I knew it was bound to happen, but did not expect it to happen this way.

I made some changes to the Mini Arrow to add a speed brake. It was essentially a 3" 45 degree inside cut of .25" width.

When I cut out the plane, I noticed some of the slots were not cut out accurately, then came the 45 degree cut I added. Horrible! Absolutely nasty!

I slowed the rate down and cut again. Still nasty. Finally I just made a test cut and started trying various settings, recalibrated both axis and even at 5% of full rate it was not producing the cut it should.

I diagnosed the problem as an X axis issue, although there is something going on on the Y axis as well, I will need to clear that up after I get the X axis working right again.

Finally, it dawned on me to pry around on the bit to see if there was any play in the gantry.

Oh YEA! Here is the culprit! The forward and backward movement of the Dremel confirms that there is excessive play in the gantry.

Reluctantly, I pull the gantry out. This of course means taking my belt mod apart! Drat! But it has to be done.

After getting the gantry out, I find the back plate holding the Dremel has come unglued at both the top and bottom of the gantry. The only thing still holding it in place is the two center tabs.

Mind you, it is not a lot of play, but it only takes a miniscule amount to cause problems!

Eric and I had discussed the use of my penny mod earlier and he was concerned the higher speeds of the gantry movement would tear off the pennies I use for bearing surfaces so that is what I was expecting to find when I tore into the gantry.

The good news is that the penny mod is working! In fact, I would highly recommend anyone doing the MK1 mod to install it if they do not have any bearings installed with other mods. For a mere 16 cents, you can take nearly all of the inherent slop out of the gantry. And at these speeds, you need for the gantry to be rock solid!

 

Yes, that was what Mark used in the videos. This is certainly not the first time I have had to reglue pieces back together. On the other hand, there have been times when I was thankful that it was glued so that I could get it back apart.

On one hand it sucks, on the other its great. Can't win. There is one place where I used epoxy and I would like to be able to get that part off now, but it is not going to budge, so what can you say.

In the long run, It all works out. As a result of tearing down the gantry one more time, I was able to go in and add a few more copper pennies to strategic places on the gantry pex pipes and got even more play taken out. I discovered missing pennies on the bottom of the gantry and added four additional ones to completely surround the top two pipe at their opening.

No doubt this will provide even more precision cuts than I was getting although I could not complain as it was before the glue came undone.

If only I could get the z axis to move faster!!! 20ipm is really the most I can get out of it and still not get an occasional miss step.

 

Well that's something I am going to have to look at very closely too...

About the Z-Gantry speed.....When I had it apart I decided to test the stepper again without any load. The screw was removed so it was just the motor.....I still couldn't get more that 20IPM.....I think some of these motors are just not quality parts. I'm a little pissed about this.....not at Mark....just the circumstances. I'm going to look into getting a replacement stepper now as well.

 

At 1/8 step, the resolution on the Z axis is (1/20 inch per turn divided by 1600 steps equals) 0.00003125 inch per step. You could start by changing the setting on your controller card to 1/4 or 1/2 step, which should multiply the IPM by a factor of 2 or 4, respectively.

 

Good information Dorsal... I will give that a shot. I also ran the gantry with no load (dremel removed) and could not get much improvement over the 20ipm range.

I am more than a little disappointed with the easy cnc steppers myself. Their control board and relay board seem to be pretty good quality although I did loose one controller chip already.

I ran my Y axis up to 1200ipm and feed rate at 400, but eventually had to back it off because I was getting some missed steps.

I think I am at 800 with feed rate of 300 now and am pretty satisfied with the performance overall.

So right now, it the Z axis that is slowing everything down. If changing the switch gives me a 4 to 1 improvement, then I think it will be a pretty respectable machine.

 

Now that my Y is moving faster, I'll have to start playing with my Z also, and that's where I'll start.