Yeah, Yeah, must be us Aussies look at things upside down...180º rotation of the plate should sort it out

 

Your'e not missing anything, they're just difficult to adjust.

Put your assemble together loosily with every thing just snug, put the edge with the standard wheels down on a flat surface with a section of cbeam inserted and push down on the standard wheels with the cbeam and tighten the standard wheels, so the standard wheels are making the same contact with the cbeam. adjust the eccentric side untill they are just touching the cbeam (just the slightess pressure will spin the wheels) and snug up a bit more. Flip it over and check the standard side is slipping the same amount and adjust if needed. Go back to the eccentric side and adjust them just a little more and tighten. Making sure the pairs of eccentric are adjusted the same. (this is where a bit of white-out or simular on the opposite side to the engraving is handy).

You will find the wheels on the mouth side of the cbeam will slip more then the back side wheels. Just make a compromise, to get it as close as possible.

Good luck and let me know if this makes sense. \ /

 

@gwandad, sounds like something is out of spec , with the most likely candidates being the wheels, plate, eccentrics or a combination. With my machine after adjustment the marked side opposite the engraving are at around the ten or two o'clock position when the eccentrics are on the top side \ / ... So defiantly contact the openbuilds part store, so they can sort it out for you.

I went though the models in fusion to see if there's a error there without finding anything that would stop you getting your wheel down in contact with some preload, so all I can think of is that it is CAM setup problem, worn or calibration machine problem at the factory.

Target size ~ 100.6 mm between centres




Table of measurements I worked out on fusion.







Sorry, I know this won't solve your problem, but loving the bush mechanic in you.

Good luck Moag

 

Sounds more precise then the standard rule on a packet of tally ho . It doesn't matter how you solve a problem, just as long as you get there.
I spent a fair bit of my life in remote very locations, with no power, radio, phone, tv, internet and they were possibily some of the best times of my life making do with what was at hand. I'm so glad you shared as it can't be fixed if you don't know its broken!
Talking about broken stuff, I think I've finally sorted out my chinese spindle and VFD after have the the wrong one voltage wise sent, then breaking the replacement VDF in my rush to get it up and running, but its replacement is up and running finially so expect a how to wire up a chinese spindle and program a VFD soonish. So I even fall down a random rabbit hole sometimes.

 

Let me know what size Spindle (water cooled or air cooled) and VFD @gwandad , I ended up with Huanyang 0.8Kw 220V 400Hz Spindle and an 1.5Kw 220v single phase to three phase Huanyang YM series Inverter... try saying that quickly. The YM series inverter is their new compact version with a handy speed control dial on the front control panel.

You will need a submersable pump if you get the water cooled spindle and some easily sourced 8mm clear PVC tubing (from memory, but I will check), a bucket, at least 2 m of good quality shielded cable for between the VFD and Spindle, a soldering iron, solder and heat shrink.

The shielded cable of suitable specs was the hardest to source by far, with most of it sold by the thousand meter roll, but you might have more luck near the big smoke. I ended up with 2m of LAPP KABEL 0036425 Multicore Screened Cable, Servo, Transparent, 4 Core, 16 AWG (1.25mm^2) from http://au.element14.com which is major over kill and at the limit of what you can terminate on the spindle size. 0.75mm^2 (19AWG) might be a better choice if rated for over 300v with good shielding. Oh you will need a standard plug and lead from the mains wall outlet to VFD (10amp); i used a old computer power lead in good nick from the tip shop.

 

Sorry for taking so long to get this together, but here's a basic guide to setup a Chinese VFD and Spindle in pdf format.
@gwandad, @GrayUK, @Kyo, @Aussie58, @Anthony Bolgar and @Mark Carew: let me know what you think and point out any mistakes
Cheers.

 

Thanks Team Openbuilds for your kind words.

Cheers Mate and glad to hear it worked out.

A couple of questions well more then a couple @gwandad if you don't mine.

Did you get a (English) manual with your HY series inverter?

Did it come with the speed dial (potentiometer not phone)?

Are the main input and output terminals different?

I've made some changes to the VFD/Spindle "How To guide PDF" as I would like to make it as complete as possible before adding it to the Tutorial section if you get what I mean. Attachment below of latest version, the bold underlined blue text are links.

So gwandad, if you go to PD177,178,179 & 189 are there any error codes stored?

Cheers Mate.

 

Thought I'd see if Fusion 360 would play fair with my setup today... Raspberry Pi with 7" Raspberry Touch Screen running bCNC, talking to a Smoothie Board, talking to DQ542MA Stepper Drivers running High Torque Steppers. Test was a simple half sphere bowl, 30mm diameter with a 6mm 2 flute carbine bit at 15000RPM in some fast growing plantation Pinus radiata.

The bottom row, was playing around with "Adaptive clearing" (man, does that take out material fast ) and the top was "Adaptive clearing with a Ramp finish path". Looks like I will have to work out how to get rid of that little nipple on the bottom, but not to bad for a first go.

 

Thought I would try something a bit more complex to cut out... curvy 3D/2.5Dee of course..., just to make things harder, near the limits of Z travel with my current setup, using some kiln dried Eucalyptus regnans Australia’s tallest trees, which is a long way short of the easiest timber to machine.

Modeled a small oval bowl 80mm long x 54mm wide x 27mm deep with a thin slightly wavy lip, lofted down about 8mm that kind of looked like this...



Thought I would see how rough a finish with thin edges using adaptive clearing would give, after some fiddling around it came up with a messy looking web of passes that looked like this...



and some stats like this...



Attached a 6mm Ball 2 flute Carbine bit with a CEL of 12mm, hanging out a little over 28mm. Whacked or I should say stuck a piece of stock to my machine, opened bCNC, clicked Home, jogged to the centre of my stock, jogged the Z down till it was just touching, zeroed out XYZ, loaded the .nc file, jogged Z up 5mm, clicked scan (all good), started the spindle (dialling it up to 22000RPM)... Held my breath and clicked Play.

- First mistake... it slowly jogs back to home then back to the cut start position (note to self, should take more notice of David the swarfer ), not to bad really as it gave the spindle some more time to warm up.

- Second mistake... when it got around to cutting the little wavy lip, it was doing some sudden changes of direction that were too extreme for my liking... I could see this when doing the cam simulations, but thought I would see what it looked like in real life

Third mistake... failed to work out maximum depth for my limit stop on Z, so hit it on the last and deepest helix plunge and all stopped with a big red error alarm... "Oh well at least I found out the limit stop works". So Turned spindle off, moved the limit switch a little higher, exited bCNC, reopened to clear the fault (rough and ready way to do it), re homed then clicked run again and watched it cut air mainly (slightly out in the Y from what I could gather) till it got to where it stopped, through my stupidity

Finished the job, not to shabby for a test run, but still lots to learn