eu 
 
Attachments3:54 AM (1 hour ago)
 
Reply

to steve, Automationtech.

Dear sir,

1) There is no such a document as Mach4 isn't ready yet nd is still under construction as well as CSMIO plugin for it.

2) Because of what I have written above there is no full support for Mach4.

We do not recommend  retrofits using Mach4 as there are many functions still not supportet or not working as they finally should.

We still recommend to use Mach3 if you don't want to wait and necessarily want to use Mach4 then use some other company motion controller.

3) Please go to terms and conditions http://www.cs-lab.eu/en/upload/pdf/Terms%20and%20conditions.pdf

4) We do not know Vital System products to well.

Regards

Wojtek Trawicki
 

CS-Lab s.c.

J.Wawak, A. Rogożynski, S. Paprocki

--------------------------------------------------------------

32

CS-Lab / Re: CMPIO/IP-S with MACH4 - new plug-in released - Servo Spindle still broken?

« on: August 06, 2015, 08:58:59 AM »

I don't have the controller installed on a machine yet because ultimately I may want to return it, but I have a scope on the differential outputs and I can see a clean square wave coming off each axis, so I'm not worried about that so much. With rewritten M3, M4, M5 and the script for the screen spindle ON button, I have reliable signals coming off the 'Spindle' axis and speed changes via G-code 'S' seem to be working correctly with the previous (.602) plug-in.

This fix does not work with the new plug-in and I am not inclined to spend any more time on this issue.

So, in fact, I HAVE a working MACH4 setup with this device. The thing looks fantastic, very high quality and it has great specs and an array of features for industrial drives. My concern is that they take 8 months to release a new plug-in that claims to solve a long standing problem when in fact it is made worse. Combine that with unresponsive tech support and it does not bode well for the long term.

Current Status: I have sent an e-mail to Automation Technologies (USA distributor that I purchased the CSMIO from) that included the a copy of the e-mail that I previously sent to CS Labs and asked for either an answer, or instructions on how to return the device. I also asked if they sell the Vital Systems product or have any specific recommendations for an Ethernet Motion controller for MACH4 that is designed for industrial drives.

I will post back whatever the result is.

Meanwhile, has anybody been successful with the new plug on and/or does anyone have configuration instructions for MACH4 with this controller?

33

CS-Lab / CMPIO/IP-S with MACH4 - new plug-in released - Servo Spindle still broken?

« on: August 05, 2015, 07:05:29 PM »

I posted this a couple days ago in the MACH4 section and got 80 reads but no responses. From that, I gather that there is a lot of interest, but nobody is actually using this controller with MACH4 yet. So I am posting here in hopes of a better response.

This question was asked of CS Labs tech support an a few days have passed with no response. Unimpressive tech support, so far.

I am doing the first of several upgrades to MACH4 and replacing K-flop controllers with the CSMIO/IP-S Ethernet controller. I was warned that the MACH4 plug In was problematic as relates to a servo driven spindle, so I was prepared for that and created a work around and reported the issue to the MACH people because it "seems' to me to be a MACH4 problem. MACH3 looks like it works fine, as expected.

The logic in selecting the CSMIO is that it is widely reported to be bulletproof with MACH3, so that provides a 'back up' plan in the event MACH4 proves to be unuseable. I looked at Vital Systems, but the cost (with all of the needed ala carte add-ons) is hard to justify. However, at this point, if the lack of support from CS Labs continued, I will return this first unit and go with Vital Systems despite the cost. Fighting with problematic equipment with no support from the manuf. is also a cost associated with any device, so the $700 CSMIO is getting more expensive by the day.

A few days ago CS-Labs released a new plug-in for MACH4 and specifically called out this issue and said it was corrected. However, so far I can not get the Servo driven spindle to work AT ALL.

I have reported this to CS-labs and hopefully they will be responsive and produce some kind of installation document. I was surprised to find none available (unless I am looking in the wrong places).

The previous plug-in is functional (with a work around for the spindle issue) and of course the option is always there to rewind all the way back to MACH3, so it's not a critical issue, but I would imagine a lot of MACH4 fans will be interested in the solution.

CS labs claim to have tested the plug-in with MACH4 v 2580, so I'm hoping I'm just doing something wrong. So, does anyone know how to correctly configure a servo drive as the spindle in MACH4

34

Mach4 General Discussion / Re: MACH4 - Modbus

« on: August 05, 2015, 06:53:56 PM »

I have a question about MACH3 TCP Modbus; I am testing a TCP Modbus device with MACH3 and while running Modbus TCP, MACH3 is using 45% to 52%of one side of a 3.2 ghz Core Intel dual processor. Jogging is sporadic with DRO updates happening in 'chunks' rather than smoothly. Turning off the Modbus results in smooth MACH3 operation and 2 to 3% CPU at idle.

Is this normal?

FYI, I was having the same issue with very high processor usage . . . Joe

Is this with MACH3 or MACH4?

35

Mach4 General Discussion / Re: Mach4 and Windows 10

« on: August 03, 2015, 02:47:21 PM »

I have Win7 Pro 64 bit on this machine and a couple of other machines with Win7 Pro 32 bit, including the CNC computers.

There is something about my development machine (64 bit) that has me on hold for the upgrade, so they say, so probably one of the less complex 32 bit machines will take the plunge first.

I'll post back the results when I have some . . . provided my internet still works . . 

36

Mach4 General Discussion / Re: Mach 4 Bug Reports

« on: August 03, 2015, 07:43:46 AM »

Setting up MACH4 with a CSMIO/IP-S

Step/Dir servo powered spindle will not start unless spindle speed is changed via G-code 'S' command or override.

Some of the problem went away with build 2580. M4 works now, but M3 does not.

Based on the behavior, I would speculate that MACH4 only sends speed change info to the plug-in, because altering the speed started the drive.

It would seem that the speed data needs to be sent to the plug-in with any spindle start command. Why M4 works and M3 does not is a mystery.

M3 used alone does not start the spindle, however the following:

S5 M3
S1000

results in the spindle starting. Prior to Build 2580, the same workaround was needed with M4, but that seems t be working now.

37

Mach4 General Discussion / Re: MACH4 - Modbus

« on: July 03, 2015, 05:21:57 AM »

I've always used bit packing with Mach3, it's been reliable, so I don't know why you say it is broken.

Peter

Interesting post made only two weeks ago by 'poppabear'

http://www.machsupport.com/forum/index.php/topic,30299.msg210714.html#msg210714

the "Bit-Of-Word" (or bit only), is broken in Briais, (in other words the Bit-packing) does not work.
for each "Bit" you want to control, you have to send the entire word for that ONE bit.

Scott 

38

Mach4 Videos / Re: Mach4 - Arduino and Touchscreen

« on: July 01, 2015, 07:02:44 AM »

In addition to using co-processors like the single chip Ethernet and possibly a Pokeys board to run the TCP Modbus (both mentioned earlier), I wanted to include a practical example of off-loading processing from the Arduino to MACH4 as also mentioned in an earlier post.

The CPU in your computer is many times more powerful than any of the little development boards. The version of LUA that is embedded in MACH4 is all Floating Point (not a good thing), but it is also extremely fast and according to statements by the MACH4 team, can run in its own thread (as was done for the PLC).

Here is the example:

The actual test case is collecting data from a pair of quadrature encoders on a Joystick and sending the data to MACH4 for use in performing jogging. The data needs considerable processing to determine basic stuff like stick position, but also more complex calculation to determine the speed at which the stick is being moved (including detecting zero stick movement) and calculating an exponential ramp based on the combined coordinate distance from center, while remaining biased toward the greater deflected axis.

This processing was being done by the DUE prior to sending the final jog speed, axis and direction data over Modbus for MACH4 to use in moving the mill table. Averaging several available benchmarks, the DUE works out to be about 7.5x faster than the MEGA at munching on FP calculations. But a 3.2ghx Intel CPU is faster . . . . .  a LOT faster.

The processing described above has been moved to LUA and the DUE  now simply collects and sends the raw data. While I did not do formal speed tests, I would estimate based on the frequency of updating the DRO with new numbers, that the speed has nearly doubled. So a 2x improvement in responsiveness over Arduino's fastest board is attainable in MACH4 at zero cost.

Food for thought.

39

Mach4 Videos / Re: Mach4 - Arduino and Touchscreen

« on: July 01, 2015, 06:48:03 AM »

I would not bother with M3 to much have you ever had a look at the PoKeys57CNC I am going to be changing over to one soon for other things just out of interest, I have not seen any new youtube vids from you lately

Looks interesting. For testing, I am going to go thru the available TCP motion controllers that are popular with MACH4. I already have the ESS up and running for testing and for my own equipment, I will be using a CSMCIO motion controller http://www.cs-lab.eu/en/produkt-2,2-CSMIOIPS___6_axis_Ethernet_Motion_Controller_STEPDIR_with_connectors.html  based primarily on my great respect for the advice of Hood, a major contributor here.

I am interested in the Pokeys 56E   http://www.poscope.com/PoKeys56E to use as a TCP Modbus 'portal' to relieve that processing from the DUE. I have the older USB version that was acquired for some past project, but I have never used it.

As far as videos, I can only post stuff of my own. Most projects are done under NDA and I would be in big trouble is I posted videos of them . . LOL!

However, there are two, possible three, new videos series coming up. The  IH mill conversion that ran over many videos will be coming back into the shop for conversion to MACH4 using the above mentioned motion controller and the setup will be getting one of the first combination InTurn™ 4th axis motor controller /Mach4 pendants as well as improved ATC software. Most of these Off-Shore mills have very flexible tables, so reinforcing the table *might* also be done . . that is undecided at this point.

I am also going to FINALLY start the 'big mill' project. That involved building an all new Bridgeport sized mill specifically to be a stable platform for the new InTurn™ ULTRA. It will have truly massive ball slides and ball screws.

The third 'possible' video series would cover the development of the MACH4 pendant . . . but I am reluctant to do that one because of the proprietary nature of it . . and the fact that I have not been collecting any video footage  . . which makes it difficult to put one togather  .

40

Mach4 Videos / Re: Mach4 - Arduino and Touchscreen

« on: July 01, 2015, 05:39:40 AM »

Its not 100% yet but im using the YunClient and Bridge libraries across to the Linux OS (openwrt) which i installed a python library called pymodbus.   .

As i said though, not 100% reliable at moment.

DazTheGas

I think this is a good track to be on and I very nearly started down this road  myself, except with the new Rasberry.  So far it has not been necessary to open the box on that project because at the moment I am looking at off-loading a lot of the processing from the Arduino DUE to MACH4 and also the GPU on the touch screen itself.

Based on my experience with a few of the Atmel processors, I thing the Yun is not up to the task or running Modbus on top of Linux and doing anything else useful at the same time, but if you can get something working and it turns out to overwhelm the Yun, you have a ready path to far more powerful processors, so you may not need to rewind very much to port everything over.

I am up to the point of doing field testing on my own TCP Modbus products on actual machines. First my own and then to my group of testers. To that end, I have installed MACH 3 and MACH4 with and ESS on a test mill and in that process I discovered an interesting anomaly with MACH3.

While MACH4 consumes between 4% and 8% of CPU power to run TCP Modbus, MACH3 is hovering between 45% and 54%  . . . doing nothing except running TCP Modbus. Screen DRO response is very choppy and jogging is jerky somewhat unpredictable. Turn off TCP Modbus and MACH3 returns to smooth operation. I am going to spend some time to see if this is fixable, but if not, then TCP Modbus will prove impractical for MACH3.