> Rest the coordinates for the bottom camera position by moving nozzle over camera lowering nozzle to board height and centering. Captured via camera position and then tried via tool position.pressed apply.
See the contradiction? If the nozzle's position is the one you want to capture, you also need to use the tool capture button:

Coordindates in OpenPnP are neutral. But all the so-called Head-Mountables on a Head have different offsets. Using the right buttons (camera or tool) is therefore important. And then also the selecting the right tool in Machine Controls is important. Note, if a camera is currently selected, it will take the Nozzle that was last selected, as tool.
It is easier to use Issues & Solutions to do that stuff. Just
enable Include Solved? and Reopen the solution.
Then read instructions extra carefully:

_Mark
--
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.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/9f97e52a-6599-4bb3-8c62-2104e077695cn%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/cc6030b9-4472-44f5-8cd7-33382fc8deb6n%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/c44c5688-0da0-40a0-a998-7e2235a7862dn%40googlegroups.com.
Mike,
I don't understand what the problem is supposed to be, exactly.
I&S is not doing something "magical". All it does is help you set the exact same settings you see, and can manually modify, in the Machine Setup.
What it does, is guiding and instructing and sometimes vision-assisting you to do it right, and - very important! - in the right order. Some things in OpenPnP are built on top of each other like a house of cards. Your head offsets and the bottom camera position are prime examples of this. Those are needed early so we can measure other things later, taking offsets and positions into account. If you change those offsets and position manually, you really, really, really need to know what you are doing. It is easy to quickly invalidate many, many settings, calibrations and captured locations are built on top of these, to make the house of cards tumble down!
Therefore: Stick with I&S, and if this process leaves you in
the rain somehow, report it here, so either we can point you
along, or we can improve I&S. If you start to trial&error
your way out of the situation, things can deteriorate quickly.
Or then be prepared to fully understand the house of cards, and
to live with its delicacy. Needs time an patience for that, hear
from you in some weeks 😉.
_Mark
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/CAJ9bMN7oK_T4-Qtqf0M3bnZQzV3NEcXp0SdmdUapnE7K_WcQ0Q%40mail.gmail.com.
Tony,
I remember this differently, but I might be mistaken.
Remember, we wanted Advanced Calibration to be additive,
rather than to just overwrite (and lose) the uncalibrated offsets:
@Override
public Location getCalibratedHeadOffsets(LocationOption... options) {
if (! Arrays.asList(options).contains(LocationOption.SuppressCameraCalibration)) {
return getHeadOffsets().add(advancedCalibration.getCalibratedOffsets());
}
else {
return getHeadOffsets();
}
}
We specifically wanted that to make it irrelevant whether Precision Head Offsets Calibration or Advanced Camera Calibration comes first, and each can now be revised independently, including IMHO setting offsets manually. For all I know it is the same thing for the bottom camera, in the absence of a head the "head offsets" are just taken as the absolute camera location.
For that purpose, in addition to the raw getHeadOffsets(),
we introduced the second Method
getCalibratedHeadOffsets(). Then we carefully revised
each call to the old getHeadOffsets()
to see if they wanted raw or calibrated. The former obviously for
the calibration itself, the latter for all other uses, including I
believe the bottom camera position. But again, I might be
mistaken.
See also this discussion (it was still "work in progress", though, disregard point 8 for instance):
https://github.com/openpnp/openpnp/pull/1297#issuecomment-953573739
_Mark
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/c44c5688-0da0-40a0-a998-7e2235a7862dn%40googlegroups.com.
Yes, I agree that this will more likely mess things up, than fix
anything. Like I said this is a "house of cards".
But on the other hand, I'm very reluctant to lock users out
completely, and imposing unconditional control by I&S and
other calibration functionality. We could easily lock ourselves
out in circular dependencies etc., especially when something is
not working right and you have to revisit stuff, find workarounds.
Sometimes, a manual calibration is still a valid fall-back, IMHO.
We made sure the Advanced Camera Calibration is a black box. It applies the tilt offsets as a relative distance, and in complete encapsulation together with the pixel warping. Anything outside the box can be treated as before. We also still have the nozzle tip calibration, that actually calibrates the precision bottom camera position each time the machine is homed or a nozzle tip loaded. These third offsets (that may also compensates nozzle tilt) are applied on top of the basic and calibration offsets.
Between those two it can become useful to be able to adjust the camera position, e.g. after changing the coordinate system reference of the machine. Example: after having tinkered with the machines X axis, or the controller homing settings, a user wanted to shift everything in X by 3.2mm to the right. So they could go through all the already captured locations and adjust them manually, including the bottom camera position and the homing fiducial. A zero-sum adjustment, no need to redo the bottom camera Advanced Calibration.
I know this is exotic, but it happens (I did that twice 😁)
_Mark
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/c736c443-bd3a-4942-b833-4df3b2b55b98n%40googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/45db45f2-c00a-7d2d-f136-63175920e048%40makr.zone.