Placement of TQFP64

[Melitonas Mikaločius]

Hello,

I cannot place the part. I Calibrated both the TOP and BOTTOM cameras, but the part is not placed on the exact position.

Attached images from OpenPNP logs. What could be wrong? How to debug?

[Marek T.]

Hi Melitonas,

Your outline drawing looks not bad... Questions:

- Other parts like just 0805 you have alligned and placed perfectly? Only with tqfp the problem occurs?

- At each probe the part is moved aside identically down-right?

- When you don't do rotation 90deg but doing placement at 0 rotation - the same effect?

- Do you have turned ON pre-rotate option?

br

Marek

[Brynn Rogers]

three of the leads are a little bent, which won't help matters, but that's all I can see...

You should bend them back carefully with a needlenose and a light touch before you solder the part down for real

[SMdude]

Check that the nozzle to bottom camera alignment is 100% correct at the start of each job.

The other thing that could cause this are errors in the offsets from nozzle to top camera.

Or, the part moves after the bottom camera looks at it.

[Marek T.]

That's why I asked if it is for every parts (then calibration) or only this (then pipeline).

Also his part is rotated 90deg to the left. If he doesn't use prerotate and have runouted nozzles - it may be this effect, that shouldn't occur then at 0 rotation.

[Marek T.]

(then calibration of ofset :-) )

[Melitonas Mikaločius]

Hello,

thank you all for the tips!

Marek:

- I tested with 1206 and did not noticed anything, but later I changed some head parts and maybe messed up something in the config  

- At each probe the part is moved randomly, I think I have problem with prat rotation and centering

- I do not do any rotation, this part is rotated in the feeder. Part rotation in the board settings and in the feeder is 0 degrees

- Will check if I have turned ON pre-rotate option

Brynn:

- I bent the leads when moving the part back to the feeder :).. many tests to come, so will leave it like this for now ;)

SMdude:

- I checked the offsets and I think they are 100% correct

I overlaid the images from the bottom camera after the rotation and from the top camera after the placement - It think the position is the same in both cameras but it is not centered correctly.

And when I do "Test alignment" in part settings it is aligned and rotated differently.. Attaching my images..

[Melitonas Mikaločius]

by I do not do any rotation I mean it is not rotated 90deg to the left

​

[Marek T.]

If it's not rotated then pre-rotate is not important in this case. Also not runout of the nozzles tips problem, small alligning rotations around few deg cant create the moving like you show.

You must confirm yourself if other parts like 1206, 0805,sot etc have the problem or not.

If they don't - don't think about calibration or offset but this part pipeline.

If they do but part is moved always the same it may be offset. Also check verticality of your nozzle moving. Maybe you do allign at high nozzle position and while lovering the nozzle change xy because axix is not ideally vertical.

If they do but part is moved randomly - it's some general mechanical problem.

You can also check if your camera distortion callibration is done well. It may be situation that in the center of camera you have ideal correction but not far from center where big parts are analysed.

[SMdude]

Change your machine speed, check/lower acceleration/stopping rates.

Sometimes at certain speeds the stepper motors will cause vibrations that cause the part on the nozzle to move.

Also, if you have poor vacuum and or vac leaks in the nozzle holder parts will easily move.

[Mike Menci]

Hello, 

Have you checked your relation Up-camera to Down-camera at reference level "0" ? 

In other words does your Up- camera point to the same spot at "0" level = Up-down View matches? 

Mike

[Mike Menci]

see this

[Melitonas]

Hello,

thank you for the suggestions. After long break I resume my PNP project.
And I still can not figure out what is the magic problem with the placing rotation :)

• My Config files: https://github.com/olablt/openpnp-cfg
• Video showing successful alignment tests: https://drive.google.com/file/d/1JKaA7FSFAz_GOMUAGQdikt8HTZxcLjx7/view?usp=sharing
• Video showing full pick and place procedure using bottom vision: https://drive.google.com/file/d/11Rabxa3Ha54t4WB0RNulA1xtoPVT9u7i/view?usp=sharing

What are the next steps to debug the problem?

regards,
Melitonas

--
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 post to this group, send email to ope...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/7a72086b-d5c9-47f4-a486-5d8c1009ab58%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Melitonas Mikaločius]

got it. the problem is in too small juki nozzle and the vibration of vacuum pump.

On Monday, January 7, 2019 at 5:57:41 PM UTC+2, Melitonas Mikaločius wrote:

Hello,

thank you for the suggestions. After long break I resume my PNP project.

And I still can not figure out what is the magic problem with the placing rotation :)