Probe multiple points around a feature as part of confirming alignment

[Steven_Hines]

Some of the double-sided PCBs I make with my Voltera have a large number of (small) holes for rivets drilled in them that make precise alignment critical.

I often spend 20-30 minutes fine tuning alignment of the top side of the PCB (the bottom is already aligned as part of drilling). What looks like “exact” alignment for two 0.7mm holes on opposite corners of the PCB – and even later when I confirm alignment by randomly selecting a few other holes – turns out to be very slightly misaligned when I get to the level phase.

The different is that while the alignment phases do only one probe at the center of the feature, the level phase probes 6 points around the circle. And often, the probe “falls into” the hole at one or more of those points due to the slight misalignment. If I were to print anyway, the ink surrounding that hole would partially fall into it and disrupt contact for the future rivet.

Perhaps one (or both) of the alignment phases could have an optional “surround the feature” probe to make alignment faster and more precise?

Thanks!

[MikeL]

Hey Steven

I’m sorry to hear you’ve having trouble getting alignment just right. When it comes to doing the height map we probe multiple points around holes specifically so we can ignore instances where the probe falls into holes.

For alignment we only use two points, one acts essentially as an anchor for the gerber while the second will dictate the rotation of the gerber. For this reasons it’s always a good idea to try and use points as far apart as possible to get the most accurate alignment. The closer the points are to each other the more likely you’ll see slight inaccuracies.

Perhaps I’m misunderstanding your point. I’m not sure how adding more points would increase speed or precision for alignment. If you could give me a bit more detail on what you mean I’d be happy to look into it. If you want to get me directly you can use the chat in the bottom right of this page or through our software. I may even be able to rope in one of our software guys to help. We’re always eager to improve the software however we can.

As a side note from my experience I haven’t found there to be much issue if ink doesn’t completely surround a hole, especially when riveting. Of course it would be ideal for that to be the case but I don’t worry too much if I don’t get perfect rings.

[Steven_Hines]

Thanks for the reply. The reason for probing around a feature would be to see how accurate the x,y alignment is in the alignment stage, before proceeding.

I often think my alignment is perfect even with a USB microscope to verify it on both axes for both holes. (The holes are on opposite corners of the board, as far as practical.) When I start the probing run and it probes around a hole, I find that the alignment is off by a few steps x and/or y. Experience tells me that if I print without re-aligning the ink may fall into the hole on one side and be separated from it on the opposite side. That often leaves the rivet with a tenuous connection to the ink.

I have partially addressed this by increasing the size of pad around the via so there is more ink. But nothing really beats accurate alignment.

Anyway, once you start the probe process, if the alignment isn’t perfect you have to stop the probe, which causes the machine to reset and then re-calibrate before it will start the re-align, adding 30 seconds to each alignment attempt.

[MikeL]

Ahh, I understand now what you mean, thanks for the clarification. Let me poke the software team and see what they think about accommodating something like this.

Automatically probing the perimeter of a hole during the check alignment step seems pretty straightforward. Though I wouldn’t be surprised if my guys came up with something clever to make things simpler all around. Ideally we would improve the alignment phase such that the verification step was less necessary. Anyway, I’ll bring this over to them and we’ll see what we come up with.

Just as a side note, when you have these slight mis-alignments do you find they’re consistent across your board? My concern would be possible issues with the calibration of your unit. I don’t expect this to be the case but it’s worth verifying.

[Steven_Hines]

As far as I can tell my machine is accurate. For instance, after drilling holes, the alignment carried over into the print for the bottom side is perfect.

And with enough effort I can fine tune alignment for the top side to be very accurate as well. But it takes a few iterations.

I think the key is that a 0.7mm hole is simply too big for the probe. When it looks to be in the center of the hole it can still be far enough away from the true center that the print ends up badly misaligned. By surrounding the hole precisely (or something better and more clever) you can be sure that you have located each feature accurately.