My tinyg did that when I first got it, but then i updated the firmware and it stopped doing that. Not sure what setting if any affects that.

 

Hi All
Just started this build .
Got the kit from France.

One thing I was not happy with was the play in the none eccentric side of the Z spacer blocks.
So I made three collars 7mm outside dia and 5.5 inside 15 mm long to fit into the holes for the V Wheel screws.
this has made the play much better and will help the Z stiffness.

Bob

 

Thanks for the suggestions with the reversing of only one of the coils. I have not had much time lately but will probably try it out tonight.

Meanwhile I see a number of new questions here:

What do people do with the OX? The main purpose for me is CNC milling of project boxes and lantern housings (see www.hauntedlantern.com and you understand why (and my avatar as well )). In the future I want to see if I can turn it into a pick-and-place machine for SMD components. That one is going to be a time consuming and challenging task but I would really like that to work. And perhaps some day I will mount a laser to the Z axis. Maybe only a moderately strong one that can cut SMD stencils for me.

Steps per inch: I have settings X/Y 40 steps per mm, Z 400 steps per mm, which applies when you have 200 steps/rev and 1/16 substepping. You can do the math to inches yourself I guess.

 

Do you know how to update the xpro?

Thank you for your reply, do you have an answer for my previous 3 questions a few posts ago? You seem very knowledgeable.......So if I go back to MM and change the shunt jumpers to 16 steps/mm and use your settings that should get be right on the money or will my machine be a little different that yours? and if it isn't the same is this statement correct? More steps/mm equal a longer movement and less steps/mm equal a shorter movement? In case I need to fine tune it.

I really really appreciate the help. This thing is driving me nuts.

EDIT: I moved one of my shunt jumpers to position 3 and removed the other one. This should be 1/16 step according to the board. I used your steps/mm settings and am not getting the correct results. I zero the machine. Enter x300 and when I measure my line I am getting roughly 224 (i don't have a metric ruler yet). thoughts? Do I keep increasing that 40 value until they match? why would our settings be different?

 

I would use the "Steps per millimeter - belt driven systems" section of this calculator: http://prusaprinters.org/calculator/

I assume you're using the "GT3" belt which I believe is 3mm pitch and 20 tooth pulleys.

 

@Pbmaster11: what @KerryH says is exactly what I did yesterday night, and with a little bit of fine tuning I got a final number of 53.38 and the measurement are now spot on. Very easy.
If you want any more info on how the Grbl set up works, you can also refer to the https://github.com/grbl/grbl/wiki/Configuring-Grbl-v0.9 looking for $100-101-102 commands.

 

This all does make sense so I am curious where I went wrong. I did assume 4mm per tooth (instead of 3) which is the reason why I ended up with 40steps/mm, but I also did measure it roughly and I am pretty sure that if the deviation was this big, I would have noticed. I guess I will measure again tonight.

Edit: Well I measured it and it was hopelessly wrong. 53.333 it is now and that works. I also got the Y axis motors cooperating now.

 

Thanks guys, I will try those settings.

EDIT: It still only travels a portion and stops... the GRBL controller still shows the machine should still be moving? my set screws are tight. I am not sure what is happening.

Any ideas on why my steppers lock up when I plug the CNC in? It just started doing that? all my leads are correct. I am using a 12v power supply. Do I need more juice? I have my current turned up on the x and y. Steppers are cold. Z axis is a little warm but that is set to a notch above the lowest setting.

EDIT 2: When I draw a 3" x 3" square the machine will move a little on the x axis (long main rails) and then stop. It then draws a 3" line on the y axis (doubled beam). then it goes back on the y to the end of the x axis then goes back to home. something is jacked up with my x axis. NEEEEEEED HELPPPPP.

 

Here's a picture of my OX. There's still some work to do getting rid of the spaghetti, though it is already far better than 10 days ago. I have decided that the part where the tools cannot come anyway, is perfect for placing all electronics. I do start to understand why some people prefer a single board solution instead of a controller and 4 drivers, though. Limit switches still have to be connected.



I was finally able to do some testing. This is g code that I generated from a (C++) program. It is running way faster than can be milled but that is remedied by just changing a constant in my program.

 

So I've disassembled and reassembled my ox quite a few times I'm becoming very proficient at it, but I would rather spend this time cutting.

I ran into an issue with the drive pulleys coming loose, I've resolved that issues and I still have a very noticeable 8-10 thousandths (.2-.25mm) of "backlash" on my X-axis. My Y-axis looks very good now, but I'm not satisfied with my X-axis.

I spent all day today converting my machine to the dual belt setup with one belt securely taped in the v-slot and still no improvement.

I believe my belts are tensioned enough, any more and I would be able to strum a tune on them.

When testing with a dial indicator, if i move in one direction the machine is accurate to .001"-.002" (.0254-.0508mm), but when changing direction it is .008-.010" (.2032-.254mm) off. This make holes very inaccurate.

Is this due to the very small amount that the belt wraps around the X-axis pulley or is this just an issue with my machine?

 

I had similar problem. turned out to be the driver chips overheating and shutting down until they cooled. solved it first by turning down the current settings to just under half way and adding a big fan and plenty of ventilation. I still need to add heatsinks to the chips but they work well for me as it is now. Carefully ( I mean really carefully ) touch the driver chips whilst it is running and if you smell burning flesh this could be the problem.

 

Hi All
please excuse me if this has been covered earlier in the discussion thread but a quick skim over the 7072 pages did not seem to locate it.
I am looking to modify a stalled CNC project with design ideas from the OX but before I jump in I am wondering how well the belt drive system handles machining of soft metals such as Brass and Aluminium. I have seen another project here where they changed it up to use ballscrews and linear rails but that would require me to buy all new extrusions
Thanks in advance

 

I have felt the board in the past and it isn't hot at all. well at least I don't think so. Would this issue just affect the x axis? the y and z seem fine.

 

Okay I have solved the dying stepper issue. the current was turned up too much. My next questions is, why do my steppers move so rough? like they are actually stepping and it is not a fluid motion. looks and sounds jabby. I have them set to 1/16" microsteps. Is there a setting in the xpro board that helps smooth out the movement?

 

I am getting the same noise, lack of smoothness. I have played with uSteps per step and quiets down at 1:1 but hearing folks use 8:1. 1:1 would require adjustments to the rest of the setup. BTW, OX, 750mm x 750mm, xPro V2, belt on x and y, gear on z.

 

Added end stops to all axis. Using for emergency stops. Tapped 5mm threads on x and y plates. Using 20mm screws for adjustable stops. For z, used t-nuts to mount in slot.

 

Hi guys. I'm still waiting for my kit (from vslot-europe) that should arrive perhaps tomorrow. Just wanted to say hi, and also that I've created an IRC channel for OX talk:
freenode.net: #cncox

Here I've only assembled the 1.5KW spindle and speed controller tonight, and that's working fine so at least the business end is sorted until the frame is

 

I'm still having an issue with my X-axis. I just tested my Y-axis via a dial indicator and it is dead on 1" (25.4mm) going one way, and then when I send it back the other direction it is only off .002" (.0508mm) over an inch.

I'm banging my head against the wall, the inaccuracy on the X-axis is unwelcome.

EDIT: added video:

 

Hello team. Here is my OX 500 x 1500 build. Started early January. At least 3 orders to the parts store and its mostly done. Modified Z axis with 1/2X10 screw. Using the GT3 belts and pulleys for X and Y. Running X and Y at 1/8 steps of 1.8deg steppers. Z set at full steps because of the finer screw. Was initially worried about the accuracy of the belts but my testing and parts have been turning out good.


I use Sketchup Make for CAD, Cambam for CAM, I wrote a custom machine control program that covers all of the g-codes that I use. Using arduino uno with custom sketch to drive the TB6600 drivers.


Few things I learned.

1. First I was not sure how to cut the v-slot. Finally used an electric miter saw with a regular carbide tip rip blade. This worked good.
   
   
2. For the GT3 pulleys. Replace the 3mm set screws with a larger 3mm machine head screw. That way you have better control when tightening it. I think its easier to get it tight without stripping it. I used a little lock tight. I got the new set screws at a hobby shop.
   
   
3. Make sure the X axis motors are mounted high. Hard to get an allen wrench into the bottom bolts after it is put together
   
   
4. So far I am having good luck with the cheap TB6600 controllers from ebay. But I took them apart to check. Found that on two the main chip was not seated well to the heat sink. Also, the only standoffs are the plastic cover. Bottom of circuit board really close to shorting on heat sink.
   
   
5. If the T-Nuts don't slide into the slot well just file the edges of the screw hole. If you force them they get stuck

Thoughts on using a dial indicator to check accuracy.

Not an expert but I figured for my setup that the xy theoretical accuracy was about .005. Now because of the size of step the controlling program has to round to the nearest step. So using a dial indicator is a little tricky. If you tell the machine to go 1 inch its going to be off as much as .005. Take into other factors the accuracy is going to be less that that. Things I wonder about are how accurate is the micro-stepping. If the motor is designed for 1.8 degrees there must be some extra loss of accuracy with micro-stepping.

Whats next.
Limit switches
Better spindle
Dust collection
Wire guides
Replace garage door opener switch with better emergency stop switch.

For those of you working on the OX now. My advice is to slow down. I'm going through post project depression now, what should I build next?

 

Now you have turned the current down, slowly increase it to see if they run smoother. I found I had to find a happy medium between overheating and noisy steppers. I run at 8 microsteps which works well and gives reasonable accuracy and speed. Fine tuning is a long process, and takes patience.

 

My kit has arrived, ordered from vslot-europe. it comes with everything (ish) but the spindle/tool/router, however, as you probably know by know, these builds have a tendency to evolve with time.

Anyway, so far a few nags with the kit. It's missing the eccentric wheel adjusters, so really, I won't be able to make a lot of progress without these. Also, the OX plates were cut a bit too big, so that will require a bit of filling, but hopefully nothing dramatic. Luckily I have a little stock of M5 bits and pieces, as I definitely needed them to get to that point already...

 

Mine also lacked the eccentric spacers. You might also want to check whether you received the shims (some people, including me, got washers instead). I also did not get the two screws for the ACME block, and one of the spacers for the Z axis motor mount was 1.5" while the others were 40mm. After some hassle it was all solved, but it did take some effort.

Meanwhile: I have done my first cutting. Eventually this needs to be done with HDPE but for a test, I took multiplex. You can imagine that the edges are a bit rought, but I am quite pleased with the result.

 

Nice, hck!

Yes I also was missing the shims, however, I had a bag of 100 A2 M5 washers, and they are /exactly/ 1mm thick, so thats precision enough in my book. Luckily I had ordered a whole lot of M5 stuff on ebay in preparation. The (good quality) lock nuts did came handy to replace the ones (hmmm..) that came with the kit.

It's definitely not a very precise kit. I got a whole bunch of 8mm screws for example, while the 5mm thick plate requite probably 10mm to get a good 'grip' on the threads on the other side. Nothing that a next-day ebay can fix tho.

 

BTW, has anyone mounted a 1.5KW spindle on an OX? I got one of these, and the 65mm mounting ring, but it's WAY too big for the v-slot mount as it is... did anyone make some sort of adapter plate?

 

Here it is: http://www.openbuilds.com/resources/makita-3709-mount-clamp-for-3d-printing.67/

 

Nice! have you posted the source for the holder anywhere? Are you happy with the plastic clamps as they are (no heat issue on the spindle?)

 

make sure that the grub screw on the pulley is tight and perpendicular to the flat on the stepper shaft. If its loose you can get some play in the pulley when changing directions.

 

Hi Scooter,
Very nice job there! I see you have some pink stuff cut out.. Is that the home depot insulation foam? Looks like a fuselage to me. Just wondering if it is indeed foam, what bit do you use to cut it out? I build my OX with laser cutter and I also plan to attach a dremel to cut out some foam pieces for RC airplanes.