10 micron repeatability is pretty amazing. Just wondering out loud... would
it make sense to mount the printer head on the spindle of a CNC mill?
Either way you do it, I don't see how you can do second and subsequent sides
without multiple registration features for placement and alignment. (It sure
would be nice to just print out 11 layer boards with micro-vias and
plate-throughs. I'd settle for just 7 layers. :-)
Volkan, I missed the mechanical and control details if you posted them. I'm
curious about the Y-axis drive, and the print head control signals and
timing. Were you able to find those on the web?
Regards,
Mike.
----- Original Message -----
From: "Volkan Sahin" <v_sahin@...>
To: <Homebrew_PCBs@yahoogroups.com>
Sent: Monday, July 21, 2008 11:11 PM
Subject: Re: [Homebrew_PCBs] Re: Homemade inkjet printer test results
>I showed these pictures to a friend
who had tried hacking an R220 and
>his reply is that it's
great, but he doubts that it is possible to get
>the
registration close enough to do double sided boards. I'm curious if
>you have any reply to that.
In the past, I used below method but it requires accurate X/Y positioning
and I'll use it for this printer also, it has a starting point accuracy of
~+/-10micron meter.
You need to have reference holes on
printer bed on center of X-axis, which requires fixed board size or
multiple reference holes. Your board also needs to have pre-drilled
reference holes. Something like below,
X-axis
|--------------------------------|
| ∗
|
|
|
|
|
|
|
|
|
| ∗
|
|--------------------------------|
∗Reference holes
Prepare your image centered at
reference holes, after printing 1st side flip on X-axis
and print 2nd side. You can also eliminate requirement of
centered image by adding offset but centering is easier.
>He uses unmodified inkjets to
print transparencies, aligns them as a
>set, then uses a
vacuum table to hold them on a pre-sensatized board
>while
exposing under an UV lamp. He says he regularly gets 5 and 5 and
>is happy to be able to do double sided.
Yes he can get
it, everybody uses this method . Personally I want to avoid stock of
photosentized boards or photo resist, developer and for me it is double work
and it can be as good as your master film but this method is
proven no doubt about it.
>I think a possible solution is
to purchase thinner than normal single
>sided stock, etch
seperatly, and then bond the two with the backing
>together
and the copper sides out. No registration problems because the
>registration is done after the etch. Someone else had
described this
>here in more detail.
It is possible, but how to avoid
air bubbles between layers without vacuum? If you use oven during
soldering bubbles can separate glued layers.
Cheers,
Volkan
[Non-text portions of this message have been removed]
------------------------------------
Be sure to visit the group home and check for new Links, Files, and Photos:
http://groups.yahoo.com/group/Homebrew_PCBsYahoo! Groups Links