I have a brushless motor spindle I made for my Phlatprinter and it works really well. Yours looks nice.

 

Hey I already started this build and I'm looking for an 80A ESC but can't find one... Are they really that hard to make? I have a bunch of high power transistors, maybe a 555 timer, bunches of high-power diodes, some copper boards, etc.

 

OK, it's a total suicide mission to make a controller. I found some 80A ESC with a bunch of bells and whistles for $17 and I paid $40 for the motor so yeah $57 sounds about right and maybe $69 with shipping. I'll just buy one for airplanes that matches my motor and then I can reuse the motor+controller for R/C airplane fun. Here's the ESC http://www.ebay.com/itm/80A-Brushle...Control_Parts_Accessories&hash=item3a5cbdbcc2

 

Also, about matching the motor, controller, and supply voltage: I noticed someone said they run theirs at 12V with no problems... Well the ESC operates at 80A continuous so 12V * 80A = 890W. Since the motor is 890kV it's operating within normal parameters and should give you 10,680RPM. I also noted the collet and shaft aren't rated beyond 10,000RPM so that's the theoretically safe limit.

 

Well if all goes according to plan, I could take some trash and cnc it into an airplane, then pull the motor off my cnc and use it as a dremel to touch up the parts, and then stick the motor in the plane and fly it. Oh and while I'm flying my plane I can put the 3D printer head on the cnc machine and let it print some plastic helicopter parts. Then when the plane battery dies stick the motor back on the cnc and make a few PCBs and helicopter parts. etc etc Oh and I promised my kid we'd make toys and robot tanks and all that...

 

Sure I am not the only one wondering where you obtained the mini shaft and chuck.

 

The shaft is on ebay from CTC tools and also on their website. There's ER11 and ER16, whatever that means.

 

Saw the link on a different page just before you sent the info, thanks anyway!

 

Excellent write up on how you built that!

BTW 890KV does not equal 897Watts. The KV rating actualyl stands for RPM per Volt of power. Most of them are 200-400watts of maximum power. IE 890KV means the motor will spin at 890RPM x 12v = 10680 RPM. (source)

Also what are you using to power the spindle? Because a 12v 80amp PSU is super spendy.

The outer casing of the motor is press fit onto the motor shaft and is actually what spins on an outrunner motor.

 

890kV=960W@100% duty cycle on 12V and 80A. You can get a 1000W power supply for $35... they are used in HP servers. I may use 2x480W($15/ea) to power mine.<Tim Taylor grunt>

 

The dude yes you are right about 12v x 80a = 960w. But you are wrong in trying to tell people that the KV rating is a wattage rating, look at the link I gave, that is the TRUE information about what KV stands for, and it can help someone pick a better motor for a spindle, or a different one if they can't get the one used in this write up.

 

I think there are a lot of misconceptions about brushless DC motors as cnc spindles. Here's how I see it:
The motors are going to use more current when there is more load. The fact that you have only 20A going to an 80A rated motor means that it will seem weak, or have no torque, when loaded. This will cause you to have to slow down the rate of cutting and could cause the motor to stop when overloaded.

There is almost no friction in these motors. They have roughly 90% efficiency. This means a 1000W motor will lose 100W to heat, so you need to get rid of 100W of heat. This is easily accomplished with a PC case fan or vacuum cleaner.

I sat down and looked at what these motors can do(based on manufacturers specs). It turns out that 960W=1.287HP. It's not as much as the 2-3HP motors used on most shop cnc mills, but it's WAY more powerful than anything I've seen for $70. I agree you wouldn't be able to run as fast as those larger motors. I think I heard people say their 3HP machines can run at 230 inches-per-minute(ipm) with slower speeds for hard wood, aluminum, etc. I wouldn't be surprised if you could run this motor at a little less than half that speed, say maybe 100ipm, assuming the power supply had 960W and the frame and stepper motors could handle it.

 

Yeah that's true. For CNC, a lower kV motor means slower speeds at max torque... This becomes important when you consider that you have to remove the heat from the material while cutting, other wise your plastic melts, wood burns, bits can weaken, etc. In steel milling, you see very low kV because:
a.) they are using fluid to cool the parts while cutting and it can't keep up with the heat transfer
b.) the torque must be high to cut material of that strength which increases friction(e.g. heat)

For a desktop CNC like the routy, I would exclude harder materials like steel. Someone already mentioned that going too high on the RPMs will burn the piece being cut(maybe 20,000RPM). I've also noticed that most collets and bits are rated to 10,000RPM with the most expensive ones going up to 18,000RPM. Anyway, I'm going with 10,000RPM for my design criteria because people are having success at that speed and it's what my parts are rated for. I can just adjust the ipm for each material and let it run day and night until the bearings give out.

 

This looks like a more plausible definition of Kv:http://en.wikipedia.org/wiki/Motor_constant

Notice that it is correctly written Kv - not KV or kV. The symbol K is commonly used to mean constant in science and engineering, and the subscript v indicates that this particular constant is a velocity constant.

So... Kv is the Motor Velocity Constant.

 

Robert, I really love that little spindle, you should make and sell the aluminum parts, for us to assembly our own motors

 

The rounded corners are not a result of the spindle speed. Either your steppers are missing steps or your machine is flexing. My guess is your gantry is twisting. You could try lowering the acceleration of the steppers to compensate. Run a few V patterns across the length of the gantry to see if there is deflection in the middle, make adjustments, run again. I see many people using an L bracket to hold the spindle and those will flex. I also see people use long Z-travel arms which can flex.

As for the burning wood, that IS due to the spindle speed. A larger diameter bit has a higher tangential velocity at the outermost point than a smaller bit. This means the bit is hitting the wood fast enough to create so much friction that the wood heats to the point of burning. Obviously the best thing to do is just reduce the speed. Other options are to cool the surface while cutting with an air blower or water shooter. Not so practical but possible.

In my post I mentioned 100ipm as a possible speed and larger motors claim 200 ipm. While these speeds ARE possible, they do burn wood and melt plastic if there is no cooling. I've created smoke just using my handheld battery powered drill on wood when I hit a knot.

It's good that you mentioned this here. Thank you for the input. I think it may be possible for us, as a community, to come up with some practical limitations of CNC machining and apply those limits to our designs and avoid over-spending on items that may not provide practical implementations. Sure, I can go buy a 120,000RPM motor but it's a waste of money because it will just melt everything and fragment my bits and send metal parts flying around the room.

 

I have been looking for DIY spindles for a while. This looks like a great build.
Questions:
1. What do you think the max. Depth of cut could be? Going through 3/8" AL is pretty impressive.
2. How did you balance the spindle seeing as the theoretical RPM is close to 11,000?

BTW, ER11 and ER16 is the size of the collet.

I am excited to get started on this build! Thank you.

 

My understanding is that there is some imbalance in these cheap motors. Another build on a different site used some kind of putty or glue as a weight and had like 2 accelerometers to find the imbalance. It's probably beyond my capabilities but it would probably extend the life of the bearings by 10% if balanced really good. Maybe not, I dunno. The fact that he used cheap bearings like 608 means the cost of replacing them won't be an issue so imbalance isn't a big deal unless it is causing vibration so bad that it affects the accuracy.

It will be difficult to find a router bit that is longer than 1" in the diameter sizes of this collet. Maybe in the 1/2" collets you can get longer bits. The max depth will be determined by your bits and then ultimately your z axis. A CNC machine is just like a drill. It can drill 6" if you have a bit that long. If you use a tiny drill bit it will probably snap easily if you drill in 6" and then try cutting sideways. I guess that's called the shear strength of the bit? As long as you don't exceed the shear strength you're good. Cut slower with longer bits and with thinner bits.

 

rotozip bits have more then 1 inch of side cutting area. But the spindle that Robert used will take 1/4inch bits, and those are available in many sizes including the bets of the betst. 65-026 Onsrud Single Flute 1/4inch SolidCarbide with 1.25inch cut depth. I would never try a cut more then probably 1/2in in depth on these style machines, I doubt they are strong enough to avoid chatter at that DOC.

 

What do those people use to cut 2" foam? I know some use a hot wire but I think I saw a 2.5" bit with 0.25" shank.

 

Check out this Calculator for what bits can do:
http://zero-divide.net/index.php?page=fswizard&shell_id=199
Also you can calculate the torque at any point on the bit Torque = Force x Length. The force comes from you stepper motors and the Length comes from your bit length between the collet and the material.

I found a pre-alpha program called the machinists math kit
http://sourceforge.net/projects/machmathkit/?source=navbar

 

ooooooh pretty. Can't wait for the collet chuck to arrive.

 

http://www.toolstoday.com/p-5202-st...er-bits-12-shank-2-flute-high-production.aspx

Part way down the page you will find 1/2 inch 2 and 3 inch long bits.

There you go, should be great in foam. Not in one of these machines though. I just don't think you will see 2+ inch DOC in 1/4in shaft bits.