advanced camera calibration: position of the up-looking camera

[Jan]

Hi all!
I just wonted to verify the position of the up-looking camera after
advanced camera calibration by placing the nozzle tip over its
configured position. I found, that the nozzle is off by quite a distance
(please see the attached screenshot). Would anyone please tell me if
that's ok or what went wrong? I've attached the log of the advance
camera calibration. (It did the calibration twice with the same result)
Thank You!

Jan

[mark maker]

I leave that one to @Tony Luken.

[tonyl...@gmail.com]

I believe that is moving the nozzle tip to the uncalibrated position of the camera.  Try using the dropdown menu (right click) in the up-looking camera view and select "Move Selected Nozzle to Camera" to see the nozzle tip at the calibrated position of the camera.

[Jan]

Hi Tony!
Sorry for the late response!
Yes, if you use the suggested function, the nozzle tip is moved closer
to the center of the camera. (see attached screenshot, I also moved it
by the indicated distance (DRO) to bring it really into center) This
leads to some more questions:
1) why does advanced camera calibration shifts the center of the
displayed image? I'm assuming any tilt != 0 would result in a Z
dependence of the position. But for what Z is the correction applied?
The function you pointed out moves the nozzle to the Z position the
camera position was recorded with. (switching advanced camera
calibration off moves the nozzle tip by (0.25,1.55,0)mm) I would assume,
that the correction is linked to that position such, that its image
stays the same.
I'd suggest to add a remark next to the position coordinates which
explains why the camera/nozzle relation is shifted now.
2) If I move the nozzle up and down, its position moves in x and y as
well. That's expected for a tilted camera/nozzle. As there is 3d
calibration in action, couldn't that be correct?
3) at which Z level will bottom vision take place? The calibration was
performed at the "Default Working Plan Z" (-6.3mm) and 2mm above. PCB-Z
is about -8.8mm (for 1.6mm PCB thickness, lower for thinner PCBs).
4) I assume, that best performance (xy-precision wise) is still
archived, if bottom vision takes place at PCB-Z. Is that still true with
advanced camera calibration in action? How can I archive this or how can
I force advanced camera calibration to take this level as the/my
preferred level.
5) Can you disentangle camera tilts from nozzle tilts? The xy-movement
provides a reference plane, nozzle tilts have a fix angle with respect
to z, while camera tilts don't as each point is connected by a straight
line with the camera center (surface of the sensor chip). I wonder
whether this divergence would be enough.
It would be great if you could rate the results of the advance camera
calibration automatically. I'd love to see some remarks if you think
like a tilt angle is undesirable large and/or the estimated location
accuracy is to low/to high. Please also provide rating results if you
think the values are good.

Jan

On 21.01.2022 18:17, tonyl...@gmail.com wrote:
> I believe that is moving the nozzle tip to the /uncalibrated/ position

> of the camera.  Try using the dropdown menu (right click) in the
> up-looking camera view and select "Move Selected Nozzle to Camera" to

> see the nozzle tip at the /calibrated /position of the camera.

> cameraViewMenu.png
> On Friday, January 21, 2022 at 7:23:17 AM UTC-6 ma...@makr.zone wrote:
>
> I leave that one to @Tony Luken.
>
> On 21.01.22 14:20, 'Jan' via OpenPnP wrote:
>> Hi all!
>>     I just wonted to verify the position of the up-looking camera
>> after advanced camera calibration by placing the nozzle tip over
>> its configured position. I found, that the nozzle is off by quite
>> a distance (please see the attached screenshot). Would anyone
>> please tell me if that's ok or what went wrong? I've attached the
>> log of the advance camera calibration. (It did the calibration
>> twice with the same result)
>>     Thank You!
>>
>>     Jan
>>

> --
> You received this message because you are subscribed to the Google
> Groups "OpenPnP" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to openpnp+u...@googlegroups.com
> <mailto:openpnp+u...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/openpnp/febb3a9c-5e94-46e3-81d7-3ef8d57a8284n%40googlegroups.com
> <https://groups.google.com/d/msgid/openpnp/febb3a9c-5e94-46e3-81d7-3ef8d57a8284n%40googlegroups.com?utm_medium=email&utm_source=footer>.

[mark maker]

Good points, I let Tony explain in detail, just a few remarks.

Re 1) "why does advanced camera calibration shifts the center of the displayed image?"

If you backtrace the light-rays that a camera receives, you have a a fanned-out cone of (reverse) light beams. From all these "rays" the calibration now picks the one that happens to be perfectly perpendicular to the machine X/Y plane. Because the rays are fanned-out at all angles, there is always one with that property (assuming your camera is not too tilted). That one ray, or rather the corresponding pixel in the image then forms the new camera center:

I explained more about that here.

https://groups.google.com/g/openpnp/c/_BqWBSMAdt8/m/rNyOVROUDwAJ

re 2) " If I move the nozzle up and down, its position moves in x and y as well."

The Advanced Camera Calibration calibrates the camera, not the nozzle. Because we always hold the part into the focal plane of the camera, i.e. we lift up the nozzle by the part height, this is also all that we need, i.e. we do not currently need 3D calibration in the bottom camera. But see (4) about nozzle/nozzle tip calibration.

If we really wanted to push this further, we would need to calibration for planarity of the nozzle tip, i.e. how tilted it holds the part, and how that tilt changes with nozzle rotation, which might start to matter for very large parts. Not sure, this would make a big difference in reality, though.

3) "at which Z level will bottom vision take place?"

If you look closely, you see there is an Auto Focus operation in preliminary bottom camera calibration, that makes sure the bottom camera Z is at the in-focus point:

https://youtu.be/md68n_J7uto?t=269

Re 4) " I assume, that best performance (xy-precision wise) is still archived, if bottom vision takes place at PCB-Z."

Yes, it is still recommended. To achieve that, make sure your camera focuses at that Z height, so (3) will nail it.

But if that is not the case for your machine, its not too bad. With Nozzle tip calibration (a whole different but equally important topic) a bottom camera Z that is different becomes less important, because we have tool specific offsets applied:

https://github.com/openpnp/openpnp/wiki/Nozzle-Tip-Calibration-Setup#what-is-the-bottom-camera-location-and-how-can-it-vary-between-nozzles

5) "Can you disentangle camera tilts from nozzle tilts? "

Tony probably could, but see (2) why this does not make much sense, and see (4) for how we covered what really matters.

_Mark

[tonyl...@gmail.com]

I think Mark pretty much covered it.  Just a couple of things:

> I also moved it by the indicated distance (DRO) to bring it really into center

Did you try rotating the nozzle (before you did the jog indicated by the DRO) to verify that offset you observed wasn't really nozzle tip runout?

>I'd suggest to add a remark next to the position coordinates which explains why the camera/nozzle relation is shifted now.

Yeah, the GUI is a bit of a mess because we're still supporting two different ways of doing things. Once we decide to "go all in" with the advanced calibration, we will probably remove a lot the legacy settings as they can be a source of confusion.

About the only other thing I'd add is that I'd still like to see compensation for Z-axis tilt be added to OpenPnP.  It should be fairly easy to measure - bottom camera calibration already measures it but it's probably better done by measuring the nozzle offsets at two different Z heights (basically do the confetti thing not only at the primary calibration Z but also at the secondary calibration Z.  I've actually tried adding the compensation to my machine by adding a Linear Transform Axis for my Y-axis (I already had one on my X-axis).  With appropriate setting of the Z and Offset factors on both of the X and Y Linear Transforms, I could get the nozzle tip to move straight up and down in the bottom camera's view.  The only gotcha is that X and Y-axis backlash compensation is getting applied after the nozzle tip moves in Z.  That's ok when moving in the positive Z direction but not when moving in the negative direction.  We'd need to make sure that any necessary backlash compensation gets applied before any negative Z motion takes place.  Mark - would that be difficult?

Tony

[mark maker]

> We'd need to make sure that any necessary backlash compensation gets applied before any negative Z motion takes place.  Mark - would that be difficult?

No. We would need to calculate the "Safe Z coordinates at given X, Y, Z", by not just replacing Z, but by approximating by using AbstractHeadMountable.getApproximativeLocation(Location, Location, LocationOption...) with the  LocationOption.KeepX and LocationOption.KeepY flags.

So the axis would counter-compensate the Z non-squareness, i.e. there is no X/Y backlash compensation taking place.

This is easy in case of the often used MovableUtils.moveToLocationAtSafeZ(). But there is a lot of hand-rolled code also doing something with Safe Z, so this would have to be refactored carefully. The code must know that the approximated X/Y coordinates at Safe Z must not be used directly (rather the original intended target X, Y).

_Mark

To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/0775eda0-62c0-44b7-9782-8ce37631c6c9n%40googlegroups.com.

[tonyl...@gmail.com]

I suspect a tricky bit is when you don't know the final Z coordinate like when probing.  Probably also need to think about manual jogging in Z.

[mark maker]

There is nothing as good as having a machine, where everything is at the same Z (i.e. withing a few Millimeters). Everybody should really try hard to achieve that.

_Mark

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/ff3dceff-e77c-4242-bd25-8da89010c90an%40googlegroups.com.