Anyone have Octoprint running?

Richy_T

I look like I'm basically there. I'm just having issues with the comms.

The printer connects fine and it'll be chatting back and forth in the background. It'll do this all day long.

Send: M105
Recv: ok T:20.9 /0.0 B:22.5 /0.0 @:0 B@:0
Recv: M105

It's when I try and do anything else, it starts hitting timeout errors. For example, trying to do anything with any of the axis...

Send: G91
Recv: relative_mode = TRUE;ok
Recv: G91

Then a long wait then

Communication timeout while idle, trying to trigger response from printer. Configure long running commands or increase communication timeout if that happens regularly on specific commands or long moves.
Send: G1 Z10 F200
Recv: ok
Recv: G1 Z10 F200
Send: G90
Recv: ok
Recv: G90

Then it's back to the M105s. Unless I have multiple commands queued and then it repeats the communication timeout stuff. I have increased the communication timeout settings but it doesn't seem like that would be the issue.

Hmm. Didn't have that issue with the temperature controls. So maybe it is just G91? I'll try some manual stuff and maybe a print.

Richy_T

Well, printing seems to be going fine. But G91 is going to be an issue as I think that's used in some dual filament stuff I want to do.

Samuel Pinches

Also, you can play with the timeouts in the octoprint settings, to see if that fixes things. It could be that some commands are not supported - the marlin build inside the A5S/A1 firmware is simply deathly old... (and so old it is un-upgradable...)

Richy_T

Well, it does seem to be doing what it should as the relative movements do occur. It just causes the timeout in the communication protocol which makes things take a very long time (possibly acceptably so for a simple filament change). I've sent an email to support so we'll see what comes out of that. I think I'll also try talking to it through repetier and possibly directly through a terminal.

Richy_T

Working with no issue today. *shrug*. Probably just as well I sent the support email to an invalid address

artair

Sadly, I'm running into these same issues, and nothing seems to be improving. I've even tried upgrading and downgrading the firmware with no luck. JGAURORA isn't answering my support e-mail (probably due to the Lunar New Year holiday in China), and Reddit hasn't produced any leads. Do you have any recommendations? If I can't fix this, I'm going to have to return the printer, which would be a tremendous downer.

Richy_T

The only thing I can think of was maybe the voltage on the Pi was low. Possibly try a different power supply, particularly if you get low-voltage warnings.

Samuel Pinches

@artair - maybe I could provide some octoprint help via teamviewer?

artair

Richy_T -- I'm using a different platform for my server (OS X on an old MacBook Pro), so undervolting probably isn't the issue on my setup. Good thought, though!

Samuel -- That would actually be incredibly helpful. Thank you! In the meantime, are there any settings you would suggest that I check on my own and test?

Samuel Pinches

@artair - well, the first thing to try is to see if you can connect to a computer running pronterface?

artair

I downloaded Pronterface and gave it a try, but it just generates the same errors as the console on OctoPrint. Both OctoPrint and Pronterface can connect with the printer and communicate in a limited fashion, but I get timeouts on both, and can't even move the X-, Y-, and Z-axis motors. Most of what I receive back from the printer is temperature stuff.

Samuel Pinches

Hmmm. Sounds like perhaps there's a printer motherboard issue of some kind. Clearly the communication from PC to USB-to-serial chip on the printer is working, but it sounds like there's possibly issues from the serial chip to the printer. All I can suggest is to try contact JGAurora and ask them for suggestion? Sounds likely that you'll have to return that - I would emphasis that to them. A comprehensive QC process should include testing the USB functionality, but somehow, I doubt that is occurring...

artair

I actually sent three e-mails to support before I (finally) got a response early this morning. The team at JGAURORA seems completely unaware of the existence of anything that isn't the "JG GREAT" version of Cura they package with their printer. Here's an excerpt from one of their responses:

"Hi,

Thank you for email.

I am sorry that i don't know OctoPrint.

Did it print with PC?

If you want to print with PC , you may see the instruction in the attachment .

But i suggest that you don't print with PC and the useb cable transering information is not steady.

For the A5S firmware , it is in the attachment.

Thanks"

I tried the firmware they sent (it's a version that's not currently available on this website) and successfully applied it. Frustratingly, it made things much worse. Pronterface can kinda-sorta connect and send basic commands, but the logs are full of errors:

"Connecting...

echo:

Printer is now online.

M105

echo:

M105

echo:

M105

echo:

M105

relative_mode = TRUE;G91

Printer stopped due to errors. Fix the error and use M999 to restart. (Temperature is reset. Set it after restarting)G0 Y10.0 F3000

G90

M114

echo:

M105

relative_mode = TRUE;G91

Printer stopped due to errors. Fix the error and use M999 to restart. (Temperature is reset. Set it after restarting)G0 Y-10.0 F3000"

OctoPrint, on the other hand, cannot connect AT ALL now. Not even a little bit. Sometimes lines come back that are full of garbage, too:

"Connecting to: /dev/cu.wchusbserial410

Changing monitoring state from "Offline" to "Opening serial port"

Connected to: Serial<id=0x11073b990, open=True>(port='/dev/cu.wchusbserial410', baudrate=115200, bytesize=8, parity='N', stopbits=1, timeout=15.0, xonxoff=False, rtscts=False, dsrdtr=False), starting monitor

Starting baud rate detection...

Changing monitoring state from "Opening serial port" to "Detecting baudrate"

Trying baudrate: 115200

Send: N0 M110 N0*125

Recv: M(-Hj%M%ZM%ZM%XJ

Recv: N0 M110 N0*125

Baudrate test retry #1"

My guess is that Pronterface is more tolerant of errors than OctoPrint, but neither of them is having a good time with this printer. I don't think this is a physical defect, either. My guess is that this is a function of the new A5S design.

The JGAURORA A5 and the A5S have different mainboards and completely different firmware. JGAURORA went with their own in-house firmware for the newer printer, too. It's becoming clear that this new configuration isn't conducive to interoperability with open standards. That's pretty aggravating, considering the whole reason I bought this printer is because my A5's heated print bed died due to bad design.

I've sent a final request to JGAURORA for any additional steps I might try, but I'm not optimistic. I think this printer is probably going back to Amazon, and I'm probably moving to the CR-20 for my next attempt. This is the third printer I've had from JGAURORA, and I never even got to see the second one. (JGAURORA's shipping department did a poor job, and the printer was damaged/destroyed en route to me. Amazon auto-returned it, but couldn't file a replacement order without my intervention for some reason. Then JGAURORA started e-mailing me, asking why I "returned" the printer that I never received in the first place.)

Sorry if I'm coming across a bit "ranty." I loved my A5 right up until it failed, and everything since then has been downright aggravating.

Samuel Pinches

That’s fair enough. We have plans for a new firmware on the A5S down the track but that hasn’t materialised yet, and I can get that you want a solution now. What’s wrong with the A5 - could it be fixed? Maybe we could look at that in a new thread.

artair

Thank you for the offer, Samuel. The A5 has what looks to be a broken bed wire that's preventing it from heating the build plate. I'm not exactly great with wiring, so I was planning to use this as an opportunity to upgrade. I may get assistance from someone who's better with a soldering iron to fix the A5, but I think I'm going to return the A5S and get a CR-20 in the meantime.

Richy_T

Have you tried with a different computer? I doubt it would be the Mac but it's always best to eliminate possibilities.

Samuel Pinches

There was a new A5S firmware released today - that's the last thing that I can think of to try.

https://www.dropbox.com/sh/e54w4d7y5aqmbf9/AADr6HbTD8-rlpdAV_vLpIOca?dl=0

artair

Sadly, that new firmware was the stuff that made things worse. (JGAURORA e-mailed it to me.) At this point, I haven't been able to print anything in weeks, so I sent the A5S back and ordered a CR-20. Thank you all for trying to help. The A5S just doesn't seem to be built with anything that isn't the "JG GREAT" software in mind.

Samuel Pinches

I'm taking a look at this stuff tomorrow. JGAurora sounds interested in getting it working.

Richy_T

I noticed some timeouts again last night. It was only the G91 commands so I've still been able to print OK.

Richy_T

Samuel, do you know if the A1 firmware is the same as the A5S? I wish JGAurora would do it properly on their website with release notes and changelogs...

Samuel Pinches

I know... there is no update for the A1 yet I've been told.... I've asked for more info but I don't expect to get it.

Richy_T

I wonder what the differences are. It's the same board and similar hardware setup so I'm not sure why it would be different.