Troubleshooting
| Topic / Symptoms | Diagnosis / Solution |
|---|---|
| Boot-Up Errors | |
| Encoder not detected Motor not detected |
* Connector not fully engaged - try unplugging the ethernet cable connecting the encoder board to the controller and plugging it back in * Bad controller port / ethernet cable / controller board - try swapping the ethernet connections and see if the problem follows the encoder/motor or the controller port. Try swapping ethernet cables to see if the problem moves with one of the cables. * These errors generally have been fixed by replacing the problem component. Send a message to anna@maslow.com to describe your issue and get a replacement. |
| Belts extend after a “Retract All”, “Extend All” series but not far enough to reach the anchors | * The Maslow thinks the frame is smaller than it is. Check FluidNC → Maslow Settings Maslow_tlX: should be a small number, say < 20 Maslow_tlY: should be about the height of your anchors Maslow_trX: should be about the width of your anchors Maslow_trY: should be about the height of your anchors Maslow_blX: 0.000000 Maslow_blY: 0.000000 Maslow_brX: should be about the width of your anchors Maslow_brY: 0.000000 * The Maslow is mis-measuring the belts. This could be caused by the magnets loose or not positioned over the encoder module. |
| Calibration | |
| General comments | Calibration is figuring out exactly where all the anchors are. It should only need to be done once and is saved in the maslow.yaml file. If you lose the maslow.yaml file (including overwriting it without saving the maslow_* values), or change your frame size, you will need to do it again. |
| The maslow from dimension is defined as the lower left anchor being at (0,0) and the lower right being at (Maslow_brX,0) | |
| The Maslow has a built in process that figures out the locations of the anchors by taking multiple measurements of the belt lengths. | |
| My sled dropped to the floor | Incorrect horizontal/vertical setting. Check that your Maslow is set to vertical orientation. Vertical (upright) is where the sled is hanging from the belts and will fall to the floor if the belts extend. Horizontal (flat) is where the sled sits on a workpiece that on the floor or a table-like frame. |
| One or more belts won’t fully retract during “Retract All” or Belt movement is uneven | • Increase current limit during retraction and calibration |
| • Adjust the motor position (what is meant by this?) | |
| • Extend and retract the belts a few times to wind the belts on the spools more evenly | |
| • If belt movement is uneven, check that the magnets are installed in the roller that is positioned over the sensor on the encoder board. (This will require opening up the arm assemblies) | |
| Unable to move errors during calibration | • Check that the width and height between anchors is accurate in the settings (Xbr and Xtr should be close to your width; Ytl and Ytr should be close to your height) |
| • it helps if the frame is very near square (measurements of the diagonal match), but it will work if the frame is not square or even close to a rectangle. | |
| • Try increasing the above settings slightly | |
| Sled falls off of workpiece during calibration | This can happen if you set the calibration size too large; try again with a smaller calibration size. |
| Arms strike the uprights during calibration | The workpiece is too large for the frame. If two arms of the Maslow hit the plastic frame trying to go in different directions, the system cannot detect this and the actual distance to the anchor will be shorter than the system thinks it is when it pulls the belts tight. The jury is still out on how badly this affects accuracy when cutting, but it will throw off calibration. |
| Z errors | The maslow instructs you to run the Z axis all the way down to have the frame against the stepper motors before you run calibration. If you do not do this, the Z distance between each of the belt arms and the anchors will be incorrect. This will result in the actual distance to the anchor being less than the system thinks it is. |
| • If you have a non-standard router, you may need to alter the Maslow_**Z values in the configuration (“Setup” → “Config”) | |
| • If you have an unusually thick spoilboard/workpiece under the sled it will make the default values (which assume 3/4 plywood spoilboard and nothing else) incorrect. | |
| • If you stack the arms in a non-standard order, the default Z values will be incorrect. | |
| Poor fitness or a successful calibration that does not produce accurate cuts. | Frame flex If the anchors move as the belts tighten, this will cause errors as the belts will be pulling in different directions at different strengths depending on where the sled is on the workpiece. |
| Error “Grid spacing is too large. Please reduce the grid size or increase the number of points” | The distance between the planned calibration points is too large, which means there is the risk that the belts will get too slack during calibration and get damaged. The maximum grid size is n * 200mm, where n = the number of grid points in each direction. |
| Software issues | |
| • Calibration takes place both on the device and in the web browser. During calibration it is important to have your web browser active for the whole session. Make sure you don’t disconnect and only access the maslow during calibration from one device. If you have multiple web browsers then they can both respond to the calibration events, causing issues. | • Capture the serial messages after calibration (success or failure!) using the “Save Serial” button. This can be very useful in the forums if asking for help on calibration issues. |
| Error Codes | |
| “What does error code: x mean?” | FluidNC error codes (may or may not help  ) Alarm and Error Codes | Wiki.js |
| Error:82 | It seems the error:82 is somehow related to using Mavericks OS (10.9.5) with Firefox 78.15.0esr, the errors are gone with High Sierra OS (10.13.6) with Firefox 115.10.0esr. More details at (Having trouble after calibration) |
| General Operation | |
| “Push to extend bug” – To extend the belts during the Extend All operation you must push on one or more of the belts rather than pull them. | Current hypothesis is that this behavior is caused by a pin on the encoder chip not being grounded. More details at (Push to extend bug?). Try soldering connections to ground or request an encoder board with the pin grounded from Maslow. For an illustrated guide to soldering the connection, see (Soldering Floating Encoder Board Pin) |
| “Extend All” does not extend far enough to reach the anchor points | If you have updated your Maslow.yaml file (probably when updating firmware), you need to enter the calibrated anchor points into the config dialog. From the Maslow main screen press “Setup”, then on the setup pop-up press “Config”. Copy the values for tlX, tlY, trX, trY, blX, blY, brX, and brY from your best calibration into the corresponding boxes and try “Retract All”, “Extend All” again. |
| One (or more) of the belts does not retract when “Retract All” is pushed | Try increasing the “Retraction Force” setting. 1300 is default; try 1500 or 1700. You can safely go up to 2500. |
| One (or more) of the belts does not retract when “Retract All” is pressed and/or Belt over-extends when “Extend All” is pressed and/or [MSG:WARN: Magnet not detected on Top/Bottom Left/Right] | Roller containing the magnet is improperly placed. Disassemble arm to ensure that the magnet is positioned over the detector on the encoder board |
| Panic with message “no heartbeat detected” | The community is still trying to figure out what is causing this. To recover, 1. power down, 2. power up, 3. click “alarm” to reset, 4. in setup, release tension, 5. disconnect the belts from their anchors, 6. retract all, 7. extend all, 8. reconnect the belts |