Skip to main content

Site Navigation

Your Account

Choose Language

How much accuracy is everyone seeing in their unit?

Hey everyone! One of the major reasons I invested in boxzy over any other unit on the market was for it's advertised accuracy which, if I'm remembering correctly, was said to be +/- 0.1mm (correct?). Not intending to put this to the test, I ordered a dial gauge from amazon to help with leveling the print bed because I wanted to be RIGHT on the money. I started out like this:

1. Zero the dial gauge.

2. attach the gauge to the empty carriage with clamps.

3. Take height measurements at: (0, 85), (85, 85), (165, 0), (165, 85), (165, 165)

(essentially at the screw locations plus some mid points)

4. Adjust and repeat until level.

between moving to each point I included "G0 Z30" and "G0 Z0" so that the gauge wouldn't drag on the surface and risk being damaged. It was between moves that I thought I'd go ahead and repeat each measurement (G0 Z30 and G0 Z0) and check the accuracy. At (0,85) it was absolutely spot on, but as the x value increased the accuracy decreased reaching a value range of .03mm at (85, 85) and .05 at (165,165). I'm wondering, has anyone else tested the accuracy of their machine? Is something just off with mine? Also, because I'm testing at the upper limit of the z axis, how much of this is being introduced by the limit switch? I would think very little given that location (0,85) was perfect. Let me know what you guys think.

Update (04/06/2016)

Master Yoda: I should have explained that better. Here's some photos:

Block Image

Block Image

The probe extends through the empty module bay, neither of the clamps touch the carriage support rods nor the frame at any time during travel.

Update (04/07/2016)

Alright, wrote a little snippet of gcode to make leveling easier with my setup:

G0 Z30 F2000

G0 X0 Y85 F2000

G0 Z5 F2000

G4 P4000

G0 Z30 F2000

G0 X165 Y165 F2000

G0 Z5 F2000

G4 P4000

G0 Z30 F2000

G0 X165 Y0 F2000

G0 Z5 F2000

G4 P4000

This removes any error experienced by the limit switches. Z axis is moved down to 30 between each move to prevent dragging the tip and pauses for 4 seconds at Z5 to give me time to write down the level values. Worked great!

As for accuracy the worst I saw was 0.4mm range between (165, 0) and (165, 165) which is a bit poor, but not unlivable. I'll consider ordering new bearings in the near future.

On another note, what's the best way you guys have found to set initial nozzle height? Up until now I've been putting a pair of sticky notes under the nozzle and then clamping down. This is because I've found that when I tighten the module clamp, the nozzle moves down ever so slightly so when it's firmly in place only 1 sticky note fits under the nozzle. Is there a better/more repeatable method? (I really hope so)

Answer this question I have this problem too

Is this a good question?

Score 1

Comments:

"2. attach the gauge to the empty carriage with clamps."

Could you illustrate your gauge clamping technique?

by

That's a nice setup!

by

Thank You for a quick update.

Rig looks good.

I was thinking of making a gauge sattle that could clamp like any other adapter.

I also like the fact that this gauge has a data output port.

(http://forums.reprap.org/read.php?159,55...)

by

I personally use the laser clamp for my nozzle height. I don't adjust it each time. You could print one fairly easily just for that purpose.

by

Hmm. That's a good idea. I'll do that, but first I have to post another problem preventing me from doing so XD.

by

Add a comment

2 Answers

Given that's you're having speed issues potentially related to a stepper pulse issue, and the step pulses coming from the arduino are what control accuracy in the motors, I'd retest once you've changed the Arduino. Movement in the X and Y shouldn't effect Z repeatability. I'll be interested to see what you come up with after installing a new Arduino. Have you tried the test with out hitting the Z limit switch; As in Z5 to Z30 and back?

Was this answer helpful?

Score 0

Comments:

I don't know if you saw the update to the laser power post, but the speed issues disappeared when I replaced the arduino. I don't disagree that there could be a pulse issue, but it would really suck if that was happening 2 arduino boards in a row. I'll give the Z5 to Z30 a try.

by

New Arduino! That's great. Shouldn't be any issues with the new one, I didn't see the update. Let me know what the result is when you move to a position without the limit switch. There shouldn't be any correlation between X Y position and Z accuracy. On your print platform are the springs toward the high side, or do you have them tightened far down? It's possible the platform is moving slightly on the bolts and springs.

by

They're not all the way down, but they're tightened pretty well. I surely hope the platform isn't moving under the light touch of a dial gauge, that wouldn't bode well for printing.

My worry is that there is play between the non-ball-screw z-rods and the bearings to the point that the accuracy losses. This makes sense (at least in my head) given how the accuracy got worse the further I measured from the Z ball screw.

by

Definitely a possibility. THK brand LMK12UU would help a lot if there is play. They are much better bearings in my opinion. Also, try checking the ballscrew clamp for the Z. It's the two bolt clamp underneath the platform. If it's too loose it will cause a loss in accuracy. Gaps should be even and bolts should be pretty dang tight. Too tight will increase turning resistance on the ballscrew. So once it starts to increase turning resistance (using your hand to turn the ballscrew from underneath with the power off), you've just gone too far, back of until it frees back up. That could help. There is a weakness to the single time probe of the limit switches too. I've asked them to add a multiple probe home function (so it probes at least 3 times), because the switches are lower rated than the machine accuracy. This would increase the accuracy margin by around 50% on the homing sequence.

by

Add a comment

Hi Boxer,

For leveling, I use a business card. I clamp it down with the card between the nozzle and the bed, then I make sure it can just slide in and out of there.

Was this answer helpful?

Score 0
Add a comment

Add your answer

Drunken Boxzer will be eternally grateful.
View Statistics:

Past 24 Hours: 0

Past 7 Days: 0

Past 30 Days: 0

All Time: 192