Bob K published a new build:

Read more about this build...

 

I have no way to calculate/simulate how the system will perform so I'm hoping for feedback from people with experience.

If I wanted a 12" Z movement I'd need a 705mm extrusion/actuator or about 2/3 of the current height. Is that enough to make you comfortable?

To make it even worse the stepper motors are about 2x the size of the one sold by OpenBuilds. I'd thought about flipping the Z stepper motors to the bottom but that would require a 170mm tall base. At least I'd have a convenient place to mount the electronics enclosure.

Bob

 

On the moving bed design, if I rotate the Z actuators by 90 degrees I'll end up with an x/y movement of about 16" x 16". That's plenty for me.



I'll go back to the moving bed model.

I have a few more days before the remaining rails/beams arrive so there's time for further refinements. I'll update the build section then.

thanks for the help - it's much appreciated

Bob

 

Rick - I bought the motors 6 weeks ago when I thought I'd be doing the big fixed bed 1.5M x 1M x 1M. I knew this was bigger than most of the CNC routers so I expected I should have more power (Tim the Tool Man syndrome alive & well here).

The other thing I liked about them is they only draw 1.8A. That allowed me to use the simpler motor drivers (that plug into my controller) which simplified the wiring.

Bob

 

The wiring comment brought up an item I've been wondering about.

I was REALLY surprised when I saw the wire sizes on the stepper motors and on some of the builds. In my world we'd be using 16 awg for a 2A load that is 1.5M away to be sure that the wiring wasn't a factor. These guys are coming in at 20 awg at the best. Are there any reports of wire size issues?

I'm going to have to go to a different connector system so I can get contacts that are small enough. Molex has an 0.062" system that has contacts that go down to 30 awg.

As long as I'm on wiring. Almost everyone has to splice the wires someplace between the controller and the steppers (etc.). I'm a little worried about the splices eventually breaking because the wires are flexed when the machine is in action (every design I've seen has at least one axis this applies to). What types of splices has the community found to be reliable? Solder & heat shrink, but type crimp connectors, 1/4" quick connects, ... ?

Bob

 

It just dawned on me what the splice answer is: Don't put the splices in the section that flexes.

I definitely have a hard time keeping things simple.

Bob

 

Justin - thanks for the feedback.

I'll use my over sized (18 awg) wires on the Nema23 steppers and the E3D heater and then probably get 20 awg for the remainder of the signals. I find that anything smaller than 20 awg is as a pain to work with.

Bob

 

RATSS!!!! - the M5 drill/tap reamed out several of the holes enough that an M5 bolt just slides in & out.

I can drill & tap them to M6x1.0. The local Home Depot has the needed bolts but they don't match the M5 bolts.

Anyone know of a source for M6 bolts that are black and have the same head as the M5 bolts?

Bob

PS - I also had a couple of problem holes on C-beams. Those I cut shorter & re-tapped.

 

Amen to the combo drill/tap in long holes. Just too many problems.

 

I just did 8 more M5 holes in a piece of V-slot using the drill then tap method. I screwed up the first two holes!

Guess I got lazy over the years and just turned the tap until I felt increased resistance, reversed the tap a 1/2 turn and then continued. That resulted in aluminum smearing around the leading edge of the tap. If thick enough that smear would remove the thread on the way out.

Things went considerably better once I went back to "rotate 1/4 to 1/3 turn, reverse for 1/2 turn" and backed the tap out completely 3-4 times per hole. On the way out if I felt increased resistance I'd reverse the tap for 1/2 a turn and then continue. I knew there would be smear if the resistance didn't drop. Usually there wasn't any smear and the smear I did get didn't cause any problems.

I've never worked with an aluminum that smeared like this one does.

I've tried getting the JB Weld down the two bad holes. Didn't seem like I got much into the. Tomorrow I'll see if I can drill & tap them.

I'm now impressed at how good a job the drill/tap did considering it is meant for use with a drill or an impact driver. Definitely not a good method for deep holes.

 

I hadn't heard of a forming tap before. Sounds like the pros like them in aluminum.

There were several mentions that hand threading using a forming tap is nearly impossible. Can it be done with a large drill or an impact driver in a home shop? Just worried that a special device is needed.

With all the issues I've had, I'll be using a forming tap if I ever need to do a non-trivial number of holes in a future product.

By any chance did I miss a resource article on machining aluminum? I expect other have had this issue.

Bob

 

I had a very pleasant surprise today. I decided to cut off the unused portion of a C-beam actuator. When I went to tap the holes I decided to use some of the cutting fluid I'd bought for metal milling. To my surprise the torque needed to run the tap in was obviously lower and there wasn't any smear on the tip of the tap. I then did a few more practice holes with the same result. In between holes I'd spray off the chips with an aerosol can of break cleaner and then I'd dip the tap into the cutting fluid.

It got me to wondering how the cutting fluid would affect the drill/tap combo. 32 holes later there were absolutely no problems. No smear on the tip and very good looking threads. I then went and ran a 15mm M5 into each hole and torqued it to the point I was afraid I was over doing it. A bolt would last 6-7 holes before I rounded off the hex in the head. I even ran a 10mm into 8 holes with the same result.

When I used the drill/tap combo I used an impact driver to drive it in without stopping. Definitely less torque needed with the cutting fluid. I did the same clean-and-then-dip-into-the-cutting-fluid between each hole.

The actual thread engagement was about 4-5mm less than the bolt length because I was putting the bolt through a cast corner brace (about 3mm thick) and there was usually a 1-2mm leadin/shoulder on the hole. I was impressed.

Cutting fluid is my new best friend.

Bob

PS - don't know if it makes a difference or not but the fluid I used was Tapmatic Natural Cutting Fluid Platinum Grade

 

Barry - where do you get the flat cable & what's the vendor name & part number?

Four of my steppers are fixed so I can use stiffer cable on them. I'll probably use stranded shielded 20 awg security cable on them. I can get that at Home Depot for 29 cents a foot.

I've ordered some 24 awg stranded shielded cat5 patch cable to play with. That should be able to handle almost everything else.

I still have one Nema23 stepper motor and a 2A PWM heater to worry about. Maybe I'll use your flat cable for these.

You can drive yourself crazy looking for cable. There should be a resource thread on this topic.

Bob

 

Newbie question - how easy/hard should it be to turn the 8mm Acme screw?

I couldn't find anything in the discussion or on YouTube that covered this.

I've got one that is easy to turn by hand, one that is moderate and three that are hard. The easy one will rotate the screw if I push on it. There's no obvious backlash on the easy one.

Bob

 

Rick - I received a M5 forming tap and #14 drills today. I did about a dozen holes and have come to these conclusions:

• cutting fluid is still needed
• a #14 drill is needed. 3/16" is too big (15mm bolt can't be torqued down) and 11/32" is too small (I was afraid I'd break the tap).
• a modest 1/2" cordless drill is enough to run the tap in

I like the drill/tap combo with cutting fluid best of all. It's messier than the forming tap but it does a great job and is easier to obtain.

Bob

 

Major problem

My Azteeg X3pro controller isn't compatible with GRBL. GRBL requires some signals to be grouped together in a single port but my hardware has them all over the place.

I briefly looked at the TinyG series. They're great for CNC mills but I don't see them mentioned in the 3D printer section so I'm assuming they're not compatible.

My dreams of having a single controller do both milling & 3D printing aren't quite dead.

Suggestions are welcome.

Bob

 

I found out that some people are running Marlin on their CNC mills.

I'm going to try Marlin for CNC milling. I'm pretty sure that you sacrifice performance but I'm so new at this that I wouldn't be able to tell.

Looks like there's activity in the TinyG/G2 camp to develop a 3D printer. Maybe that'll bear fruit by the time I outgrow my current controller.

Bob

 

Got my Y axis moving today. YEAHHHHHH!!!!!!

Movement is only at about 25mm/s. BOOOOO!!!!!

Off I go to learn more about stepper motors.

So far I've learned that the big boys use 48V and higher and like to shell out big $$$.

One thing I don't have is an idea of how much torque my application needs. I just ran across a post that looks promising: http://www.openbuilds.com/threads/48vdc-or-24vdc-for-stepper-motors.6734/#post-34750

Bob

 

I did my first cut today!!!

Just a shallow pocket in some scrap wood but it proves that the Marlin printer software can be used for milling. I had to manually edit the SketchUcam g-code before Marlin was happy with it. That could get painful on complicated pieces. I'll probably play with other g-code generators to see if their output's any more compatible.

I've got the system running at 40mm/s on the X and Y axis. That's sort of consistent with the torque curves I got today for my steppers. The curve stops at 5 revolutions a second with the power at about 20% of the low speed torque.

I was really surprised that the torque curves for 24V and 48V operation were so close to each other. For half the curve they're within 20% of each other. Probably because, at 17mH of inductance, this motor wants to be operated at about 130V.

I still have a lot of learning to do on stepper performance in a system. If I try to push the feed rate above 40mm/s it starts to shake, it chatter loudly and the actual speed slows down considerably. What I don't understand is:

• If the rate is too high, why does it slow down instead of stopping? If I push the feed rate high enough then the steppers just chatter and don't move (to be expected). I'm just surprised there's such a wide transition zone. I'm really surprised that it's not loosing any steps when its chattering. Maybe the motors are starting and stopping very rapidly (high speed torque not enough to move the bed but the startup /low speed torque is enough to move it).
• When the systems starts to chatter (and move slower) I can press on the carriage with all I can yet the speed doesn't change. If the chatter indicates it's getting to it's max speed then that implies that it's also getting close to zero available/extra power which implies that my putting a lot of force opposing the movement should have a very noticeable effect. If the motors are starting and stopping very rapidly then that would explain why there is apparently lots of power available.

Another thing I'm scratching my head over is the g-code feed rate is set to 5,000mm/min but the observed rate of movement was about 2,400mm/min (40mm/s). Dropping the feed rate in half slows the system down by half. If it had something to do with the number of micro-steps not matching the steps/mm then my move lengths would be way off.

Oh well, those are mostly curiosity questions since it appears I have the system fairly close to maximum performance and the performance is more than enough to mill aluminum and mild steel.

On to building the print head!!!

Bob

PS - should have put a tarp under it before the cut. My shop vac just didn't get as much of the chips as I expected during the cut.

 

What are people doing to set the X/Y/Z zero points and/or how are they getting the work piece squared away to the machine?

I see a few people using the conductive touch block for the zeroing the Z axis. That looks reasonably easy to do repeatable.

I'm scratching my head over the X & Y methods.

Right now it looks like I'll start off just making sure the work piece is big enough to overlap the boundaries and let the machine cut off all the excess.

But what do I do when I need to modify an already machined piece? With my "skills" I know I'll be wanting to salvage some of my early attempts.

Bob