Eyeball Engineering

[Duuuuuuuude]

Don't know what I did last night, but this morning when I fired up mach, everything seemed to be working right. After a few dry runs I bravely chucked up an engraving bit, took a big step back, and put the spoil to the spoil board.

[Robbob]

zip zip

YAY

[rob_b]

Beep beep...

I still enjoy watching a cnc do its magic! Now what are you going to make?

[Duuuuuuuude]

Don't know yet. I could either fart around or make headway towards actually producing something. Either way I need to edumacate myself more with cut2d.


I also carved a roadrunner into some plexi, and found out that if I turn the spindle speed up too high, it'll shut down and drag the bit around until it breaks. That wasn't cool.

[Duuuuuuuude]

Oh my freaking gourd. I fire it up again after lunch and I'm back to making tiny roadrunners again.

[Robbob]

Draw a square - Cam it in Lazy and try MeshCam trial ....... see if both write the correct gcode.

Lazy was wrong scale for me ... had the same exact problem. Touring car shock tower was the size of a 3mm nut. Used Meshcam and it was perfect. Cut2D nailed it.

[Duuuuuuuude]

Turns out I didn't dl lazycam. Still not sure why the roadrunner sporadically generates to scale.


Brought up some of the wizards and cut circles and pockets and they are pretty damn close to being right. The 20mm round pocket I cut came out to 19.88mm. A little tweaking should make that right I think.

The other thing I noticed that needs ironing out is that it won't repeat patterns accurately. Before the roadrunner shrunk again, I engraved it into some plexi. I ran the program again to clean up the boogers, and it was just a hair off on one axis. So I did it 4 more times, each time was just a bit further off than the last. Initially I figured that I should be able to repeat it enough times that at some point I wouldn't hear any contact whatsoever. Never happened though.

[Duuuuuuuude]

Got a part drawn up, converted to gcode, and took it for a spin on the spoil board, then went for broke and bolted down a sheet of aluminum. I've got some things to learn about feed and speed.

[KingGee]

Originally Posted by Duuuuuuuude

Got a part drawn up, converted to gcode, and took it for a spin on the spoil board, then went for broke and bolted down a sheet of aluminum. I've got some things to learn about feed and speed.

Recently been there. Found and downloaded a free trial of G WIZARD and learned a lot as well. Tooling manufacture specs are pretty close if they list router data. Router was ripping along and after the mess it actually worked out decent. 120ipm at 20000 rpm is crazy/scary to watch. A light air blast at the tool is definitely needed to keep everything happy. Working on a system to mist coolant while collecting it and not misting the entire router right now.

[rob_b]

I use this free S/F calculator...
FSWizard - Free Advanced CNC Speed and Feed Calculator

I also use the paid version on my phone...

[Duuuuuuuude]

So I've run 4 parts so far. Minor tweaks between each, mostly feeds and speeds, and a direction change too. On the last I swapped the 2 flute for a 1 flute upcut bit. Way better finish and less noise too. Top/right is #1, bottom/left is #4.

[Duuuuuuuude]

Originally Posted by KingGee

Recently been there. Found and downloaded a free trial of G WIZARD and learned a lot as well. Tooling manufacture specs are pretty close if they list router data. Router was ripping along and after the mess it actually worked out decent. 120ipm at 20000 rpm is crazy/scary to watch. A light air blast at the tool is definitely needed to keep everything happy. Working on a system to mist coolant while collecting it and not misting the entire router right now.

I was thinking about sitting my machine in a big, shallow tub with a pump and doing a simple flood setup.

[KingGee]

I thought about that too. Since my spoil board is mdf I'm gonna make a plexiglass tub to set parts in with a vice mounted in it and recess it into the mdf so i dont loose any of my z axsis. Congrats on finally getting it to work. Once you get the hang of it it's so addicting to just cut random stuff.

Thanks for the other calculator rob_b. My trials about to expire and I'm not paying what they want.

[Duuuuuuuude]

Progress has been good enough to do a small run of dewangers.




Running 9 ipm @ .0125" doc with a single flute endmill. I did some experimenting last night and gradually increased the feed rate up to 15 with no noticeable change in finish until I suddenly heard the router struggle and break a bit. Currently I can run about a dozen parts @ 9 ipm before the endmill is dulled enough to get noisy and chattery.

Today I'm ditching the mdf spoil board because it's soaking up whatever lube I throw down and changing the surface height as I cut (derp) and replacing it with a sheet of 1/4" pvc. Bought some 1/2" wide double sticky tape to hold everything down and its working awesome as long as I don't try to use wd40 as lube (derp #2).

Since it's going to be as hot as balls today, I'll be spending as much time as I can inside recreating all of my crudely hand drawn parts diagrams into actual cad drawings, and maybe stick a sheet of delrin down and see what I can get out of it.

[JohnRobHolmes]

What is your spindle speed? You will probably find a 3 or 4 flute to be longer lived if you keep it cool and lubed. Chips look good from here.



A flood setup wouldn't be my first choice, but if you don't have air supply then it might be the best for you. You ever get an air compressor in a location where it could run late and not bug the fam? If that isn't an option, the coolant I sent does great in flood between with about 12% mixture. Top off with 5% solution as the water evaporates. It would typically last 6 months in a machine running 80 hour weeks, so your limitation will be if it goes rancid or if your tub needs cleaning.



I wonder about your machine losing steps. Do you have a homing routine, or any rapids in the program? Always the same direction and axis that looses steps?

[Duuuuuuuude]

Exact spindle speed is unknown, I'm guessing its at 9000 or so. That was another thing I was tweaking last night.

I could get away with running a compressor at night in the summer months when school is out, but it'd be a no-go the rest of the year. I've contemplated building a box for it right outside the back door so that it's out of the garage completely. Right now I'll just have to be content with occasional chip evacuation via the shop vac and squirting parts with a spray bottle.

I've still got a little bit of the stuff you sent, as well as some coolmist. Don't really care for the coolmist at all, though I'm probably not using it like it was intended.

The ESS resolved the missing step/stalling issue completely. I'm 99% certain it was a crummy signal feed through the pci port. Didn't matter which direction, or which axis. The only way to make it not happen pre-ess was to go very, very slow.

[rob_b]

Glad you are finally making parts! You could build a diy fog buster with little effort. I can get you a drawing of mine if you are interested. From what I read on the internet (haha) a low pressure flood isn't all that great as it does little to actually clear chips. What you really want is to blast the shit out of them at high pressure and volume. Obviously, the later needs a full enclosure and waterproofing of the electronics.

This is a nice setup. Lol

https://instagram.com/p/2CeMoqC0yN/

[Duuuuuuuude]

Yeah, the more I think about it, a fog system would be best. Flood would be easier, but as you said, messier.

Drawing up a wheel fixture tonight. That should be interesting.

Got some delrin cut this afternoon with mixed results. Eventually I ended up running 30ipm @ .0125doc and had a good finish in all but two areas and I'm not sure why. Kinda thinking the backlash was letting it bounce around a bit. There was also an issue with getting tape/adhesive residue stuck to the endmill and making every cut thereafter crummy.

Two other issues that have come up: the "fan" thing on top of the motor is starting to slip. Its just a press fit on the motor shaft, but the vibrations have worn it out a bit and it sounds like a badly spun bearing...lol. The other is the hole in the frame for the x axis support bearing is sloppy and is letting the bearing vibrate around inside it. Not sure how to fix that one. Might have to see if I have any brass shim stock laying around that I can wedge in there...


First production part off the router...

[Duuuuuuuude]

It's looking like light and fast cuts are the name of the game with this machine. Ran some prototypes for a new part tonight and settled into 18ipm @ .008doc through 6061. Surface finish isn't super awesome, but the chips are good, vibration is down, and there is very little heat generated at the part. A few hours in the tumbler will clean up a good deal of the surface imperfections anyway, so no worries there.

I glopped some ca glue on the fan to lock it onto the shaft, and it worked for a while, but eventually rattled loose again. Thinking about ditching it and using a pc fan instead, wired in parallel with the spindle motor so I don't need to run another set of wires up to it.

Some quality carbide single and double flute endmills will be in my mailbox tomorrow. Gigitty.

[BigSki]

Congrats on keeping your cool and seeing this thing to the end.