Liteplacer program crashes regularly

[JuKu]

> When I move the gantry left, the image moves to the right, and when I move it up, the image moves down.

This is most likely the issue. The orientation mrandt shows is the right one.

> The dot is about 7mm to the right and up from the XYZ location. Is that okay?

Yes. As long as it is in view and not negative (so limit switches are not triggered when trying to move on top of the dot), you are fine.

> From the above image, can you tell if this is likely to be an issue later on?

Yes, I'm afraid so. My table is smoother than that and worked like captha for the hole recognition. I'm now using sticker rulers and extending the sticker paper under the tape holes.The downside is that I need to decide up front which positions are for black and which for white tapes. Obviously, white background and white component tape don't work that well.

> but when I alter the zoom obviously the size of the box changes. Is this taken into account automatically

Yes. There is display zoom in the lower right corner, which only zooms in to the image the system is seeing. With that of, the display shows what the image processing sees. If there is measurement zoom, that is taken into account. I noticed you are using 0.7 as zoom, which makes the image smaller, decreasing resolution. Try 1.0 or disabling it. (But as noted, the issue is in camera alignment.)

Mrandt is doing my support job very well.

[mrandt]

...just trying to help if I can, Juha

Cheers
Malte

[tek]

http://i.imgur.com/ZULD99z.jpg

http://i.imgur.com/L40vJLW.jpg

Please also check your camera orientation (moving gantry left should move the image to the left, moving it away from you should move image to the bottom).

When I move the gantry to the right, the XY drawing moves to the left of the display. When I move the gantry away from me, the XY drawing moves to the bottom of the display. So, this is wrong? I don't understand how that's possible or how to fix it. I feel like I'm missing something blindingly obvious.

I'm not sure what material the surface is I'm afraid. We were looking for something that would provide ESD protection, and I dont have details of the purchase available.

Again thanks for your continued support.

[mrandt]

Judging by your pictures, the camera is oriented correctly.

If you move the gantry away from you (Y+), then the crosshair in the image should move in the direction that the Y arrow on your paper is pointing. I guess one could say, the image itself moves downwards. Like scrolling up in a window

Similarly when moving the gantry to the right (X+), the crosshair in the image moves in the direction of the X arrow.

I think you homing mark is huge - mine is about 1mm in diameter. Also, it seems rather far to the right... Maybe it's worth playing with that?

Make the mark smaller and put it a bit closer to the place, where your camera is when machine is at 0,0,0 coordinates. This way, you could also do away with any measure zoom below 1, which in fact shrinks the image. Try detection without measure zoom at all, or use a small positive value - like 1.5 or 2.0.

Also try the following:
1. Go to "Basic" tab
2. Click "Home XYZ" button to move machine to mechanical 0,0,0
3. Switch to "Setup Cameras" tab
4. Select down looking camera - can you see your homing mark and is it near the cross, preferably a bit to the top right?
5. Click "Homing" - "To Display" and switch on Circle Detection - is your homing mark detected?
6. Click "Homing" - "Measure" - you should get a corresponding message with the relative coordinates of the circle in the message window at the bottom left.

Does a full home cycle work afterwards?

If not, I think it would help Juha diagnose, if you attempted a full homing cycle and copied the complete output from the message window box on the lower right. Just click it, hit Ctrl + A (select all) and then Ctrl + C (copy to clipboard).

[JuKu]

> When I move the gantry away from me, the XY drawing moves to the bottom of the display. So, this is wrong? I don't understand how that's possible or how to fix it.

Your motor is wired incorrectly. Turn power off and swap the wiring on one of the motor coils. In other words, swap A1 and A2 on X motor on TinyG. http://www.liteplacer.com/wiring-the-motors/

[tek]

I feel like I'm being told conflicting info by both of you. Mrandt, the crosshair follows the paths that you describe above. But JuKu you tell me the wiring is wrong?

F5 sends gantry left
F6 sends gantry right
F7 sends gantry away
F8 sends gantry toward
F9 rotates needle anticlockwise
F10 rotates needle clockwise
F11 raises Z
F11 lowers Z

Surely if the wiring was wrong, this would not be the case?

__

I've made a smaller homing mark for testing with the white paper at about 2mm diameter, and followed your instructions (I've been following the same instructions previously)

http://imgur.com/a/c3l1S
Note: at part 3 the circle detection was flickering slightly this time. That being said it has failed before while highlighting the circle solidly.

That is the process. Still ended with the same error as usual.

I did already collect the log details for JuKu in an email yesterday.

[mrandt]

I cannot comment on your e-mail discussion with Juha

But if F5 moves left and F7 moves forward (away from you), wiring should be OK I guess? Also, if it was incorrect, I would assume the mechanical homing would not work.

Circle detection should work reliably though. Make sure the mark is as round as possible. I just used a laser printer to print a solid circle on a piece of white paper which I then taped to the table.

If the orange circle "flickers", try to adjust the "Threshold" filter value - this usually helped in my experiments.

What is the result of clicking "Measure" in Camera Setup?

[JuKu]

Sorry, I got confused. Gantry moves away, drawing goes down is of course correct. The image matches what the gantry is doing, and that is as it should be.

Your image shows "measure" result as X:4.725, Y:4.2. If you move the gantry there, is the circle in the crosshair?

[tek]

Sorry, I got confused. Gantry moves away, drawing goes down is of course correct. The image matches what the gantry is doing, and that is as it should be.

Your image shows "measure" result as X:4.725, Y:4.2. If you move the gantry there, is the circle in the crosshair?

Found the issue. When you asked what "measure" showed vs actual positioning of the crosshair, it turned out everything was moving double the distance needed and I hadn't realized. I put a ruler down and used shift + f6 and instead of moving 1mm it would move 2mm, but X would still show as being X:1.0.

I hadn't even noticed the Step angle option up until this point, but it was suggested I change it from 0.9deg to 1.8deg and the next time I tried to home it it found the circle instantly. I guess I knew it would be something simple that I'd overlooked. Pretty embarrassing. Oh well.

Thanks so much to both of you for your patience in helping me with this problem. No doubt something else won't work later on so I'll be sure to come back to bug you some more

[JuKu]

The embarrassment is on me, and it is not very small. I assembled part of a batch using 1.8deg/step motors and shipped that out. I couldn't know which customers had wrong motors, so I tried to contact everyone from that batch. Either the message never reached tek or I never got the answer (and so assumed that kit wasn't one of the faulty ones). Anyway, the error is mine and I apologize for the extra trouble. I'm arranging the replacement.