Except for the fact that he's already pointed out that he does NOT have a mill to CNC.
Gerry
UCCNC 2017 Screenset
http://www.thecncwoodworker.com/2017.html
Mach3 2010 Screenset
http://www.thecncwoodworker.com/2010.html
JointCAM - CNC Dovetails & Box Joints
http://www.g-forcecnc.com/jointcam.html
(Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)
I should probably stop skimming threads so much...
-Andy B.
http://www.birkonium.com CNC for Luthiers and Industry http://banduramaker.blogspot.com
Just a few words of wisdom.....
Don't buy anything until you have a design to work from, not just in your head. After you think you have it finalized, wait a week or so because you are bound to think of something that you want to change. When you think you have it nailed down, share the design here. It may save you money and headaches. If you are too much in a hurry to do this, don't build a CNC, buy one because you won't be happy with what you end up with. Haste makes waste.
oh vey, thanks for all the wise words of wisdom guys. im going to put this on hold and look into a cnc mill option. maybe ill get lucky and find one locally for cheap! time to browse those auctions....
Ok so for less I can get a manual mill with power feeds and a DRO. It will save time then how im doing it now, but i dont see myself getting those nice round contours on the inside of the aluminum spacer. (im terrible at etch n sketch)
what bugs me is I really cant grasp WHY "wood" cncs are so much different then "Metal mills"???? I understand rigidity seems to be a huge factor. but lets say it was possible to achieve a very rigid machine and rule this out.
next issue seems to be the router/motor because they spin too fast and dont have enough torque. So IF that is the case (correct me if im wrong) cant I just replaced the router with something that spins at a slower speed and has more torque? I am almost positive i came across some guys build (4'x8'+) and he mounted a full motor on the gantry and looked like he configured some contraption to hold the bit... now that combined with an end mill bit, they are still doing the same basic functions... xyz axis... So is the real issue the router IF the machine is deemed rigid enough?
EDIT: did more searching. would a cnc spindle motor be more effective? (assuming i can find one with lower RPMS and more tq/hp) or maybe a electric motor running pulleys to a chuck/spindle assembly like drill presses? Also do you guys think the Super PID would be beneficial? seems like it only turns routers down to 5000... mine only goes to 8000.. is it worth it to slow the router down ~3000 rpm?
PLEASE keep in mind i do NOT i need insanely low tolerances, I am NOT a machinist. nor will i be making things with such tight tolerances. it is NOT going to be a work horse like the "production" like i feel you guys are thinking of.
Im sorry for the emphasis and redundancy. I just feel like my idea of a "good finish and production" is completelyyyyy different then most people here... I'm sorry if i misused those terms.
This is a picture shown on solsylva's site that was done by one of the 10"x9" machines. which really isnt all that rigid. this finish is completely acceptable for my needs, so I'm having a hard time believing i cannot achieve what i want to do with a steel constructed "wood" style cnc table....
(I don't mean this in an angry or argumentive tone if thats the way its coming off)
Would love to hear more of your counter arguments and rebuttals for these!
If you grab the collet on a decent sized mill, and push and pull as hard as you can, it might move a few thousands of an inch. Do it on the Solsylva and it might move 1/8".Not only do you not want any movement here, you don't even want any vibration.what bugs me is I really cant grasp WHY "wood" cncs are so much different then "Metal mills"???? I understand rigidity seems to be a huge factor. but lets say it was possible to achieve a very rigid machine and rule this out.
Cutting forces required to cut wood with a router are probably less than 10% of the force required to make a similar cut in aluminum. Mills will have spindles with much larger bearings to handle the much higher loads placed on them.
You mentioned using 1/2" steel tubing? I think most would recommend at least 2"x2". Again, going back to rigidity.
As for that pic, I wouldn't expect to see a finish that good. I've seen a lot of pictures here of aluminum cut on routers, and the quality of that cut would probably in the top 5% of all the work I've seen. Anything I've seen compaable to that was done on a much more rigid machine. Also, be aware that with all things being equal, a 10"x9" machine will be far more rigid than a 25x25 machine.
Gerry
UCCNC 2017 Screenset
http://www.thecncwoodworker.com/2017.html
Mach3 2010 Screenset
http://www.thecncwoodworker.com/2010.html
JointCAM - CNC Dovetails & Box Joints
http://www.g-forcecnc.com/jointcam.html
(Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)
Brian Oltrogge's Platform CNC cutting aluminum with a high degree of finish.
If you examine Brian's machine you can see how the structure differs in practice from the Solsylva which might give you a direction for your own needs. Obviously the principles across various cnc machines are the same but the execution is what ultimately changes the performance characteristics. It's clearly possible to produce your product with a cnc router.
Chris
Get a mill. There is a reason most people are not using one or are recommending you go a different direction.
Do you really want to spend your time and effort only to end up with a part that is just average or the tolerances are loose? Your customers won't. Even if they are racers. G0704 for about a grand and with your machining skills, you could build Hoss kit in no time. Use it manually til then. You will enjoy the results much more.
A lazy man does it twice.
It's pretty much all rigidity. High spindle speeds are ok if your mill has the feed speed and cutting force necessary to maintain a good chip load. You also need the rigidity at that point too. If you check on youtube, you can find plenty of vids of Al being cut at 18k rpm or whatever but, they all have flood coolant, are moving pretty darn fast, and are very rigid machines.
To take an aggressive cut at those types of speeds takes a very rigid machine.
The advice you're getting here is based on many years of collective experience. I've only been doing this a couple years but have yet to hear someone say "my parts are finished too quickly". It may seem like not a big deal to have to wait four hours for a part rather than one but that though will change when the bills start rolling in.
To summarize, if Al is what you want to mill, build, buy, or convert a mill for Aluminum. Don't build a mill for wood.
-Andy B.
http://www.birkonium.com CNC for Luthiers and Industry http://banduramaker.blogspot.com
There are a lot of people with wood routers and mills on the forum here. Perhaps you can find someone within 100 - 200 miles and drive over to see what they are able to do with their system.
I will tell you, it was a real eye opening experience for me to go from my normal wood working experiences to visiting my brother's garage with a bridgeport clone mill and do some work on Al. The forces involved and dramatically different cutting speeds were pretty astounding.
If you watch people like Tryally (Luiz Ally) on YouTube you will buy a Sherline. There are some people who can get great results with those machines. In no way am i knocking them or their talents and knowledge. Either way it is going to cost you more than you thought. Being the self professed tool junkie you claim to be, you will far exceed your budget every time you see Hoss, Simpons36 or countless others doing fun and exciting things with their machines. The more bases you cover in the beginning, the less time you will spend redoing it later. If you are in the Houston area, I could empty my garage a bit!
Hit it on the head there. Can you mill aluminum on a DIY wood cnc? Sure. I'm planning on doing a few pieces myself. Would I do it that way if I had a mill? No. I've seen used mills on CL go for $1000, and that included the DRO. Add some steppers and away you go. The only thing that stopped me was the $ to move it and no way to get it into the basement workshop. Much cheaper and easier to convert a used manual mill to CNC than it is to build a wood CNC from scratch that would only do the job @ 10% capacity, if that.
Well, in a sense, it is a lot easier to build a 10 x 9 machine more ridgidly than a 25 x 25, simply because there is less unsupported areas.
As to metal mills versus wood routers, you only have to look at a few to see the differences. Most metal mills have either dovetail or box ways; most CNC wood routers use linear bearings of some kind. While many higher end CNC machining centers use linear bearings, the rail profiles are much larger and most likely have a higher preload rating than on a wood router. Most CNC metal mills are made of cast iron, for its mass, ridgidity, and vibration damping. Most wood routers (with some exceptions) are made of steel tubing and are rated for light, occasional metal routing.
While you could take a mill spindle and mount it on a router gantry, the assemblies are usually pretty heavy, and you'll have to design around that. It might be better, if you go a gantry route, to make the gantry fixed, which would allow you to use heavier materials.
As to the SuperPID, I use it for cutting aluminum, but even then I keep the router at 9000-13500rpms. I don't know if you really want to have a SuperPID pumping juice into your router so it can keep 5000rpm while milling metals. With routers you really want to spin them faster and use tooling designed to work with that. Maybe one way would be to get a Sherline spindle, and drive it with a router with a SuperPID with a 3:1 or 4:1 belt ratio. Even then, with the preload adjusted, I don't know how much the Sherline spindle will like spinning continuous at 7500-10000rpm.
As to using steel tubing: It's not a trivial thing to bolt linear rails onto them. Especially with preloaded profile linear rails; if they're not mounted to a milled or ground surface, they can bind and cause premature wear.
One way around the limitations of CNC routers is to employ software that will allow constant tool engagement. It will allow to to cut deeper and faster than a router normally can. The problem is that this software can cost more than than even some entry CNC mills. You do get very nice surface finishes however. I don't see many DIY routers out there that can cut 1/8"doc at 72ipm... but that's the software. I had updated the software since this video, and the nmachine does not make the full retracts between roughing levels anymore like it does here:
[ame=http://www.youtube.com/watch?v=JVpKLBXvtSs&feature=plcp]DIY CNC Router, cutting BMX chain sprocket, Part 1: Pocketing - YouTube[/ame]
I've looked into the mill option. money isnt the issue for purchasing one since ive found used bridgeports for around 1300 (which is less then building the cnc table) but they are manual (with DRO)
Now the next problem is, i was doing some searching last night and a lot of comments i read discouraged turning a manual mill into a cnc.... so that shied me away.
The G0704 is a nice option but it seems a bit small. the G0705 would be more fitting but it already is almost at its max capacity on my first part (my part is 7"x9") and im not sure it the hoss kits will work with this one..
See this is what i dont get, everyone says not to use a wood cnc because it has xyz problems. but if those are addressed why cant it be done?
rigidity- add extra bracing and make it out of stronger material
speed of router- super PID/variable routers (unless those are still too fast? but i doubt it since the part ive shown was done by a variable speed router.....)
cooling/lubrication- add a coolant pump
router spindle flex- upgrade to a sturdier spindle motor
this is what im having a hard time grasping. all the "flaws and problems" seem like they can be address, just no one is addressing them... Now if you guys were saying it just isnt cost efficient, then i can completely understand that... but to say its not capable makes me wonder after seeing videos like Brian Oltrogges =/
I definitely havent ruled a milling machine out yet..
^^ cool stuff! checked out a video but right off the bat, that machine's envelope is too small for me.. The part i want to make is 7"x9"
And now the more i think of it. as much as i dont want to admit it. the bridgeport might be ruled out due to transportation. unloading it.... looking into the bench top/smaller mills.
Just because the part is 7" x 9", it could still be done on any machine mentioned if you know how to reposition the part. But no reason to start working around to begin with. Regarding tool paths and the like. All of the software I have used allows controlled entry of a few different techniques, ramping, plunging, helical, tangential etc. After a little use of G Code you should be able to recognize the retraction of the Z and edit it out.
Regarding moving, you should be creative enough. The Bridgeport can be disassembled and broken down into manageable parts. Some beer, patience and a few strong friends and you are there. You wouldnt regret it. Pieces of pipe/round bar can move the world. I cant imagine why anyone would say converting a manual mill is not good unless they are spoiled by working around $200,000.00 VMC's or the like. If the screws didnt have enough resistance, gear the steppers down. Sacrifice a bit of rapid speed for holding torque. The final result will always cut wood and do metals wonderfully. The same wont be said for making the little machine do big work. My first machine was a Smithy 3n1 (that is another story). I moved that machine by myself with the pipes and an engine hoist. Didnt disasseble a thing. The bridgeport wouldnt be that easy but?
It depends. For profile cuts I'm limited just like everyone else. But for pocketing I do have an advantage. I do intend to do a thread on just this topic in the near future when I have some free time. Basically with traditional poceting toolpaths the machine will have to slow down going into corners. While this obviously slows down cutting, this is not the best thing for the tool. Complicating matters is that the tool plows into corners, and the tool engagement doubles. If you're climb milling, the tool will pull into the corner and you're left with a witness mark in the corner. With the toolpaths I use, the bit never plows into a corner; rather it will arc in and nibble at it at close to max feedrate. (Actually I could have decreased the machine time on this program another 25-30% since I had full retract on wach pass, and didn't turn off helical entry on the finish passes in the pockets. I also didn't have the max feed set between cuts. When set right it almost looks like a "retrograde" movement on a watch.) The radial depth of cut will also be reduced slightly with each pass into the corners to ensure tool engagement remains constant. Also, these movements are all G2/G3 arc commands which produce smoother movement and faster feedrates. But basically, the key with these toolpaths are keeping a constant load on the tool, which allows you to run at as close to constant feedrate as possible. In fact with commercial milling centers, the only "speed limit" to these toolpaths are detemined by the max feedrate of the machine, since the spindles are very powerful and the machines very ridgid and high powered servos are used. But the benefit for lighter-duty machines is the control of tool engagement.
The big rub is the cost. I was fortunate to "buy in" with a used seat, and upgraded. I do generate enough income with my machine where I could justify the investment (which was probably over two times what your machine budget is, and more if I bought the software outright.)
-Andy B.
http://www.birkonium.com CNC for Luthiers and Industry http://banduramaker.blogspot.com