11 Apr
2013
11 Apr
'13
2:50 p.m.
A couple of folks on the VCForum talking about making new flip chip boards:
http://www.vintage-computer.com/vcforum/showthread.php?36190-New-flip-chip-…
11 Apr
11 Apr
2:32 p.m.
On 4/11/2013 11:50 AM, MikeS wrote:
A couple of folks on the VCForum talking about making
new flip chip
boards:
http://www.vintage-computer.com/vcforum/showthread.php?36190-New-flip-chip-…
I saw an earlier mention of making flip chips which this thread on
vintage-computer brought back to mind. Someone suggested or built a
straight 8 out of "mini" flip chips with surface mounted components.
One poster there mentions the con to that is that you can't use the
original modules if you were to change the form factor.
However it would be more feasible to make a backplane if one selected
the right edge connectors in PC form rather than wire wrap, and the
overall device would make for another type of "spare time gizmos" type
of artifact if done that way and replicated.
Lots of work, but it would be better in some ways than doing an original
form factor 8. There are Pros to that too, but obtaining supplies /
parts are less in your control if you go with an actual 8 and try to
match the original (using mixture of old and new parts) and also
presents some problems too (wire wrap backplane would be fun).
With my record of not doing much of any projects I won't be doing this
one any time soon, but it would be great to see something like this happen.
worth noting on the thread, the flip chip boards were going to be cut
out of clad stock, not made up by a board house, and the poster was all
ready with gallons of Ferric Chloride and a resist pin to make his first
board. and $20 each for the boards in that form. I would think having
the boards made up for you by the board house would come in for 1U Flip
chips cheaper than that.
also, do board houses / DEC / HP have any copyright rights to enforce
over the edge card form anymore? Something like the trouble you run
into taking an old IBM manual to a Staples and getting grief about
copying it? Some will some are assholes.
Jim
3:07 p.m.
On Apr 11, 2013, at 2:32 PM, Jim Stephens <jws at jwsss.com> wrote:
worth noting on the thread, the flip chip boards were
going to be cut out of clad stock, not made up by a board house, and the poster was all
ready with gallons of Ferric Chloride and a resist pin to make his first board. and $20
each for the boards in that form. I would think having the boards made up for you by the
board house would come in for 1U Flip chips cheaper than that.
Oh yeah. For 2-sided boards? Sure. I know Andrew Lynch's S100 boards
run $20 each, and they're solder-masked and silkscreened as well, not
to mention about 3 times larger than a flip chip. And you don't even
need to turn your hands yellow.
I gave up ferric chloride etching in high school. I couldn't get the
yields up high enough. I was always tempted to try the photoresist
presensitized stuff, but it was EXPENSIVE, so I never did.
- Dave
3:34 p.m.
On Thu, Apr 11, 2013 at 03:07:27PM -0400, David Riley wrote:
I gave up ferric chloride etching in high school. I
couldn't get the
yields up high enough. I was always tempted to try the photoresist
presensitized stuff, but it was EXPENSIVE, so I never did.
I tried it a little and had no luck getting consistent exposure.
The pens don't cover too well either (but they'll do in a pinch).
The Datak direct-etch dry transfers work *great* but it's a hell
of a lot of work for each board.
Haven't touched any of this in ages -- these days I'm a germ farmer
(septic tank). I'm not sure a gallon or two of FeCl3 would be enough to
sterilize it but there's no way I'm risking it just to save a few $$$
from oshpark.com (which kicks ass) or batchpcb.com
(which used to until
they botched a web-site upgrade last September and still
can't be bothered
to fix it).
John Wilson
D Bit
6:34 p.m.
On 4/11/2013 12:34 PM, John Wilson wrote:
On Thu, Apr 11, 2013 at 03:07:27PM -0400, David Riley
wrote:
My understanding was that they charged more by the hole than by size.
Also double sided if the layout was doable that way would make the board
order from a real board house way cheaper. Andrew's latest SCSI board
is $12 and area wise not that much less than these flip chip cards being
discussed. In the quantity that a mass buy for multiple builds of
straight 8's would scale to, I'd think the price would get pretty
reasonable.
I knew the fellow who did the card tabs here in Orange County, Stu
Phillips for MDB, Dilog and Emulex, but he died a few years ago, and I
think his company changed hands.
Otherwise I'd look him up and try to get his dies. The dies cost, the
runs and material don't. Helped that he knew some tool and die makers
who owed him to get into the business in the first place.
If anyone can trace the Stu Phillips, and you can point me at whoever
might still control it, I could ask about the dies, as I doubt they were
discarded.
Jim
I gave up ferric chloride etching in high school.
I couldn't get the
yields up high enough. I was always tempted to try the photoresist
presensitized stuff, but it was EXPENSIVE, so I never did.
I tried it a little and
had no luck getting consistent exposure.
The pens don't cover too well either (but they'll do in a pinch).
The Datak direct-etch dry transfers work *great* but it's a hell
of a lot of work for each board.
Haven't touched any of this in ages -- these days I'm a germ farmer
(septic tank). I'm not sure a gallon or two of FeCl3 would be enough to
sterilize it but there's no way I'm risking it just to save a few $$$
from oshpark.com (which kicks ass) or batchpcb.com (which used to until
they botched a web-site upgrade last September and still can't be bothered
to fix it). John Wilson
D Bit
12 Apr
12 Apr
1:15 a.m.
On Thu, Apr 11, 2013 at 03:34:17PM -0700, Jim Stephens wrote:
My understanding was that they charged more by the hole
than by size.
Also double sided if the layout was doable that way would make the
board order from a real board house way cheaper.
Oshpark.com and batchpcb.com charge *only* by size, which is great for tiny
SMT stuff but gets expensive quickly for flip-chips or dual/quad modules
which have to be big no matter how sparsely populated they are. I've had
small-volume prototype stuff done all over (very good prices in Macau) but
for stuff in big enough volumes to sell, I've been doing everything in the
US since the EPA may be largely useless but at least it eventually puts a
stop to *really* needless pollution, and I'm worried about what happens to
waste in less regulated countries. And in the US, getting any kind of real
area for $12 a pop wouldn't be easy...
I knew the fellow who did the card tabs here in Orange
County, Stu
Phillips for MDB, Dilog and Emulex, but he died a few years ago, and
I think his company changed hands.
Oh man I'm so sorry to hear that! I've seen his name stamped on the back
of tons of handles and if I ever found the contact information I swear I've
got for him around here somewhere, I was going to order a run of "D Bit"
handles, if he was still making them. He did really nice work.
Otherwise I'd look him up and try to get his dies.
The dies cost,
the runs and material don't. Helped that he knew some tool and die
makers who owed him to get into the business in the first place.
I've never held any 3D-printed output in my own hands and the pictures I've
seen vary from "suspiciously perfect" to "looks like shredded wheat",
so I
don't know if it's dumb to ask if 3D-printed flip-chip handles could handle
the violence of doing their job? Same question about DEC-style edge connectors
and/or obsolete Mate-N-Loks (either way the contacts aren't much of a problem
to dig up but the shells are long gone).
Otherwise, what really goes into making a die? I've got an el cheapo Sieg X2
mini-mill and lots of free time ... and there are lots of plastics companies
here in western Mass ... yeah another dumb question.
John Wilson
D Bit
3:16 a.m.
Otherwise
I'd look him up and try to get his dies. The dies cost,
the runs and material don't. Helped that he knew some tool and die
makers who owed him to get into the business in the first place.
I've never
held any 3D-printed output in my own hands and the pictures I've
seen vary from "suspiciously perfect" to "looks like shredded wheat",
so I
don't know if it's dumb to ask if 3D-printed flip-chip handles could handle
the violence of doing their job? Same question about DEC-style edge connectors
and/or obsolete Mate-N-Loks (either way the contacts aren't much of a problem
to dig up but the shells are long gone). Otherwise, what really goes into making a die?
I've got an el cheapo Sieg X2
mini-mill and lots of free time ... and there are lots of plastics companies
here in western Mass ... yeah another dumb question.
John Wilson
D Bit
2:57 a.m.
On Thu, Apr 11, 2013 at 11:32:33AM -0700, Jim Stephens wrote:
Someone suggested or built a straight 8 out of
"mini" flip chips with
surface mounted components.
That would be me :) While I applaud the effort to build a functioning
replica. That is way to much work for me. Doing a "modern" version with
small SMT components would allow for more freedom and it would be fun to
see how small and how power efficient it could be made.
It definitely can be done:
http://blog.makezine.com/2006/09/22/homemade-mt15-computer/
Unfortunately, Dieters original page is gone :( Anyone have a copy?
Lots of work, but it would be better in some ways than
doing an
original form factor 8. There are Pros to that too, but obtaining
supplies / parts are less in your control if you go with an actual 8
and try to match the original (using mixture of old and new parts)
and also presents some problems too (wire wrap backplane would be
fun).
Spot on.
At the moment I'm at the stage where I probably don't realize the amount
of work that needs to be done, which is a good thing :)
Cheers,
Pontus.
3:17 a.m.
On Fri, Apr 12, 2013 at 08:57:46AM +0200, Pontus Pihlgren wrote:
It definitely can be done:
http://blog.makezine.com/2006/09/22/homemade-mt15-computer/
Unfortunately, Dieters original page is gone :( Anyone have a copy?
Oh, it had just moved:
http://www.6502.org/users/dieter/mt15/mt15.htm
/P
8:18 a.m.
On 4/12/2013 12:57 AM, Pontus Pihlgren wrote:
On Thu, Apr 11, 2013 at 11:32:33AM -0700, Jim Stephens
wrote:
Wayback machine gives only until 2007.
http://www.6502.org/users/dieter/index.htm
Someone suggested or built a straight 8 out of
"mini" flip chips with
surface mounted components.
That would be me :) While I applaud the effort to build a functioning
replica. That is way to much work for me. Doing a "modern" version with
small SMT components would allow for more freedom and it would be fun to
see how small and how power efficient it could be made.
It definitely can be done:
http://blog.makezine.com/2006/09/22/homemade-mt15-computer/
Unfortunately, Dieters original page is gone :( Anyone have a copy?
Lots of work,
but it would be better in some ways than doing an
original form factor 8. There are Pros to that too, but obtaining
supplies / parts are less in your control if you go with an actual 8
and try to match the original (using mixture of old and new parts)
and also presents some problems too (wire wrap backplane would be
fun).
Spot on.
At the moment I'm at the stage where I probably don't realize the amount
of work that needs to be done, which is a good thing :)
Cheers,
Pontus.
11 Apr
11 Apr
4:53 p.m.
On 4/11/13 11:50 AM, MikeS wrote:
A couple of folks on the VCForum talking about making
new flip chip boards:
http://www.vintage-computer.com/vcforum/showthread.php?36190-New-flip-chip-…
does this mean I can sell off my R series modules for $20 ea?
12 Apr
12 Apr
5:51 a.m.
I think thats the chap who wants to build a straight 8 " from the ground
up", including the core, hes bought thousands of actaul cores !!, now thats
going to be a projrct and a half
DaveH
----- Original Message -----
From: "MikeS" <dm561 at torfree.net>
To: <cctalk at classiccmp.org>
Sent: Thursday, April 11, 2013 7:50 PM
Subject: New Flip Chip PCBs
A couple of folks on the VCForum talking about making
new flip chip boards:
http://www.vintage-computer.com/vcforum/showthread.php?36190-New-flip-chip-…
-----
No virus found in this message.
Checked by AVG - www.avg.com
Version: 2012.0.2240 / Virus Database: 2641/5737 - Release Date: 04/10/13
4119
days inactive
4120
days old
11 comments
9 participants
participants (9)
-
aek＠bitsavers.org -
bfranchuk＠jetnet.ab.ca -
dave.g4ugm＠gmail.com -
david＠hheng.plus.com -
dm561＠torfree.net -
fraveydank＠gmail.com -
jws＠jwsss.com -
pontus＠Update.UU.SE -
wilson＠dbit.com