[hudsonlighting]

I'm having problems with milling my PCB's. I mill both FR4 and aluminium back PCB's but upon testing with a continuity tester I find that all my tracks are effectivley all connected together.
I have double checked the milling goes through the copper clad section with a magnifing glass. I have even gone over the boards deeper and deeper re testing the continuity as I go.
It does get to a point where the continuity stops but at that point with a 60 degree engraving bit the isolation tracks are near 2mm wide. Does anyone else have this problem.
I am starting to think it could be small fragments from the engravings but its a perfect short where I thought there would be some resistance in this case.
I have been able to make some boards but my sucess rate must as low as 5-10%. The last board I made I had to go around it for 2 hours with a stanley knife and it was only 2 long tracks.
Any Ideas?

[eSilviu]

I'm having problems with milling my PCB's. I mill both FR4 and aluminium back PCB's but upon testing with a continuity tester I find that all my tracks are effectivley all connected together.
I have double checked the milling goes through the copper clad section with a magnifing glass. I have even gone over the boards deeper and deeper re testing the continuity as I go.
It does get to a point where the continuity stops but at that point with a 60 degree engraving bit the isolation tracks are near 2mm wide. Does anyone else have this problem.
I am starting to think it could be small fragments from the engravings but its a perfect short where I thought there would be some resistance in this case.
I have been able to make some boards but my sucess rate must as low as 5-10%. The last board I made I had to go around it for 2 hours with a stanley knife and it was only 2 long tracks.
Any Ideas?

1. you can test for shorts with a high curent power source (PCB will start to heat ar the short-circuit point)
2. My results are always good. I use 45 degree, 0.1mm V bits from China, with isolation of 0.15-0.2mm. Depth of mill is always 0.11-0.13mm. Deeper will just wear the bit faster.

[chuck99z28]

I use a low power microscope and have seen what you're talking about.

You can get small chips that short out the traces.

Try scrubing the board with some lightly abrasive cleanser.
Don't be afraid to scrub hard. Use a backing board to prevent bending thin pcbs.

I use a product call "Bar Keepers Friend". But any of that powdered scrubing cleanser should work.

Wash off with a large volume of water. You're trying to scrub out the fragments and wash them away.

Compressed air is best for drying and blowing out the tracks.

Maybe slow down your ipm too. or run the same g-code twice. Once to mill, once to clear the track.

There is a how to video on my site. The video is on the intial cleaning, but you need to do this between every step almost.
Especially if you're plating through holes.
http://www.voodooengineering.com/ind...g-the-pc-board
My site isn't completed but there is a page on cleaning. I'm working to document each step but stuck working on something else right now.

[Zig]

Hi Hudsonlight,

it is a problem we all face.

The poor swarf clearance feathered copper edges and residual shorts are caused in part by the ductile nature of copper and certainly by inferior cutting tool.

Change Your tool to one with a negative ( down cut ) cutting edge.

Go away from D bits, they are a general purpose engraving tool not a PCB structuring tool.

Use somenthing like precise bits or one of the other sources which supply like tool. Pay particular attention to the shape of the leading ( cutting ) edge of the tool.

The cutting edge must lean into the material as it cuts it so that the copper laminate is sheared downwards into base material.

This way copper will be captured against base material sheared and removed.

D bit has a vertical edge and it may give you acceptale result BUT NOT FOR LONG> It will fail to shear the copper as soon as it looses initail keennes of the edge and you will deform copper without breaking it away.

Copper in its pure annealed form is extremely ductile and very difficult to cleanly cut. The net outcome is ofcourse excessive use of the stanley knife.


You may also try using a scourer ( of the honey i burnt the lunch let me scrub the pans variety) to scrub the surface of the laminte PRIOR TO MILLING.

Yhis will tend to work harden the copper sheet and produce a cleaner cutting outcome.


If all the above works out for You, You owe me a beer.

P.S. My preference is for German made LPKF tools but I use Precise EM2E8-0625-60V bit. Remarkably similar performance at a fraction of the cost. Precise ought to send me a beer as well for such a wonderful write up...JUST KIDDING ( doesn't hurt to ask though huh? )

http://www.precisebits.com/applications/pcbtools.htm and no .. there is no pecuniary interst at the moment

[TDA]

P.S. My preference is for German made LPKF tools but I use Precise EM2E8-0625-60V bit. Remarkably similar performance at a fraction of the cost. Precise ought to send me a beer as well for such a wonderful write up...JUST KIDDING ( doesn't hurt to ask though huh? )

It might be bad by the time you got it. But if you are ever in Colorado I'll buy you one.

:cheers:

Also, (and on topic) Zig is right. You need a tool that will leave a clean cut. Try to find something that has actual cutting edges and not just a ground flat.

[hudsonlighting]

Well guys I'm still having the same problems. Well not with the FR4's, with these I can guarantee that they will be fine every time I do one. But the aluminium backed ones are a different story. I purchased both coated and non coated bits from precise bits and have even ensured my machine runs around the tracks 3 times each time 0.01mm deeper than the last pass and also the same depth as the first pass. I have actually found that the copper becomes attached to the aluminium which if it happens in more than one place makes the board completly useless.

I have also scrubbed the boards before and after machining.

Any ideas?

[hudsonlighting]

Out of interest what speed and feed rates does anyone use? I am currently doing 550mm per minute at 22k rpm

[chuck99z28]

I don't know if it would help you or not.

But when I mill lexon or acrylic I spray the board with wd40.
It helps a lot.

[Zig]

Hudson,

A number of issues might be in play.

Try no to touch aluminium with your tool. Aluminium tends to bind to the tool and may cuse the tool to cut imporperly.

Try not to penetrate the dielectric layer at all.

How do You control the tool penetration and how do You maintan flatness of the laminate while cutting?

I have made some copper backed laminate and machined it without the kind of problems You are seeing.

The dielectric was formed by two sheets of 0.1mm prepreg ( the adhesive used in laminating FR4 multilayer boards).

I can supply You with metal backed laminate made to Your spec.


EDIT: You might want to bring Your feed rate down to 300 mm/minat the spindle speed You are cutting at.