[zephyr9900]

(If this subject has come up before, I'm sorry, but searching for "PCB" brought no results)

I want to build a small router to machine PCB material. Not just isolation routing, but actually machining 1/2 way and all the way through the material for model railroad trackwork bases. I have two BSA polymer-nut leadscrews and a supply of nice NOS 10mm Thomson profile linear rails and bearings (both ironically now part of Danaher) squirreled away for just this use.

I am worried about the abrasive dust generated in machining the FR4. My first inclination is go with a "tall wall" Momus-style gantry and keep all the works well above the cutting. But the arguments for the better stiffness of a moving-table bridge design are well-made. However, I have yet to see a well-enclosed (bellows cover or elsewise) moving table mechanism.

My working area does not need to be that large. 4" x 12-14" is all I need. I will be using a Proxxon IB/E (which I already have as my Tormach auxiliary spindle, retrofitted with semi-ceramic nose bearing) as the spindle, or maybe spring for one of Paul Jones' CNConabudget spindles. I will be specifically using a 1mm carbide cutter for the whole process. I do have a decent supply of surplus 1/2" and 1" aluminum tooling plate for the structure, as well as a nice piece of 2" square x 1/8" wall square aluminum tubing which may play well into the bridge or gantry cross-structure.

Do any of you have direct experience with machining PCB which would lead to useful advice?

Thank you,

Randy

[awerby]

I'd look for another material if I were you. Have you looked into Paperstone? It's a phenolic-paper product that machines well and is very hard and stiff.

That fiberglass-reinforced board is hard on cutters, bad for machinery and injurious to your health. Raising the gantry won't make the problems go away. Normal PCB routing is bad enough, but you're really asking for trouble if you insist on removing a lot of board, rather than just skimming off the copper and drilling some tiny holes. That said, a fixed bridge and moving table design would be better than a moving gantry; it's stiffer, the weight of the board isn't a problem, and the extra footprint of such a small machine isn't going to take up much space.

If you insist on routing all that fiberglass board, make sure to protect the machine's sliding parts with bellows, invest in some diamond-coated tooling, and get yourself on the list for a lung transplant now - it takes a while to find a donor...

Andrew Werby
www.computersculpture.com

[zephyr9900]

Thank you for your reply, Andrew. In the UK they have phenolic-based PCB material available, which would be ideal, but I have not seen it here in the US. The copper face (single-sided PCB is fine) is essential because that is what the rails are soldered to.

When I have done limited prototyping on the Tormach, I have hand-held a spot vacuum tip next to the cutter to suck up the swarf but I would make a woodworking-style vacuum foot for this machine. I have seen a hookah-style arrangement in the past that sucked through a water jar to trap most of the dust before it even got to the vaccuum (on which I would use a plaster-dust bag of course).

Randy

[louieatienza]

I haven't done PCB routing yet, but I too would recommend the moving table/fixed gantry since the travels are very small. I have one here made of 3/4", 7/8" phenolic, milled t-slot extrusions, aluminum tooling plate, profile rail... the work cube is 14.4 x 8.5 x 5.5. I am selling it to offset the cost of a new mill I am purchasing.

http://www.cnczone.com/forums/cnc_wo...er_mill-8.html

I've cut aluminum up to 1" thickness for stuff like new nut carriers for my larger CNC, GEEX and G10/FR4 up to 1/2" thickness, with no problems. I do have a dust shoe but phenolic dust really clogs the filter of the vacuum. I really need to build a cyclone separator....

[zephyr9900]

Thank you for the reply, Louie, that is a good-looking machine. Did you ever fit bellows to protect the table leadscrew and bearing rails? I still haven't found any moving-table machine with one fitted. I guess another possibility would be a window-shade cover like on the Z axis of an Accuslide lathe conversion Z axis.

Randy

[wizard]

(If this subject has come up before, I'm sorry, but searching for "PCB" brought no results)

I want to build a small router to machine PCB material. Not just isolation routing, but actually machining 1/2 way and all the way through the material for model railroad trackwork bases. I have two BSA polymer-nut leadscrews and a supply of nice NOS 10mm Thomson profile linear rails and bearings (both ironically now part of Danaher) squirreled away for just this use.

Sounds like a nice start. Do look up those 10 mm slides to see if they are the ones that can only take a load in the downward direction. Some small linear rails will not work properly if used improperly.

I am worried about the abrasive dust generated in machining the FR4. My first inclination is go with a "tall wall" Momus-style gantry and keep all the works well above the cutting. But the arguments for the better stiffness of a moving-table bridge design are well-made. However, I have yet to see a well-enclosed (bellows cover or elsewise) moving table mechanism.

The dust as you know is very abrasive and darn right nasty on tooling. Because it is dust it will get every where even with a good dust extraction system. The good dust extraction is needed obviously but you need to go industrial with a PM (preventative maintenance) plan. In a nut shell your PM program needs to be aggressive with cleaning your machine. That means whipping down the lead screws and linear rails after every use.

As far as the lead screws and slides go you can run them dry or lubricate them often. Either way wear or just getting crude up will will cause you grief. I'd opt for a very light grease or oils lube that is applied often to attempt to flush out crude before it becomes a problem. The problem is if you get a little bit if garbage in the bearings it can lead to the balls skidding which can very quickly ruin a slide.

My working area does not need to be that large. 4" x 12-14" is all I need. I will be using a Proxxon IB/E (which I already have as my Tormach auxiliary spindle, retrofitted with semi-ceramic nose bearing) as the spindle, or maybe spring for one of Paul Jones' CNConabudget spindles.

Shoot for a Z axis saddle that you can bolt different spindles to. In the end once you have a machine you will want to use it for many things. To that end I'd shoot for a Square (14 x14") machine as in this size it will still be a bench top machine.

I will be specifically using a 1mm carbide cutter for the whole process. I do have a decent supply of surplus 1/2" and 1" aluminum tooling plate for the structure, as well as a nice piece of 2" square x 1/8" wall square aluminum tubing which may play well into the bridge or gantry cross-structure.

Sounds like you are well ahead of the average build here.

Do any of you have direct experience with machining PCB which would lead to useful advice?

Thank you,

Randy

Actually no experience at all with PCB except for a few kits in the youth worked on by hand. In any even I only think you have a couple of problems to address. One is dust which everyone has mentioned already. The next is PCB material mounting, here you may seriously want to think about a vacuum solution to hold the PCB material down. The vacuum hold down should be seen as a supplement to mechanical clamping.

As to the question of fixed gantry I believe that wins hands down over a moving gantry solution. Again in a machine this size the extra bench area isn't a big deal. More importantly I see it as an easier approach that yields a high precision machine with less effort. Considering you have a mill already and a fine selection of materials you ought to be able to build a very precise machine that literally bolts together.

[louieatienza]

Thank you for the reply, Louie, that is a good-looking machine. Did you ever fit bellows to protect the table leadscrew and bearing rails? I still haven't found any moving-table machine with one fitted. I guess another possibility would be a window-shade cover like on the Z axis of an Accuslide lathe conversion Z axis.

Randy

I just tape plastic on the front and back edge of the t-slot table, the same thick clear vinyl they use to cover tables, books, etc. Heavy gage shower pan liner works well too. I don't bother on the gantry or carriage since they're above and the dust shoe covers that. Anything else gets pushed away by the sweeps on the bearing blocks. I really wouldn't overthink it too much. I cut 3/8" and 1/2" phenolic with no protection on any rails or screws on my large machine, and the dust collector works fine, I just wide down the rails once in a while:

Milling 3/8" G10 Garolite, Home Made CNC Router, High Speed Toolpaths - YouTube

[zephyr9900]

Wizard, thank you for your reply.

Do look up those 10 mm slides to see if they are the ones that can only take a load in the downward direction. Some small linear rails will not work properly if used improperly.

The slides are Thomson AccuMini with 0.03C preloaded trucks. They have both good load and moment capability (though I won't rely on the latter).

The dust as you know is very abrasive and darn right nasty on tooling.

That has been my big fear. I have located a source for single-sided phenolic PCB in an appropriate size here in the US so that is less of an issue now. I'll rough-cut the blanks to width on my sheet metal shear.

As far as the lead screws and slides go you can run them dry or lubricate them often

I'll probably give BSA a call and talk to an applications person. Given I got them from disassembling a piece of existing equipment the replacement cost would be many times my original cost...

The next is PCB material mounting, here you may seriously want to think about a vacuum solution to hold the PCB material down.

That is a good idea. I do have a piece of 10mm thick Metapor that I bought years and years ago for a possible vacuum-forming stage that might be better utilized here.

As to the question of fixed gantry I believe that wins hands down over a moving gantry solution

I'm still up in the air about that, Wizard. But with tall walls it's more of "overhead crane" than "gantry". The main drawback to gantries that people mention is the racking. But I don't have a problem building a "shear plate" that is 6 inches or so wide (deep?) (for an 18-inch span) into the X axis to alleviate that. My linear bearings and leadscrews are long enough to do that and still have a good working area in Y. But having both X and Y axes mounted to a common base is a very good way to go.

Randy

[zephyr9900]

I just tape plastic on the front and back edge of the t-slot table, the same thick clear vinyl they use to cover tables, books, etc. Heavy gage shower pan liner works well too.

I see what you mean, and also about the dust collector shoe. I guess I'm actually pretty well off there because I know my workpiece will always be a consistent thickness and my depth of cut will never be more than 1/16" so I can optimize the collector shoe fingers/brush pretty well. And thank you for putting up all those good videos. They are very informative themselves.

Randy

[zephyr9900]

The slides are Thomson AccuMini with 0.03C preloaded trucks. They have both good load and moment capability (though I won't rely on the latter).

It is a little frustrating that Thomson doesn't sell hole plugs for the 10mm rail, nor rail tape to cover the exposed holes (I've asked them). I'm thinking of trying 2-mil Kapton tape (which I have found in a 5/16" width) for rail tape. From what I gather, I need to run the trucks off the rail onto a "keeper" while I install the tape after degreasing the top of the rails. I need to find out anyway because I'm using the same rails and trucks on my CNC lathe mechanism, of which I'm about to do a 2d generation. That's where I've learned about getting crud in the rail mounting holes.

Randy (easier to reply to my own post than edit the first one)

[louieatienza]

I'm still up in the air about that, Wizard. But with tall walls it's more of "overhead crane" than "gantry". The main drawback to gantries that people mention is the racking. But I don't have a problem building a "shear plate" that is 6 inches or so wide (deep?) (for an 18-inch span) into the X axis to alleviate that. My linear bearings and leadscrews are long enough to do that and still have a good working area in Y. But having both X and Y axes mounted to a common base is a very good way to go.

I think however for a small machine, it's a lot easier to make fixed gantry uprights stiffer. Since they don't move, it's a lot easier to add more mass and size, which should help stabilize the cut. Another benefit is that a moving table is likely not as heavy, so performance should be better. Also with a moving table the bearing blocks and leadnut carier can be placed optimally to mitigate racking. But yes, if you decide to do a moving gantry, I agree that having a wider transverse plate can keep the gantry legs in perfect relation to each other.

It is a little frustrating that Thomson doesn't sell hole plugs for the 10mm rail, nor rail tape to cover the exposed holes (I've asked them). I'm thinking of trying 2-mil Kapton tape (which I have found in a 5/16" width) for rail tape. From what I gather, I need to run the trucks off the rail onto a "keeper" while I install the tape after degreasing the top of the rails. I need to find out anyway because I'm using the same rails and trucks on my CNC lathe mechanism, of which I'm about to do a 2d generation. That's where I've learned about getting crud in the rail mounting holes.

They make stainless steel tape for DYMO-style label embossers, though the narrowest I know of is 1/2". Dymo 325-00 1/2" Non Adhesive Stainless Steel Tape - Free Shipping

As to the bearing rails... Usually from my experience, if there are two bearing races on the rail, you shouldn't remove the block from the rail without the "keeper". Most all 4-bearing-race rails I've seen have removable blocks, though you should still be careful replacing them, and in the same way they were removed if you have a matched set.

[wizard]

Wizard, thank you for your reply.


The slides are Thomson AccuMini with 0.03C preloaded trucks. They have both good load and moment capability (though I won't rely on the latter).

I wouldn't call their load capacity all that great. For a PCB router that probably won't be an issue. Thomson is a little strange with their ratings, but in most cases the small single row linear slides can't handle large moments on the cars and some can barely handle force in an upward direction.

That has been my big fear. I have located a source for single-sided phenolic PCB in an appropriate size here in the US so that is less of an issue now. I'll rough-cut the blanks to width on my sheet metal shear.

You will still have fine dust that gets into the slides, it just isn't as abrasive.

I'll probably give BSA a call and talk to an applications person. Given I got them from disassembling a piece of existing equipment the replacement cost would be many times my original cost...

I've been trying to score linear motion parts from some salvaged equipment but so far nothing really useful, consider yourself lucky. At best I've collected some junk, a couple of motors and the like.

That is a good idea. I do have a piece of 10mm thick Metapor that I bought years and years ago for a possible vacuum-forming stage that might be better utilized here.


I'm still up in the air about that, Wizard. But with tall walls it's more of "overhead crane" than "gantry". The main drawback to gantries that people mention is the racking. But I don't have a problem building a "shear plate" that is 6 inches or so wide (deep?) (for an 18-inch span) into the X axis to alleviate that. My linear bearings and leadscrews are long enough to do that and still have a good working area in Y. But having both X and Y axes mounted to a common base is a very good way to go.

Randy

Well what you do with the gantry is up to you. The way I see it, a fixed gantry on a PCB mill doesn't create huge space issues and should be easier to build to a high tolerance and keep the machine in alignment.

[wizard]

I think however for a small machine, it's a lot easier to make fixed gantry uprights stiffer. Since they don't move, it's a lot easier to add more mass and size, which should help stabilize the cut.

This is true but I still see that the big advantage is being able to align the machine once and have it stay aligned for a very long time. It should also be far easier to get that alignment right.

Another benefit is that a moving table is likely not as heavy, so performance should be better. Also with a moving table the bearing blocks and leadnut carier can be placed optimally to mitigate racking. But yes, if you decide to do a moving gantry, I agree that having a wider transverse plate can keep the gantry legs in perfect relation to each other.

One big benefit is that the rails can be put under the table thus shielding them from the direct deposit of dust and shavings. You have to be careful about the load you put on the linear rail cars but again being a PCB mill this shouldn't be a problem with a rational design.

They make stainless steel tape for DYMO-style label embossers, though the narrowest I know of is 1/2". Dymo 325-00 1/2" Non Adhesive Stainless Steel Tape - Free Shipping

As to the bearing rails... Usually from my experience, if there are two bearing races on the rail, you shouldn't remove the block from the rail without the "keeper". Most all 4-bearing-race rails I've seen have removable blocks, though you should still be careful replacing them, and in the same way they were removed if you have a matched set.