I’m planning to cut a Sandwich panel, which is made out of 46mm thick polystyrene, with 2mm fibreglass sheets on both sides (50mm total thickness).
The panel is 2.4m x 2.4m I am planning to make my frame bigger than the panel. My question is by how much. I’ll extend the belts with Dynema rope (that does not stretch). How much longer should I make the belts?
PjMaslow wrote:
I¢m planning to cut a Sandwich panel, which is made out of 46mm thick
polystyrene, with 2mm fibreglass sheets on both sides (50mm total thickness).The panel is 2.4m x 2.4m I am planning to make my frame bigger than the panel.
My question is by how much. I¢ll extend the belts with Dynema rope (that does
not stretch). How much longer should I make the belts?
play around with this frame calculator
http://lang.hm/maslow/maslow4_frame.html
move the anchors until you have your panels in the green (or almost in the
green), then it will tell you how much belt you need.
currently, the maslow only supports rigid belt extensions, but if you have the
anchors raised so the belts are almost flat, the difference should be minimal.
David Lang
Ok so I have followed your advice from the other thread and installed solid extensions rather than wire rope cables.
I got the machine calibrated first time. Very happy.
It was the end of the day, and I needed to pack it away. When returning the next day, I extended the belts, hooked them into the holders, and applied tension. Got “Ready to Cut”. Very happy!!!
Hit down to “JOG”. Machine started unspooling all the straps. non-stop. All of them. Very Sad!!!
Tried twice more with the same results.
Please help!!!
Can you put a copy of your maslow.yaml file and a copy of the maslow serial log file here?
How long are the fixed extensions? What did you use.
Which version of software, firmware have you loaded?
Unconfirmed 901407.crdownload (6.8 KB)
Im running Version: v1.15. My solid extentions are 800mm. I used 30x30mm square tubing.
Default Belt end is the black bit at the end of each belt (30mm)
You will need to change the belt End Extension from
beltEndExtension: 30.000000
to the distance from your anchor point to where the belt goes into the belt end. Mine is almost exactly 1Metre, yours will be around 800mm
OK thanks for the fededback.
What I ended up doing was recalibrating the machine, and it kept giving me issues. Some of the belts were giving too much slack when it moved.
I noticed that my extension arms were bobbing up and down when they took up tension. I thought this might cause issues, so I ensured they were fixed and did not move up or down when tension was applied.
This seems to have fixed the issue.
I just made it move along my first drawing. Very happy.
Thank you for the support.
Did you change the Belt End setting?
No. It seems to be working. Do I still need to?
Quick question with regards to the Z axes.
I believe fully down is zero. So when i put the bit in it will be in the up position. When it starts it goes down all they way. How should it work when you start the cut? I presume it should stay up until you get the the first plundge?
PjMaslow wrote:
I believe fully down is zero. So when i put the bit in it will be in the up
position. When it starts it goes down all they way. How should it work when
you start the cut? I presume it should stay up until you get the the first
plundge?
That will depend on your CAM software, some setups tell the system to home,
which takes the Z axis to zero, not what you want.
what CAM software are you using to create the gcode?
For example, in fustion 360 there is a grbl postprocessor that does homing, and
a maslow postprocessor that is basically identical execpt it skips the homing.
David Lang
I would expect it not to have an accurate idea of where your anchor points are or that it would roll out too much belt. On reflection it doesn’t really know, it just pulls each belt tight, has no way of knowing how much is still internally spooled.
With regard to the Z axis - Z stop is the lowest the machine is capable of and is where the Z axis is all the way down (as when calibrating/Finding Anchor points) Z Home is where you set the bit just clearing the work surface prior to cutting. Your CAM software will have a safe setting you can set (I use 2-4mm) which lifts the bit when moving and not cutting, then depth of cut lowers it into the material to cut.
Ok Thank you Ill set the Belt end today and give it a bash.
I use SketchUcam in Sketchup. Ill try again and keep you posted. Thanks again for the feedback.
OK, I changed the Belt end distance and recalibrated it.
Did a dry run, and it worked well up to a point, then stopped mid-cut. Log attached. Please let me know why it’s doing that.
Maslow-serial (4).log (13.7 KB)
I also see that it always goes down to zero on the Z axes when it starts so I will still need to fiure out what in my Gcode is causing it to happen. Would it be as simple as to remove that line of code or will it be more complex?
Window Test 4.nc (10.9 KB)
I see from your serial log your Fitness is 0.61, that is on the low side with only 17 measurements. I think you need to increase the calibration grid and rerun Find Anchors. I don’t think it has a very good idea of where your anchors are.
Do you need 1200 to reliably retract belts? Try lowering it. If you can get it done to 900 that would help.
Generally, Calibration threshold does not need to be higher than retract threshold current.
I am guessing you were not actually cutting? Setting that depth (18.75mm) is pretty ambitious if you were.
Ok I removed my extensions just to see if I can get it working, as I think they are causing the issues.
Recalibrated and it worked fine. Did an aircut and it worked well, but stopped close to the end. “lost signal for more than 20 seconds.”
How do i reconnect and continue with a job once that happens?
Just completed my first cut. All went well.
Got the Gcode sorted by just changing the Z axes at the start to a higher level. Fairly easy.
Im still not sure about the calibration as my cut is not 100% to scale. As you can see in the photo my circles arent 100% round.
Would it help to increase the calibration grid?
PjMaslow wrote:
Would it help to increase the calibration grid?
usually the larger the grid the better.
David Lang
Do you have the debug setting which prints the motor currents turned on? I’ve seen that cause connection issues because the machine is sending so much more data
The last cut went well and the machine did not stop.
