Newbie-Questions

Hi,

New owner here. I got a second hand Maslow 4.1 in the UK. It seems to have had very little use. I have managed to run the calibration and I am using it on the floor. I have a few connection issues but have managed to make it move and cut (but not what I want).

Could someone help me iwth a few points? :

  1. Do you need to set the Z home everytime you use it? I had a cut fail part way through (it just crashed) and when I retried to restart the bit was pushed all the way down and it started to try and cut two boards thickness in one go instead of the 6mm I asked for.
  2. Do you have to retract and extend everytime you turn it on? I.e. you can’t leave it it setup and turn it on? Every time I turn it on it wants me to go through that again.
  3. I did the calibration with a router bit in and at a random Z height ( I had missed the bit about removing it). What difference will that make if I set the Z separately? Is that why my cut went weird on the restart?
  4. Does it run autominously if it loses wifi? Is that why my cut may have failed?

I’m using a 4.1 with v1.19 firmware.

Sorry for all the questions.

Regards,

Francis

boredofchoosingnames wrote:

  1. Do you need to set the Z home everytime you use it? I had a cut fail part
    way through (it just crashed) and when I retried to restart the bit was pushed
    all the way down and it started to try and cut two boards thickness in one go
    instead of the 6mm I asked for.

whenever you do a retract/extend cycle, you need to run the router all the way
down and set Z-stop. otherwise the machine may have the wrong idea of where it
is.

Z-home is an offset from Z-stop, intended to let you set the end of the bit at
the surface of your material and use that as your starting point

  1. Do you have to retract and extend everytime you turn it on? I.e. you can¢t
    leave it it setup and turn it on? Every time I turn it on it wants me to go
    through that again.

you should not need to, if it does a clean shutdown (power off after finishing
something and being idle for a few seconds), it should remember where it was.

  1. I did the calibration with a router bit in and at a random Z height ( I had
    missed the bit about removing it). What difference will that make if I set the
    Z separately? Is that why my cut went weird on the restart?

it means that your calibration is incorrect and the anchors are not where the
machine thinks they are.

in the calibration, it is taking measurements of the belts and calculating where
the anchor points actually are, If the router was not all the way down, then
it’s calculations are going to be wrong (because the anchors are closer than it
calculated them to be)

  1. Does it run autominously if it loses wifi? Is that why my cut may have failed?

it is supposed to keep going if it loses wifi.

I¢m using a 4.1 with v1.19 firmware.

good, that is the current version

Sorry for all the questions.

questions are good.

David Lang

2 Likes

Thank you David,

I’m going to ease the anchor points which seem a bit tight and rerun the calibration/z-stop and z-home etc.

I have had some odd behaviour but am getting the hang of it.

Below is the first cut of shame….

I’m mildly impressed it managed to more or less cut two boards in one pass. Might have dulled the bit but I stopped it fairly quickly.

Regards

Francis

1 Like
  1. After a crash you need to check the Z stop and Z home values are correct Z stop is with the Z axis all the way down, then click on Set Z-Stop under config menu. You can do this without removing bit by putting a block under the Maslow to lift it so the bit doesn’t foul the surface. This will also set the Z home to 0, so you adjust that to where the bit just misses the surface of you work piece and set Z home.
  2. You should be able to leave it setup and power off. On power up you have to Apply Tension to get to Ready to Cut. I usually move it to the centre before releasing tension and powering off as I have found it can move when the belts aren’t approx. equal on power off.
  3. The machine identifies it’s location by extending and counting rotations driven by the belt. It expects the Maslow to be as low as possible to get an accurate map of it’s position, It compensates for the Z height during operation so yes, it is important that Find Anchors is done with Z right down.
  4. It will continue to complete the job after losing browser connection. Work is being done to allow reestablishment of control from a new browser instance.

Did it come with V1.19 loaded? Can you paste a copy of your serial file here.
Until you get a feel for the machine and how it operates, I suggest you run some GCode files a few times without a bit in the machine, i.e. air cut

2 Likes

Ian,

Thank you for the reply. I’ll try and recalibrate today. The machine came with the firmware updated by the seller, but it has been about three weeks since I bought it and have had time to set-up and print the anchors. I had some odd disconnections so I updated it to the newer firmware that was released in the interim.

Since writing, I had some issues retracting also, top right and bottom right belts wouldn’t go in. I had to extend before I could get them to retract correctly. I am wondering if this is what caused the failed cut if it is binding. I guess trying the calibration will flush that out.

Attached is my current serial output.

Regards,

Francis

Maslow-serial (3).log (3.3 KB)

So I had another go at calibration and it failed several times. Also on attempting to retract I had an issue with top right and bottom right. Neither retracted and I had to extednm then retract again to get them to pack away. Serial log attached.

Is it worth extending the belts to their full extent to release any built-up tension in the reel?

Regards,

Francis

Maslow-serial_failed_calibration.log (96.3 KB)

So the retraction was set at 1300, upping to 1500 allowed full retraction. I appreciate this maybe masking a symptom but I will try calibration again.

Upping the force to 1500 did not work. I am still getting two cables that require multiple attempts to retract.

Maslow-serial_failed_calibration_with1500.log (99.2 KB)

boredofchoosingnames wrote:

Upping the force to 1500 did not work. I am still getting two cables that require multiple attempts to retract.

Then you are going to need to disassemble the arms and make sure everything
moves freely

make sure the spool turns freely on each half of the arm and that the gear on
the motor is not loose or rubbing against the plastic.

David Lang

1 Like

I agree with David, a lot of us have sanded and lubricated where the spool runs within the arm. A silicone lubricant is recommended. You need to ensure that the spool turns easily on each half. As the parts are moulded, there is a slight taper to allow them to release from the mould so one side will be tighter than the other. While you have it apart, I would do all the arms. You might find a slight bulge where the nuts and bolt heads have bulged the plastic slightly.

2 Likes

Ian / David,

Thank you for your comments. I have had a very long night disecting my machine. I do not have silicon lubricant, any one have a link on Amazon to the suggested version?

Anyway things I found:

  • lock nuts where there shouldn’t be holding the router clamp on (real pain to remove)
  • One motor had a the gear tight against the plastic (one of my non retracting arms)
  • One motor had no grub screw on the gear (a working arm)
  • One motor had the grub screw but it was loose (another of my non retracting arms)
  • Bad fitting cable guard. I think I have improved it.
  • As far as I can tell, the arms were added in the wrong order. I labeled them as I took them off and when I installed them following the build guide, they have ended up pointing to different positions.

I did also remove the cable from the spools and they freed up fine but I suspect I will be taking apart to lubricate again. I will try a calibrate in the morning and see.

I think arm error probably happened when the previous owner did the upgrade. He did say he had not used it after the upgrade so probably didn’t notice a problem.

Francis

When you say you removed the cable from the spools, do you mean the belts? I hope you didn’t take them right off, the should be glued into the spool and nearly impossible to remove. The spools need to spin freely on each half of the arms. They weren’t originally designed to be lubricated, so if they are spinning freely it will probably be OK.
I ended up putting an extra bearing on the motor side to stop the drive screw damaging the plastic (not designed that way) and left off the grub screws after I found two of them floating around the first time I pulled the arms apart.
The arm order is less important, so long as the Z offset in the maslow.yaml file corresponds to the relative height of each arm.

Ian,

No, I did not disconnect the cable, just unroll it to free it up and check the reels would move.

Slightly worrying that all the issues I found, you think may not be a problem :laughing:.

I’m going to run a Z stop and calibrate shortly and will report back.

Francis

boredofchoosingnames wrote:

  • lock nuts where there shouldn¢t be holding the router clamp on (real pain to remove)

I use the lock nuts everywhere. they are a pain to remove, but it’s more painful
to have nuts falling out.

I also switched to different bolts that are much easier to deal with deeper
heads so they don’t strip and smaller heads that better fit the holes

  • One motor had a the gear tight against the plastic (one of my non retracting arms)
  • One motor had no grub screw on the gear (a working arm)
  • One motor had the grub screw but it was loose (another of my non retracting arms)

fairly common problems.

  • Bad fitting cable guard. I think I have improved it.
  • As far as I can tell, the arms were added in the wrong order. I labeled them
    as I took them off and when I installed them following the build guide, they
    have ended up pointing to different positions.

that would make a significant difference. The Z offsets for each arm being wrong
will ruin your calibration attempts.

I did also remove the cable from the spools and they freed up fine but I
suspect I will be taking apart to lubricate again. I will try a calibrate in
the morning and see.

keep us informed.

David Lang

The bit that I said wasn’t a problem was the order of arms, with the proviso that the corresponding Z values in the maslow.yaml file matched the actual order.
The other problems you found are significant.
Lubrication maybe necessary, but they were originally designed without any.

Evening all,

Thanks for the responses.

I did mange to get it calibrated after much faffing and fiddling. Took about 3 goes and on the last one it calibrated really quickly.

By that point it was too late to do too much more. Will try a dry run tomorrow.

Still having issues with retraction but it is not as far as I can tell due to excessive tension.

If I issue the retract command it will pull back 2-3 belts. The remaining one wont move. I can then do extend and it wont move either.

If i then push the loose piece of belt back in, and give it the lightest of touches to pull back out it will extend really easily. (Hence I don’t think tension)

I will try and upload a video of this behaviour. It feels more like a sensing issue than a tension issue.

Francis

What you are doing is moving the roller and this revolves the magnet which the sensor picks up and allows you to draw out the belt. When the belts are retracted, you are doing the same thing, but it is harder initially because the belt is tight.
What value do you have as your retract force (set up, Config). You need to set this to the lowest value that lets you retract all belts reliably. It is used in the Find Anchors.

1 Like