Those circles are intention in the cad, little dogbones I added.
The striping is smooth to the touch actually, I’m not sure why it’s so much darker. But definitely I’ll order up some single flute upcuts, the one I’m using right now is a compression bit but I want to start with what’s going to get me a good working machine first, so I’ll grab the single flutes.
When I get back home, I’ll try to run things a bit slower and set my bit size back to what it actually measures and see if I can get some different results. Do you have a recommended doc/feed/speed for 3/4 mdf with a single flute upcut? Basically a good known good starting point for this job. I’ve got too many unproven variables I think I’m fighting through
Compression bits only work at full depth of cut, you can not step them down…and they significantly increase the load as they bring all the material to the center instead of out of the cut. Those work best when you are not slotting, basically as a finishing bit not a hogging out bit. They are made for material that chips, not MDF.
Single flute upcut is the one you want there.
Make sure you get a bit that can handle 3/4" I am not sure any of mine do.
Start at the milling basics settings, from there you can increase. This all depends on what tolerances you are trying to keep. If you really want to keep all your MDF cuts under 0.1mm (4 thousandths of an inch) tolerances you will need to slowly increase your MRR with each test cut. Your finishing pass with be critical, might even want to do two of them.
Sorry, I can’t figure out multi quoting in here lol.
There are two, I think the lighting/shadows make it only look like one dogbone. I’m not home right now or I’d grab a top down.
Hmm yeah they are definitely very visible.
Ah good to know with the compression! I never got an explanation about that, I bet that is the culprit for the visible lines.
Wow I’m not sure how I missed that page, I must not be very observant. I’ll start there and work my way up. I don’t necessarily need it tighter than 0.1mm more or less just trying to get an understanding of how to tighten things up without the cheating I’m doing by oversized the bit in cam. But I’ll start with your milling basics page and see what happens there, should clear up any issues I’m having or confirm there’s something in CAM I don’t have right. I’m realizing switching to another CAM will be difficult as they don’t all post process for klipper.
Thanks so much for your help, fingers crossed I can get it working!
I concur with the other advice given, and want to mention that you really should not have to misrepresent your bit diameter to get the desired result. Going forward, you want to always be able to list your bit diameter as it actually is, and have the CAM to work right.
Estlcam offers a free trial. From their site:
“…you can test Estlcam for free. The free version has full functionality… Just download , install and try…”
If you don’t have the same CAM issue there, it would seem to indicate the CAM setup in Fusion is somehow flawed.
I downloaded it to give it a shot but haven’t done anything yet as I know I’ll be fighting gcode given I’m running Klipper. Might have to do some manual manipulation to give it a test if it comes down to it though. I’m going to work through Ryan’s suggestions first and see how it goes! Agreed, misrepresenting is plain wrong
While the cutting passes should be at a depth of cut suited to the bit and the chip load, and presumably could often be less than the thickness of the material, ideally the finishing pass should be at the full depth of the material in a final single pass.
In other words, for a finishing pass, the depth of cut is the full depth of the thickness of the material.
The only difference would be what percentage of material is still left to be cut off, that is being taken by the finishing pass, in a very light amount of shaving the material.
In reality, there is slop in the collet and the cutter is never perfect to technically an outside dimension is typically a tiny bitt too small and an inside diameter is a tiny bit too large. If that is ever backwards something is really wrong. If that is exaggerated, you are pushing too hard or some other parameter is a little off.
This is tiny, but real. I am not trying to say you can’t get it perfect just if you have an error that is the way it should be.
So funny you guys mention this, the biggest change was when I switched from climb cutting to conventional. In Climb Cutting my perimeter was too large and inside was too small. When I switched to conventional is when things went the other way for me. Maybe with slower speeds and what @BingBadaBoom you’re saying, climb cut for rough passes and conventional for finishing, that might clean things up.
My 1/8 single flute upcut arrives tomorrow, so I think I’ll slow things down, switch back to climb cutting and use the speeds and feeds @vicious1 you have in your basics page. Set a baseline and go from there. When it comes to 3/4 MDF, what is a good number to pursue in the future for a well oiled machine? Would it be more realistic to start upping the DOC until things get dicey or feed? Also, the basics page doesn’t mention spindle speed, but I think I see most people hovering around the 2 setting for the Kobalt and Makita routers for about 12000 rpm I think I read?
EDIT: I should learn to read better, the basics does mention increasing the DOC instead of speed as i progress.
As far as slop… are you sure that your core is appropriately tensioned to the rail?
Two bolts on the bottom can be tightened to squeeze the bottom rail bearings together and set the core tension on the rails. You should not be able to rock or rattle the core on the beam. (You also really don’t want that too tight either. All skate bearings touching the rail is the key. It takes tiny adjustments to go from loose to too tight. Take your time.)
If your core can move at all, your dimensions are all suspect at that point.
I’ll double check that this morning. I tried shaking things the other day and it seemed sturdy (after I remember to turn the motors back on) but it was only a quick sanity check… I’ll really put eyes on it this time, thanks for the suggestion!
My bits should be here in the next hour or so, so I plan on putting some work in today!
I did a lot of cutting little box joints today and what I’ve found was:
The core to the gantry is rock solid, no wiggle there. The only amount of play was in the direction of the belts (is that backlash?) but it wasn’t very much and I had to put a good amount of force into it.
Right off of the basics page, everything was flawless, just really slow. I found leaving a stepover for the finishing pass made things so much worse, maybe I’ll try it again with conventional.
DOC I shouldn’t be pushing more than 50% of the bit diameter, that is when “slotting” which is what I think I’m doing by cutting pieces out like I am. Along with that, WOC I’m learning plays a role into things
As long as my DOC stays at 50% of the diameter, I’m having no problem upping my speeds back up to 25mm/s as long as I bump the spindle speed up as well. I found that I can hear when it want’s more speed and as long as I don’t ever hear a chattery change it tone when it’s making it’s cut, it’s going to cut well and exactly at it’s mark.
I’m within a tenth of a millimeter now which is pretty cool, this has all been with a single flute upcut (which wanted faster spindle speeds as well I found which makes sense).
I feel like a lot of this is against what I keep hearing people say to do, but if it works it works? I’m thinking I’m going to try my hand at another torsion box after a few more cuts to double confirm things are where they need to be. My Harbor Freight dust collector arrives Wednesday so hoping to have my new table setup and ready for it.
Appreciate any and all of you who are actually reading all these posts of me rambling on about my problems
When I say faster spindle speeds, I think I took my router to setting 4 today when I was moving 1500mm/m. I have no way of being certain of the actual rpm but it sounded like it was enjoying the cut. When I was running at the slower speeds suggested in the basics post you have (480mm/m if I recall), router setting 2 seemed the most comfortable.
Climb cutting just isn’t happening. Even with finishing passes it just never made it back to it’s dimension, even with a 480mm/m finishing pass, I guess that’s how much deflection is going on?
With conventional, things stayed really accurate. I ran my roughing passes .3175mm outside my mark followed by a finishing pass and everything stayed within a 10th of a millimeter. So being able to rough at 1500mm/m and 3.5mm doc and then do a single finishing pass at 480mm/m I had spot on dimensions.
I’m not sure if I call this a sweet spot for me right now cutting out mdf or practice some more, play with some other settings first. Not sure what else to play with based on my findings.
Well I’m cutting a new table! Feel good about what I’ve seen so far. I also got a new toy recently, the larger Harbor Freight dust collector and man what a difference. So much quieter and so much air flow than the shopvac. Now the poor shopvac can go back to being a shopvac.
The dust collector is setup for dual 4" hoses and even though I stepped it down to 2.25" shopvac hose, and then again to a 2" hose at the router (don’t yell at me guys) it’s still doing better than the shopvac was.
That being said, is going full 4" to the boot the most realistic answer? Is there a more realistic answer? Would like to do the reasonable thing here.
Maybe should have done this as a DM, so my apologies up front.
I don’t have an opinion on your question, I mean I could, it would be worth 2 cents.
But with your new toy, probably to early to tell, but…
Have you seen the need or thought about doing the impeller upgrade?
Do you have any thoughts about losing the bag, and going with a MERV style filter?
Have you measured the db level?
Sorry for being so inquisitive, but I’ve been in conflict for a bit on what I want to do for a better DC system. Correctly have a dust deputy, and love its performance with the shop vac, could stand to do without the screaming shop vac. Currently working on a slow build LR3, and anticipating longer run times, figured the shop vac will be updated to something better quickly.
I have the same big HF dust collector, and I have not done an impeller upgrade (don’t know much about it), but I did swap out the air bag filter for a nice tall pleated-filter canister (aka the MERV style filter). I also used my LowRider (back then it was a v2) to create an MDF + buckles adapter to connect a regular large trash can (with regular bags) to the bottom. I pre-cut some “view slots” in the side of the can so I visually see how much saw dust / chips are in the bag, so I know when to change it.
