How to do controlled depth milling?

[Shai]

I'm trying to make automatic feeders for PNP using two PCBs as the "walls" of the feeder. I was thinking it would be great to mill into the PCB walls partially to guide the SMT tape. Is it possible to export gerbers and upload to a service like JLCPCB.com so they can automatically mill a certain depth rather than milling all the way through?  

I know I can probably get a CNC and do it myself, but was wondering if a service provider can be directed to do it somehow? Any ideas on how to do this in Eagle? See attached examples of countersunk PCBs I found online.

[you...@ciciora.com]

Those look like multi-layer boards where one layer was routed before they were laminated together.

Just my guess...

- Steve

[Shai]

Turns out PCBWay has an option for custom z-axis milling and counterbores. However, I'm not sure how to specify it in an Eagle file. If anyone knows or has experience with these types of features, please share. I will ask them as well.

[Rene Rettkowski]

Hi, i have done some z milled pcbs. With altium i used a separate layer and defined it as z milling with specified depth in the manufacturing instructions. Pcbs were made by another PCB factory but any manufacturer should understand. Regards Rene

Shai <shai...@gmail.com> schrieb am So., 14. Apr. 2019, 21:37:

Turns out PCBWay has an option for custom z-axis milling and counterbores. However, I'm not sure how to specify it in an Eagle file. If anyone knows or has experience with these types of features, please share. I will ask them as well.

--
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/528ddfb6-7d95-494c-8af3-d53bfdfa65a2%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Shai]

Hi Rene,

For the Z-milling, did you draw a line for the tool to follow and specified width/depth or made a contour outline and specified depth only?

[Rene Rettkowski]

I drew a line, same as routing for panels. A full polygon should work fine for bigger cutouts or z milled areas. You should check the minimum toolsize in the design rules to avoid addional costs. But i would not define the tool, just dont go below the minimum for guiding rails.

Shai <shai...@gmail.com> schrieb am So., 14. Apr. 2019, 21:53:

Hi Rene,

For the Z-milling, did you draw a line for the tool to follow and specified width/depth or made a contour outline and specified depth only?

--

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/29f46d40-8822-440e-b984-ccde7591d58a%40googlegroups.com.

[Shai]

I was thinking of making 0.5mm deep tracks on both sides of the PCBs to guide the SMT tape. There will also be a spring on top to keep it down as SMT tapes vary in thicknesses of course. Would anyone object to having 0.5mm deep tracks? Is it enough or too little for SMT tape?

-Shai

[John Plocher]

On Sun, Apr 14, 2019 at 2:39 PM Shai <shai...@gmail.com> wrote:

I was thinking of making 0.5mm deep tracks on both sides of the PCBs to guide the SMT tape. 

Thoughts:

If the whole tape will slide thru this PCB channel, then it depends on what the component heights are, doesn't it?  0603 dandruff is "thin", SOT's are thicker, SOIC's moreso, Electrolytic caps start to get on the fat side, much thicker than a 1mm guide will allow.

If you are only thinking of grabbing the edges of the tape, with the components hanging down between the guides, this might work, but could the thickness differences and friction characteristics of paper tapes -vs- plastic -vs- splice tapes -vs- fiberglass bite you?

Could the gap between the bottom and top PCB's be an issue if an edge gets pinched in it and binds...?

... or, am I misunderstanding your intent and got my assumptions all wrong? :-)

  -John

 

[Shai]

John,

You're right that all of that has to be taken into consideration. This is why I plan to make the slots maybe 0.05 - 0.1mm deeper than necessary on both sides to accompany tolerance in tape width and make them wider than necessary with the addition of a flat spring that will keep the tape pressed down so it doesn't bounce up. There definitely has to be some give in the feeder to allow the tape to advance.

At 50 feeders, PCBWay estimates it will be $7 per feeder with z-axis milling and countersink holes. Total width of feeder assembled is 11mm, which saves a lot of space!

See attached screenshots.

[bert shivaan]

Can you show the pick location?

--
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/f1e84d7c-19f5-4eb9-9e6b-ecec978d8b1a%40googlegroups.com.

[SMdude]

What if, rather than getting the pcbs milled at depth, get some extra pcbs made to the profile of your groove and attach them to your main pcb on either side. Might be cheaper and you can get some more support(thicker) on the tape.

[John Plocher]

Aha!

I was visualizing the other orientation, with PCB top and bottom, not left and right...   Duh!

Last things that come to mind now are the potential difficulty in feeding new tape thru the assembly and whether or not 0.5mm on each side is deep enough to keep the tape in line.  The latter might be easiest with a couple of inboard full cut PCBs glued (or heck, soldered!) to the ones you show, giving potentially 1.6mm * 2 additional support. 

  -John

[Michael Anton]

Please find a copy of the EIA 481 Taping standard, so you use appropriate tolerances.  The width of the tape can vary by +-0.3mm, so I'd say 0.5mm grooves would not provide much support if the tape happened to be on the narrow end of the specification.

As John suggested, using strips of PCB material to support the tape, and soldering them together would probably work better.

[Shai]

The problem with stacking PCBs is it is more expensive, more labor intensive and you can't align them properly unless using a dowel. Doesn't seem worth it at that point.

[Michael Anton]

Perhaps looking at thicker PCBs, and deeper slots would be an alternative.

If the slots have to be at least 8.3mm wide, to account for tape tolerance, but the tape is ends up as 7.7mm wide (now this might not be possible for 8mm tape, as it is specified as +0.3 -0.1mm, but wider tape is +-0.3mm), then you may not have any support on one side, assuming the tape is not centered.  This could be even worse if the PCBs are not perfectly flat, and that they are not held together accurately enough, meaning the slot could be wider than planned, leading to jamming.  If the nozzle ends up hitting the part a little hard during pickup, there is a pretty good chance that the tape will be pushed into the slot too.

[John Plocher]

Here is the tape guide for my chmt48 , which may give you a visual reference for clearances and tolerances...

Note the cover pulls by my finger, the curved spring provides pressure under the pick point and the washer under the screws captures the outgoing tape in a groove similar to what you are experimenting with...

  John

--

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/3871396a-8499-47c1-963e-41624295bcbd%40googlegroups.com.