Searching \ for '[OT] PCB milling' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/pcbs.htm?key=pcb
Search entire site for: 'PCB milling'.

Exact match. Not showing close matches.
PICList Thread
'[OT] PCB milling'
2000\03\28@144714 by Jerry Merrill

flavicon
face
We are considering buying a table-top PCB prototype machine.  These produce
single and double sided boards by milling (rather than etching) away
unwanted copper.

Anybody wish to share their experiences with these devices?
Costs, limitations, strengths?


Jerry Merrill

spam_OUTjerrymTakeThisOuTspamtech-tools.com
http://www.tech-tools.com
FAX: (972) 494-5814
VOICE:(972) 272-9392
TechTools
PO Box 462101
Garland,  TX  75046-2101

2000\03\28@150131 by Andrew Kunz

flavicon
face
Limitation was it wouldn't do fine traces (under 20 mil) accurately.  There was
always the whisker of trace from the one side of the cutter.

Boards which are not perfectly flat would be cut unevenly.  Made for REAL
problems.

AP Circuits turned out to be a better deal in the long run.

If you're looking at spending that much money, get a Kepro system instead.

Andy










Jerry Merrill <.....jerrymKILLspamspam@spam@TECH-TOOLS.COM> on 03/28/2000 02:45:59 PM

Please respond to pic microcontroller discussion list <PICLISTspamKILLspamMITVMA.MIT.EDU>








To:      .....PICLISTKILLspamspam.....MITVMA.MIT.EDU

cc:      (bcc: Andrew Kunz/TDI_NOTES)



Subject: [OT] PCB milling








We are considering buying a table-top PCB prototype machine.  These produce
single and double sided boards by milling (rather than etching) away
unwanted copper.

Anybody wish to share their experiences with these devices?
Costs, limitations, strengths?


Jerry Merrill

EraseMEjerrymspam_OUTspamTakeThisOuTtech-tools.com
http://www.tech-tools.com
FAX: (972) 494-5814
VOICE:(972) 272-9392
TechTools
PO Box 462101
Garland,  TX  75046-2101

2000\03\28@150805 by Scott Dattalo

face
flavicon
face
On Tue, 28 Mar 2000, Jerry Merrill wrote:

> We are considering buying a table-top PCB prototype machine.  These produce
> single and double sided boards by milling (rather than etching) away
> unwanted copper.
>
> Anybody wish to share their experiences with these devices?
> Costs, limitations, strengths?

I'd recommend against it unless you're doing RF and need sculpt special pcb's. I
personally haven't used one, but at a previous job we had one and found that the
time spent making a board is more expensive than sending out for a really cheap
proto-board. (you can get over night turns with out solder mask and silkscreen
very cheaply - depending on the size, etc). Another problem is that through
holes are not plated.

Scott

2000\03\28@155949 by Don Hyde

flavicon
face
Our company has one, and it's a real lifesaver.  We mill lots of prototype
boards.  The machine is not cheap, I think ours cost between $5000 and
$10000, but we mill enough boards that it saves a lot of money and time.

It takes a while to mill a board, so you don't want to make production
quantities.  Around here, it's common to send out for etching if there are
10 to be made, but the 2- or 3-turns needed to get the design right are much
quicker with the milling machine.

You can mill boards from the same CAD files you will send to a board house,
but the milled boards still aren't EXACTLY like etched ones.  You still need
to test an etched one before you commit to volume production.  This is
especially true if you have RF stuff like we almost always do.

If you're a glutton for punishment, supposedly you can mill multilayer
boards by milling two or more boards and then gluing them together with
epoxy, but as a practical matter, they're limited to double-sided boards.

The milling machine is the greatest thing ever made for whipping up a
professional and finished-looking prototype to impress a potential big-$
customer.

It's fabulous for one-of-a-kind test fixtures.  Especially when you realize
6 months later that you need another one just like it (just hope you didn't
lose those old CAD files!)

> {Original Message removed}

2000\03\28@162415 by Larry G. Nelson Sr.

flavicon
face
I used a T-Tech unit and routinely did 8 and 8 with no major problems. Be
sure to replace the setscrew holding the bit in place every few boards if
you don't want it to loosen up and cut too deep. I am looking to buy a used
one for myself. Tooling can add up but shop around and you can save money.
Invest in the proper VAC also or you will have problems with a possible
fire in your shop VAC since the copper dust works thru the shop VAC filter.
3M with the proper filter will do. Also it is a noisy process so be
prepared to find it a home away from everyone.

Larry


At 02:58 PM 3/28/00 -0500, you wrote:
>Limitation was it wouldn't do fine traces (under 20 mil) accurately.
There was
{Quote hidden}

<@spam@PICLISTKILLspamspamMITVMA.MIT.EDU>
{Quote hidden}

Larry G. Nelson Sr.
spamBeGoneL.NelsonspamBeGonespamieee.org
http://www.ultranet.com/~nr

2000\03\28@163448 by Sean Breheny

face picon face
I was under the impression that AP circuits used milling equipment, though
(at least in the cheap proto 1 service). IS theirs just SO much more
expensive that it overcomes the problems?

Sean

At 02:58 PM 3/28/00 -0500, you wrote:
>Limitation was it wouldn't do fine traces (under 20 mil) accurately.
There was
{Quote hidden}

| Sean Breheny
| Amateur Radio Callsign: KA3YXM
| Electrical Engineering Student
\--------------=----------------
Save lives, please look at http://www.all.org
Personal page: http://www.people.cornell.edu/pages/shb7
TakeThisOuTshb7EraseMEspamspam_OUTcornell.edu ICQ #: 3329174

2000\03\28@165530 by Plunkett, Dennis

flavicon
face
I have been watching with bated breath the posts on this one. Yes these
machines will do the job but after the experience of using one I have my
reservations. It would seem that unless you can drive a truck between the
pads and components the reliability of the board will be in doubt. It can do
8thou spacing, but the number of bits that you go through are quite large.
The accuracy of double sided boards is sloppy to say the least. Also you
will have too limit the number of vias on the board and where they are
placed (Not a good idea to put them under a surface mount IC!) You will
spend lots of time removing unwanted wisks of copper and will often be
caught out by the "Un routed copper area". Use boards that are pretinned and
use the Mulitcore through hole plating system. In all these items are slow
and very old technology that has not kept pace with the current IC
footprints. Often the software to drive these things is not very user
friendly, and the consumerables are large and the units require losts of
love and care (Alignment) before the results are anywhere near good.
Realy thou, these are great for a single sided board as you can route a silk
screen layer on (Can do that on Vero!) and have a good looking board in a
shortish period.

To sum up, these are OK, but will limit the technology and your board layout
methods

Dennis




> {Original Message removed}

2000\03\28@190855 by Brian Kraut

picon face
No, they use etched boards.  It is a WHOLE lot cheaper and faster to etch them
than to mill them one you get above a quantity of around 1/2 a board.  That is
after you have the equipment to do it.

Personally, I think a home built PC board drill for the holes and photo etched
pre-sensitized boards is the best route for gotta have it today protos.  You
can laser print the artwork and etch a complicated board much better and faster
than milling it.  Now if you could just etch those damn holes.

Sean Breheny wrote:

{Quote hidden}

2000\03\28@191930 by TIM

flavicon
face
-----Original Message-----
From: Brian Kraut <engaltEraseMEspam.....EARTHLINK.NET>
To: EraseMEPICLISTspamMITVMA.MIT.EDU <RemoveMEPICLISTEraseMEspamEraseMEMITVMA.MIT.EDU>
Date: Tuesday, March 28, 2000 5:09 PM
Subject: Re: [OT] PCB milling


yes..this is so true ...J.I.T.  secenerio...today i need it ,,so how does
one do the plated thru hole? hummm


>Personally, I think a home built PC board drill for the holes and photo
etched
>pre-sensitized boards is the best route for gotta have it today protos.
You
>can laser print the artwork and etch a complicated board much better and
faster
>than milling it.  Now if you could just etch those damn holes.

2000\03\28@195248 by John Orhan

flavicon
face
Hi all,
On the subject of PCBs, I've heard the X Y plotters are the way to go. Any
thoughts?

               {Original Message removed}

2000\03\28@203103 by l.allen

picon face
> Hi all,
> On the subject of PCBs, I've heard the X Y plotters are the way to go. Any
> thoughts?
>
Been there... done that.. got the T-shirt... and get my
boards done with a PCB manufacturer now.

Getting the pens to work reliably, the ink right, the speed
of drawing right and slightly altering the Roland dxy1200
to take PCBs was a trial but not impossible. It was just
that if you didnt keep using it, the ink would go off, the
pen would take half an hour of fiddling to get it flowing
right and then there was etching and drilling....enough
said (broke about 6 X 0.5mm drill bits just getting this
setup right).
Seemed a good idea at the time since we had a plotter
anyway but it was only feasible if you were using it
almost daily.

Now with new chemical/safety regulations in NZ the
keeping, using and disposing of Ammonia Persulphate is
a huge hassle.
_____________________________

Lance Allen
Technical Officer
Uni of Auckland
Psych Dept
New Zealand

http://www.psych.auckland.ac.nz

_____________________________

2000\03\29@043807 by rubenj

flavicon
face
>
> yes..this is so true ...J.I.T.  secenerio...today i need it ,,so how
> does one do the plated thru hole? hummm
>
>
> >Personally, I think a home built PC board drill for the holes and
> >photo
> etched
> >pre-sensitized boards is the best route for gotta have it today
> >protos.
> You
> >can laser print the artwork and etch a complicated board much better
> >and
> faster
> >than milling it.  Now if you could just etch those damn holes.

This is exactly what we do and get pretty good results too.

1. Take the gerber output from the CAD and print it out on a
transparent film (the type that is used for hand made ink drawings)
with a good laser printer (at least 600 dpi) with GC-Prevue. I also
use GC-Prevue to make groundplanes and a small centerhole in every
pad which makes it a lot easier to drill. Print the topside reversed
so the printside of the film gets closest to the board when exposing
it with UV-light.

2. Use adhesive tape to fix the top and bottom side films together
(if double sided). Remove the protective coat from the pre-sensitized
board and fix the films on the board with transparent adhesive tape
(preferably where there is no traces). If you have made some
reference pads outside the edges of the board it is easier to get the
topside aligned with the bottomside.

3. Before exposing the board make sure the UV-lamp has been lit for
about half an hour. Until we discovered this we got very uneven
results, sometimes over-exposed, sometimes underexposed. I'm still
not sure if it had to do with the lamp, but now it works ok almost
every time.

4. Place the board under a non UV-blocking piece of glass that
presses the film against the board. Make sure the glass is clean.

5. Expose both sides of the board. With our lamp and distance to the
board, an exposure time of 9 minutes is good.

6. Develop the board in a solution of one teaspoon natriumhydroxid
(not sure if this is the right word, translated from swedish) per 250
ml water. When you clearly can see the traces appear the development
is ready. Rinse the board in cold water.

7. Etch the board. Use a solution of 3 parts water, 1 part 30%
hydrochloric acid, 1 part 35% hydroperoxide (vdteperoxid in swedish).
Use protective goggles, gloves and a well ventilated area. Use a
glass or plastic container. The solution can get pretty hot.

8. Rinse the board and remove the rests of the photoresist (on the
traces and groundplanes) with thinner or another solvent.

9. If the size of the board is important, now is the time to trim it
to the exact size.

10. Put a solderable protective coating on both sides of the board.
Make sure it is completely dry before the board is handled.

11. Drill the board. Always use high speed hardened drills. Use new
drills frequently. The centerhole in the pad is very useful when
drilling the board. Especially when drilling double sided boards
because its not so easy for the pad on the bottomside of the board to
come loose from the board if you push the drill too fast or have a
dull drill.

12. Make the trough plated vias. I drill these vias (the ones with no
components in them) with a 0.8mm drill and drive a slightly larger
pin through the hole (with pliers and a hammer), cut the pin as close
to the board as possible on both sides and solder it to the board.
For the pin I use one pin cut of from a pin header. The through
plated holes which have components in them are soldered on both sides
where possible. If not possible an extra hole can be drilled adjacent
to the component.

I have recently used this method for a board with a SX52 chip on it
(0.65 mm or 0.0255 inches spacing) with good result.




==============================
Ruben Jvnsson
AB Liros Elektronik
Box 9124, 200 39 Malmv, Sweden
TEL INT +46 40142078
FAX INT +46 40947388
RemoveMErubenspam_OUTspamKILLspampp.sbbs.se
==============================

2000\03\29@095737 by Jerry Merrill

flavicon
face
The other Engineers were pushing to get one of these for quick prototyping.
We knew they would not work for production and that vias would be  a
problem, but the ads would lead one to believe they were capable of 2 or
even 3 traces between pads.  If we limited ourselves to 1 trace between
pads, it seemed doable - particularly with the amount of surface-mount
parts we are using.

My concern was that we would end up spending a lot of Engineering time
doing custom layouts for the machine, baby-sitting the thing, and then
troubleshooting cats-wiskers and solder-bridges.

We do not want to get into the chemical mess of etching our own and I think
our time could be better spent elsewhere.

Based on these responses, I'll axe the purchase request and put more $ into
the prototype budget for quick-turn protos.

Lots of real-world feedback! Thanks.


At 01:58 PM 3/28/00 , you wrote:
>Limitation was it wouldn't do fine traces (under 20 mil) accurately.
There was
{Quote hidden}

Jerry Merrill

RemoveMEjerrymTakeThisOuTspamspamtech-tools.com
http://www.tech-tools.com
FAX: (972) 494-5814
VOICE:(972) 272-9392
TechTools
PO Box 462101
Garland,  TX  75046-2101

2000\03\29@095950 by mike

flavicon
picon face
>-----Original Message-----
>From: Brian Kraut <EraseMEengaltspamspamspamBeGoneEARTHLINK.NET>
>To: RemoveMEPICLISTKILLspamspamMITVMA.MIT.EDU <PICLISTSTOPspamspamspam_OUTMITVMA.MIT.EDU>
>Date: Tuesday, March 28, 2000 5:09 PM
>Subject: Re: [OT] PCB milling
>
>
>yes..this is so true ...J.I.T.  secenerio...today i need it ,,so how
>does
>one do the plated thru hole? hummm

don't RS or Digi-Key sell then by the bagload ;-) if only..
but there are through hole pins, made by Harwin IIRC..
--
Life is dark and empty for the PIC developer without VirtualBreadboard
go look at it...  http://www.virtualbreadboard.com

2000\03\29@131649 by Michael Rigby-Jones

flavicon
face
part 0 5707 bytes
<P><FONT COLOR="#0000FF" SIZE=2 FACE="Arial">We have one here, and although it is capable of good results, even for RF stuff, it is very slow, and you really do need to babysit it.&nbsp; A slight glitch in the programming and it snaps the (very expensive) milling bits for a passtime.&nbsp; It does do pretty good single sided boards though, such as the type one might use at home for a PIC project :o)</FONT></P>

<P><FONT COLOR="#0000FF" SIZE=2 FACE="Arial">Mike</FONT>
</P>
<UL>
<P><FONT SIZE=1 FACE="Arial">{Original Message removed}

2000\03\29@161236 by Dale Botkin

flavicon
face
I have done this.  In 1992, in fact.  I used an HP 7475, a gutted and
cut-off HP felt tip plotter pen with a slightly trimmed artist's .5mm
permament market slid through the HP pen barrel.  I hear you can now get
HP pens with permanent ink, I don't think they existed then.

I had to tape the PCB material to paper and use 1 or 2 inch/sec plot
speed, but it made beautiful, perfect plots and great boards in half an
hour or so, including etch time.  I never even attempted a double-sided
board, of course.

Dale

On Wed, 29 Mar 2000, John Orhan wrote:

> Hi all,
> On the subject of PCBs, I've heard the X Y plotters are the way to go. Any
> thoughts?
>
>                 {Original Message removed}

More... (looser matching)
- Last day of these posts
- In 2000 , 2001 only
- Today
- New search...