Those are very fast speeds. But it isn’t really fair, since there was no load. The available torque drops at higher speeds. What youve measured is where that torque drops below the torwue needed to move the gantry. But not where that torque drops below what you need to mill.
One thing I’ve done to add a constant load to the bit is to tie some string to the bit, and then drape the string over the side of the table and hang a weight of a specific size (0.5kg or 1kg). Then the gantry has to move on its own, and also lift that weight (which is constant force, ignoring friction). That might give you more realistic top speeds.
You did up the firmware speed limits for moves and travel right?
I think that is close to what I was finding with no loads, but with a router and hose on board. Cut those down significantly and you have some safe rapid travel speeds to use.
I changed to a 24V power supply today. I was able to get to 40m/min on the Y axis with the spindle attached, but no hose. The motors didn’t skip at that point and still seemed to have more than adequate torque. I didn’t test any further because if one of the motors started skipping, the gantry would jump off the rail and things could get broken. Sometimes enough is enough. I won’t be cutting at this speed anyway. Honestly I moved away from the table during the tests, as it didn’t feel very safe to sit next to CNC moving around like an F1 machine
40m/min is a lot. To be honest even 20 is a lot. I haven’t decided what speed I will use for the finishing passes, but I will probably concentrate more on acceleration settings to make sure all the movements are nice and smooth and accurate.
Then I did another test with a 20mm stepdown into the foam. I went all the way up to 9m/min, then I could hear the spindle slow down and chunks of foam started to break off. The machine was pushing it no problem, so it seems that my chinese 500W spindle is the bottleneck. I think it is rated 10k rpm, so a faster spindle/router would be a good idea. In any case, I won’t be cutting at this speed either. Somewhere between 3 and 5 m/min is probably realistic.
During the speed tests the current didn’t go up more than 1.6A on the 24V PSU. The motors were almost cold, the drivers warm but not too hot. I had the stepper power set to 1250 mA.
Below is a picture of LR3 cutting through the foam at 8m/min with 20mm stepdown. I wanted to be able to remove large quantities of foam quickly, and it seems that LR3 can do a very decent job here )
BTW the faster I cut, the more dust is thrown around the machine. The MGN rails were all covered in snow today and I already had to lube one of them to keep it moving smoothly. I will try to put the vacuum back on and see if I can come up with a better way to contain the dust.
Yes, Z travel was limited to 200mm in the default firmware, and mine is 230, so I had to recompile anyway to get the full Z range. Now I just use M201/M203 commands to set the limits, not sure if that was possible with the original firmware or not.
I need to buy a longer vacuum hose and I think my machine is ready to mill the first board!
Not distance, speed. I have the firmware hard limited on max speed, max travel, and max accelerations. I have it set to a max of 50mm/s with you getting 666mm/s you would need to make a lot of changes including acceleration or the machine is not big enough to get over 200 or so. I have 24V speed tests in another thread.
Both of these sound high. Your power supply was drawing 1.6A at what voltage, meaning before or after the brick? What steppers do you have that 1250 didn’t melt the mounts?
Honestly I don’t remember what settings I changed exactly, it was a few weeks ago. I went through Configuration.h and Configuration_adv.h and changed the settings that were related to table size, Z travel and possibly something related to feed rates, not sure.
I still have default max feedrate unchanged in my firmware:
They are not melting the mounts In fact they are as close to melting the mounts as your finger is close to melting your left mouse button hehe. A little bit warmer than my hand but that’s about it. The stepper drivers are way, way hotter! But still seems to be in the OK zone.
I have a “lab” PSU with knobs for adjusting voltage and current and it also displays voltage and current in real time. When the motors were driving around at full speed, I saw the current jump to 1.6A for a few milliseconds, but most of the time it was staying around 1A for idle and 1.2-1.3A while moving.
The reason I started experimenting with speed settings was that… well, surfboards are big. With the default speed settings it would probably take ~6 hours to mill one side of a board… kinda boring. Especially when the finishing pass is just driving around removing 1-2mm of foam with almost zero resistance… why not speed it up if the machine can handle it?
Absolutely. We always need to double check when big numbers get mentioned. People used to say 1000mm/s+ not realizing they never got past 50 from my limiting.
Another update about my progress: I tried milling the full size board. OK, where do I start… milling foam is obviously more complicated than I thought.
I had to calibrate the stock and make a flat surface at 100mm. I used a 30mm flat surface milling bit with 2 flutes. The endmill makes bigger foam chips instead of powder that comes from the burr. The chips fly everywhere and stick everywhere. Super sticky! The chips stuck to the X bottom bearings which later started jamming the X movement. I had to stop half way, because X motor started skipping.
Since I didn’t need the endmill to go deep for this pass, I raised the spindle, installed the dust shoe as close to the surface as possible with a hope that it will contain the dust. Honestly - not much help. Slighly better, but the bearings were still covered in foam chips.
OK so its a short pass, I paused it a couple of times, cleaned things up, eventually got the surface done.
Next was the pocket roughing pass. The burr cuts much nicer and the dust stays around the cutting area. Life is much easier BUT the bearings are still covered in foam. As the bearings roll around the X rail, they compress the foam and form some kind of material similar to paper, which then rolls around the bearings creating a mess. Basically I had to babysit it and vacuum the bearings manually every minute or so. Boring stuff, but it works.
In the middle of the first pocket cut, my chinese 500w spindle proudly passed away after an outstanding service life of about 5 hours. Honestly I don’t think it liked the dust all that much. I liked the chinese spindle because it was so silent. Oh well, now it’s even more silent! Luckily my stock doesnt seem to be damaged, so I am off to the store to buy a makita.
I think in the end I will need to design a custom dust shoe and make sure it also sucks the air away from the X bearings. The question about whether MGN rails will jam because of the dust/powder is still open.
For what you are doing, you need long bristles. I designed that shoe for close wood cutting. You have a very unique build, That I wish I could replicate but I am too far from the ocean to start shaping boards. Add some long bristles to the shoe to get them all the way down to the cutter. you can glue or screw them on to the shoe I designed if you do not want to redesign. That should take care of more then 90% of the mess. Maybe -Amazon.com
You can also add a sweeper to the lower rail to cover the bearings and sweep the rail as you move you have inches to spare on each side. I do think some bristles will make it so this is not needed.
Any chance you could build some sort of shield/wiper around the bearing so that foam is kept away from the bearing-to-rail interface and the foam dust gets wiped away as the carriage moves across?
I actually ordered these a few days ago! Still in the mail. But that got me thinking: If the bristles are long (say 100mm, the longest I could find) and the dustshoe radius is not so long (say 30-40mm) then there is a chance that the bristles will reach the endmill rotating at high speed and then… well, i will probably need a new dustshoe )) Or not?
From there I thought I would have to design a dustshoe that is wide enough to ensure that the bristles never reach the endmill. That probably means R=100mm give or take a few mm. That seems kinda big ) But if that’s what it takes, why not.
Sailplanes have wipers across the length of their wings to wipe away stuck flies and mosquitos and increase aerodynamic efficiency. If that can be done, I am sure some sort of wiper for LR3 bearings can also be invented!
Not sure Can you feel the static build up anywhere? If I get near my dust bucket I get zapped when I touch the ground, SO I need to ground that thing as well. I wanted to even print the dust shoe out of conductive material just to get it to ground more. I am becoming a ground-a-holic.
I think it all comes down to the parts you are cutting. Along the sides is where you will find out. flat surfaces are easy. That is one of the things I would want to see in person!
I had a bunch of static in my vac hose and it was causing the controller to reset and ultimately killed a built in driver. . It is one of those weird things that you don’t expect.
Each piece of foam doesn’t create a ton of static, but a couple of million pieces can add up.
It is my impression that an anti-static hose will have a continuous piece of metal as the spine (and maybe they mix up some special plastic, IDK). Do you have to ground that spine to make it work?
The easy way to ground a vac hose is to just use some ground wire (like from a romex) and put it through the hose, and ground it on one side. If even a few volts start collecting at one end, it will leak or short to the wire instead of being able to build up to the point of being dangerous. As long as you aren’t sucking in big big chunks, the wire won’t cause clogs.
If the spine is already the ground wire, then I think you just have to ground it.
Would static be reduced by terminating last 6-8 inches of regular vac hose with rigid aluminum ductwork (usually used for dryers), grounded and non sparking? Also, ensure router body is grounded?
Random ideas… positive air flow along the rails, or bristles fastened to core, or motorized bristles even rotated/oscillated by small motor. Bonus points for making a Core Addon with rail cleaning bristles mechanically driven by x axis belt as the Core moves.
In fact they are as close to melting the mounts as your finger is close to melting your left mouse button hehe. A little bit warmer than my hand but that’s about it. The stepper drivers are way, way hotter! But still seems to be in the OK zone.
. It is one of those weird things that you don’t expect.