TLDR: Have we got a lemon? Any help much appreciated

Troubleshooting
1

First some Background:

Some small ‘m’ maker friends and I invested in the Maslow 4.1 Kickstarter.

We had a busy year so we only assembled it a few months ago and very soon discovered we had one of the faulty boards that could’ve control one of the stepper motors.
This was swiftly sorted by Bar (for which many thanks).

Being in Australia the DeWalt router is not a sensible option and we have not yet invested in a spindle however Josh has machined a piece of aluminium into a workable pen holder.

We will move to a spindle if and when we have got the machine working reliably. We are dedicating our Wednesday afternoons to this.

To date we had successfully drawn the Maslow Doorsign that elsewhere was used to educate. And we had successfully drawn a line drawing (2D) of a WM graphic on an A3 sheet following Orob’s excellent guide to generating GCode from Inkscape.

Since then we have been unsuccessful at getting a 3D graphic drawn. (Note, by 3D I mean using layers in inkscape to create the various depths of cut but changing pen colours to represent them on paper.) For this the canvas was set to 1400 x 700 and a corresponding roll of paper was attached to the spoil board.

I am posting this because we need some help.

What follows is a log in the from of notes I made in progressively more detail as we struggled. While I do have some other serial logs but I shall attach the latest at the bottom.

Maslow issues

2026/08/19

Maslow stopped in the middle of a job.

Went through a series of turn off and turn it on again to no avail.

Eventually went home with all the belts still extended

At home I tried to retract belts.

No response.

Discovered that the “applied tension belt traction limit” was set at 300 and could not be reset from the quick menu. Went into configuration in Fluid NC and still see 300.

Try to reset to 1200

Response was:

“Set failed !Error 503: server not responding.”

The Setup button was greyed out and did nothing.

In Options > Apply tension belt limit and changes to belt Retraction Limit and Extend Distance just do not stick.

I was able to get the Maslow onto my WiFi which we had not been able to do at Josh’s. However I was not able to update to the new firmware.

At some intervening time I was able to retract the belts but I appear not to have made any notes.

2026/09/02

Back at Josh’s.

Manually updated firmware to 1.24.0

Replaced Index & YAML

Extended all, applied tension and the Maslow failed to attain Ready to Cut. Apparently still Taking Slack. Release Tension is the only green option. But ‘Maslow is not ready to move’ keeps coming up.

So

Power off, on and reload maslow.local.

Setup is available again.

We Retract all … : State Retracting. Release Tension is again the only green option.

So

Reload maslow.local

Power off, reload maslow.local.

Previous Serial log has vanished.

Setup not available

Reload

State:Retracted

Extend all

State: Extended

And because we were so frustrated we cycled through again

Retract all

State:Retracted

Extend all

State: Extending

So

Reload webpage

Setup provides choices.

Apply tension

State: Taking Slack

So

Reload webpage

State: Taking Slack

So

Reload webpage

State: Taking Slack

Apart from Stop the only button available is Release tension

Exploration occurs. Question; What is Belt retraction limit? (mm) If extend distance is 1200 could 300 be the issue? (NB conflation with current limit which in turn is elsewhere referred to as Retraction Force).

Changed it to 1000 in FluidNC Config Items.

Won’t save to YAML (Server not responding).

Reload.

Nowt

Reboot

Reload

Action button changed to red Alarm

Reload

State: Unknown

Action button still red Alarm but Setup is available.

Available options are Retract all, Extend all & Release tension.

Clicking Alarm changes button to Idle

State is still unknown.

(/Maslow_Apply_Tension_Belt_Retraction_Limit has saved to 1000 (which is within 200mm of /Maslow_Extend_Dist of 1200))

Release tension

State: Idle

Retract all

State:Retracting

Reload webpage

State: Taking Slack

So

Reload maslow.local

It doesn’t.

So

Stop and Reload maslow.local

It does.

State: Retracted and Setup is again available.

I’ve missed my volleyball in the park and my collaborator has gone for a meeting.

Shall I have one last try?

Can I even extend all the belts on my own?

Extend all

State: Extended

(Find anchor locations has become an option for the first time in a few sessions)

Apply Tension

State: Taking Slack

Only available (green) option is Release tension.

I try clicking Apply Tension again.

[MSG:INFO: Cannot take slack until belts have been extended] (which they had).

However the state is now Taking Slack.

I try clicking Find Anchor locations perhaps for a bit of sport.

[MSG:INFO: Cannot start Find Anchors until the belts have been extended] (which they are)

I try Stop

[MSG:INFO: Unable to determine machine position from belt lengths]

I decide to go home

[MSG:INFO: don’t you flipping dare]

I try release tension

[MSG:INFO: as above]

I rediscover a Restart FluidNC button.

State: Unknown

Action button Idle and Setup is available.

Available options are Retract all, Extend all & Release tension.

Retract all

State: Retracted

I pack up and go home.

State: Frustrated

Can anyone offer any suggestions or do we have a failure in the board?

Maslow-serial-26-09-02a.log (3.2 KB)

2

Hi walcarpit,
I am on the Central Coast NSW and have been using the Maslow 4.1 for a while now. I think I was the first to get the 4.1 update as it arrived in Australia about 2 weeks before Bar got the 1st kits in America.
Some things to check:
Have you run the Find anchors since upgrading to 1.24?
What is your retract tension value?
Can you copy your maslow.yaml file and post it here please.
Are you using a horizontal or vertical setup?
Serial log is not preserved over restarts - you have to save it before closing browser.

This limits how much belt is retracted before a confirm popup is provided. It is used to pause belt retracting when it has accidently not been terminated on anchor point. You can safely ignore it for now.
Until you have a successful Find Anchors (or manually enter valid Anchor Points) the Maslow can’t operate successfully.
Initial procedure

  1. Power up Maslow and connect via browser
  2. Click Setup (the shortcut key on Main menu (Retract, Extend etc.) is principally used after a successful Find Anchors)
  3. Click Retract
  4. When belts are fully retracted click Extend.
  5. Connect belts to anchor points: Bottom Left (BL) and BR are anchors are closest to the power connector, TL and TR are closer to the type C USB connector, i.e. Power connector points down.
  6. Click on Find Anchors.
    Maslow will do a short upper belt release to work out if it is horizontal or vertical, it will then do an initial search for Anchor points. The first 6 points will have fairly slack belts, from then on the should stay fairly tight as the Maslow goes around 3 Rectangles, stopping at an increasing number of points on each side. It should then return to the middle, and if successful, go to Ready to Cut.
1 Like
3

Thanks for the reply, Ian, good to connect with company in this great southern land, although I have read some of your posts elsewhere here.

I shall work through your questions but first, if our belt extend distance is set at 1200, what should our belt retraction limit be? Currently FluidNC>Config Items> /Maslow_Apply_Tension_Belt_Retraction_Limit is 1000

Have you run the Find anchors since upgrading to 1.24?
I tried and it failed to respond. It will be done as soon as it allows.

What is your retract tension value?
This is tricky to answer. Elsewhere I read that lower is better but we found ourselves making progress when we bumped it up. There appear to be three places to set this (Options>Apply Belt Tension Limit where it is apparently set at 1000 and in Setup>Config>Retraction force where it is apparently set at 1400 and in FluidNC>Config Items> /Maslow_Retract_Current_Threshold where it says 1400).

Can you copy your maslow.yaml file and post it here please.
I presume the file is on the board itself. If that is no longer identical to one that we installed, which I guess is the one that actually has the data that I just relayed above. If I have necessarily have an exact copy on this computer please tell me how to access it. If I have to retrieve it from the Maslow board, please teach me how to do that.

Are you using a horizontal or vertical setup?
Horizontal. Currently on the tray of my ute (elsewhere known as a pickup truck) with the anchor points at the four corners of a full sheet of ply. I am planning to contribute a design to the fora about collapsible frames but until we are confident at controlling the beast that remains an engineering concept.

thanks again,

Walter

4

It may be confusing but it Belt Retraction limit is the distance the belt will travel after Apply Tension is applied before it stops and asks if you want to continue. It is there in case you forget to connect one or more belts to anchors. You can safely ignore it for now.
Retraction force (same thing as Maslow_Retract_Current_Threshold) value is only used when running Find Anchors and Retracting the belts. The lowest value you can run and still get all belts to retract is the optimum.
To upload a file from the Maslow to your computer, go to FluidNC pulldown select the 1st green button (Manage Local Files) then click on the down arrow next to the file you want to transfer. It should download to your downloads folder where you can drag it across to here.
You will probably need to expand your anchor points to get a decent work area.
I have to go out for a couple of hours. I’ll check to see if you have communicated when I get back.

5

Thank you for the clarifications, they help heaps.
I used an iPad to transfer the yaml and here it is:

maslow.yaml.html (7.0 KB)

6

you could try setting these values to match what you said was your setup is 2400mm x 1200mm (Sheet of ply on the ute).
kinematics:
MaslowKinematics:
tlX: 139.199997 Change to 0
tlY: 3230.600098 Change to 1200
tlZ: 100.000000
trX: 4308.899902 Change to 2400
trY: 3230.699951 Change to 1200
trZ: 56.000000
blX: 0.000000
blY: 0.000000
blZ: 34.000000
brX: 4340.399902 Change to 2400
brY: 0.000000
brZ: 78.000000
beltEndExtension: 30.000000
armLength: 123.400002
beltToothSpacing: 1.998800
encoderTeeth: 22.000000
maxSegmentLength: 5.000000
fixedZ: false

7

Thank you Ian,

We changed those settings in the Fluid config. We also followed everything else that you suggested including running Find Anchors which went through some 42 positions.

Whichever one or all, it worked.

The Maslow successfully drew 18 rectangles with half of them doubled to represent a bevel.

It is not a lemon and I am merely mortified that I let a typo into the subject.

Thanks again.

1 Like
8

Hi Walter
Fixed your typo. As a point of interest what did the Anchor Points end up being?

1 Like
9

I’m guessing TL: 1173.342 TR: 1160.330 BL: 1160.781 BR: 1158.118 answers your question.

I saved the Maslow serial log after the find anchors completed. 56 Waypoints to be precise.
I copied this from the end of it:

[MSG:INFO: Find Anchors fit: rms=0.579mm max=1.732mm perAnchor=[0.570,0.600,0.527,0.615]mm converged=1]
[MSG:INFO: Find Anchors recompute complete: tl=(-0.523,1189.088) tr=(2344.653,1182.357) brX=2346.565 SSR=76.373 points=57]
[MSG:INFO: Find Anchors set Maslow_Extend_Dist=1262.156 (TR-BL avg=1312.951, TL-BR avg=1315.556, extension=153.400)]
[MSG:INFO: Find Anchors state reset]
[MSG:INFO: Requesting state change from Finding Anchor to Ready To Cut]
[MSG:INFO: Synchronizing motor positions for READY_TO_CUT:]
[MSG:INFO: TL: 1173.342 TR: 1160.330 BL: 1160.781 BR: 1158.118]
[MSG:INFO: Succeeded]
[MSG:INFO: Find Anchors complete]

So I’m guessing that TL: 1173.342 TR: 1160.330 BL: 1160.781 BR: 1158.118 answers your question.

It was interesting to watch the ply sheet bend as the belts pulled on the corners. It’s not something that worries us at the moment and in spite of that the Maslow retraced lines pretty accurately.

1 Like
10

That is a good thing to be concerned about. If there is significant flex during the calibration process it’s going to lead to inaccurate measurements which will throw things off

11

These show the amount of each belt extends to reach the centre point.

These are the calculated anchor points
tlX=-0.523, tlY=1189.088, trX=2344.653, trY= 1182.357, brX=2346.567, brY=0,
blX=0 and blY=0.
Recommend you have some sort of rigid cross brace under the ply to mount your anchors on. This will increase the effective area your Maslow can work on and be much more accurate. A frame you can drop into the ute which extends out the back (so the belts don’t foul the cab) could work.
See the following for some ideas
Maslow 4 Documentation | Documentation and resources for Maslow 4 CNC machine