Repeatable placement issues

Started by trev, July 28, 2019, 05:30:30 PM

Previous topic - Next topic


Hi all

Does anyone have trouble with components losing their bias data?

Usually if I have placed components on a board, then exit out and adjust a bias, sometimes thats when things can go wrong. Also it was recommended to me to re-center the camera position at least once in the morning and then again after lunch. I did that and it threw a few components off. Strange thing was it was just a random couple. Maybe re-centering resets biases sometimes?

Also I wondered if when I load a board and use auto fiducial, sometimes it upsets some component biases. I dont have enough proof but wondered if a board was off by a 2mm say and the software auto corrects, maybe that causes some calculation overflow...

I feel sometimes I am chasing my tale. I have to just sit there a nudge components to ensure they flow.

Does anyone get repeatable placement on say a batch of 50 boards or does every other board end up needing a tweak? If it needs tweaks then so be it but it would be great to know the limits.

Also when I try to change biases I get screen corruption pretty much every time. Is there any specific video card/motherboard hardware or driver setup that limits this. I was told I am running an old software and that later versions fixed it but I am using the latest version from this forum.

Any help would be appreciated as the machine seems to be running well other than me failing to get biases set firmly.




Thats one bit I've never had problems with, once the bias is set, it sticks...
I run both small batches of a few dozen, and larger runs of 150 boards at a time
I use auto fiducial and sometimes have to manually move the head when a boards fiducials are ouside the limit..
but I never noticed a problem.
using the later windows 98 with USB support with the latest version of the software as far as I know



Thanks, Chris.

I find that once it's in a groove it's fine. Its repeatable but sometimes changing the odd thing can upset it. Am sure its software. I wondered if my PC is under spec'd and maybe its memory issues.



By "screen corruption" do you mean the well-known thing where the screen display becomes offset, or something else?

Incidentally the quickest way to recover the former is to load any other job, ignore the feeder error messages, then reload the job you were doing. This avoids the re-homing delay when starting RVPlace



I have found so far that if I do not zoom in or upset the default view when entering that screen it seems to not throw a wobbler and ccorrupt. As soon as I zoom in on an area the offset appears.

Placement is now repeatable. I think its just getting familiar with the machine and a friend helped me check a few things. I think the tool pickup position was not quite right which was not a good thing mechanically for the machine. Also it wasnt always picking components up cleanly - possible due to tool position stuff. I did find that once I had the components picking up, I had to lock the x and y on the component because occasionally it would just drift off position. It was strange because it would sometimes just start missing the pick up point on a tape. As soon as I locked X and Y I never had any issues again. Placement was pretty good for 0603.

Also the auto fiducial would align off position sometimes. Think this was due to the threshold not being great. I might make Mike's LED board for that camera and see if that makes it more reliable. But if the auto fiducial doesnt align quite right then this obviously would create component position skew.


the tool pickup position is critical as any error there can cause the odd step to be lost
It was also a contributory factor to me losing z axisposition early on... a badly worn rack and it caused the machine to jump a tooth.

have a look at the method I used to set the positions using a couple of steel balls, it actually works !

    Mike's Forums »
    Versatronics RV series Pick & Place machines »
    RV Hardware »
    setting tool positions