Scratch Build – Complete Posthuman

I'll add that the chips should be treated like a hazardous material. Tiny chips are wickedly sharp, and can get into your skin.

 

Had to dig quite deep to find one (not at the shop currently), but here's an old one. A lot has changed since then, but the milling head is still the same:


As for the material, it is meant for hobby use and just about the only thing I can get off the shelf around here without having to deal with minimum order quantities and quotes...
An adaptive clearing strategy seemed to work quite well, as soon as the cutter was properly in there and not trying to ramp or bore.
I'll definitely keep your advice in mind next time I shop for material though.

 

Hmm... I have three observations/suggestions:

1. Never mill in a drill chuck. Ever. They just can't hold the forces. Especially a big 12mm endmill like in the picture.
Plus they have usually a lousy run-out, causing the endmill to rattle.
Apart from the non existing precision, the rattle can cause the chuck to open and then you have a loose rotating endmill flying in your direction...
Or even the entile chuck comes of the MK holder and flies in your direction
Get yourself an MK2 collet holder and a few collets. At AliExpress just a few bucks. Safety first mate


2. Have you checked the brushes?


3. If you have/want to replace it, I have a spare motor for free if you are interested.
I have a Optimum BF20, which is quite similar. But it has only 850W and is rated only for 3000 rpm.
So a bit weaker than what you have today.
It has maybe 20 hours milling time under the belt + 10 years in the cabinet under the mill...
It was the first thing I exchanged. The entire electronics are just crap on these machines...



But the best thing, if you have to replace it anyway, is to convert it to a 3-phase motor.
I can give you a few tips if you want to go that route...

 

Quite a nice setup you have there, I was actually looking at that exact mill as well when I got my BF30 3 years ago.

1. I kinda expected a comment about that This picture was when I first got it set up (almost 2 years ago) and the ER tool holders hadn't come in yet, I never milled much with it. Surprisingly the runout is actually really good, at least better than the Chinese ER holders.

2. Kinda, I opened the motor up and had a look, but didn't see anything. I also don't know what I'm looking for though, so I don't know what I was expecting. Just remembered that I was rolling on the camera when I did, so here is a screengrab, maybe it tells you something:


3. Thank you, that's a very nice offer, though seeing that I am wishing for more rpm as is, going from a 4600rpm motor to 3000rpm might not be ideal. If I'll have to replace it, I was considering either modifying the head to go from a gearbox (I never use the low gear) to belt-driven, giving me the option for more rpm. I'd also have to replace spindle bearings at that point, as I doubt they would take a liking to increased rpm, so it's quite involved. My other consideration was to completely ditch the stock head and go with a 3.5kw water cooled spindle, possibly even with atc. That would also be quite expensive and I might miss being able to go below 5k rpm and still have torque.

 

Ahh...good to hear that you use the collets.
Cause even when the no load run-out of the chuck is good (and that is what you measure) the business end of an endmill in a chuck is so far away from the spindle bearing, that there is a considerable bending at high feed rates.
Just pull it by hand and you will notice a flex.

Difficult to say something, indeed. Just the wire of the bottom brush in the pic looks kinda burnt and somehow smaller in gauge in the middle, where it is dark grey instead of copperish
But then it is hard to tell and the discoloration can be just graphit dust from the brush...


Ditching the head sounds like a nice little project of its own.
But are you sure you need 3,5 kW?

 

Could you use a servo motor on the gearbox?

 

The slight burning might be from when I accidentally turned it off a bit too suddenly and something shorted, it blew the breakers, but everything was fine after. Not sure though.
As for the 3.5kw, not sure, I'd probably be fine with less, however, the price difference was not too big on the quote I got. I'll have to decide on the exact power if I actually decide to go for it.

I don't see why not. I'd have to go ahead with a full bearing replacement to be able to leverage higher speeds.
What are the main benefits making a servo worth the extra money? Rigid tapping isn't really a big priority for me.

 

I was thinking you might keep your low speeds and not have to do much rebuild, but I'm half running it by Dan. I'm not sure what would happen to your gearbox in that setup.
...Imagine getting the gearing running beyond it's rated load. --100,000 RPM or something mad like that.

 

Back in the workshop. I decided to ignore the spindle for now and continue using it, just watching out not to drive it too hard. I even managed to get some descent machining done on the keyboard plate. The adaptive toolpath for the 4mm endmill worked great and I had no issues. Rest machining with a 2mm endmill also worked decently, but the machining goods did require two sacrifices. Good thing the extra endmills I ordered just came in. Finally the contour was a complete disaster. I went back to a 4mm endmill and tried to take just a light pass at a descent feedrate, but no dice. Two more sacrifices and something moved, so when I finally got once around it was offset a bit. Decided to then just use a jigsaw to cut it out instead. Slotting in this material definitely is no fun. It just gums up the endmill and snap.







...and I just got started with machining

Remember how I said I just ignored the motor? Well not quite. I kept checking the power draw and also the temperature. While the surface temp probably doesn't mean much, after half an hour of use, it was up to 95C, which I thought is a bit much. So it let it cool a bit and added a fan. The fan helped a lot with surface temps, but I doubt it did much inside. Just kinda firing my IR thermometer into the vents the highest reading I got was 125C, toasty!
After two hours of cycle time I did notice the motor was sounding and feeling a bit worse, and there was a suspicious amount of black dust everywhere around it. So I took it apart again and this time even I was able to quickly see that something was not right. Remember the brushes, how they looked fine just a few days ago? Well now one of them is less than half the size, while the other one completely fell apart. The front bearing also didn't feel too good, there was some noticeable friction in one spot. So I decided to take it apart further. That was easier said then done. I don't have a gear puller, so screwdrivers and a hammer were my only option. Since I presumed the bearing dead anyhow, I didn't hesitate to hit it with a hammer. The gear was a much tougher story though, only moving slightly until it was stuck for good. The only thing that ended up working was leveraging the higher thermal expansion of the brass compared to the steel shaft. In other words much fire and prying finally got it off.
Funny enough the bearing feels super smooth now, but I'm pretty sure I just introduced enough slop for it to not bind up any longer. Now I'm faced with the decision weather I should replace the bearings and brushes in the hope of repairing the motor, or give up on it completely. Replacing them would also mean to get the old bearings off the shaft, which might be a bit difficult as well. I watched quite a few YT videos about Grizzly 0704 (practically the same machine) conversions lately and I'm definitely tempted to upgrade to a belt drive with a more powerful motor and maybe even a power drawbar. I just kinda need a mill to make the parts needed for the conversion. I'm also assuming that a more powerful motor will probably not have the same mounting holes and shaft diameter, so I can't drop it in and then convert it later. Looks like I have lot's to think about.







Let me know if you have any genius insight into this.

Also, let me know if all this CNC talk is a bit too off topic for a buildlog and I should make a new thread somewhere else.

 

Don't worry - I talk even more about CNC in my log currently
Some info how to repair a motor is always useful...

When I converted my BF20 I ditched the gears and replaced them with pulleys.
The 15mm HTD 3M belt is more than strong enough to deal with the power. At the occasional events of crash, it never tore. The motor stops, because the VFD shuts down due to over-current...
So no worries there. And if you go beyond 2kW, just use the 5M profile... Maedler has a nice little tool for timing belt dimensioning on their web site...

And then you have the full flexibility in terms of shaft dia, axle distance, etc...
It also runs quiter than a gear system.

The downside is, you have to make an adapter plate - so you have to rebuild your motor first

 

Thanks for the tips. Surprised that such a small belt is enough, I would have definitely chosen a much larger one (at least wider), but that would have been based only on gut and no actual experience or calculations.
Do you have any suggestions for the motor & vfd? I'd want around 2kw so I can keep it single phase. I do have 3 phase, I would just have to either run a long extension cord to the washing machine, or have an electrician wire it to my machine / do it myself illegally. (You didn't hear the last one) Then again, with a 2kw motor, the compressor, and the steppers all on the same breaker I might also run into a current limit...

Do you have any idea what caused the motor brushes to disintegrate? Would a shot bearing be enough for that, or do I run the risk of it happening again right away when I replace the bearings and brushes?

 

Are you sure you re-assembled them correctly?
The other one looks kinda skewed to me. As if the pressure from the spring pushed with an offset...

As for motor and VFD, I use a regular 1-phase industrial motor and Fit-E VFD from EMK.

https://shop.emz.de/jsp/epctrl.jsp?lng=1&pri=emz&mod=emz000034

The 2-pol motors are rated around 3000 rpm, but with the max frequency of the VFD it spins at 4000 rpm without any complaint. I wouldn't go higher with this milling head anyway because of the bearings...

Mine is a size 71 with 0.55 kW.
You'd be surprised how powerful they are. It is far stronger than the old 850W DC motor.
With the chiploads I am doing in aluminum (max 20 mm² cut cross section @ 1200 mm/min feed @ 4000 rpm) no issues.
And if I need more torque I can still half the revs with the mill head (same like yours).

2 kW you only need if you have a high rev HF spindle.
Torque is a function of power and revolution ( M = P / (2 x (Pi) x n) )

So a 0.55 kW motor at 4000 rpm has the same torque than a 2.2 kW spindle at 16000 rpm (1.3 Nm)
If the spindle is rated at a higher rpm (say 24000 like most Chinese ones are) it has even less torque, because the VFD would push only 1.5 kW to such a spindle at 16000 rpm...

 

The brushes had an angled face when I took the motor apart the first time, I'm assuming that's due to some play in the holder pulling it slightly skew. Though I wouldn't say that it's impossible I put them in the wrong way. I think I did it right, but could be wrong.

If I go to belt drive I'd want to either replace the bearings (you're right, they definitely won't do more than 4k, even that might be pushing it, as they get warm at just 2k already), or at least build everything so I can add a 1:2 pully ratio in the future giving me 6-8k rpm. I find myself doing lot's with small endmills, so having more rpm directly results in shorter cycle time. But what you're saying makes sense and if you can get that kind of performance out of your .55kw I should be more than fine with 0.7-1kw, even if I double the speed.

I've also been toying with the idea of getting a router style machine with a quick spindle for the small and soft material work. This idea is further fueled by the very limiting 20cm of Y travel my current machine has. Add an endmill and some linking moves on either side and a simple distro plate is pushing the work envelope. It would cost me roughly the same to get a Killerbee from RatRig (work area of roughly 750x500 on the one I'd get, build wise on the better end of hobby routers, with linear rails and leadscrews (not ballscrews), but still hobby), as reworking my milling head with everything I want on it (like a pneumatic drawbar with TTS toolholders). I'm currently on the waiting list for that machine, so I have like a month to make up my mind about it. In then end I'll probably do both, but for now I only have the budget for one of them.

 

Make a rough adapter plate with a hack saw and drill.

 

True, that could actually work. One guy I saw used a motor mount made from random wood scraps until he machined the proper one. Since it's belt driven, it doesn't need to be perfect as long as it allows tensioning. Typing it out like this really makes this the obvious choice

 

Well, these limitations were exactly the reason why I decided on a router too.
And once you have that, the need for high rpm on your current rig becomes less important.

If you go with the 1:2 ratio, 0.75 kW is probably your sweet spot.
You will have still enough torque at higher rpms and if needed you could still reduce the ratio with the mill head gear set back to 1:1

Because these 1-phase industrial type motors are cheap, but big and heavy. So a 1.1 kW type would give your stepper motor on the Z-axis some issues.
Even with the 0.55 kW type I felt more comfortable with a counter weight behind the axis...

The Killerbee looks interesting.
I find the portal sides a bit flimsy for this size of machine, but I definately like the 4 carrier design at each side. This should improve rigidity a lot...

 

Well, if they end up being the weakest link, I can always make thicker ones. Though being steel instead of aluminum, I imagine the leadscrews or more specifically their nuts being the first issue. But then a cheap set of Chinese ball screws should be a fairly easy replacement.

Thanks to cheapskate's suggestion I also found that getting a Chinese AC servo doesn't cost much more than an industrial motor and vfd. Do have any idea how their power compares to dc motors? I'm looking at either a 0.75kw or 1.5kw (anything in between costs the same as a 1.5kw). Their advantage would also be smaller size, giving me more options for a potential automatic drawbar. Not to mention all the nerdy things you can have them do. I found a bunch of guys on YouTube adding them to their G0705s, one of them even used a 3d printed motor mount until he was able to use it to make a metal one, I definitely like that idea. As far as I could see, they mostly use 1.5kw, though who knows how much power they actually use.

 

If your stepper can handle the extra weight, go with the 1.5 kW version.

Too much is never enough

 

yea, my steppers are definitely strong enough, they're stupidly overkill. I could probably lift up my machine if I strap the head to the ceiling . I chose motors that were too small on my first router build many years ago, so I overcompensated with extra long Nema 34 motors that are the size of the stock spindle motor.

 

Don't. The biggest issue I had with my build was a warped ball screw. It makes a wavy surface on the cuts. It's chinese, after all.
The screws on my Taig are awesomely solid. Keep them oiled and it will take a long time to develop slop, at that point it's a matter of tightening the nut, or replacing it. That's why the nut is a softer material.