Resolved: Cooldown and Calibration Issues

It occurs to me I don’t know where to find that information. Where would I look? Thanks

1 Like

I actually just listen. She whispers it softly to me. :wink:

3 Likes

And here I thought that was “shave and a haircut” at the end of calibration :smiley:

2 Likes

That’s going to be just awful :wink:

3 Likes

OK, now we’re tied. I turned my GF back on, had a second browser running and both were reading “cooling down” still. Killed them both and started GFUI again. Now I’ve got the long scan with the head parked at the home position. Using your fix

Cancelled, opened the lid, All printed well.

At the end of the print, the head homes, no calibration, time shows 00:00 at top right, and I get he normal “keep lid closed” message. I hit cancel after a minute, GF showed ready and normal calibration ran.

Things are getting better if not quite back to normal yet. :slight_smile:

2 Likes

Pretty bold of them to start the version number with a “1”.

On it’s first boot it, prior to any updates from the mothership, it was v0.0.1-959.

2 Likes

That sounds about right.

Where you getting this version number from man? I’m sitting here wishing, so that’s my first clue you are toying with my naivete…

1 Like

Console port on the control board.

I’ll work up a quick How-To when I get a minute for those that like to live dangerously.

5 Likes

Ahh, I was suspecting that’s what you did and starting to search around mine with a flashlight. 'Cause I couldn’t find diddly about versions in the API.

For the so-inclined, BTW, some interesting endpoints include /api/users/machines, /api/designs, and /api/materials. That red print preview is actually rendered via JavaScript from the .puls file, so everything you’d need to write a viewer for them is right there. Hypothetically, of course.

5 Likes

Thanks Scott, please do. it would go a long way toward easing my anxiety when s*it goes sideways :no_mouth:

2 Likes

Good find. That certainly could be useful.

Of course.

Another forum regular has made a good amount of progress on analyzing the make up of the puls file, including writing a simple decoder. Maybe he’ll reveal himself and his findings…

A quick and dirty pick apart shows that it won’t be that difficult to cobble together a simple package to provide at least basic use should problems arise with the cloud model. It wouldn’t have super bells and whistles and would be missing a lot of the refinements that are securely living in the cloud, but you wouldn’t be left with a paperweight.

7 Likes

I’m having this issue tonight too. Either sticks at Cooling Down (it’s probably 70 degrees in the room) or recalibrates after every print.

1 Like

Yes I can decode puls files. They have a header with name value pairs that set things like the motor modes and current, fan speeds for air assist, exhaust, intake, filter exhaust, filer intake, etc, and the waveform frequency, which is 10kHz. After that it is a raw uncompressed waveform with one byte per sample at 10kHz. That means the 3 hour limit comes from having a 100MB buffer.

Each byte has step and direction bits for each motor plus a laser enable bit, making 7. The top bit indicates the lower 7 bits are a laser power PWM value instead.

5 Likes

Thank you @hansepe for letting us know about this, and thanks all for the conversation. It helped us track down a bug which we’ve now resolved. Would you let us know if you see this or something similar again? If you do, we’ll go find the next issue!

2 Likes

Thank you for the reply.

I’ll not be able to test on my GF until tomorrow night thanks to work. If anything crops up I’ll start a new thread. :slight_smile: