Primo Squaring Questions Issue

Truing Primo

I have been up and running for a while with PRIMO recommended build size and larger steel pipe size with V1 bought parts (Except for 3D printed from a friend).

I guess what I am wondering is where to approach squaring and it is reasonable to achieve near perfect sub mm precision here? I already spent a lot of time on squaring but:

  • my round cuts are taller then they are wider. Off by maybe up to 1.0 mm ish. Grows with size? Likely squaring…
  • my holes tallest dimension (Vertical) is cutting smaller then spec as well. Likely steps issue?? Software?

I am not looking forward to re-squaring. So many places and I don’t have a good idea where to go now. I have already spent a LOT of time getting it as close as I could. My bases were still like 1mm or perhaps less off. I am a slight parallelogram which may be my issue.

It is really tricky with these self centering base holes as I did not drill PERFECT placed holes in the table. Is it worth modifying the bases to have L slots and flat tops at the corners and reprinting to give me a way to nudge my imperfect drills? Or is there enough slop elsewhere to fix this? I tried “adjusting the core” by turning the truck bolts, but this does nothing as far as I can tell. Its tough to feel “just tight” with these lock washers. I have tightened down most bolts to just snug but I need to check again. I have re-tightened down the belts which had seemed to have loosened. I have not really touched the core bolts yet (beyond those trucks) but nothing seems loose. I am using strong pipe-clamps for the endstops.

I was using a 25mm circle to evaluate with a .05 - .15mm tolerance (wider) pocket both via ESTLCAM. I was longer SSW up to NNE direction (like this: /, see pics). XY zero @ bottom left. Gaps are up to 1mm. But I suppose there may be some inconsistency here which may lead me to something being loose. I am not sure if my evaluating and playing is doing this or its just a sloppy cuts. Takes a long time to evaluate. The 25.15mm pocket is in hardwood being filled with MOP circle inlay.

I am dealing with essentially 4 mating parts that all need to be 1/4". a (1) Template cut with 4 holes 1/2" using (2) drill bushings with exactly .25 inch holes, (3) the hardwood part itself being cut with matching 4 holes, and (4) - 1/4" pins inserted to bring everything into alignment. This is how I discovered the holes are smaller then spec in the TALLEST dimension; about .0235-0.24 inch. Specified as 6.35mm in Estlcam. I could probably add some tolerance to the 1/4 cut holes to fit the too small, I just don’t like the sloppiness of it and I wonder if adding tolerance also must “scale” as designs get bigger and smaller.

I don’t know I am going to move on with my project and make do with the old school methods and I can troubleshoot this on the side as I go. SOmethings don’t matter precision wise, something’s are critical.

Not sure where to begin I guess but ill have to fit it in. Some pics.




Arrow in pics on circle is Y North.
As you can see i’ve already done a BUNCH of testing. That is how I discovered my software workflow was flawed and learning about Geometric bounding box setting in Inkscape and redrawing designs and importing into ESTL and verifying with ruler. So I now know its not the software. Except my parts are smaller than intended so I don’t really know its not software but its not likely software. I am way over thinking now… :slight_smile:

Thanks all

Possibly useful here

And I guess I n Ed to revisit
https://docs.v1e.com/electronics/dual-endstops/

I find so many description ambiguous and confusing while trying to figure this all out. Without contextual explanations it’s hard to evaluate…especially those gcode descriptions. “Set the current position” and “adjust the endstop offsets” are so vague without background or context and the detail matter. What does auto-home mean technically? If it includes the Z axis bad things happen. And does dual end stops mean 1 on X and 1 on Y or does it mean on both X min and X max and also on Y min/max? Yet more to learn and read…

In the V1 context, “dual endstops” is a feature that uses multiple endstops on the X and Y axes to enable “auto squaring” of those axes to each other. This mean that there’s an end stop switch/sensor for each of the motors that drive the X and Y axis. This requires each motor have a dedicated stepper driver and requires enabling of some settings in the firmware. V1 pre-built firmware with this option enabled indicates this by the presence of the word “dual” in the file name.

The end stops are at the same end of travel for a given axis. I believe X1 min, X2 min, Y1 min, and Y2 min are the most common locations. but you could have either axis (or both) with endstops at the max end. It is critical that the switches match the motors, i.e. X1 end stop must be tripped by motion of the X1 motor and so on. With this configuration, you can make offset adjustments in software so that even if your physical end stops don’t exactly make the machine square, the X and Y axis will be perfectly perpendicular after running the X/Y homing procedure as the homing routine will add an appropriate micro-adjustment before “locking” the axis motors for matching movement. This takes some experimental tweaking since each machine build ends up a bit different. The video linked earlier in this topic walks you through setting the offsets to dial-in the perpendicularity of your machine.

Is there any play in your z axis, x, or y axis? My primo ended up with a cracked core and a cracked truck and an EMT clamp was broken, so there was some bearing play between the core and one of the pipes as well as some play at the truck. If you are off in only one direction, you might check for any slop or deflection in your machine between the cutter and the bearings out to the trucks. Now, what I was cutting did not require the precision you seek, so I just wrenched it tighter and kept at it, but I am now replacing everything that touches the pipes.

thanks i’ll have to look at that again. I did crack my core at one point. I used some support s and epoxy rather then reprinting. So I suppose that could be an issue I need to check thanks.