Hi guys, I’ve just assembled my machine. Got it all connected up. I had to use an earlier firmware version as the latest on was being flagged by my anti virus. But anyway, got it working was able to set it up on my wifi and was going through the calibration steps…
step 1 retract all, only one retracted, clicked it again and a second one retracted, but the two others won’t budge (am I meant to be able to pull the tracks in and out because they don’t move at all, but wasn’t sure if that’s meant to be).
Step 2 was to go into the config, but I didn’t have that button in the setup window, I only had retract, extend (which wouldn’t extend anything), calibration and test I think.
So I thought I’d update it to the latest firmware, managed to do that but now I can’t find it on my wifi and can’t connect via it’s IP address. (some more details on this, maslow.local was working before the firmware update, and the blinking blue light tells me the IP address, but both those options is now just a forever loading webpage, nothing happens). Once I told it and got it to connect to my wifi, I could no longer detect it’s wifi to join. Presume that’s normal.
Also when I plug it in, theres a low constant whirling noise… have tracked it down to the two Z axis motors. They’ve become very hot so have unplugged them. If I plug them back in they immediately start making noises, but it doesn’t look like they are turning the shafts.
I’ve started dismantling it to see if I can find a reason why it wont retract or extend and the motor shaft is just impossible to turn by hand (presumably the gearbox is making it difficult), so I guess that’s why I can’t move the tracks in or out manually. But 2 did move via the software so maybe this is normal.
Have reassembled it now, as I think about it, I remember that none of the motor shafts were turnable when I assembled them, so this must be how they are meant to be.
Update:
Following some advice in another thread and trying to reset the ESP and the software… running FLUIDTERM came up with:
Grbl 0 [FluidNC v0.75 (Maslow-Main-e2ef1bb2-dirty) (wifi) ‘$’ for help]
[MSG:ERR: Configuration is invalid. Check boot messages for ERR’s.]
[MSG:INFO: Encoder connected on Top Left]
[MSG:INFO: Motor detected on Top Left]
[MSG:INFO: Encoder connected on Top Right]
[MSG:INFO: Motor detected on Top Right]
[MSG:INFO: Encoder connected on Bottom Left]
[MSG:INFO: Motor detected on Bottom Left]
[MSG:INFO: Encoder connected on Bottom Right]
[MSG:INFO: Motor detected on Bottom Right]
[MSG:INFO: Current z-axis position loaded as: -56.000]
[MSG:INFO: Starting Maslow Version 0.76]
Update:
Oh Yay! clicked install-fs.bat and now I can get back into FluidNC via the IP address. This seems like progress.
Update:
Ok so I can now connect. I can move the Z axis up and down, but retract or extend does nothing. (and I can now see the config box in the setup menu which is encouraging.)
Update:
If I increase the retraction force to 1800 (would be nice to know what these numbers mean, and what a safe limit is!) they do all retract. Which is great. But I now cant release / extend them. So it’s all looking tidy now but somewhat useless if I can’t loosen the tracks (I think people call them belts).
Hi Bar! Thanks so much, following this I’ve managed to get 3 moving and now I understand how to do it they move quite freely.
But one, the uppermost motor, I can move it only a few mm by rocking it and now I can only get it to extend if I push the belt back in! It seems to release the motor and give me a little more length each time.
Yes that last motor is strange, no matter how hard I pull (and I’m being a lot more aggressive with it now than I was) I can’t get it to release and extend.
The rock method you suggested does give me maybe 5mm of movement and I can only get it to release if I push it back in! That triggers the motor somehow and it releases maybe 10mm of belt.
Feels like something is backwards here but cant think what.
Unfortunately I know exactly what is going on there. There is a known but rare issue (100% my fault) where the encoder starts working backwards. The solution is to solder one of the pins on the encoder board to ground which forces the encoder to only work in the correct direction.
If you feel up for modifying your encoder like that it should fix the issue, and if not I can modify one for you and send it to you ASAP. We’re still not sure why it happens to some boards and why it happens when it happens, but the fix seems to work reliably.
Hi guys, don’t understand what’s going on. Picked this up again this evening and turned it all on to double check which was the offending motor as getting to the motor boards is a total disassembly job… and they now all work fine! Don’t understand how this could have fixed itself so am somewhat nervous. But will keep going for now. Fingers crossed.
This is the behavior of the “floating” pin. Sometimes it will work in the correct direction, and then randomly switch. It is still worth doing on all 4 boards in my opinion (not that my opinion is best in this group of much smarter people )
And yes, I feel that pain. Be careful not to strip bolts. you can if you are good, solder through the hole once you take the bearings off so as not to have to unbolt the motors. I just saw a post recently about this… I took mine apart and it was really difficult and I ended up using a dremel to cut slots in the top of 2 bolts to get one of my arm’s motors back off again.
To be honest, in just 3 minutes of retracting and extending one of the belts is now so incredibly tangled that I’m going to have to dismantle it all anyway. I have no idea how a belt can twist itself like this, simply remarkable.
Strongly recommend something in the design needs tweaking to stop this happening for the M4.1 because wow. What a nightmare.
Ok so you think do all of them? Good to know as I would have only done the faulty one.
)