UR201 sudden movement

[johu]

Hi there,

when using the UR201 wireless pendant I sometimes notice a unexpected movement of the axis.
When it happens it's always the axis which is active/selected on the UR201.

My last issue was with:
- the X-axis selected
- MPG mode on Velocity 1%

I was jogging the X-axis in position. Suddenly the axis moved a big (~5m) distance with medium speed.
I am certain I was not moving the dial.
The speed of the axis was apporximately max speed of 1% movement.
The axis jumped to apporximately +5mm and then back to it's original position.
Fortunately the mill was above my workpiece.
But this same issue has also happend several times while I was probing the Z-axis, and damaged my workpiece and tool.

Is there a log facility to view all the input commands UCCNC receives?


UCR201 Plugin V1.8 beta
MPG filter constant: 5
MPG speed multiplier: 20

[cncdrive]

Which UCCNC version are you using?

[rabbit88]

I have seen a very similar issue to this using 1.2116 (latest test version) where an unexpected long motion happens. I have been pretty lucky
in that the couple of times its occurred my tool has been clear of fixtures etc, but its fairly unsettling... I did see this with the previous version
as well. Its pretty rare, I suspect I have only seen it 3 times in total.

I have also noticed when I connect or disconnect a usb flash stick to load a program sometimes it kills my UCR-201 completely
and I end up with 'no connection to software' type message on the UCR201 display. Previously it would throw a few usb library errors then
continue on just fine, now I do not get these errors with 1.2116 but the UCR-201 goes offline completely from time to time.

Exiting from UCCNC and restarting resolves this, so it does not appear to be hardware or pendant ?

The random large motion for me does seem to happen after probing tool length, I have run some pretty long (over an hour) programs with no
issues at all, this may just be luck, but it feels related to probing.

I have the vague impression that my motion may not have been axis aligned, but I cannot say this for sure (I just
lunged for the estop rather than taking too much notice of what was happening)..

I was for a long time using an xhb04 wireless pendant and never had this happen, I changed to the UCR201 because the XHB was causing some odd
issues with resetting coordinates during jog for probing in the middle of a program.

Hardware is a uc400eth

[johu]

[UCCNC]
Software: 1.2114
Firmware: 1.037
Hardware: 1.000
API: 2.1880
Device type: UC300ETH_5LPT

[UCR201 Plugin]
Version: 1.8 beta
MPG filter constant: 5
MPG speed multiplier: 20

[UCR201 Wireless Jog Pendant]
(With the emergency button pressed and holding down F1 and F5)
UCR FW: V1.005


Is there a log facility available?

[Matt]

I'm a new UCCNC user and have also been also running into concerning phantom movements since adding the wireless UCR201. I'm soon about to give up on using it in fear of the damage it could cause. On an earlier table (unrelated to UCCNC) I also ran into unpredictable movements with an XHC pendant and Mach 4. I always assumed it was wireless interference because there's quite the mess in my local air. On my old setup it was always an axis overshooting or continuing forever despite the MPG no longer being turned. You could recreate the symptoms on the old XHC pendant by changing the axis being jogged while using the MPG to generate the pulses. But now that similar weird things are happening with completely different hardware... I'm not yet entirely sure what's going on.

With the UCR201 it just randomly sends my Z axis toward Z- without notice. Only way to stop it is hitting an estop. It's so far always been when the wireless pendant is being moved (not buttons being pressed or the MPG wheel being turned, just the device itself being moved around in the air). I have suspicions that the pendant is reporting back regularly to the dongle and think messages are getting missed. I'm back to questioning if it's again some sort of interference. My hypothesis is that the frequencies being used are not transmitting each pulse from the MPG but instead reporting "mpg start/stop + cw/ccw" events along with an integer after 100ms or so which the software then uses to make assumptions about what happened on the pendant. I'm only assuming this because it seems like it would be easy to infer most of this on the dongle but also if any of the data is missing or incorrect I'd expect the behaviour we're seeing. As far as I can tell nothing about the protocol it's using is documented publicly so this is really a wild guess with so little to go on.

[UCCNC]
Is there a log facility available?

I dream of a day where this is available because I otherwise quite like the software. It has a lot going for it but any sort of verbose logging seems to have been omitted from UCCNC. I'm hopeful they will notice the value and interest in this. Personally I want a verbose log of every I/O so I can work backwards through a series of events to determine what actually happened. Right now I need to rely on camera footage and screen recording (if I happen to remember to leave the diagnostics page open) to even stand a chance at understanding what took place.

[cncdrive]

What is your UCR201 firmware version number is?

[Matt]

What is your UCR201 firmware version number is?

Thanks for the reply, again! It's nice to see developers actively keeping an eye on things like this. Thank you for that.

UCR201 V1.005
UCCNC 1.2115
UCR201 Plugin V1.8 beta

I had another scare just now, just minutes before posting this reply. I was a mere 4 feet or so (line of sight) when I commanded Y to move Y+ in 1.0 increments by approximately 12 clicks on the UCR201 MPG. The Y axis continued to move indefinitely toward Y+ reaching at least twice that distance before being caught. This time it was possible to correct the movement by turning the MPG CCW 1 click,and not requiring an estop. This resulted in Y stopping and moving Y- exactly 1 as requested. Might be worth noting that this does not mess up the actual machine or work coordinates. All that reports correctly. Perhaps my fingers left the MPG between detents which I'll try to be more mindful of. It just kept moving along as though I was still spinning the MPG.

One other oddity is that when upgrading from UCCNC 1.2113 to 1.2115 I believe channel settings somehow might have changed themselves or at the least reverted to a default. I could have sworn I had that all set to CAN/US C-5 (from back when I was playing around with the idea of interference) and today I find it is set to C-1. What I can't explain about that is how the pendant also changed to C-1 without intervention so this last part might be a figment of my imagination.

Hey, while you're here I'm just curious... does the pendant have a way to update the firmware on it should you later need to update that? Even if it's a serial connection inside the device itself. Curiosity gets the best of me. I do very much enjoy the rapid updates. I miss a physical selection knob for changing axis or distance but eventually I'll get the hang of it. The rapid updates more than make up for it at the moment.

[cncdrive]

Matt,

It is impossible that the pendant itself changed it's RF channel, but if you upgraded the UCCNC software and installed a new profile file then the Channel number on the PC/plugin side changes to the default.
The UCCNC plugin is capable to upgrade the UCR201 firmware via the RF link.

And it is important that you set the US RF channel if you are in the US, because the EU channels are the Taxi radio channels in the USA, so then the pendant could interfere with the Taxi radios if you set it to EU channels.

[Matt]

The UCCNC plugin is capable to upgrade the UCR201 firmware via the RF link.

Happy to hear this is possible! I've got no idea what is planned but just seeing the changes from 1.2112 to 1.2115 makes me think you're actively working to make the product better and better.

And it is important that you set the US RF channel if you are in the US, because the EU channels are the Taxi radio channels in the USA, so then the pendant could interfere with the Taxi radios if you set it to EU channels.

I've been using the US channels to try and stay compliant. Thankfully even if I was on the other I'm located so far in the middle of nowhere that I'm many miles away from any active taxi operations. Taxi isn't a word in the local vocabulary. You can't even get a pizza or groceries in my local area. But, there is a fair bit of traffic on ~900MHz in the area from inexpensive weather stations, antique PTP/PTMP links and all that fun stuff.

I didn't look it up yet but do you happen to have a chart for what exact frequencies your US channels use? I wouldn't mind looking at what local frequency is the least noisy and switching over to that. Also, what's the 1-15 option that shows up when you hold the lower left and lower right buttons?

[Matt]

This trick is getting old, fast. Again today I had an axis runaway on me while using the pendant. This time it was Y (and a slave C axis). I was slowly jogging the Y axis in Y- direction using 0.001 increments to dial in my part. As I re-positioned myself for a slightly better view it randomly charged full speed to the Y- end of the table. Thankfully the software limits stopped it or it would have smashed the homing sensors right off the table. It was also sheer luck that my work piece was on the other side or that would have been the end of a probe. I feel like I should buy a lotto ticket today.

I don't know if I can continue to use this pendant. This is getting dangerous and potentially very expensive when it does this again with an endmill or probe. I don't have a clue why it's happening or how to prevent it (aside from unplugging the pendant).