July 12, 2017 at 1:26 pm #38123
At the brach today, have to get a shot of it all tomorrow. Tried doing the same test at a corner to minimize conduit flex. Results were barely any better.July 12, 2017 at 1:29 pm #38124
That’s what we are getting at, it shouldn’t be flex. lets see the whole thing, what you are seeing there is far from normal. something is wrong. If it was a ziptie issue your zip ties would have to be flexing a ton. I am chasing issue that are so small they are hard to measure, yours is like a 1/4″.July 12, 2017 at 4:23 pm #38134
42-ish mm to the end mill. I have it pretty much as far in as it will go, might get a couple mm more, but it will start hitting the flutes. I also need enough hanging out for the dust collection. With it how it is right now, about half the end mill will stick out the bottom.July 12, 2017 at 5:39 pm #38139
If you push it at one of the x or y motors, then you’ll eliminate a lot of culprits. If you do it where the belt is 4x longer on one side, then if it’s the belt, then it will stretch 4x longer in direction and not the other. Be sure to test the 4:1 and 1:4 to eliminate differences in zip ties.
Ryan, those ties are much tidier than I’ve seen before. I want to go check my low rider out now.July 12, 2017 at 6:25 pm #38141
Looking at those pictures I can see the Y axis on mine has a loose belt, the just out of focus ziptie is too round. I don’t want to touch it though it is printing so perfectly right now.July 12, 2017 at 6:58 pm #38143
The video is soo bad…helping me is going to be a challenge. Phone camera is broken so I have to use one of those 2012 digital cameras, super grainy. I will make a cutting aluminum shot tomorrow when the lighting is good.
You can’t really make out anything in that video. Unfortunately I think I am suffering from one of my classic, “sum of everything” problems where there isn’t one cause. It can move linearly more across the Y than the X. This is likely from the x conduit being a lot longer so the gantry can actually bend it more. It still bugs me though that my steppers can actually move with the belt though. Just checked, none of the pulleys are slipping off the stepper shafts.
I may just put this to rest a little and wait until I make whatever the mpcnc 2 ends up being. Brushless spindle rpm controlled, stainless steel, hex hardware, better belt system, more perpendicular, and whatever new designs there might be.
Just my opinion, if there is an mpcnc 2, one of my requests would be to do some sort of upgrade to the belt system. That or include it if there is a pro version. This is really only an issue found on smaller machines that have such little conduit flex.July 12, 2017 at 7:00 pm #38144
The lowrider is 9’long and uses the exact same setup.
Take some pictures it doesn’t have to be video. I just need to see it all.July 12, 2017 at 7:09 pm #38145
What in particular do you need pictures of? Tried the best I can…sun is setting. I don’t think it is even my zipties anymore. This is so hard to explain. Hey if you drive down 7 hours, maybe you can take a look :D. Jk, but if you saw this irl you would probably know whats up in like 5 minutes.
In those videos I am pushing pretty hard, harder than the material I cut will ever push back with. Honestly if you need to work on more important stuff that is fine by me. Though if you really do think what I have here is really abnormal then be my guest 🙂 All my corner legs are JB welded so they are really stiff. The workbench they sit on is also quite stiff.
Attachments:July 12, 2017 at 11:12 pm #38161
Dui, ni shuo de duiParticipant
I think what you describe is normal, I have the very same problem, though I have quite a bit less play compared to your first video. I think this is a limitation due to the belt system, not much can be done here. I tried tightening my belts like crazy, at which point the gantry becomes quite hard to move, entirely removed the zip ties, but it still has some flex.
The only thing I have not tried yet it to swap those tiny belts to some wider ones, but I don’t think I will, because it’s a lot of work for an uncertain result.
In my opinion, the only sure way around this is to change the belt system for a threaded ball bearing system or at least a threaded rod, but it it quite a lot of work and the Z axis itself will probably never be rigid enough anyway.
I’ve just purchased a 3040 CNC frame, should receive it tomorrow. I’ll use it for aluminum milling and keep the MPCNC for softer materials.
Maybe you could try to tigthten a bit more your bearings, it will reduce a bit the natural movement freedom of your machine and can help to lower the backlash a littleJuly 13, 2017 at 12:13 am #38162
I have been planning to try using PU belt instead of the usual rubber ones. I have both types in 6mm and 10mm width and everything required to mount it. Only thing missing is time.
Upgrading to 25mm stainless hydraulic tubing (found some at a local scrapyard) seems to have reduced the flex a lot.
I have only made a couple cuts in alu with partial success, broke a bit the first time 😀July 13, 2017 at 1:54 am #38165
Someone has tried threaded rod, but backlash must suck. Maybe I can try to covert to ballscrews. Pricey but man that would be sweet. You would have to mount the steppers somewhere on the outer railJuly 13, 2017 at 9:05 am #38180
I think what you describe is normal, I have the very same problem, though I have quite a bit less play compared to your first video.
So not normal to have that much play?
You should not be able to wiggle it that much by the tip of the cutting bit, that is not normal. If both of you are having this issue we should work it out, you both want to make major upgrades to the design (and Dui has already). There are tons of these out there, I know it can be a lot better without any modifications.
Kevin, I just want a picture of your whole machine in a single frame. Preferably a few from different angles.July 13, 2017 at 10:46 am #38188
Took some sexy shots with a dslr. They’re actually so sexy that they’re about 6mb each, all over the upload limit 🙁 imgur to the rescue
Much improved from the beach sand level graininess of the other pictures. My zipties barely move, to be honest I don’t know if I can really get it much better. It may actually be my conduit since that one axis flexes more than the other. Shame though cause even with .5mm stepover 4mm doc adaptive clearing, my cuts are pretty undersized. I tried a 20mm pocket and got like 19.5mm. That is a little too far off for my liking. I really want to be able to cut at a higher doc though, cause beating up the poor tip of the endmill with low doc slots ruins the cutters so fast. Not to mention when the tip is gone you can’t plunge the material anymore, might as well throw the bit out.
Let me explore a finishing pass though. All this deflection is gonna happen no matter what, so maybe a conventional finish pass is what I need 🙂 I will upload a vid later on youtube of some of my sketchy cuts. I think I basically just reached the limit of the machine. I mean for something that weighs like 20lbs and costs $400 canadian pesos, I am pretty happy. I am curious to see how dui’s 3040 goes, that was actually what I was going to get originally last summer.
On the more conservative settings that many others seem to use my cuts are coming out great. Maybe I should stop complaining…July 13, 2017 at 10:57 am #38191
You absolutely must use a finishing pass, even on wood. I thought those numbers were with a finishing pass? Every cut other than HD foam needs a finishing pass, always. rough out all the material and then a finishing pass. If you are within .5mm in aluminum with a 3’axis on a roughing adaptive pass you will be spot on with a finishing pass, adaptive takes bites out of the path walls a finishing pass makes a cut along the path.
So after watching the shaking video you put up, again, and again, you do not have the steppers energized, there is no flex, your rollers are moving correct? I have been so focused on the center gantry when I looked out at the steppers they are moving. so either you have about 1/4-1/2″
of belt movement or the steppers are powered off and not holding. How exactly did you make that video, what steps did you take to energize the steppers?July 13, 2017 at 11:03 am #38194
They were energized for sure. I use repetier host for loading my gcode I make from f360. I just rapid it like 1mm x and y, so that way they engage and they hold their place. They make the buzzing sound we all are not fans of. In that video I was really moving the bit. Like I said, I doubt my material will ever lash out on me that harshly without everything chattering like crazy. Didn’t know a finishing pass was so important….oops. Well my new endmills are going to be delivered in an hour or less so let me make the videos and see what kind of numbers I come up with. What radial stock do you go with? The fact I am undersizing with remaining stock turns off in f360 makes me think, do I even need the offset? I suppose I can just make it do a contour around the pocket.July 13, 2017 at 11:29 am #38200
An idea of where you are at, and keep in mind still with only a roughing pass, there is nothing that says you can’t do two finishing passes as well. you are in the ballpark of “rough-standard” machining, The price is ever increasing on that scale.
You are cutting a full 4mm deep and .5mm wide in aluminum, that was unheard of for this machine 6 months ago before we all learned the benefits of adaptive/trochoidal milling. My video was made a year ago and my business exploded when people saw this, .3mm DOC at 5mm/s. You are doing at least ten time the material removal with your cut.
I am not sure what you are expecting but I just wanted to kind of show you where your at and I still think it will be much much better with proper CAM.
July 13, 2017 at 11:51 am #38201
Ohh..my god. The finish pass. This is a game changer lol.
Just tried 5mm doc .5mm stepover! Did a big slot in 5 minutes. The results are nothing but amazing. The finish pass is incredible! It left a beautiful finish and left my tolerances better than anything I had ever seen on my machine. Final width is within .05mm across the whole pocket. It really works! Nominal 20mm, the pocket is about 19.96mm. I got to video this, nothing but amazed. I got a slight bit of chatter but, seriously 5mm doc, .5mm stepover, single flute, 10mm/s. It was really going at it but it managed to get the job done 🙂
Edit: my calipers have the ability to rack a little, so I don’t trust that tolerance. Still, no doubt I am in my .1mm goal.July 13, 2017 at 12:30 pm #38209
There we go I knew something was not right…July 13, 2017 at 12:34 pm #38210July 13, 2017 at 12:45 pm #38216
Pics or it didn’t happen
I was literally scrolling down to say this, when I ran into those pics.
That looks awesome.July 13, 2017 at 2:47 pm #38235
I’m still in the process of printing parts, so I can’t speak from experience, but I do have a question regarding the zip-ties. The ones in the photos above look like they are really tight around the posts, but I don’t think that would necessarily mean they were pulling the belt tight enough. If I compare those to the ones in the build instructions, the instructional photos always show one end tight to the post and the other with a small gap between the end of the belt and the post where the zip-tie has been tightened to almost form a rectangle between the post and belt. Could this be causing slop in the belt on Lionkev’s machine? (Attached image from assembly instructions)
Attachments:July 13, 2017 at 2:48 pm #38237
Wow, that is a pretty cut, very impressive.July 13, 2017 at 2:53 pm #38238
Mike it turns out is was a CAM issue not a ziptie thing.July 16, 2017 at 10:22 am #38463
Need to do a little better with my shooting, but you get the idea. Once I square up my tool the slight chatter should go away. On my finishing pass when the bit is sliding across the walls, it chatters, so clearly a rubbing issue not a lack of rigidity per say.
1 user thanked author for this post.July 16, 2017 at 10:33 am #38465
That looks great. Much faster than I was imagining.July 16, 2017 at 11:59 am #38471
That video makes me want to try cutting alu again, my MPCNC might be a bit to big though. Better build another one 😀July 16, 2017 at 12:11 pm #38473
Pretty nice video, I might need to tune up my machine a little bit. All that troubleshooting paid off!July 16, 2017 at 7:07 pm #38554
Dui, ni shuo de duiParticipant
Great success Kevin, congratulations!
I think I might try with a different bit, it might be the issue with my attempts. I thought the one I bought was for aluminum, but apparently it is for steel. And it turns out that milling aluminum with steel tooling is not good at all…
The results you got are pretty impressive, and speed seems to be quite decent,
Keep up the great work, would love to see a few actual parts coming out from your machine!July 21, 2017 at 4:35 am #39028
I never could imagine a machine like this could get almost mirror like surface finish in alu, guess the secret is in the cam software and applying your cuts wisely 🙂July 24, 2017 at 9:46 pm #39276
Welp…lets just say the success was short lived. Just tried doing some real cuts in some very thick aluminum plate, 3/8 inch. That is practically the length of the entire flute of my end mill I use. It went about as bad as you could imagine. 🙁 I lost 3 brave end mills today, 12 dollars just grew legs and walked out of my garage.
Okay, for real parts and not just pockets, the story changes. Just letting everyone know out there, if you plan to contour a shape, do NOT do it as a slot all the way around. You can get away with this if your part is rather thin, I would say 6mm or less in thickness. Otherwise you got what I just got, a mess. What happened was eventually the slot got so deep that the chips were pretty much stuck even with a vacuum. The bit started recutting and things got hot quick. Eventually it got so bad that the whole end mill got clogged with molten aluminum. Everything started chattering and I thought my the dw660 was gonna explode lol.
I also get this terrible chatter that actually broke 2 of my bits at VERY specific parts of my adaptive pockets like in the video I posted. It is so disappointing because everything sounds like its going really well with aggressive settings and then all of a sudden, this horrible chatter begins and the bit breaks. It’s very important to note that this happens when the cutter is NOT removing much material (for example, a light and slow finish).
That really bugs me, I am not sure what to do here. Dui is right in saying that the mpcnc for aluminum is sensitive. At this point I feel I am back to square one 🙁 Nothing seems to have gotten the results I wanted quite yet. The biggest problem for me is that chatter I said earlier. I will upload a video of it if I can’t solve it.
What I am going to do now is
1. move up my workpiece even further to the point where the bottom of the z conduit almost falls out.
2. Try to get some air blast going. Seriously. The chips are so nice but they just get recut cause I can’t get them out of the way.
4. Slot using trochoidal which in my opinion doesn’t make much sense but worth a shot https://www.youtube.com/watch?v=1X7G8pyiHac
3. Praise our lord Ryan Zellars
You must be logged in to reply to this topic.