Teething problems- calibration, connection drops and retraction issue

Hi all,
So I received the replacement control board which solved the issue of only one z motors working , so .. so far so good !! Thanks Bar for sorting .

Ok , I followed the calibration routine and got as far as setting up if the calibration routine , all went well until the WiFi connection dropped just after the calibration process started .. the WiFi connection keeps dropping so I have to refresh the browser continuously… and as it’s the pc doing the calculations etc this could mean I’ll never get off the ground ..
so after restarting the Maslow and connecting to Maslow WiFi , I pressed release tension button and took the CNC off the frame , tried to retract all belts , and one caught on something temporarily, I freed it but when the other 3 belts had fully retracted , the one that got caught stopped retraction 20cm too soon … and I can’t seem to get that to resolve itself…

Any advice on this would be appreciated :nerd_face::+1:

1 Like

ok, update,
manged to get the 4th belt to fully retract, so started the calibration again , unfortunately the connection to Malsow wifi kept on dropping , so the calibration halted a number of times .
I know there’s an option to add the maslow to an existing network so ill give that another go … I shyed away from doing that TBH, as the maslow didn’t join last time with firmware v1.4 , instead it created a new network with the same name as the existing network , hopefully firmware 1.5 has fixed that issue ? . :sweat_smile:

This is probably the best thing to do.

There are two different settings, one sets the name of the wifi network that maslow will create and the other one sets the name of the wifi network that maslow will connect to.

1 Like

ignuf wrote:

so I tried to get the maslow to join exising network but now have the issue
where I get some sort of login window ZTE… requesting a password , can’t
find it so far , is there a default passowrd ?

that should be the password for the network you are connecting to.

the default password for the maslow hosted network is 12345678

David Lang

2 Likes

I had constant dropouts when connecting to the maslow directly too, connecting to it via my home wifi has mostly rectified this. I also had to make sure the wifi modem was close enough to the maslow for a good signal, when it was further away I still had some issues

2 Likes

Deadset wrote:

I had constant dropouts when connecting to the maslow directly too, connecting to it via my home wifi has mostly rectified this. I also had to make sure the wifi modem was close enough to the maslow for a good signal, when it was further away I still had some issues

There are at least three problems happening here

  1. the maslow 4 has a very poor wifi signal (could be a combination of where the
    antenna is and the noisy electronics around it)

This is why your access point needs to be close

  1. the maslow does not give you access to the Internet. As a result, many
    devices decide that you really don’t intend to use this connection and
    disconnect after failing to ‘login to the captive portal’ disconnecting you and
    breaking calibration.

There were some patches early on to try and trick the devices into accepting the
hotspot, we may need to revisit this as it may have only worked for some
devices, or they may have done something to break the detection.

  1. some devices decide that if you are idle, you can be disconnected (breaking
    calibration)

David Lang

2 Likes

Thanks Bar and DLang ,
I’ve got the Maslow connected to the network and that seems to be a more stable connection now …that said, it still drops the connection every now n again , can the usb-c port be used to attach a cable directly to a PC?

Started the calibration process,
The Maslow has completed the 7x7 grid and is calculating currently….

Some additional questions if that’s ok?..

Does the calibration process, include the z axis at various heights ?
I’m wondering if not, then could it be contributing to the accuracy issue brought up on another thread in the forum ?.. if I raise the z axis , even a small amount, while the tension is still applied it does feel like it’s having an affect on the whole assembly..

TBH, It kinda feels like the four corner axis should be independently mounted directly to the base so that they’re not subject to the z axis raising up and down.

Thanks again and regards
G

thanks Deadset, i’ll give it another go :+1:

update,
hey guys ,
so the calibration is acting strange now…
the Lower Right motor isn’t pulling the belt tight like it was earlier this week… ive taken a short video so will try to upload

calibration should be done with the Z all the way down. once it knows where the
corners are, it then accounts for the extra belt needed due to the Z height.
This is why the z offset values need to be set correctly.

I agree that having the arms mounted to the sled rather than the router is a
more solid approach and have done some work towards the firmware supporting that
variation of the machine but have not had time to build a variation.

David Lang

2 Likes

Hi ignuf,

That is exacly the same problem I ran into.
Eventually managed to calibrate the maslow.
I also did think the bottom right belt was a little bit loose.
But with the first cut the bottom right motor did fully unwind itself. Even after the point it will wind up back again the other way around.


Managed to rewind it carefully with the bench power supply far enough so the maslow could take over.
Polarity issue is not possible. Retract en extend do work the correct way around.

My maslow is now ready for a calibration again. Maybe higher retraction/calibration force will solve my issue.

Raymon

Edit. Picture taken after +/- 120cm of cut (4ft)

2 Likes

Hope the next attempt works out ok Raymon :crossed_fingers:,
I’ve had to re initiate the calibration so many times so far and a new issue seems to emerge each time , definitely keeping my brain occupied :sweat_smile: worth the perseverance :nerd_face:

Here’s a video of the calibration issue, the lower right belt seems to sometimes tighten and other times not… if the belt is loose, and u look at the motor , it’s seems to be making very small rotations .. all seemingly a little random.. then the calibration stops and the pc starts to do the calculations, fitness values are tiny..

Maslow calibration

made a few more attempts to calibrate but the lower right corner constantly seemed to act randomly, sometimes tightening or not tightening

try increasing the calibration current setting, that will make it pull harder
when trying to retract the belt during calibration.

David Lang

2 Likes

Managed to make it all work with 1200 retraction and calibration force!

Did two plunge cuts in the beginning due to incorrect homing I think.
But after that it went smooth. This is actually my first cut with my own generated gcode. :flexed_biceps:
Only a little wobbly in the corner but that is maybe a dull bit combined with a too big depth of cut..

2 Likes