Mote Controller Just Stops After Some Usages

Support
1

Hello, all.

I sent some questions in to Pimoroni, and some technically speaking complaints and they helped me. They replace the item that arrived damaged, and they answered my questions.

However, more serious problems developed with my Mote and after some email exchanges they started dropping hints and pushing me to this place. As much as I did not want to sign up just for this problem I did so because the Pimoroni guys were so nice to me that I had to fulfill their wishes of moving my support needs here. If anyone can help me with my problem it would be most appreciated.

Last year I built a device that uses 4 Mote sticks controlled by the Mote controller and a Raspberry Pi. It worked well, better than expected, for a while, then it stopped working. It was so dead it seemed obvious that the Mote controller failed. So I bought a new one. Turned out that reformatting the Pi SD card and reinstalling Raspbian fixed the problem. For both controllers.

Then it happened again. A new, from scratch, Raspbian with newly installed Pimoroni Mote libraries was the only way to fix it each time.

When it mattered, when it really, really, mattered, it happened again. When the device was used in public and it reflected on me and many people were depending on it working, the same thing happened. It spontaneously gave up. Embarrassing timing for sure.

The failure makes it so that the Mote controller will not even show up under lsusb check and such. The Mote dies. It no longer exists as far as the Pi is concerned, with no warning nor error message other than being unable to find anything called “Mote”. No specific error message. Slash and burn is the only method for recovery then.

Has anyone ever experienced anything like this? Is there a way to keep the Mote going? Has anyone ever actually done that?

Is there a super specific version of Python I should be using for my script perhaps? I cannot even think of a way to diagnose this. My searching for how to determine what version of the Mote Python libraries has did not yield an answer. It was installed summer of 2018.

Mote controller, 4 sticks and apparently Raspbian GNU/Linux 9 (stretch), Python 2.7.13. From the terminal: “python-serial is already the newest version (3.2.1-1).”.

I just want to be able to use a Python script (or other easy language) to run the Mote controller on a Raspberry Pi without it just stopping to exist after some (small) random number of uses.

Anyone have this problem and can share experiences? Anyone run a Mote without issue?

Thanks for reading my post. Help is definitely appreciated.

2

Did you ever try the mote phat? I’ve found that far more reliable over the years. The pi it runs on has been kept up to date with dist-upgrade and the HomeKit code from Sandy’s tutorial works pretty well.

My anecdotal experience with pi over the last few years would suggest that raspbian does struggle with complex usb devices. Anything that can be done through GPIO seems to behave better.

3

Let me start off by admitting I don’t own any mote devices. That being said, one drawback, if you want to call it that, with a Pi is power failures or power bumps. And or improper shut downs etc. Both can cause a corruption of the data on the SD card. Once that happens weird things start happening.

I feel for you issue. I have several Pi’s running Motion Eye OS. Every once in a while one will just stop working? No warning, they just lock up. I have to power cycle them and sometimes they are fine and some times they aren’t. Only way to fix the ones that aren’t fine is to reflash the card with a backup image.

4

Davros wrote “Turned out that reformatting the Pi SD card and reinstalling Raspbian fixed the problem. For both controllers.”

For me that means the actual problem is with the pi not the mote controllers.

I’ve got a couple of mote controllers and a mote phat. I’ve never had any problems with them at all. I had a strange problem with a dodgy RGB pixel on a mote strip, but that wasn’t a controller problem. The mote controller is effectively a specifically hard-wired and hard-coded Arduino Leonardo using a ATmega32u4 (which is very similar to the Flotilla dock, but that’s no use for mote strips!). If you pull the power to the mote board and then power it up again, it justs starts listening for mote serial commands, the pi doesn’t download anything to it at startup. You would have to do something pretty wrong as an end-user to damage the mote firmware flashed during production. As you managed to get it to work again with a new raspbian install, the controller isn’t damaged, for some reason, the detection and/or comms wasn’t working from the pi. It could be a conflict with something else you’ve installed on the pi. It could also be an irritating “order of installation” issue where you’ve accidentally stopped it working by installing something which takes over the USB port for itself and stops the mote hardware/software from being detected and communicating.

The mote phat hardware/software works a bit differently, but it still pushes out the same RGB data to the LEDs on the mote strips.

How good is your power supply? Is it managing to properly power the pi, mote controller and all the mote strips?
Do you use the second power socket on the mote controller to boost the power at all, or is it all done over the USB socket with one cable from the pi?
Did you know you don’t need a pi, you can also use the mote controller with a pc/mac? You could try running your mote that way to see if you have any similar issues. Note that you might have to work out how to drive the mote controller from your pc with python or something similar, but it isn’t rocket science.

5

Thank you for the responses, all.

Assumed this was a software thing since the controllers work sometimes, which is why I asked about specific Python versions, etc.

I thought I had a good power source going but after reading some of the above replies it got me thinking.

The Pi had a beefy USB power pack attached to it. I thought the power port on the Mote was inconsequential as it is just there to make the Mote sticks brighter and is USB based so it cannot go above 500 milliamps anyway. It had a low amp power pack attached to it and I replaced it.

I took a super beefy USB power pack that can do at least 2.4amps per port and used it to power both the Pi and the Mote controller.

The system booted up and lit up. I had not touched it since the last failure. Power into the Mote was the only difference.

Yet to really test it or use it for more than a few minutes, but the fact it booted up and the Mote lit up tells me this is the problem area.

Will update when I do a full test.

Not sure which power pack I used but I have two super beefy Anker ones. I think they are Powercore series. One does at least 2.4 amps and the other 3 amps per port. I can find out if anyone wants to know.

I did not know there is a Mote phat. That is intriguing.

Thank you to all of you for these suggestions.