Homebrew_PCBs

Hi,

I went thru the list and found some comments on bits for etching.

It seems like the box of 50 burrs from Harbor Freight are the best
bang for the buck, IF they last.

has anybody used them ?

any reports on how well they last ?

Dave

In a message dated 1/12/2003 12:39:07 PM Central Standard Time,
dave_mucha@... writes:

> It seems like the box of 50 burrs from Harbor Freight are the best bang for
> the buck, IF they last.
>

Dave: You didn't mention if these are solid-carbide, diamond-"plated", or
what. Carbide holds up well against such as FR-4 ("glass-epoxy"), Diamond
would "forever", if the diamond were the ONLY material touching the
glass-fibers, but that "diamond plated" stuff is just that: They
nickel-plate diamond-grit onto a HSS spindle, so as soon as the nickel
outer-"skin" is worn off (within first two seconds of use!), there is little
"holding" the diamond bits on the steel. AND, there is plenty of steel
(nickel-plated or not!) exposed to the glass-fibers, between the rather OPEN
grit-area, at the start!

I found out LONG ago that even the "new wholesale" price of solid-carbide
drill-bits (often more than $2.00 per bit!) is cheaper "per hole successfully
drilled" than ANY HSS drill-bits, even if you can resharpen them yourself
(every twenty holes?), and have a "chuck" capable of gripping HSS bits, which
are NOT made with "One Size Fits All" 1/8" dia. shanks as are carbide bits.

A carefully-zealous "browsing shopper" can find BOXES of useful-sizes of
solid-carbide bits offered in the various mail-order catalogs such as JDR
Microdevices and JAMECo, etc. I don't know that their current pricing is
like, but I have bought several boxes of 50 "reconditioned" bits of size #57
or #65, etc., for less than $1.00 per bit! NEVER throw away the 1/8" shank
when the "drill-part" breaks off, as that is VERY useful "tool-making stuff"!
All you need is a little diamond-wheel on a little grinder-motor, some
imagination, and minimal skill!

I do not "machine-etch" PCB's (easier to chemically-etch!!!), but I note
there are many "of us" who do, and I wonder why no one has offered that these
busted, "useless" (NOT!) solid-carbide drill-shanks would be ideal for the
tooling on a machine-etching job! Ya just have to grind an "engraver's tip"
on the pointed-end of the shank (where the drill-part left for a galaxy far,
far away)! True, one must have a nice small lathe (Unimat, Sherline, Taig,
etc.), if not a "bit holding fixture" and a diamond-cup-wheel grinder rig,
and some wisdom and skill. Some may-well know a "machinist-friend" who
grinds his own tooling, who will do these as a favor. Might be a "project"
for some ambitious soul! Others might find some grinding-shop willing to DO
this "for us all" at a reasonable price, mail-order---we just supply the
shanks/blanks. Can someone with more wisdom on this point comment???

Jan Rowland, old troll








[Non-text portions of this message have been removed]

Hi Jan,

yes, the box of 50 solid carbid, 'reconditioned' bits in various
sizes.

My Old Scott nameplate engraver uses a half flute single edge for
cutting and for milling, that seems like the simplest way. and you
can do that by hand if you have a diamond wheel.

The mechanics about what you pointed out is that the fiberglass does
not creat chips and leave the cutting zone, it powders and eats away
at soft stuff in the area. I have had to drill cindeblock on
occasion and used standard HSS drills. probably lasted about as long
as HSS in PC boards.

With my limtied metal working experiance, I have firmly come to the
conclusion that you 'can' do the impossible, but it takes time and
effort. you 'can' regrind after every few holes or just buy the
correct bits in the firstplace.

I too have a box full of punches and scribes made of broken tooling.
can't beat good carbide for putting your mark on parts.

I was wondering how well these reconditioned bits work ? why they are
sooo cheap ? are they undersized ? bad units? factory rejects? how
can so many people sell them ? maybe it is a PCB factory waste
product from the far east. Sombody has to make all the PCBs for our
TV's and Radio's.

Dave

--- In Homebrew_PCBs@yahoogroups.com, JanRwl@A... wrote:
> In a message dated 1/12/2003 12:39:07 PM Central Standard Time,
> dave_mucha@y... writes:
>
>
> > It seems like the box of 50 burrs from Harbor Freight are the
best bang for
> > the buck, IF they last.
> >
>
> Dave: You didn't mention if these are solid-carbide, diamond-
"plated", or
> what. Carbide holds up well against such as FR-4 ("glass-epoxy"),
Diamond
> would "forever", if the diamond were the ONLY material touching the
> glass-fibers, but that "diamond plated" stuff is just that: They
> nickel-plate diamond-grit onto a HSS spindle, so as soon as the
nickel
> outer-"skin" is worn off (within first two seconds of use!), there
is little
> "holding" the diamond bits on the steel. AND, there is plenty of
steel
> (nickel-plated or not!) exposed to the glass-fibers, between the
rather OPEN
> grit-area, at the start!
>
> I found out LONG ago that even the "new wholesale" price of solid-
carbide
> drill-bits (often more than $2.00 per bit!) is cheaper "per hole
successfully
> drilled" than ANY HSS drill-bits, even if you can resharpen them
yourself
> (every twenty holes?), and have a "chuck" capable of gripping HSS
bits, which
> are NOT made with "One Size Fits All" 1/8" dia. shanks as are
carbide bits.
>
> A carefully-zealous "browsing shopper" can find BOXES of useful-
sizes of
> solid-carbide bits offered in the various mail-order catalogs such
as JDR
> Microdevices and JAMECo, etc. I don't know that their current
pricing is
> like, but I have bought several boxes of 50 "reconditioned" bits of
size #57
> or #65, etc., for less than $1.00 per bit! NEVER throw away the
1/8" shank
> when the "drill-part" breaks off, as that is VERY useful "tool-
making stuff"!
> All you need is a little diamond-wheel on a little grinder-motor,
some
> imagination, and minimal skill!
>
> I do not "machine-etch" PCB's (easier to chemically-etch!!!), but I
note
> there are many "of us" who do, and I wonder why no one has offered
that these
> busted, "useless" (NOT!) solid-carbide drill-shanks would be ideal
for the
> tooling on a machine-etching job! Ya just have to grind
an "engraver's tip"
> on the pointed-end of the shank (where the drill-part left for a
galaxy far,
> far away)! True, one must have a nice small lathe (Unimat,
Sherline, Taig,
> etc.), if not a "bit holding fixture" and a diamond-cup-wheel
grinder rig,
> and some wisdom and skill. Some may-well know a "machinist-
friend" who
> grinds his own tooling, who will do these as a favor. Might be
a "project"
> for some ambitious soul! Others might find some grinding-shop
willing to DO
> this "for us all" at a reasonable price, mail-order---we just
supply the
> shanks/blanks. Can someone with more wisdom on this point
comment???
>
> Jan Rowland, old troll
>
>
>
>
>
>
>
>
> [Non-text portions of this message have been removed]

In a message dated 1/12/2003 5:04:39 PM Central Standard Time,
dave_mucha@... writes:

> I was wondering how well these reconditioned bits work ? why they are sooo
> cheap ? are they undersized ? bad units? factory rejects? how can so
> many people sell them ? maybe it is a PCB factory waste product from the
> far east. Sombody has to make all the PCBs for our TV's and Radio's.

Dave: GOOD questions, and ones I think worthy of discussion "here"! I have
wondered this myself, and I have had to rely only on my "imagination" for
answers, as it is not SO earth-shaking a quest that'd I'd "get in the car and
drive about to the several PC-houses within 50 miles, to inquire"!

They work very well on MY "CNC" drill, so long as the machine makes NO
attempt to "move to next hole" before the quill is 100% UP and still. And,
for hundreds of holes, it seems, before dull. Actually, I have never
"dulled" a carbide bit, yet! I have also used them in a vertical mill (used
as "drill-press") to drill HSS with success. Carbide is AMAZINGLY tough
stuff!

The word "recoditioned" merely means "re-sharpened"! I do NOT think they are
"bad", "rejects", or "far east waste". I think they are simply "sold cheap"
to those who scarf-up such, and re-sharpened, as the PCB-houses probably
drill almost-ONLY in "stacks", so that re-sharpened bits would be just TOO
short to be useful for that application. MY guess.

Yes, I am GUESSING that the houses that drill those PCB's for "our TV's and
computers" go through MILLIONS of these bits (ever see a PCB machine drill,
and CHANGE ITS OWN BITS? Wizardry! Four or five holes per second, through
four STACKS up to 7 boards deep! 100,000 and up RPM air-drills! So,
mind-bending as it is, I guess there ARE sources for those "boxes of fifty" I
have seen offered by, oh, three or four mail-order houses. Yes. Whew!

Surely there is SOMEone "in here" who is "professional" or has a sister who's
husband's cousin works for a PCB-house who can TELL us how far-off I am with
these guesses??? We'd ALL like to know?

Jan Rowland


[Non-text portions of this message have been removed]

We have a PCB layout guy here at work that worked most of his life at a PCB
house before working for us. He says that all those 'reconditioned' bits are
because the PCB industry needs such incredibly tight tolerances in hole
width, that after a certain number of holes, the drills are out of 'spec'
and can't be used. They aren't that dull, they're just a bit too narrow for
precision use. They stop using bits if they are actually out of spec or not
- there's just a certain limit (number of holes I guess) and after that
limit, the bits are discarded. They can be cheaply re-sharpened and sold to
people like us, so throwing them out isn't all that expensive for the PCB
house as they get some money back for their investment. My friend said that
it's much more expensive (because of an increase in the amount of 'errors'
in a board) to use a bit that might be out of spec than to just use new bits
every x holes. I couldn't believe the tolerances he was talking about in the
higher end PCB shops. These guys really have their processes down.

*Brian

> Dave: GOOD questions, and ones I think worthy of discussion
> "here"! I have
> wondered this myself, and I have had to rely only on my
> "imagination" for
> answers, as it is not SO earth-shaking a quest that'd I'd
> "get in the car and
> drive about to the several PC-houses within 50 miles, to inquire"!

We always drilled a maximum depth of five times the drill diameter.
This was the general rule of thumb for carbide bits. Most of the time
we drilled a stack of panels three deep and only one deep if they
were multilayer boards. We sharpened our bits only two or three times
and then sold them off. Most bits need to be resharpened after about
2000 to 2500 hits.

When drilling multilayers we always used new bits and never resharps.
Our bits were used first for multilayers, resharpened and then used
on two layer panels for a couple of cycles and then junked. Many
shops now days are only doing multilayer work so they put out tons of
bits that have drilled 2500 or less holes. These are the bits to get.
For single sided and hobby boards you can just about drill until they
break although the dull bits will make a larger burr.

The original auto tool changers were "pods" located along the front
of the drill table. The idea was for the machine to dump the bit and
then pick up the next one etc. They almost never worked correctly and
many shops just put their machines in a manual tool change mode. Many
newer machines have large capacity "cassette" type tool holders and
these work pretty good, much better than the original pod type.
I found that the machines would break more bits by crashing them into
the tool pods. The bits almost never break during drilling.

In the "manual" mode when a tool run was complete the machine table
would move to the "home" position and stop. The spindles had a
dynamic braking circuit to stop them in about four or five seconds.
These spindle bearings were so good that a new spindle might spin for
two minutes after the power was removed so the braking circuit was
needed.

The spindles were driven by a three phase frequency converter and the
dynamic brake circuit put 7 volts DC across one of the windings to
stop the spindle rotation. All of these spindles used .125" collets
and the plastic ring on the bit was used to set the depth. These
rings were pressed on so that the tip of the bit to the top of the
ring was set for .81". The Z axis of the machine could then be set
for a precise drill depth. Hope this info answers some of the bit
questions.

Tom

twb8899 wrote:

>

[deleted other interesting paragraphs]

> When drilling multilayers we always used new bits and never resharps.
> Our bits were used first for multilayers, resharpened and then used
> on two layer panels for a couple of cycles and then junked. Many
> shops now days are only doing multilayer work so they put out tons of
> bits that have drilled 2500 or less holes. These are the bits to get.
> For single sided and hobby boards you can just about drill until they
> break although the dull bits will make a larger burr.
>

Thanks Tom for sharing your knowledge with the group, although I have couple of
questions.

What is different about the holes drilled in multilayer boards that make it you
cannot use re-sharpened bits ? Is it because the bore is left smoother with a
new drill bit and sometimes that makes a more reliable copper connection
(electroless plated) to the inner layers ?

I read that they etch back the epoxy to make sure "drill smear" does not cover
the exposed copper foil of inneralyers inside the holes. The process is called
"desmear" and consists of three steps. An epoxy swell (a water miscible organic
solvent), epoxy etch (hot potassium permaganate/hydroxide), followed by
neutralize (sulfuric acid/ammonium persuflate). Is this what you used to do ?

I would imagine that after epoxy desmearing that it wont matter what mess is
left behind after drilling with old & blunt bits.

Adam

In a message dated 1/15/2003 2:29:33 AM Central Standard Time,
twb8899@... writes:

> Hope this info answers some of the bit questions.

Wow! Thanks, Tom! That is VERY useful to us hobby-folk!

Yeah, I watched a 4-quill CNC PCB-drill with "pods" change its own bits, and
all SEEMED to be fine, but I just could NOT help thinking, as I walked away,
"This CAN'T be so reliable ALL the time?!!!" GOOD to hear it isn't!

Interesting you say the quill motors are electric. I was told they are
PNEUMATIC, to run at 100,000 RPM! Hmmm...

Jan Rowland





[Non-text portions of this message have been removed]

Adam,

Everything you mentioned about etch back and desmearing is correct.
In our process we used new drill bits for multilayers to mainly limit
the amount of smearing. There could have been a requirement for this
in the MIL-55110P specification as well. Some MIL requirements would
require stuff like that. A new sharp bit helps big time on
multilayers since you need a good sharp cut through the inner layer
copper. These multilayer panels were so expensive that the bit cost
was a small fraction of the overall cost. It just wasn't worth the
risk drilling with used bits or resharps. The multilayer pre-preg
bonding materials have improved over the years so it's probably not
such a big deal any longer but we all know how hard it is to break
old habits.

When we went to the newer activated palladium catalyst in 1990 we no
longer needed desmearing for commercial multilayers. We only did a
slight etchback with straight technical grade (92.8%) sulfuric acid.
Desmearing was still performed on military boards though because of
the MIL specifications.

Tom



--- In Homebrew_PCBs@yahoogroups.com, adam Seychell
<adam_seychell@y...> wrote:

>
>
> twb8899 wrote:
> >
> [deleted other interesting paragraphs]
>
> > When drilling multilayers we always used new bits and never

resharps.

> > Our bits were used first for multilayers, resharpened and then

used

> > on two layer panels for a couple of cycles and then junked. Many
> > shops now days are only doing multilayer work so they put out

tons of

> > bits that have drilled 2500 or less holes. These are the bits to

get.

> > For single sided and hobby boards you can just about drill until

they

> > break although the dull bits will make a larger burr.
> >
>
> Thanks Tom for sharing your knowledge with the group, although I

have couple of

> questions.
>
> What is different about the holes drilled in multilayer boards that

make it you

> cannot use re-sharpened bits ? Is it because the bore is left

smoother with a

> new drill bit and sometimes that makes a more reliable copper

connection

> (electroless plated) to the inner layers ?
>
> I read that they etch back the epoxy to make sure "drill smear"

does not cover

> the exposed copper foil of inneralyers inside the holes. The

process is called

> "desmear" and consists of three steps. An epoxy swell (a water

miscible organic

> solvent), epoxy etch (hot potassium permaganate/hydroxide),

followed by

> neutralize (sulfuric acid/ammonium persuflate). Is this what you

used to do ?

>
> I would imagine that after epoxy desmearing that it wont matter

what mess is

> left behind after drilling with old & blunt bits.
>
> Adam

Jan,

The big professional drilling machines all use electric spindles but
some have air bearings for rotation. These air bearing spindles have
no ball bearings and therefore no metal to metal contact. Everything
spins on an air cushion with almost no run out. Very small holes can
be drilled with this type of spindle. The spindles are electric
driven to obtain the torque needed especially for larger holes.

All of my machines had ball bearing spindles, however, the spindles
did slide up and down in a cushion of air for very fast action. The
XY table also rides on an air cushion against the granite table.
My favorite machine was an Excellon EX-200 Driller/Router. This was
considered a "small"(6500 lbs!) machine. It had three spindles and
could drill or route three stacks of panels up to 12" x 24". It also
had an optical scope for digitizing. We used this machine for all of
our engineering and prototype work. The original specifications said
this machine could drill 400 holes per minute. This was probably true
for a .1" grid drilling only one deep. Our average drilling rate was
around 150 holes per minute when drilling three panels deep.

I shut off the auto tool changer mode since it just wasn't reliable
(ask any Excellon service tech!). Several types of spindles were
available but I used the 60,000 rpm drill/route spindles on the EX-
200 and 80,000 drill only spindles on the other machines. The air
requirements for these machines was about 20 cfm at 90 psi.

Tom

--- In Homebrew_PCBs@yahoogroups.com, JanRwl@A... wrote:
> In a message dated 1/15/2003 2:29:33 AM Central Standard Time,
> twb8899@y... writes:
>
>
> > Hope this info answers some of the bit questions.
>
>
> Wow! Thanks, Tom! That is VERY useful to us hobby-folk!
>
> Yeah, I watched a 4-quill CNC PCB-drill with "pods" change its own
bits, and
> all SEEMED to be fine, but I just could NOT help thinking, as I
walked away,
> "This CAN'T be so reliable ALL the time?!!!" GOOD to hear it isn't!
>
> Interesting you say the quill motors are electric. I was told they
are
> PNEUMATIC, to run at 100,000 RPM! Hmmm...
>
> Jan Rowland
>
>
>
>
>
> [Non-text portions of this message have been removed]

I am surprised at the air consumption and that they would use
electric motors for drilling. the motors would be much bigger than a
similar power air motor.

Dave




--- In Homebrew_PCBs@yahoogroups.com, "twb8899 <twb8899@y...>"
<twb8899@y...> wrote:

> Jan,
>
> The big professional drilling machines all use electric spindles

but

> some have air bearings for rotation. These air bearing spindles

have

> no ball bearings and therefore no metal to metal contact.

Everything

> spins on an air cushion with almost no run out. Very small holes

can

> be drilled with this type of spindle. The spindles are electric
> driven to obtain the torque needed especially for larger holes.
>
> All of my machines had ball bearing spindles, however, the spindles
> did slide up and down in a cushion of air for very fast action. The
> XY table also rides on an air cushion against the granite table.
> My favorite machine was an Excellon EX-200 Driller/Router. This was
> considered a "small"(6500 lbs!) machine. It had three spindles and
> could drill or route three stacks of panels up to 12" x 24". It

also

> had an optical scope for digitizing. We used this machine for all

of

> our engineering and prototype work. The original specifications

said

> this machine could drill 400 holes per minute. This was probably

true

> for a .1" grid drilling only one deep. Our average drilling rate

was

> around 150 holes per minute when drilling three panels deep.
>
> I shut off the auto tool changer mode since it just wasn't reliable
> (ask any Excellon service tech!). Several types of spindles were
> available but I used the 60,000 rpm drill/route spindles on the EX-
> 200 and 80,000 drill only spindles on the other machines. The air
> requirements for these machines was about 20 cfm at 90 psi.
>
> Tom
>
>
> --- In Homebrew_PCBs@yahoogroups.com, JanRwl@A... wrote:
> > In a message dated 1/15/2003 2:29:33 AM Central Standard Time,
> > twb8899@y... writes:
> >
> >
> > > Hope this info answers some of the bit questions.
> >
> >
> > Wow! Thanks, Tom! That is VERY useful to us hobby-folk!
> >
> > Yeah, I watched a 4-quill CNC PCB-drill with "pods" change its

own

> bits, and
> > all SEEMED to be fine, but I just could NOT help thinking, as I
> walked away,
> > "This CAN'T be so reliable ALL the time?!!!" GOOD to hear it

isn't!

> >
> > Interesting you say the quill motors are electric. I was told

they

> are
> > PNEUMATIC, to run at 100,000 RPM! Hmmm...
> >
> > Jan Rowland
> >
> >
> >
> >
> >
> > [Non-text portions of this message have been removed]

In a message dated 1/18/2003 6:17:44 AM Central Standard Time,
twb8899@... writes:

> Several types of spindles were available but I used the 60,000 rpm
> drill/route spindles on the EX-200 and 80,000 drill only spindles on the
> other machines. The air requirements for these machines was about 20 cfm at
> 90 psi.

Tom:

TWO points: 60,000 and 80,000 RPM? Were these "high frequency" motors, or
"brush motors"? TEFC, I would assume!

WOW, 20 CFM at 90 PSI! That would keep a 10 hp compressor running possibly
over 50% duty-cycle! Wouldn't it?

Jan Rowland


[Non-text portions of this message have been removed]

Dave and Jan,

Yes, these big machines took lots of air to run them. The air
requirements are because of the air bearing table and spindle
sleeves. Each axis glides on a cushion of air in these systems.

Our air source was a 15 HP, 60 CFM rotary screw compressor. It would
carry all three of our machines but it was a rare occasion when more
than two machines were running at the same time.

The spindle motors have high frequency three phase motors. There is a
frequency converter in the machine that drives a high voltage
transistor output stage at about 150 volts. The frequency is
controlled by the CPU and displayed on the CRT screen as spindle
speed.

If you ever get the chance to watch one of these Excellon machines
run you will certainly be amazed. Try to watch even one hole being
drilled and ten more will be drilled before your eyes can focus on
one... these machines really scream.

Tom

--- In Homebrew_PCBs@yahoogroups.com, JanRwl@A... wrote:
> In a message dated 1/18/2003 6:17:44 AM Central Standard Time,
> twb8899@y... writes:
>
>
> > Several types of spindles were available but I used the 60,000
rpm
> > drill/route spindles on the EX-200 and 80,000 drill only spindles
on the
> > other machines. The air requirements for these machines was about
20 cfm at
> > 90 psi.
>
> Tom:
>
> TWO points: 60,000 and 80,000 RPM? Were these "high frequency"
motors, or
> "brush motors"? TEFC, I would assume!
>
> WOW, 20 CFM at 90 PSI! That would keep a 10 hp compressor running
possibly
> over 50% duty-cycle! Wouldn't it?
>
> Jan Rowland
>
>
> [Non-text portions of this message have been removed]

In a message dated 1/19/2003 4:29:47 PM Central Standard Time,
twb8899@... writes:

> If you ever get the chance to watch one of these Excellon machines
> run you will certainly be amazed. Try to watch even one hole being
> drilled and ten more will be drilled before your eyes can focus on
> one... these machines really scream.

Yes, Tom, I have SEEN a 4-quill drill stacks of 7-high blanks, and the "peck
rate" was two to five holes per second (I am not an f-meter, myself, so I
could not tell exactly how fast). But I could have SWORN the quill-motors
were air-driven. Probably the operators had no clue, themselves!
Definitely a mind-bending thing to watch! $$$$!!!!!!!!!

High-F, huh? Lessee, if 400 Hz. can make 24,000 RPM (woodworking machinery
with which I had a tiny bit of hands-on), then 80,000 would require 1.33 kHz.
"AC" 3-phase. WOW! Guess that's no big deal electronically, but the EMI
must be SUMMIN to deal with!

Well, appreciate the info! Jan Rowland


[Non-text portions of this message have been removed]

JanRwl@... wrote:

> In a message dated 1/18/2003 6:17:44 AM Central Standard Time,
> twb8899@... writes:
>
>
>
>>Several types of spindles were available but I used the 60,000 rpm
>>drill/route spindles on the EX-200 and 80,000 drill only spindles on the
>>other machines. The air requirements for these machines was about 20 cfm at
>>90 psi.
>
>
> Tom:
>
> TWO points: 60,000 and 80,000 RPM? Were these "high frequency" motors, or
> "brush motors"? TEFC, I would assume!
>
> WOW, 20 CFM at 90 PSI! That would keep a 10 hp compressor running possibly
> over 50% duty-cycle! Wouldn't it?
>
> Jan Rowland
>

I haven't seen these motors running my self, but I know they carry a
hefty price tag (secondhand AUD$2500) and suck a lot of juice. I
learned this when I was shopping around for a PCB spindle for my home
made manual drilling machine. They need a variable frequency variable
voltage 3 phase power supply (voltage is proportional to speed because
of back EMF). I think the guy selling them said they run up to 250V at
full RPM. They can use precision ball bearings under 100,000 RPM and
air bearings for higher speeds. I did not buy one of course. Check the
following link.

http://www.westwind-airbearings.com/

Adam