TECHNET Archives

February 1998

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
"Gary Crowell Sr." <[log in to unmask]>
Reply To:
TechNet E-Mail Forum.
Date:
Tue, 10 Feb 1998 11:10:27 -0700
Content-Type:
text/plain
Parts/Attachments:
text/plain (112 lines)
Elstgeest Sander wrote:

> We are currently looking at PCB milling/drilling systems for producing
> our
> prototypes. Etching is, due to environmental issues, impossible. Does
> anyone
> have any experience on this subject, especially in Europe?  Two
> manufacturers we
> are looking at are LPKF and T-Tech.
>
> TIA,
>
> Sander Elstgeest
> design engineer, NKF electronics
>

We have an LPKF system here that is in use daily, producing a dozen or
more small tester adapter cards per week.  It is invaluable in our
applications, where even a fast proto house would be a day or more
behind our turnaround time.  Your Mileage May Vary.  The key is in
determining if any milling system is really cost effective for your
application.

Even a relatively small board can take an hour or more to mill, so if
you are planning on doing more than a small quantity of any single
board, a fast turn might be faster.  If you are only doing a couple of
boards a month, you'll probably never amortize the cost of the system,
unless you can figure in a large time-to-market savings.  With the
initial cost of the system, the board blanks, milling tools, and the
man-hours expended to run it, I doubt that any individual board cost
compares favorably with a fast-turn pcb.

Of course there is the plated thru-hole problem.  There are add-ons to
both systems that automatically fill the via drill holes with a
conductive ink/epoxy/whatever.  I believe the advertisements claim the
result is a <200 milliohm via, but when we tried it here, the result was
more on the order of 600 milliohms.    Maybe the result improves with
practice, we only gave it a short test.  I'd like to hear the results
from anyone who is using one of these systems.  Anyway, our techs have
gotten the procedure of filling the vias with short soldered wires down
to an art, so it doesn't take long.  But there again, adding on a
half-hour of soldering to each board can make the fast-turn plated-thru
pcb look even more attractive in terms of time and manpower cost.

On very small boards (a few square inches), we have been able to
routinely do 5 mil lines, but don't expect a 100% yield there.  You can
do the same on larger boards, but expect an even  lower yield as the
chance of a milling depth variation increases over the larger area.
Fine line boards also require more careful inspection and touch up, in
addition to the time required for a careful set-up. At 10 or 12 mils,
its not so critical, just set it & go.

Also remember of course, you're not getting any solder mask, so the
copper is going to tarnish and look awful after a few days.  And, it
takes more care in soldering to avoid bridges.  No silk screen either,
although you can mill text on open areas.

The real difficulty comes in the problem of layout restrictions.  Since
you have to solder both sides to get thru-hole continuity, you have to
prohibit traces connecting to a pin on the component side that would be
underneath the component when it is placed.  For example, with a DIP
socket you would be unable to solder to its pins underneath it on the
component side, so all traces to that socket's pins would need to route
on the solder side.  This limitation means that if you were prototyping
a pcb for later production as an etched pcb, you would either be
imposing the milled limitations on that layout, or would have to plan on
doing a complete new layout.

All of the above sounds like I'm pretty down on it, but like I said at
the start, our LPKF has been invaluable in our application where we do a
dozen or so boards a week, typically of 3 or 4 different designs.

Which system?  I've looked at the specs on both the LPKF and T-Tech web
sites, and they both seem to have about the same mechanical
specifications, as far as positioning accuracy, speeds, etc.  It looks
like LPKF has a wider range of models available, so you might be able to
choose a bed size that is less expensive if that's all your application
needs. Other than that, I haven't compared costs.

Another group here at this company has a T-Tech.  I haven't seen it, but
they've been over to see our LPKF and agree that they're pretty much the
same.  They bought the T-Tech due to their familiarity with the its
software from an previous employer, but they hinted that it looked to
them like the LPKF software was more capable and easier to use.  That's
just a quick opinion of course, not based on an objective comparison.

The software is important, and it might well be the best means to
differentiate between the two systems.  The LPKF software seems to be
very capable and versatile.  Its not utterly intuitive, and a user would
greatly benefit from some prior knowledge of Gerber files.  It took me
most of a day to dope through it to get my first good board, with no
help (and later finding out I was using an old version of the manual);
but once you get the procedure down, its pretty easy to use.  Expect to
waste a few boards at first.

I'd like to hear other opinions too, and I hope this one is useful.

Gary Crowell
Micron Technology

################################################################
TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe:   SUBSCRIBE TechNet <your full name>
To unsubscribe:   SIGNOFF TechNet 
################################################################
Please visit IPC web site (http://jefry.ipc.org/forum.htm) for additional information.
For the technical support contact Dmitriy Sklyar at [log in to unmask] or 847-509-9700 ext.311
################################################################


ATOM RSS1 RSS2