[amyers]

those powder cup things look awesome hoss, i'm gonna pick up one of the craftsman powdercoaters on ebay so put me down on the list if you start selling them please.

[zerodegreec]

Got my BF20 up and making chips. What is a reasonable tolerance to achieve? Stock screws and repeatable calibration and backlash tests. Seem to be off by 0.02 to 0.1mm on x and y.

I know it is machine specific but hopped for a general"reasonable" range.

Plus for some reason my machine cuts parts ~1 to 2mm further into the x+ than the stock was zeroed. This one is making me pull my hair out. We are talking the entire part... 😊

[SCzEngrgGroup]

An honest +/-0.1mm is about as good as you should hope to get for the typical conversion. Getting significantly beyond that takes a LOT of time and/or a LOT of money, and/or a LOT of dumb luck.

Regards,
Ray L.

[SCzEngrgGroup]

An honest +/-0.1mm is about as good as you should hope to get for the typical conversion. Getting significantly beyond that takes a LOT of time and/or a LOT of money, and/or a LOT of dumb luck.

Regards,
Ray L.

[outsider787]

So +/- 0.1mm for a conversion using the stock screws.
What about if a little more money gets put into the machine to upgrade to C7 cheap Chinese ball screws/ball nuts?
What should one hope for accuracy then? How much more accuracy will the extra $200ish get you?

adrian.

[hoss2006]

Got my BF20 up and making chips. What is a reasonable tolerance to achieve? Stock screws and repeatable calibration and backlash tests. Seem to be off by 0.02 to 0.1mm on x and y.

I know it is machine specific but hopped for a general"reasonable" range.

Plus for some reason my machine cuts parts ~1 to 2mm further into the x+ than the stock was zeroed. This one is making me pull my hair out. We are talking the entire part... ��

.003 - .004 backlash is pretty typical for the stock screws, you can get that below .001 with ballscrews.
Were the calibration tests repeating a good distance or were they all over?
How long is your part?
Care to post a pic and some gcode, the error may be there.
It's not dumb luck when you take the steps to dial in your machine.
Hoss

[hoss2006]

those powder cup things look awesome hoss, i'm gonna pick up one of the craftsman powdercoaters on ebay so put me down on the list if you start selling them please.

Thanks, I'll add your name and let you know.
Hoss

[hoss2006]

So +/- 0.1mm for a conversion using the stock screws.
What about if a little more money gets put into the machine to upgrade to C7 cheap Chinese ball screws/ball nuts?
What should one hope for accuracy then? How much more accuracy will the extra $200ish get you?

adrian.

The chai ballscrews can get the backlash down to under .001 (some nuts need repacked for this)
Accuracy of .002-.003 is achievable without too much effort, get the backlash down, warm up the machine before you run, do calibration and backlash tests and put it to work with good machining practices, practice will make perfect-ish.
Hoss

[Tex-VA]

Thanks, I'll add your name and let you know.
Hoss

Hey Hoss, Please put my name on the list too. Tex

[Kibo]

Is there a way to tell whether it's worthwhile to pursue repacking the chai ballnuts before assembling the machine?

[zerodegreec]

Well I found my "issues" First off when I pumped out the G-Code I did not change the tolerance of HSMxpress. My finishing passes were set at 0.1mm (well that would do it!)
After making some changes and tightening up the tolerance I can hold 0.01 throughout the part (as accurately as I can measure with a decent quality caliper).

Hoss are your tolerances you gave in Metric or Imperial?

And why was my X+ so far off???? take a look at THIS!
I felt like I was going in circles trying to find where my measurements were so inconsistent. I re-checked my calibration and backlash and to my supprise I was off my 0.0127mm to 0.025mm on my Y and X respectivly from last night... Odd. I did each axis 3+ times after I re-set everything and it was ban on every single time. And I am using a couple 1,2,3 blocks setup with a very rigid indicator setup.

[Mad Welder]

Hoss it's imposible to keep up with your progress, but again excellent work with the powder coat gun fitting..in time I'll get a 3D printer, but then I'll have to read all your 3D printing thread(nuts)...loved the bit too about the camera stuck to your forehead with double sided tape......

maybe velcro would be better

[Fastest1]

Is there a way to tell whether it's worthwhile to pursue repacking the chai ballnuts before assembling the machine?

You could measure the backlash to some degree with a fixture of some kind. Assembly will involve a break down or 2 of the machine most likely. Especially if it is your first build. You will have time to measure backlash on the actual machine, order balls and repack while doing other tasks like wiring, plumbing, enclosure , tool changer, PDB. You get the point.

[Fastest1]

Hoss it's imposible to keep up with your progress, but again excellent work with the powder coat gun fitting..in time I'll get a 3D printer, but then I'll have to read all your 3D printing thread(nuts)...loved the bit too about the camera stuck to your forehead with double sided tape......

maybe velcro would be better

He hasn't proved he is using either attachment technique yet. Maybe a big terry cloth headband with camera attachment. As much as he gets accomplished he has got to be sweating profusely!

[hoss2006]

Hey Hoss, Please put my name on the list too. Tex

Sure thing, I'm making a new 12x12x12 printer so that should help with production.
Hoss

[hoss2006]

Well I found my "issues" First off when I pumped out the G-Code I did not change the tolerance of HSMxpress. My finishing passes were set at 0.1mm (well that would do it!)
After making some changes and tightening up the tolerance I can hold 0.01 throughout the part (as accurately as I can measure with a decent quality caliper).

Hoss are your tolerances you gave in Metric or Imperial?

And why was my X+ so far off???? take a look at THIS!
I felt like I was going in circles trying to find where my measurements were so inconsistent. I re-checked my calibration and backlash and to my supprise I was off my 0.0127mm to 0.025mm on my Y and X respectivly from last night... Odd. I did each axis 3+ times after I re-set everything and it was ban on every single time. And I am using a couple 1,2,3 blocks setup with a very rigid indicator setup.

Well that would explain it.
Is it too late to ask for your money back on that probe?
I was using imperial units.
Hoss

[hoss2006]

Mad and Fastest, in my last video I used a camera "tripod", put a 1/4-20 stud in the end of my flexible mag base.
Hoss

[Fastest1]

Nothing like KISS! Resourcefulness!

[zerodegreec]

Yep they are sending me a new one with the option to return for refund if I am not satisfied (great people at CNC4PC.

Funny how I was lucky to hold 0.5mm before the co version (dummy behind the wheels) and now I am frustrated to no have 0.01 or better.

[SCzEngrgGroup]

Yep they are sending me a new one with the option to return for refund if I am not satisfied (great people at CNC4PC.

Funny how I was lucky to hold 0.5mm before the co version (dummy behind the wheels) and now I am frustrated to no have 0.01 or better.

You do realize the tip of that probe is adjustable to get it centered properly, and you MUST center it when you first receive it, and periodically thereafter to maintain accuracy? They don't come perfectly centered out of the box. EVERY probe of that style requires centering before use.

Regards,
Ray L.