[EE]: Comments wanted on PCB prototyping idea.

1 view
Skip to first unread message

Byron A Jeff

unread,
Dec 7, 2002, 2:50:49 PM12/7/02
to PIC...@mitvma.fight.spam.mit.edu
After watching all the discussion on baking PCBs in toaster ovens and watching
the slow decline in reliability of my heavily used wirewrapped thermostat
board I realize that at long last I'll have to move to PCB prototyping.

While clearly board houses a clearly the right option for finished product,
the turnaround time and costs are prohibitive for hobby prototyping. And while
the toaster oven concept is intriguing, it's unclear if reproducable, reliable
boards can be produced with the technique.

So I want the throw out an idea that covers the en masse production of every
aspect of through hole PCB production. The traditional toner tranfer is handled
per Carl Ott's document:

http://users.rcn.com/carlott/toner_transfer_exp.pdf

The two issues remaning are drilling and soldering. After thinking through the
thought of putting together an CNC style drilling machine (and subsequently
abandoning the idea) I realized that a predilled copper clad PCB is the ticket.
Something like the boards in the lower left of this Digikey catalog page:

http://dkc3.digikey.com/PDF/T023/V5/0865.pdf

Pricey for sure ($9 US for a 3x4.5 board) but fits the ticket.

As for soldering, I forsee dipping the assembly in a solder pot. These units
seem to have a permanent home on Ebay. Here's a perfect example I just came
accross:

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=1922309798

May be worth the $60 shipping.

So the process:

* Generate layout in Eagle.
* Laser Print to toner transfer
* Laminate onto predrilled copper board. The real interesting thing is that
these boards are double sided, so with a few vias both sides can be used.
* Etch.
* Populate
* Dip

Whatcha think? Nothing in the process needs to be done on an individual basis
including drilling...

BAJ

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsub...@mitvma.NotNeeded.mit.edu


Josh Koffman

unread,
Dec 7, 2002, 3:09:42 PM12/7/02
to PIC...@mitvma.spammers.are.scum.mit.edu
Well, I like the predrilled idea. This of course as I sit here and have
to "touch up" some of my holes because on a 40 pin dip, 20 holes have to
line up right or the socket won't fit in. The only downside I see will
be aligning the resist to the board. Especially if you plan to run
traces between pins...even a little bit off and you have an IC pad in
the middle of nowhere, and a trace that contacts a pin where it
shouldn't. I guess the only other downside is the cost of the boards
(ouch). Might be cheaper (at least for the first run) to design a copper
clad predrilled board in Eagle, then use the free PCB service (first
time only) to get a few of them to play with. Cool product though, I'd
never seen it before. Having never used a solder pot, how do u hold the
board while you dip it? Do any of the components have a tendency to move
away from the board?

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
-Douglas Adams

Byron A Jeff wrote:
> The two issues remaning are drilling and soldering. After thinking through the
> thought of putting together an CNC style drilling machine (and subsequently
> abandoning the idea) I realized that a predilled copper clad PCB is the ticket.
> Something like the boards in the lower left of this Digikey catalog page:
>
> http://dkc3.digikey.com/PDF/T023/V5/0865.pdf
>
> Pricey for sure ($9 US for a 3x4.5 board) but fits the ticket.

--


http://www.piclist.com hint: To leave the PICList

mailto:piclist-unsub...@mitvma.Cut.mit.edu


H. Carl Ott

unread,
Dec 7, 2002, 3:56:59 PM12/7/02
to PIC...@mitvma.take.this.out.mit.edu
Hi Byron,
Well I've made more the a few homebrew PCBs.
(picture of what I'm working on this afternoon.
http://users.rcn.com/carlott/at-tnc_smd.jpg )

So I'll throw in some comments.
Actually making the boards is not that bad using toner transfer.
You are right, the drilling is a bit of a pain. The soldering is not
really that big of a deal for me unless we are dealing in production
quantities.

But I'm not sure predrilled copper clad is really going to help much.
That's interesting material at digikey. I used to make prototypes with a
similar vector product 10 years back. You use pad cutters /x-acto and a
dremel to isolate copper.
But for me the main problem is that .1" hole spacing. Too many parts
are not on that spacing. It also seriously restricts your trace routing,
and it does not help much with SMT or rf designs.

SMT seems to be the main motivation for the toaster oven interest. I
used to use a lab hotplate for a similar purpose, but I found I could hand
solder faster. BGA might force me back to a hotplate or toaster oven.

Solderpots are good for low density through hole designs with a
soldermask (without the mask you get too many shorts). You populate the
board, dip it in flux and the carefully dip it in the solderpot. Again,
this won't help with SMT.

What I really want is a cheap user friendly cnc drill/engraver to enhance
my home prototyping . Under $500.00. This might help with drilling, but I
really want it to engrave panels and make clean cutouts in enclosures. I
could probably do it for $1000 to $1500 with a sherline and one of those
stepper retrofits. Just a bit pricey for me at the moment.

BTW, I love any discussion on improving home prototyping methods.

Regards,
-carl


At 02:55 PM 12/7/2002, you wrote:
>After watching all the discussion on baking PCBs in toaster ovens and watching
>the slow decline in reliability of my heavily used wirewrapped thermostat
>board I realize that at long last I'll have to move to PCB prototyping.
>
>While clearly board houses a clearly the right option for finished product,
>the turnaround time and costs are prohibitive for hobby prototyping. And while
>the toaster oven concept is intriguing, it's unclear if reproducable, reliable
>boards can be produced with the technique.
>
>So I want the throw out an idea that covers the en masse production of every
>aspect of through hole PCB production. The traditional toner tranfer is
>handled
>per Carl Ott's document:
>
>http://users.rcn.com/carlott/toner_transfer_exp.pdf
>

>The two issues remaning are drilling and soldering. After thinking through the
>thought of putting together an CNC style drilling machine (and subsequently
>abandoning the idea) I realized that a predilled copper clad PCB is the
>ticket.
>Something like the boards in the lower left of this Digikey catalog page:
>
>http://dkc3.digikey.com/PDF/T023/V5/0865.pdf
>
>Pricey for sure ($9 US for a 3x4.5 board) but fits the ticket.
>

>As for soldering, I forsee dipping the assembly in a solder pot. These units
>seem to have a permanent home on Ebay. Here's a perfect example I just came
>accross:
>
>http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=1922309798
>
>May be worth the $60 shipping.
>
>So the process:
>
>* Generate layout in Eagle.
>* Laser Print to toner transfer
>* Laminate onto predrilled copper board. The real interesting thing is that
> these boards are double sided, so with a few vias both sides can be used.
>* Etch.
>* Populate
>* Dip
>
>Whatcha think? Nothing in the process needs to be done on an individual basis
>including drilling...
>
>BAJ

--


http://www.piclist.com hint: To leave the PICList

mailto:piclist-unsub...@mitvma.CounterSpam.mit.edu


cdb

unread,
Dec 7, 2002, 7:58:38 PM12/7/02
to PIC...@mitvma.extra.mit.edu
Sounds like a good idea, but aren't those boards just like Veroboard
- the copper is in strips?

colin
--
cdb, bod...@optusnet.Spammers.Are.Scum.com.au on 08/12/2002

Light travels faster than sound. That's why some people appear bright
until they speak!

--
http://www.piclist.com hint: To leave the PICList

mailto:piclist-unsub...@mitvma.RemovE.mit.edu


cdb

unread,
Dec 7, 2002, 8:00:28 PM12/7/02
to PIC...@mitvma.removethis.mit.edu
Another option is to use large surface mount components where ever
possible and save drilling that way.

colin
--
cdb, bod...@optusnet.Bogus.com.au on 08/12/2002

Light travels faster than sound. That's why some people appear bright
until they speak!

--
http://www.piclist.com hint: To leave the PICList

mailto:piclist-unsub...@mitvma.Cut.mit.edu


PicDude

unread,
Dec 7, 2002, 8:02:41 PM12/7/02
to PIC...@mitvma.cut.mit.edu
BAJ, I hadn't seen the digikey boards before -- nice, but $$$.

For me, drilling has always been the most painful part of the
process. I do however have this idea to use one of the PCB
proto houses to supply me with boards that are completely
drilled on a grid of 0.1" centers. But it would be completely
copper clad still. IIRC, $33 for 10" x 15" from 4pcb.

When I need to build a circuit, the design is laid out so that
the component leads fall on the 0.1" grid, then the board coated
for photoetching, and etched in the usual way. I'm not sure how
the toner-transfer method would work with the holes already
drilled, which is why I figured on the photo-etching process.

Soldering by any method, but I choose the manual way. Not
familiar with how the solder pot, wave soldering, etc works.
Anyone wants to summarize the options will get in my good
books. :-)

Cheers,
-Neil.

--
http://www.piclist.com hint: To leave the PICList

mailto:piclist-unsub...@mitvma.RemovE.mit.edu


PicDude

unread,
Dec 7, 2002, 8:42:54 PM12/7/02
to PIC...@mitvma.extra.mit.edu
cdb scribbled:

>
> Another option is to use large surface mount components where ever
> possible and save drilling that way.
>
> colin


Actually, in my past life back in the Caribbean where
PCB drills etc weren't available easily, I used to spread
out the pins of regular 0.3" DIPs and other components
and surface mount them. A great advantage of this is that
components can be mounted on both sides, but I'd have to
jumper wires around the edge for connections.

Cheers,
-Neil.

--
http://www.piclist.com hint: To leave the PICList

mailto:piclist-unsub...@mitvma.Spammers.Are.Scum.mit.edu


Byron A Jeff

unread,
Dec 7, 2002, 10:15:18 PM12/7/02
to PIC...@mitvma.spammers.are.scum.mit.edu
On Sat, Dec 07, 2002 at 03:59:27PM -0500, H. Carl Ott wrote:
> Hi Byron,
> Well I've made more the a few homebrew PCBs.
> (picture of what I'm working on this afternoon.
> http://users.rcn.com/carlott/at-tnc_smd.jpg )
>
> So I'll throw in some comments.
> Actually making the boards is not that bad using toner transfer.
> You are right, the drilling is a bit of a pain. The soldering is not
> really that big of a deal for me unless we are dealing in production
> quantities.

I struggle with soldering. It's one of the reasons I've stuck with wirewrap
so far.

>
> But I'm not sure predrilled copper clad is really going to help much.
> That's interesting material at digikey. I used to make prototypes with a
> similar vector product 10 years back. You use pad cutters /x-acto and a
> dremel to isolate copper.

They still sell that stuff. Called Verobard. But each strip is isolated from
the others so you have to populate a bunch of jumpers and use pad cutters.
I want to eliminate that task be etching....

> But for me the main problem is that .1" hole spacing. Too many parts
> are not on that spacing. It also seriously restricts your trace routing,
> and it does not help much with SMT or rf designs.

Neither which are on my design requirement. If SMT were the target then the
holes are unnecessary.

The only through hole parts I use that don't fit on a 0.1 grid are large
components like high power semiconductors and inductors. I can live with
that.

>
> SMT seems to be the main motivation for the toaster oven interest. I
> used to use a lab hotplate for a similar purpose, but I found I could hand
> solder faster. BGA might force me back to a hotplate or toaster oven.

You could solder hundreds of pins in 10 minutes? Tell me more!

>
> Solderpots are good for low density through hole designs with a
> soldermask (without the mask you get too many shorts). You populate the
> board, dip it in flux and the carefully dip it in the solderpot. Again,
> this won't help with SMT.

SMT isn't on my radar. With WW I've always used through hole components, so
I'm totally comfortable with them.

>
> What I really want is a cheap user friendly cnc drill/engraver to enhance
> my home prototyping . Under $500.00. This might help with drilling, but I
> really want it to engrave panels and make clean cutouts in enclosures. I
> could probably do it for $1000 to $1500 with a sherline and one of those
> stepper retrofits. Just a bit pricey for me at the moment.
>
> BTW, I love any discussion on improving home prototyping methods.

Well I appreciate your document. It has inspired me to think about a setup...

BAJ

--
http://www.piclist.com hint: To leave the PICList

mailto:piclist-unsub...@mitvma.Cut.mit.edu


Byron A Jeff

unread,
Dec 7, 2002, 10:21:37 PM12/7/02
to PIC...@mitvma.counterspam.mit.edu
On Sat, Dec 07, 2002 at 03:43:27PM -0600, Josh Koffman wrote:
> Well, I like the predrilled idea. This of course as I sit here and have
> to "touch up" some of my holes because on a 40 pin dip, 20 holes have to
> line up right or the socket won't fit in. The only downside I see will
> be aligning the resist to the board. Especially if you plan to run
> traces between pins...even a little bit off and you have an IC pad in
> the middle of nowhere, and a trace that contacts a pin where it
> shouldn't.

I'm going to spend some time with Eagle trying to figure out if reasonable
designs can be mapped onto the format.

> I guess the only other downside is the cost of the boards
> (ouch). Might be cheaper (at least for the first run) to design a copper
> clad predrilled board in Eagle, then use the free PCB service (first
> time only) to get a few of them to play with.

Doesn't seem cost effective. Olimex is one of the cheapest and fastest
services around and a double sided 12.6"x7.8" is $104 USD. That's about
6 of the smallest digikey boards which would be half the cost.

> Cool product though, I'd
> never seen it before. Having never used a solder pot, how do u hold the
> board while you dip it? Do any of the components have a tendency to move
> away from the board?

No clue. I'm looking for advise on this procedure.

BAJ

--
http://www.piclist.com hint: To leave the PICList

mailto:piclist-unsub...@mitvma.RemoveThis.mit.edu


Byron A Jeff

unread,
Dec 7, 2002, 10:23:39 PM12/7/02
to PIC...@mitvma.bogus.mit.edu
On Sun, Dec 08, 2002 at 11:03:47AM +1000, cdb wrote:
> Another option is to use large surface mount components where ever
> possible and save drilling that way.

Addressed in my initial post. How can SMT components be cheaply and reliably
applied to a board?

This was proffered as an alternative to SMT.

BAJ

--
http://www.piclist.com hint: To leave the PICList

mailto:piclist-unsub...@mitvma.Extra.mit.edu


Byron A Jeff

unread,
Dec 7, 2002, 10:25:46 PM12/7/02
to PIC...@mitvma.nospam.mit.edu
On Sun, Dec 08, 2002 at 11:02:28AM +1000, cdb wrote:
> Sounds like a good idea, but aren't those boards just like Veroboard
> - the copper is in strips?

I don't think so. I'm pretty sure they are solid copper clad with holes drilled
so that a pristine board has all copper connected...

BAJ

--
http://www.piclist.com hint: To leave the PICList

mailto:piclist-unsub...@mitvma.AntiSpam.mit.edu


H. Carl Ott

unread,
Dec 7, 2002, 11:43:46 PM12/7/02
to PIC...@mitvma.counterspam.mit.edu
At 10:18 PM 12/7/2002, you wrote:


> >
> > SMT seems to be the main motivation for the toaster oven interest. I
> > used to use a lab hotplate for a similar purpose, but I found I could hand
> > solder faster. BGA might force me back to a hotplate or toaster oven.
>
>You could solder hundreds of pins in 10 minutes? Tell me more!

>tup...
>
>BAJ


Well, I solder pretty fast, but the thing to remember is that with SMT
you still have to apply solder paste and place the parts. This takes time.
I found that with standard designs of mine (a few soic and a couple dozen
discretes) that the quickest way was to put solder on one pad of all the
parts on the pcb. Then attach the parts by reflowing that pad with a
soldering iron and positioning the part with tweezers. When all the parts
are down, solder the rest of the pins.
No paste to apply, and you combine the parts placement with the soldering.

I did not even mention that solder paste is expensive, and has a somewhat
limited shelf life.

With high pin count SOICs or BGAs this is really not the way to do
it. But so far, it works for me.

One of the reasons I switched over to SMT was to stop drilling all those
holes. It's really not all that hard to solder SMTs, and there are other
benefits...

Regards

-carl

--
http://www.piclist.com hint: To leave the PICList

mailto:piclist-unsub...@mitvma.RemoveThis.mit.edu


cdb

unread,
Dec 8, 2002, 12:12:17 AM12/8/02
to PIC...@mitvma.cut.mit.edu
Well here is someone who advertises in an electronics NG he has
designed some prototyping type SMT boards

http://protoboards.theshoppe.com/index.html

These look OK a bit pricey but then he probably can't produce them in
thw quantities required to get the price down.

Colin
--
cdb, bod...@optusnet.RemovE.com.au on 08/12/2002

Light travels faster than sound. That's why some people appear bright
until they speak!

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


PicDude

unread,
Dec 8, 2002, 12:26:31 AM12/8/02
to PIC...@mitvma.no.spam.mit.edu
Seen these on Ebay as well. Lately I have needed to build
with some SMT components, but there are still many thru-hole
components, so a PCB permitting a mix of components would be
nice.

> Well here is someone who advertises in an electronics NG he has
> designed some prototyping type SMT boards
>
> http://protoboards.theshoppe.com/index.html
>
> These look OK a bit pricey but then he probably can't produce them in
> thw quantities required to get the price down.
>
> Colin

--

Charles Craft

unread,
Dec 8, 2002, 1:10:47 AM12/8/02
to PIC...@mitvma.no.spam.mit.edu
On Sat, 7 Dec 2002 15:59:27 -0500 "H. Carl Ott" <te...@SI.NoThiS.RR.COM> wrote:

> Hi Byron,
> Well I've made more the a few homebrew PCBs.
> (picture of what I'm working on this
> afternoon.
> http://users.rcn.com/carlott/at-tnc_smd.jpg )
>

I've seen polygon areas (ground plane) meshed/screened/webbed out like your
board before.

What's the purpose? Isn't etchant life determined by the amount of copper
dissolved in it and wouldn't this add a lot to it? Is this done on RF boards
only?

thanks
chuckc

Josh Koffman

unread,
Dec 8, 2002, 1:06:30 AM12/8/02
to PIC...@mitvma.spammers.are.scum.mit.edu
I think you misunderstood a bit. One of the mail order houses (possibly
4pcb) has a special where you can get your first order of 3 boards or so
free. James posted on it under something like [EE]: Free PCBs, or
something close. I thought it would allow you to try out this system
without having to pay big bucks for those PCBs from Digikey.

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
-Douglas Adams

Byron A Jeff wrote:
> > I guess the only other downside is the cost of the boards
> > (ouch). Might be cheaper (at least for the first run) to design a copper
> > clad predrilled board in Eagle, then use the free PCB service (first
> > time only) to get a few of them to play with.
>
> Doesn't seem cost effective. Olimex is one of the cheapest and fastest
> services around and a double sided 12.6"x7.8" is $104 USD. That's about
> 6 of the smallest digikey boards which would be half the cost.

--

Matt Pobursky

unread,
Dec 8, 2002, 1:29:11 AM12/8/02
to PIC...@mitvma.notneeded.mit.edu
On Sat, 7 Dec 2002 18:59:42 -0600, PicDude wrote:
> For me, drilling has always been the most painful part of the
> process.  I do however have this idea to use one of the PCB proto
> houses to supply me with boards that are completely drilled on a grid
> of 0.1" centers.  But it would be completely copper clad still.
> IIRC, $33 for 10" x 15" from 4pcb.

Advanced PCB (or almost any other board house) will nick you for a
healthy charge for excessive drilling density if you do that. IIRC,
their maximum drill density for the $33/ea. proto PCB's is 40 holes/sq.
in. -- anything more is an extra charge option. Drilling time is their
most expensive step in the manufacturing process.

Matt Pobursky
Maximum Performance Systems

PicDude

unread,
Dec 8, 2002, 1:45:46 AM12/8/02
to PIC...@mitvma.take.this.out.mit.edu
That figures. And I also found out that they don't allow
panelizing or mixed circuits on a single board (for making
up some other boards I need).

I have a coupon from PCB-Pool in the UK for a free proto
PCB. According to the rep, there are really no limitations
-- anything that I can put on the board is fine, and they'll
route all the boards for free as well. And they accept
Eagle .brd files directly. I might have to give them a try.

Cheers,
-Neil.

Jim

unread,
Dec 8, 2002, 1:44:07 PM12/8/02
to PIC...@mitvma.no.spam.mit.edu
>
> I've seen polygon areas (ground plane) meshed/screened/webbed out like
> your board before.
> ... Is this done on RF boards only?

Anything built for *true* RF shielding is going to be solid
coppper - and not some sort of 'pattern' that allows a
small amount to leakage to occur from one side to the
other ...

I spent the last couple of weeks working over some boards
designed for use in a system in the 2.4/2.6 GHz area - maximum
use of milled aluminum 'chassis' along with milled compartment
separators (between mixers, amps and filters) was the order
of the day with *tons* of non-thermally isolated vias in and
around power-Amp parts!

Not a shred of meshed/screened ground plane in sight.

I'm sure there are other reasons for this mesh approach
though (and if I were able to think today I might come
up with a few!) besides the 1) visual appeal aspect or that
it 2) requires slightly less 'plating' (or solder) after the
etching step 3) reduces the temperature 'sink' effect when
soldering in the vicinity of a PTH to ground (I had some
probs on those 2.4 GHz boards while doing re-work even
though the vias (to ground) were of the "thermally-isolated"
style!) ...

RF Jim


----- Original Message -----
From: "Charles Craft" <chuc...@MINDSPRING.NoThiS.COM>
To: <PIC...@MITVMA.Delete.MIT.EDU>
Sent: Sunday, December 08, 2002 12:16 AM
Subject: Re: [EE]: Comments wanted on PCB prototyping idea.


> On Sat, 7 Dec 2002 15:59:27 -0500 "H. Carl Ott" <te...@SI.RemoveThis.RR.COM> wrote:
>
> > Hi Byron,
> > Well I've made more the a few homebrew PCBs.
> > (picture of what I'm working on this
> > afternoon.
> > http://users.rcn.com/carlott/at-tnc_smd.jpg )
> >
>
> I've seen polygon areas (ground plane) meshed/screened/webbed out like
your
> board before.
>
> What's the purpose? Isn't etchant life determined by the amount of copper
> dissolved in it and wouldn't this add a lot to it? Is this done on RF
boards
> only?
>
> thanks
> chuckc
>

Peter L. Peres

unread,
Dec 8, 2002, 1:44:03 PM12/8/02
to PIC...@mitvma.delete.mit.edu
On Sat, 7 Dec 2002, Byron A Jeff wrote:
*>> What I really want is a cheap user friendly cnc drill/engraver to enhance
*>> my home prototyping . Under $500.00. This might help with drilling, but I
*>> really want it to engrave panels and make clean cutouts in enclosures. I
*>> could probably do it for $1000 to $1500 with a sherline and one of those
*>> stepper retrofits. Just a bit pricey for me at the moment.

This may be true but a hand drill is all you really need. It even works
without a stand. The secret is to get a drill that fits in the palm of
your hand. I modified a Johnson DC motor (3A,12V) by adding a mandrel.
Similar results can be had with an 'engraving' tool (the rotating kind).
You hold it like a pencil while resting your hand on the board and drill.
It works great, even in FR4, but phenolic and plastified cardboard (blue)
works better. The only drill guides needed are the holes in the pads. To
drill large holes predrill with the 0.8 or 1mm drill as above then use the
right diameter drill in a proper drill with the board jigged.

Peter

Josh Koffman

unread,
Dec 8, 2002, 2:04:37 PM12/8/02
to PIC...@mitvma.spammers.are.scum.mit.edu
Do you find that you are able to get the holes perfectly vertical?

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
-Douglas Adams

"Peter L. Peres" wrote:
> This may be true but a hand drill is all you really need. It even works
> without a stand. The secret is to get a drill that fits in the palm of
> your hand. I modified a Johnson DC motor (3A,12V) by adding a mandrel.
> Similar results can be had with an 'engraving' tool (the rotating kind).
> You hold it like a pencil while resting your hand on the board and drill.
> It works great, even in FR4, but phenolic and plastified cardboard (blue)
> works better. The only drill guides needed are the holes in the pads. To
> drill large holes predrill with the 0.8 or 1mm drill as above then use the
> right diameter drill in a proper drill with the board jigged.

--

Jim

unread,
Dec 8, 2002, 2:12:58 PM12/8/02
to PIC...@mitvma.take.this.out.mit.edu
chuckc wrote:
> > I've seen polygon areas (ground plane) meshed/screened/webbed out like
> > your board before.

I think I found the reason! From the last page of Henry Ott's paper
on using "the toner technique":

- - - - - - - - - - - - - - - - - - - - -

Laser printers really don t want to print large areas of black.

Hold the output of a test sheet up to the light and you will see
that the black is not black at all but actually gray.

The toner transfer system faithfully transfers this non-solid toner
to the copper, and when you etch you get some pitting.. On really
humid days this problem seems to get even worse. I get around this
by not having solid ground planes, instead using a hatch pattern
for the plane.

If anybody else has experienced this problem and has a solution, I d
love to hear about it.

- - - - - - - - - - - - - - - - - - - - -

RF Jim

Peter L. Peres

unread,
Dec 8, 2002, 2:15:04 PM12/8/02
to PIC...@mitvma.fight.spam.mit.edu
On Sun, 8 Dec 2002, Josh Koffman wrote:

*>Do you find that you are able to get the holes perfectly vertical?

No but they are near enough (I use a larger drill than prescribed for
pins). Alignment is good even for 64 pin DIL monsters and connectors. You
have to hold the drill firmly and vertically at the start then it works by
itself. It is not very tiring. Wear latex gloves to avoid FR4 fiber
implants into your skin and putting fingerprints on the bare copper.

Peter

Robert Rolf

unread,
Dec 8, 2002, 3:03:53 PM12/8/02
to PIC...@mitvma.fight.spam.mit.edu
Jim wrote:
> chuckc wrote:
> > > I've seen polygon areas (ground plane) meshed/screened/webbed out like
> > > your board before.
>
> I think I found the reason! From the last page of Henry Ott's paper
> on using "the toner technique":
> - - - - - - - - - - - - - - - - - - - - -
> Laser printers really don t want to print large areas of black.
>
> Hold the output of a test sheet up to the light and you will see
> that the black is not black at all but actually gray.
>
> The toner transfer system faithfully transfers this non-solid toner
> to the copper, and when you etch you get some pitting.. On really
> humid days this problem seems to get even worse. I get around this
> by not having solid ground planes, instead using a hatch pattern
> for the plane.
>
> If anybody else has experienced this problem and has a solution, I d
> love to hear about it.

You print your artwork twice and transfer it twice.
This also has the advantage of greatly reducing pin holes from toner
flaking off (if the darkness is set too high). It IS tricky to
get the alignment perfect, but it does work with great patience.

You could also try printing twice onto the transfer film.
Unfortunately most printers don't have reliable paper pickup, so
you need to hand feed the sheets get better (but not perfect)
alignment.

More modern laser printers don't have a problem with large black
areas (at least my HP LJ4s, 5, & 6's don't).

Robert

Jim

unread,
Dec 8, 2002, 3:22:43 PM12/8/02
to PIC...@mitvma.we.hate.sp.am.mit.edu
Do you have any recommendation as to which HP
laser printer works out best?

There is consignment shop near here that carries
quite a selection of HP printers - all I currently
have are dot-matrix and ink-jet!

I read-over Ott's paper on toner transfer method and
visited the http://www.dynaart.com website and read
about their http://www.dynaart.com/E.DTF/E.DTF2.html
"UNIVAC" - Vacuum Frame and heat-gun method (with a
Reynolds "Cooking Bag" placed over the transfer paper
during application of the heat from the gun) and this
looks like the way for me to go ...


RF Jim

----- Original Message -----
From: "Robert Rolf" <Rober...@UALBERTA.NoSpam.CA>
To: <PIC...@MITVMA.Spammers.Are.Scum.MIT.EDU>
Sent: Sunday, December 08, 2002 2:09 PM
Subject: Re: [EE]: Comments wanted on PCB prototyping idea.

H. Carl Ott

unread,
Dec 8, 2002, 4:22:33 PM12/8/02
to PIC...@mitvma.nospam.mit.edu
At 08:28 PM 12/8/2002, you wrote:

>This may be true but a hand drill is all you really need. It even works
>without a stand. The secret is to get a drill that fits in the palm of
>your hand. I modified a Johnson DC motor (3A,12V) by adding a mandrel.
>Similar results can be had with an 'engraving' tool (the rotating kind).
>You hold it like a pencil while resting your hand on the board and drill.
>It works great, even in FR4, but phenolic and plastified cardboard (blue)
>works better. The only drill guides needed are the holes in the pads. To
>drill large holes predrill with the 0.8 or 1mm drill as above then use the
>right diameter drill in a proper drill with the board jigged.
>

>Peter

My main problem with a hand drill, is breaking of the carbide drill
bits. I was never able to drill more then 10 holes with a .027 bit before
it went snap. A very small drill like you suggest might improve the issue,
but small drill press is really the best. I'm lucky to have a small Dumore
jewelers drill press. It's wonderful. But even a Dremel in a drill press
adapter is okay.

Pilot holes in the pads (annulus / donut or whatever your cad package
calls it) really helps alignment. Be sure to use them where possible.


-carl

H. Carl Ott

unread,
Dec 8, 2002, 4:36:04 PM12/8/02
to PIC...@mitvma.antispam.mit.edu
At 02:26 PM 12/8/2002, you wrote:
>Do you have any recommendation as to which HP
>laser printer works out best?
>
>There is consignment shop near here that carries
>quite a selection of HP printers - all I currently
>have are dot-matrix and ink-jet!
>
>I read-over Ott's paper on toner transfer method and
>visited the http://www.dynaart.com website and read
>about their http://www.dynaart.com/E.DTF/E.DTF2.html
>"UNIVAC" - Vacuum Frame and heat-gun method (with a
>Reynolds "Cooking Bag" placed over the transfer paper
>during application of the heat from the gun) and this
>looks like the way for me to go ...
>
>
>RF Jim


I tried a HP LJ-4 and still had the pitting problem. I had people do
test prints for me on a few printers and never really saw one that had good
solid blacks.
Might be the reason this product exists:
http://www.laserbuddy.com/ It does not solve the toner transfer
problem (second hand information).

It has been a couple of years, so newer printers may have fixed the
problem. But since I don't want to go out and replace my tank-like HP
LJ-III. It does not help me too much

That dyna-art vacuum box is certainly interesting. But I'm not sure it
will fix the problem. You really want something to not just melt the toner
but to get it to reflow slightly.
I'd like to hear some first hand experiences with it.
It is cheaper then the super-fuser dyna-art sells.

-carl
.

Peter L. Peres

unread,
Dec 8, 2002, 4:59:38 PM12/8/02
to PIC...@mitvma.fight.spam.mit.edu
On Sun, 8 Dec 2002, H. Carl Ott wrote:

*> My main problem with a hand drill, is breaking of the carbide drill
*>bits. I was never able to drill more then 10 holes with a .027 bit before
*>it went snap. A very small drill like you suggest might improve the issue,

Definitely. And I do not use carbide drills, I use plain HSS and change
them often. They are not so expensive and hard to break with a small
drill.

Peter

*>but small drill press is really the best. I'm lucky to have a small Dumore
*>jewelers drill press. It's wonderful. But even a Dremel in a drill press
*>adapter is okay.

I agree 100%. But a cheap drill press is worse than no drill press (too
much play = broken bits and inaccurate holes).

Peter

Josh Koffman

unread,
Dec 8, 2002, 5:14:21 PM12/8/02
to PIC...@mitvma.removethis.mit.edu
Their website points to a review by Don Lancaster. In it he says that it
won't work for toner transfer because the spray gets on both the toner
and non toner part of the paper. He does mention something called "Cross
Linking" that sounds interesting. Anything ever come of it? The article
is available at: http://www.tinaja.com/glib/resbn68.pdf

Josh

--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
-Douglas Adams

"H. Carl Ott" wrote:
> Might be the reason this product exists:
> http://www.laserbuddy.com/ It does not solve the toner transfer
> problem (second hand information).

--

H. Carl Ott

unread,
Dec 8, 2002, 5:37:47 PM12/8/02
to PIC...@mitvma.bogus.mit.edu
That's an old article, But it must of been the source of my 'second hand'
information about laserbuddy being unsuitable for toner transfer.
Just for the hell of it I ordered a can of the stuff a few minutes ago. I
figure it's worth doing a couple of tests myself.
All it really has to do is chemically melt the plastic toner, it might
just be acetone in aerosol. But if there's other stuff in the spray it
might screw things up.

The crosslinking is intriguing, but I never heard of any developments.

-carl

H. Carl Ott

unread,
Dec 8, 2002, 5:48:03 PM12/8/02
to PIC...@mitvma.delete.mit.edu
At 12:04 AM 12/9/2002, you wrote:

>Definitely. And I do not use carbide drills, I use plain HSS and change
>them often. They are not so expensive and hard to break with a small
>drill.

Yep, I'm sure the snapping problem is much less with HSS bits.
But can you find HSS bits smaller then .040? (wire gauge 60). That's a
bit large for some of the parts I use, but I can see it working for most stuff.


>*>but small drill press is really the best. I'm lucky to have a small Dumore
>*>jewelers drill press. It's wonderful. But even a Dremel in a drill press
>*>adapter is okay.
>
>I agree 100%. But a cheap drill press is worse than no drill press (too
>much play = broken bits and inaccurate holes).
>
>Peter

I agree with you there.
I've been lucky to find some good but cheap equipment at hamfests and
fleamarkets over the years.
I'm sure anybody could something to fit their budget on ebay.

Hmmmm, actually searching on drill press on ebay gets over 250 hits. Now
I wonder about vertical mills.....


-carl

PicDude

unread,
Dec 8, 2002, 6:57:31 PM12/8/02
to PIC...@mitvma.bogus.mit.edu
I've used commercial copying services (eg: Kinko's here
in the US) for transparencies for photo-etching, with
better results than I could get with a laser printer.
But it's not consistent -- I've had pale grey results
as well.

A somewhat related problem I've seen is that they print
it to paper and actually "photo" copy it which sometimes
results in a slight distorted image. Holding both
transparencies over each other proves this. Even on a
drill press, the drill hole may line up with the pad on
top, but not perfectly on the bottom.

Cheers,
-Neil.

> Laser printers really don t want to print large areas of black.
>
> Hold the output of a test sheet up to the light and you will see
> that the black is not black at all but actually gray.

> ...
>
> RF Jim

Josh Koffman

unread,
Dec 8, 2002, 6:59:30 PM12/8/02
to PIC...@mitvma.we.hate.sp.am.mit.edu
Cool, good luck with your trials. Keep us posted!

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
-Douglas Adams

"H. Carl Ott" wrote:
> Just for the hell of it I ordered a can of the stuff a few minutes ago. I
> figure it's worth doing a couple of tests myself.
> All it really has to do is chemically melt the plastic toner, it might
> just be acetone in aerosol. But if there's other stuff in the spray it
> might screw things up.

--

PicDude

unread,
Dec 8, 2002, 7:17:58 PM12/8/02
to PIC...@mitvma.no.spam.mit.edu
H. Carl Ott wrote:
> ...

> jewelers drill press. It's wonderful. But even a Dremel in a drill press
> adapter is okay.
>

Earlier this year, I picked up a Dremel drill press for this same
reason, but found it too wobbly. Pre-punching the holes with a
small punch seemed to help quite a bit, but it's a tedious
process still.

Cheers,
-Neil.

Richards, Justin P

unread,
Dec 8, 2002, 8:49:46 PM12/8/02
to PIC...@mitvma.notneeded.mit.edu
<snip>

While on the subject I once used a prototype board where the traces
looked a bit like this ...

o o o o o o o o o o
| | | | | | | | | |
o o o o o o o o o o
| | | | | | | | | |
o o o o o o o o o o
o-o-o-o-o-o-o-o-o-o---+
o-o-o-o-o-o-o-o-o-o-+ |
o o o o o o o o o o o o
| | | | | | | | | | | |
o o o o o o o o o o o o
| | | | | | | | | | | |
o o o o o o o o o o o o
o-o-o-o-o-o-o-o-o-o-+ |
o-o-o-o-o-o-o-o-o-o-o-+
o o o o o o o o o o
| | | | | | | | | |
o o o o o o o o o o
| | | | | | | | | |

My art is not too flash but basically it had 2 continuos traces that
snaked their way from one end of the board to the other under the IC for
- and +. Having + and - bus running the entire board was great. Also
there were segments of trace with three pads, ie after inserting an IC
there would be an additional 2 pads/holes to connect other components
to.

The stuff was perfect for what I do, you could really pack the
components in fast and with very few jumpers. Just cant seem to find it
any more and it had to come from either Dick Smiths, Jaycar or
Altronics. If anyone in OZ knows where I can get this stuff from I
would be really really pleased. All I can get now is vero strip i think
it is called, where the traces are silver looking. The stuff I am after
was copper looking traces.

Regards Justin

Ashley Roll

unread,
Dec 8, 2002, 9:40:03 PM12/8/02
to PIC...@mitvma.we.hate.sp.am.mit.edu
Hi Justin,

Jaycar still has something like it, catalogue numbers HP-9556 and HP-9558..
not sure if it is the exact same stuff, but it has the rails..

Cheers,
Ash.

---
Ashley Roll
Digital Nemesis Pty Ltd
www.digitalnemesis.com
Mobile: +61 (0)417 705 718


> -----Original Message-----
> From: pic microcontroller discussion list
> [mailto:PIC...@MITVMA.NoSpam.MIT.EDU]On Behalf Of Richards, Justin P
> Sent: Monday, 9 December 2002 11:53 AM
> To: PIC...@MITVMA.NoThiS.MIT.EDU
> Subject: Re: [EE]: Comments wanted on PCB prototyping idea.
>
>

> <snip>
>
> While on the subject I once used a prototype board where the traces
> looked a bit like this ...

[ snip ]

> My art is not too flash but basically it had 2 continuos traces that
> snaked their way from one end of the board to the other under
> the IC for
> - and +. Having + and - bus running the entire board was great. Also
> there were segments of trace with three pads, ie after inserting an IC
> there would be an additional 2 pads/holes to connect other components
> to.

--

Katinka Mills

unread,
Dec 8, 2002, 9:48:29 PM12/8/02
to PIC...@mitvma.nothis.mit.edu
> -----Original Message-----
> From: pic microcontroller discussion list
> [mailto:PIC...@MITVMA.Fight.Spam.MIT.EDU]On Behalf Of Richards, Justin P
> Sent: Monday, 9 December 2002 9:53 AM
> To: PIC...@MITVMA.We.Hate.SP.AM.MIT.EDU
> Subject: Re: [EE]: Comments wanted on PCB prototyping idea.
>
>
> <snip>
>
> While on the subject I once used a prototype board where the traces
> looked a bit like this ...
>
>
>
> o o o o o o o o o o
> | | | | | | | | | |
> o o o o o o o o o o
> | | | | | | | | | |
> o o o o o o o o o o
> o-o-o-o-o-o-o-o-o-o---+
> o-o-o-o-o-o-o-o-o-o-+ |
> o o o o o o o o o o o o
> | | | | | | | | | | | |
> o o o o o o o o o o o o
> | | | | | | | | | | | |
> o o o o o o o o o o o o
> o-o-o-o-o-o-o-o-o-o-+ |
> o-o-o-o-o-o-o-o-o-o-o-+
> o o o o o o o o o o
> | | | | | | | | | |
> o o o o o o o o o o
> | | | | | | | | | |
>
If it is the one I think it is it used to be sold by Tandy before they
dropped all decent electronics stuff.

<http://www.radioshack.com/product.asp?catalog%5Fname=CTLG&category%5Fname=C
TLG%5F005%5F014%5F003%5F000&product%5Fid=276%2D168>

Is the closest I can see on the main RatShack website.

Regards,

Kat.

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.427 / Virus Database: 240 - Release Date: 6/12/2002

Jim

unread,
Dec 8, 2002, 10:04:52 PM12/8/02
to PIC...@mitvma.antispam.mit.edu
> reason, but found it too wobbly. Pre-punching the holes with a

In the past I've 'pre-loaded' the linear bearings in these
wobbly presses by using as MANY rubber bands as required to
make one side (the movable piece) hug the other (the stationary
rail) on an old Sears drill press designed to hold a hand
drill ...

RF Jim

Jim

unread,
Dec 8, 2002, 10:12:58 PM12/8/02
to PIC...@mitvma.take.this.out.mit.edu
Carl, I came across this interesting tidbit concerning
Laser Buddy in this issue:

http://www.tinaja.com/glib/hackar4.pdf


Laser Buddy Document Spray by Buddy Products offers Bakerizing in
a can. You can spray this glop on any laser printed output and it
will get blacker, smoother, and more durable.

The resoution appears to go up and text gets slightly bolder.

Their spray is mostly methylene chloride, acetone, and some isobutyl
acetate. Works like a champ. But use this one outdoors only.

Watch out for fingerprints. No, it won t help direct toner printed
circuit transfers much. Because of residues.

- - - - - - - - - - - - - - - - - - - - - - - -

RF Jim


----- Original Message -----
From: "H. Carl Ott" <te...@SI.CounterSpam.RR.COM>
To: <PIC...@MITVMA.CounterSpam.MIT.EDU>
Sent: Sunday, December 08, 2002 4:41 PM
Subject: Re: [EE]: Comments wanted on PCB prototyping idea.

> That's an old article, But it must of been the source of my 'second hand'
> information about laserbuddy being unsuitable for toner transfer.

> Just for the hell of it I ordered a can of the stuff a few minutes ago.
I
> figure it's worth doing a couple of tests myself.
> All it really has to do is chemically melt the plastic toner, it might
> just be acetone in aerosol. But if there's other stuff in the spray it
> might screw things up.
>

> The crosslinking is intriguing, but I never heard of any developments.
>
> -carl

--

PicDude

unread,
Dec 9, 2002, 1:44:12 AM12/9/02
to PIC...@mitvma.bogus.mit.edu
Have you folks seen these "snap-apart" boards for SMT prototyping?
Really nice, but a bit pricey I think.
http://www.beldynsys.com/

Cheers,
-Neil.

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


Roman Black

unread,
Dec 9, 2002, 2:17:34 AM12/9/02
to PIC...@mitvma.extra.mit.edu
H. Carl Ott wrote:
>
> At 02:26 PM 12/8/2002, you wrote:
> >Do you have any recommendation as to which HP
> >laser printer works out best?

> That dyna-art vacuum box is certainly interesting. But I'm not sure it


> will fix the problem. You really want something to not just melt the toner
> but to get it to reflow slightly.
> I'd like to hear some first hand experiences with it.
> It is cheaper then the super-fuser dyna-art sells.


Hi Carl, don't use a laser printer!! It's not the
best way. I have a friend that uses a different
method and i've seen his boards.

Use a PHOTOCOPIER. You can print the artwork on
any laser or bubblejet, then go to a copy house
and photocopy on DARK. A photocopier puts down
a much heavier toner layer than a laserprinter.
Then iron on the PCB and it's ready for etching.

If you use good qual laser paper (fluffless)
for the first artwork sheet and the photocopy
they come out very nice. :o)
-Roman

Katinka Mills

unread,
Dec 9, 2002, 2:28:44 AM12/9/02
to PIC...@mitvma.removethis.mit.edu
> -----Original Message-----
> From: pic microcontroller discussion list
> [mailto:PIC...@MITVMA.Bogus.MIT.EDU]On Behalf Of Roman Black
> Sent: Monday, 9 December 2002 3:15 PM
> To: PIC...@MITVMA.Cut.MIT.EDU
> Subject: Re: [EE]: Comments wanted on PCB prototyping idea.
>
>
> H. Carl Ott wrote:
> >
> > At 02:26 PM 12/8/2002, you wrote:
> > >Do you have any recommendation as to which HP
> > >laser printer works out best?
>
> > That dyna-art vacuum box is certainly interesting. But I'm not sure it
> > will fix the problem. You really want something to not just
> melt the toner
> > but to get it to reflow slightly.
> > I'd like to hear some first hand experiences with it.
> > It is cheaper then the super-fuser dyna-art sells.
>
>
> Hi Carl, don't use a laser printer!! It's not the
> best way. I have a friend that uses a different
> method and i've seen his boards.
>
> Use a PHOTOCOPIER. You can print the artwork on
> any laser or bubblejet, then go to a copy house
> and photocopy on DARK. A photocopier puts down
> a much heavier toner layer than a laserprinter.
> Then iron on the PCB and it's ready for etching.
>
> If you use good qual laser paper (fluffless)
> for the first artwork sheet and the photocopy
> they come out very nice. :o)
> -Roman

I would have to disagree. I use a Brother HL-1440

I set it to graphics mode 1200x600 dpi. I print on to tracing paper which is
50GSM (the heavier stuff works better I am told, but mine does not distort).

I have made ~25 PCB's with this method and never had a problem (I use a
professional uv artwork box) and Kinston (IIRC) pre sensitised PCB's.

Regards.

Kat.

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.427 / Virus Database: 240 - Release Date: 6/12/2002

--

Roman Black

unread,
Dec 9, 2002, 3:17:32 AM12/9/02
to PIC...@mitvma.removethis.mit.edu
Katinka Mills wrote:

> > Hi Carl, don't use a laser printer!! It's not the
> > best way. I have a friend that uses a different
> > method and i've seen his boards.
> >
> > Use a PHOTOCOPIER. You can print the artwork on
> > any laser or bubblejet, then go to a copy house
> > and photocopy on DARK. A photocopier puts down
> > a much heavier toner layer than a laserprinter.
> > Then iron on the PCB and it's ready for etching.
> >
> > If you use good qual laser paper (fluffless)
> > for the first artwork sheet and the photocopy
> > they come out very nice. :o)
> > -Roman
>
> I would have to disagree. I use a Brother HL-1440
>
> I set it to graphics mode 1200x600 dpi. I print on to tracing paper which is
> 50GSM (the heavier stuff works better I am told, but mine does not distort).
>
> I have made ~25 PCB's with this method and never had a problem (I use a
> professional uv artwork box) and Kinston (IIRC) pre sensitised PCB's.


Yes but you have to do a UV exposure, at cost
of time and possible exposure error, then you
have the (quite high!) cost of pre-coated PCBs,
then the hassle and cost of developing, all before
you can etch.

My friends system requires one print of artwork
on relatively cheap paper, and one photocopy
followed by simply ironing on a blank PCB. This
is much cheaper and quicker. I can't see why anyone
would want to bother with pre-coated PCB cost
unless you are doing really fine tracks and NEED
a photographic quality system??
-Roman

Katinka Mills

unread,
Dec 9, 2002, 3:30:41 AM12/9/02
to PIC...@mitvma.counterspam.mit.edu
> -----Original Message-----
> From: pic microcontroller discussion list
> [mailto:PIC...@MITVMA.RemoveThis.MIT.EDU]On Behalf Of Roman Black
> Sent: Monday, 9 December 2002 4:14 PM
> To: PIC...@MITVMA.Fight.Spam.MIT.EDU
> Subject: Re: [EE]: Comments wanted on PCB prototyping idea.
>
>

Your last line said it, I do need high quality as most of my designs use an
AVR Mega 128 in QTP package.

But the Artwork box was free, and the cost of pre coated PCB's over blanks
from altronics is not that much (esp when you go to computronics to buy the
pre coated PCB's ;o)

Regards,

Kat.

---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.427 / Virus Database: 240 - Release Date: 6/12/2002

--

cdb

unread,
Dec 9, 2002, 6:30:31 AM12/9/02
to PIC...@mitvma.no.spam.mit.edu
I refer you all to the extremely expensive Euro700 (sorry my keyboard
ain't with it yet) project from Elektor last year - same one that is
supposed to be able to do minor milling, original design was for a 3
drill station hole/engraving borer.

Colin
--
cdb, bod...@optusnet.NoThiS.com.au on 09/12/2002

Light travels faster than sound. That's why some people appear bright
until they speak!

Bob Ammerman

unread,
Dec 9, 2002, 7:23:19 AM12/9/02
to PIC...@mitvma.removethis.mit.edu
I used to have an LJ4MP (postscript version of LJ4P). This printer had great
blacks.

I currently use an LJ2100M. This one is not nearly as good on the blacks (in
fact I was quite disappointed in it after being spoiled by the LJ4P).

Bob Ammerman
RAm Systems

Peter L. Peres

unread,
Dec 9, 2002, 2:03:43 PM12/9/02
to PIC...@mitvma.extra.mit.edu
On Sun, 8 Dec 2002, H. Carl Ott wrote:

*>At 12:04 AM 12/9/2002, you wrote:
*>
*>>Definitely. And I do not use carbide drills, I use plain HSS and change
*>>them often. They are not so expensive and hard to break with a small
*>>drill.
*>
*> Yep, I'm sure the snapping problem is much less with HSS bits.
*> But can you find HSS bits smaller then .040? (wire gauge 60). That's a
*>bit large for some of the parts I use, but I can see it working for most stuff.

This is a metric country. I use 1mm or 0.8mm drills. 1mm I can buy at any
DIY shop, 0.8 is $pecial order.

Peter

PicDude

unread,
Dec 9, 2002, 2:22:44 PM12/9/02
to PIC...@mitvma.we.hate.sp.am.mit.edu
Local electronics company here in Austin, TX has these drills for
a couple bucks each (IIRC). I got 1.0mm, 0.8mm and 0.7mm bits
from them just a few months back. There website does not cover
everything they have in stock, so it's better to call.
( http://www.tinkertronics.com ).

Cheers,
-Neil.

Peter L. Peres mumbled:

Spehro Pefhany

unread,
Dec 9, 2002, 2:48:05 PM12/9/02
to PIC...@mitvma.spammers.are.scum.mit.edu
At 08:38 PM 12/9/02 +0200, you wrote:


>This is a metric country. I use 1mm or 0.8mm drills. 1mm I can buy at any
>DIY shop, 0.8 is $pecial order.

Try the kind of supplier that feeds machine shops and such like.

I use KBC tools, which has HSS 0.80mm drills for about 45 cents US ea.
(the smallest sizes, 0.35mm and 0.30mm cost a lot more).

M.A. Ford solid carbide PCB drills with standard 0.125" shank for
collets are more, about $2.50 US. But two (because you might break
one rather than wear it out) of those costs about the
same as an envelope of 10 of the HSS bits.

I got a nice assortment of 50 solid carbide drills and mill bits and other
things (mostly smaller than 1mm) from another supplier for about $15, with
color-coded plastic collars and a storage box.

Unfortunately, the HSS small drills are usually only jobbers length below
about 1mm (or 3/64), rather than the preferable (IMHO) stubby screw machine
length.

>Best regards,

Spehro Pefhany --"it's the network..." "The Journey is the reward"
sp...@interlog.Cut.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

Chris Loiacono

unread,
Dec 9, 2002, 3:16:24 PM12/9/02
to PIC...@mitvma.fight.spam.mit.edu
Believe it or not, Digi-Key has 'em down to somewhere near 20 mils....
Also, I have been able to find .007 in dia HSS drills from several standard
machine tool suppliers.

BTW, they're not 'bits'. If I am not mistaken, 'tool bits' are used in
machines where the part moves, or rotates and the cutting tool is
stationary.

CL

> *> Yep, I'm sure the snapping problem is much less with HSS bits.
> *> But can you find HSS bits smaller then .040? (wire
> gauge 60).

--

William Chops Westfield

unread,
Dec 9, 2002, 3:48:21 PM12/9/02
to PIC...@mitvma.spammers.are.scum.mit.edu
There is, of course, a vast number of suppliers on eBay and similar of
"used" carbide drills in PCB-appropriate sizes. These are bits that some
CNC drilling operation has used for the "appropriate lifetime" (a certain
number of holes) and discarded. They almost certainly have a great deal
of life left, especially at hobbyist levels of RPM, feed rate, and visual
feedback...

BillW

Bob Barr

unread,
Dec 9, 2002, 3:50:25 PM12/9/02
to PIC...@mitvma.cut.mit.edu
On Mon, 9 Dec 2002 13:16:20 -0600, PicDude wrote:

>Local electronics company here in Austin, TX has these drills for
>a couple bucks each (IIRC). I got 1.0mm, 0.8mm and 0.7mm bits
>from them just a few months back. There website does not cover
>everything they have in stock, so it's better to call.
>( http://www.tinkertronics.com ).
>

They've changed their name to Bantam Electronics.

The new web site is : http://www.bantamei.com/ (It's redirected for
now but that may change in the future.)


Regards, Bob

Dominic Stratten

unread,
Dec 9, 2002, 4:12:48 PM12/9/02
to PIC...@mitvma.remove.mit.edu
Rapid Electronics in the UK sell a box of 10 used Tungsten Carbide drill
bits for #12.70+vat

They are reground so they are sharp again but this leaves them slightly
shorter than a CNC drilling machine is expecting. Useful for hobbyists but
not much good for a machine that is expecting a 38mm long drill bit.

These actually come from "European PCB manufacturers"


----- Original Message -----
From: "William Chops Westfield" <bi...@CISCO.We.Hate.SP.AM.COM>
To: <PIC...@MITVMA.CounterSpam.MIT.EDU>
Sent: Monday, December 09, 2002 8:52 PM
Subject: Re: [EE]: Comments wanted on PCB prototyping idea.

> There is, of course, a vast number of suppliers on eBay and similar of
> "used" carbide drills in PCB-appropriate sizes. These are bits that some
> CNC drilling operation has used for the "appropriate lifetime" (a certain
> number of holes) and discarded. They almost certainly have a great deal
> of life left, especially at hobbyist levels of RPM, feed rate, and visual
> feedback...
>
> BillW
>

Brendan Colven

unread,
Dec 9, 2002, 4:14:45 PM12/9/02
to PIC...@mitvma.antispam.mit.edu
> There is, of course, a vast number of suppliers on eBay and similar of
> "used" carbide drills in PCB-appropriate sizes. These are bits that some
> CNC drilling operation has used for the "appropriate lifetime" (a certain
> number of holes) and discarded. They almost certainly have a great deal
> of life left, especially at hobbyist levels of RPM, feed rate, and visual
> feedback...
>
> BillW

Exactly, I recently picked up a set of 50 assorted drills/rasps that had
1/8" shanks to use in my regular drill press. I've only broken one and it
was because I moved the PCB before I lifted the bit (safety glasses are a
must). The set was ~$27 CDN from www.leevalley.com (no affiliation) item #
"92W02.22" . They're based in Canada but I think they have a warehouse in
the US and can ship internationally.

I didn't happen to get a #66 bit but I did get a #65 and some #67s. Close
enough for me.

Brendan

Spehro Pefhany

unread,
Dec 9, 2002, 4:47:44 PM12/9/02
to PIC...@mitvma.remove.mit.edu
At 04:26 PM 12/9/02 -0500, you wrote:
>Exactly, I recently picked up a set of 50 assorted drills/rasps that had
>1/8" shanks to use in my regular drill press. I've only broken one and it
>was because I moved the PCB before I lifted the bit (safety glasses are a
>must). The set was ~$27 CDN from www.leevalley.com (no affiliation) item #
>"92W02.22" . They're based in Canada but I think they have a warehouse in
>the US and can ship internationally.

Thanks for posting the item number especially. I couldn't find it under
Hardware->Fasteners-> Drill bits, I should have looked under
Carving->Power carving!

Their web site has a ways to go, but they deal with online orders
efficiently IME.

>Best regards,

Spehro Pefhany --"it's the network..." "The Journey is the reward"

sp...@interlog.We.Hate.SP.AM.com Info for manufacturers: http://www.trexon.com


Embedded software/hardware/analog Info for designers: http://www.speff.com

--

H. Carl Ott

unread,
Dec 9, 2002, 4:54:10 PM12/9/02
to PIC...@mitvma.counterspam.mit.edu
Interesting,
And pretty much as expected. Lancaster has spoken a lot about bakerizing
toner in the past. Usually with heat. I'm sure this stuff is just doing the
same thing chemically. I'm not sure what residues are left, but when I get
a can of the stuff in I'll do some tests. Probably won't work, but we'll try.

Part of the issue is that we want to remelt the toner so that it's a
continuos film, just a bunch of overlapping dots.
Just heating the toner melts it, but it does not want to flow, instead
it just beads. The spray must lower the surface tension of melted toner and
improves the wetting action, also probably causes the toner to migrate into
the paper, might be a potential problem for us.

Heat with pressure causes the toner to flow, but the paper backing acts
like a buffer, reducing the amount of reflow. I might try to experiment
with sending the board with the toner transferred onto it but the paper
removed back through the laminator with some sort of high temp non-stick
plastic sheet to get the toner to reflow. The problem here, is too much
pressure and the toner melts and smears all over the place.

All of this is really for something that is only a minor problem with the
whole toner transfer process. The pitting on the large copper areas is
mostly an cosmetic matter.

-carl


At 09:18 PM 12/8/2002, you wrote:
>Carl, I came across this interesting tidbit concerning
>Laser Buddy in this issue:
>
>http://www.tinaja.com/glib/hackar4.pdf
>
>
>Laser Buddy Document Spray by Buddy Products offers Bakerizing in
>a can. You can spray this glop on any laser printed output and it
>will get blacker, smoother, and more durable.
>
>The resoution appears to go up and text gets slightly bolder.
>
>Their spray is mostly methylene chloride, acetone, and some isobutyl
>acetate. Works like a champ. But use this one outdoors only.
>
>Watch out for fingerprints. No, it won t help direct toner printed
>circuit transfers much. Because of residues.
>
>- - - - - - - - - - - - - - - - - - - - - - - -
>
>RF Jim

--

H. Carl Ott

unread,
Dec 9, 2002, 5:11:08 PM12/9/02
to PIC...@mitvma.removethis.mit.edu
Roman,
It's not really a photo-copier versus laser printer issue.
Any toner based system is going to to use the same basic principle to put
the toner on the paper.
Whether the image gets on the drum from a laser or from a mirror
lens/optic system does not really matter.

However, there is a large variation it the effectiveness of the different
toner engines out there.
I'm sure there are photocopiers out there that do a much better job then
my aging HP LJ-III. I'm also sure there other laser printer out there that
do a superior job. But you just have to test them to see what works.

My laser printer puts plenty of toner on the paper (even with the
contrast dial in the middle).

It's just those large solid block areas. You can see this easily by
doing a test printout and holding it up to the light. The edges of the area
will be black, but as you move towards the center of the filled area it
starts turning grey. I suspect it's an issue with the printer having
problems with transferring large areas of charge on the imaging drum. All
of these systems are primary designed to print/copy text. Which they do
quite well.

As it is the system works well, it could just use a little improvement.

Regards,
-carl

At 06:14 PM 12/9/2002, you wrote:

>Hi Carl, don't use a laser printer!! It's not the
>best way. I have a friend that uses a different
>method and i've seen his boards.
>
>Use a PHOTOCOPIER. You can print the artwork on
>any laser or bubblejet, then go to a copy house
>and photocopy on DARK. A photocopier puts down
>a much heavier toner layer than a laserprinter.
>Then iron on the PCB and it's ready for etching.
>
>If you use good qual laser paper (fluffless)
>for the first artwork sheet and the photocopy
>they come out very nice. :o)
>-Roman

--

William Chops Westfield

unread,
Dec 9, 2002, 7:32:02 PM12/9/02
to PIC...@mitvma.counterspam.mit.edu
Whether the image gets on the drum from a laser or from a mirror
lens/optic system does not really matter.

Thank you. I was about to make the same claim.


However, there is a large variation it the effectiveness of the
different toner engines out there.

There is also significant variation in the actual composition of toner.
(Reading the company MSDS's can be entertaining!) I wonder if there's
something simple that can be corolated with "success" in the toner transfer
PCB making process? (And could people who are doing well with it please
tell us exactly what type of printer (or copier) they're using?)

BillW

Jim

unread,
Dec 9, 2002, 7:46:11 PM12/9/02