Searching \ for '[EE] Markers for marking PICs?' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/devices.htm?key=pic
Search entire site for: 'Markers for marking PICs?'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Markers for marking PICs?'
2010\04\08@130539 by Vitaliy

face
flavicon
face
We have several projects here using the same PIC. We'd like to mark them
with color dots to be able to easily tell the difference. Is there anything
out there that will survive the RoHS reflow oven?

Vitaliy

2010\04\08@132649 by PICdude

flavicon
face
Check mcmaster.com, item # 2862T61.  Allegedly will withstand temps of  
800 deg-F or 2000 deg-F (depending on color).

Cheers,
-Neil.


Quoting Vitaliy <spam_OUTpiclistTakeThisOuTspammaksimov.org>:

> We have several projects here using the same PIC. We'd like to mark them
> with color dots to be able to easily tell the difference. Is there anything
> out there that will survive the RoHS reflow oven?
>
> Vitaliy
>
> -

2010\04\08@140843 by Dwayne Reid

flavicon
face
At 11:05 AM 4/8/2010, Vitaliy wrote:
>We have several projects here using the same PIC. We'd like to mark them
>with color dots to be able to easily tell the difference. Is there anything
>out there that will survive the RoHS reflow oven?

We use liquid ink pens.  The ones we have are silver and gold but I
have seen others.

The ones we have use a tip that is very similar to those that are
used for liquid ink pen plotters - there is some kind of a stiff wire
that runs down the center of the tip barrel.  It has a weight on it
and can be heard to slide back and forth as you shake the pen.

The paint dries quickly and you must keep the pen capped while not in
use or the tip clogs.

I'll take a look when I'm down in the shop and get manufacturer and
model numbers.  However, I *think* they come from Home Depot.

dwayne

--
Dwayne Reid   <.....dwaynerKILLspamspam@spam@planet.eon.net>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
(780) 489-3199 voice          (780) 487-6397 fax
http://www.trinity-electronics.com
Custom Electronics Design and Manufacturing

2010\04\08@143619 by Vitaliy

face
flavicon
face
Dwayne Reid wrote:
>>We have several projects here using the same PIC. We'd like to mark them
>>with color dots to be able to easily tell the difference. Is there
>>anything
>>out there that will survive the RoHS reflow oven?
>
> We use liquid ink pens.  The ones we have are silver and gold but I
> have seen others.
>
> The ones we have use a tip that is very similar to those that are
> used for liquid ink pen plotters - there is some kind of a stiff wire
> that runs down the center of the tip barrel.  It has a weight on it
> and can be heard to slide back and forth as you shake the pen.
>
> The paint dries quickly and you must keep the pen capped while not in
> use or the tip clogs.
>
> I'll take a look when I'm down in the shop and get manufacturer and
> model numbers.  However, I *think* they come from Home Depot.

Dwayne,

Mfr/model number would be greatly appreciated!

Vitaliy

2010\04\08@152436 by Vitaliy

face
flavicon
face
PICdude wrote:
> Check mcmaster.com, item # 2862T61.  Allegedly will withstand temps of
> 800 deg-F or 2000 deg-F (depending on color).

They only have two choices of colors: white and yellow (black is not really
a choice for marking chips). Purchased both, plus a box of their regular
markers. Will do some tests & report.


2010\04\08@153836 by PICdude

flavicon
face
I had marked (pun most certainly intended) these in the past as I  
intended to get a couple for a similar purpose, so a report would be  
nice.  I'm sure you can come up with a coding system of two or more  
dots (like binary) to have a number of different identifiers.

FWIW, I was thinking of having a number of small stencils made up  
(like component stencils), and running these markers over the chips to  
put text on them.

Cheers,
-Neil.



Quoting Vitaliy <piclistspamKILLspammaksimov.org>:

> PICdude wrote:
>> Check mcmaster.com, item # 2862T61.  Allegedly will withstand temps of
>> 800 deg-F or 2000 deg-F (depending on color).
>
> They only have two choices of colors: white and yellow (black is not really
> a choice for marking chips). Purchased both, plus a box of their regular
> markers. Will do some tests & report.
>
>
> -

2010\04\08@155834 by PICdude

flavicon
face
Hmmm... I missed this one too (16305T41), which is also available in  
green and blue.

BTW, black would be a nice stealth way to mark on the chips, as the  
chips are actually a dark shade of gray.  Reminds me of my web  
development days -- I'd put debugging info (on beta versions) right on  
the screen in an inconspicuous area with white text on a white  
background.  The users wouldn't know it's there, but our team could  
click and drag in the area to highlight the text and make it visible.  
But I digress.

Cheers,
-Neil.


Quoting PICdude <.....picdude3KILLspamspam.....narwani.org>:

{Quote hidden}

>> --

2010\04\08@165822 by Philip Pemberton

face
flavicon
face
Dwayne Reid wrote:
> We use liquid ink pens.  The ones we have are silver and gold but I
> have seen others.

In which case, the silver and gold markers that Pilot make may well work.

They're usually found in stationery shops in packs of two (one silver
pen, one gold pen) and the silver ones are good for marking ICs, storage
boxes, and so on. The gold ones.... not so much.

They do take a while to dry, though :-/

--
Phil.
piclistspamspam_OUTphilpem.me.uk
http://www.philpem.me.uk/

2010\04\08@172630 by Vitaliy

face
flavicon
face
"PICdude" wrote:
> FWIW, I was thinking of having a number of small stencils made up
> (like component stencils), and running these markers over the chips to
> put text on them.

Hm, that's an interesting idea. I think using a sponge might work even
better? A tiny silkscreen fixture would be perfect.

Vitaliy

2010\04\08@174302 by MCH

flavicon
face
How about a plain ol' stamp? (the kind you use for a return address)

Joe M.

Vitaliy wrote:
> "PICdude" wrote:
>> FWIW, I was thinking of having a number of small stencils made up
>> (like component stencils), and running these markers over the chips to
>> put text on them.
>
> Hm, that's an interesting idea. I think using a sponge might work even
> better? A tiny silkscreen fixture would be perfect.
>
> Vitaliy
>

2010\04\08@180326 by Olin Lathrop

face picon face
Vitaliy wrote:
> Hm, that's an interesting idea. I think using a sponge might work even
> better? A tiny silkscreen fixture would be perfect.

Something isn't making sense.  If you want the labels to survice reflow,
then that means you are distinguishing the firmware in the PICs at assembly
time.  That means you are programming the PICs before assembly or getting
them programmed that way.

What about in-circuit programming?  Your products are complex enough so that
they obviously need to be tested after assembly.  Usually adding ICSP to
that operation costs very little, allows a much shorter lead time between
new firmware and shipping it, and doesn't obsolete existing PICs when the
firmware changes.  Most of my customers have done this analisys at some
point, and most of them program the PICs as part of the production test and
calibration they have to perform anyway.  Some even load different firmware
just for the testing, then load the operational firmware once the hardware
has been verified.

I know ICSP isn't for everyone, but it seems the predominant way to do it
except for very high volumes, for good reason.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\04\08@181233 by PICdude

flavicon
face
Quoting Vitaliy <@spam@piclistKILLspamspammaksimov.org>:

> "PICdude" wrote:
>> FWIW, I was thinking of having a number of small stencils made up
>> (like component stencils), and running these markers over the chips to
>> put text on them.
>
> Hm, that's an interesting idea. I think using a sponge might work even
> better? A tiny silkscreen fixture would be perfect.
>
> Vitaliy
>
> -

2010\04\08@181650 by PICdude

flavicon
face
Depends on whether the ink can handle the reflow temps though ... (OP  
requirement).  Not sure the paint/ink in these paint markers would be  
able to be poured into a stamp pad and not dry up within minutes though.



Quoting MCH <KILLspammchKILLspamspamnb.net>:

> How about a plain ol' stamp? (the kind you use for a return address)
>
> Joe M.
>


2010\04\08@182324 by Alex Harford

face picon face
On Thu, Apr 8, 2010 at 10:05 AM, Vitaliy <RemoveMEpiclistTakeThisOuTspammaksimov.org> wrote:
> We have several projects here using the same PIC. We'd like to mark them
> with color dots to be able to easily tell the difference. Is there anything
> out there that will survive the RoHS reflow oven?

I don't follow the logic.  I assume you want to differentiate between
PICs because they have different firmware preprogrammed.  So once it's
on a board you know what PIC is on there, as you can inspect the PCB
to determine the product.

Is this so you can ensure the right PIC was installed onto the PCB?
If the manufacturer installs the wrong parts on your boards, wouldn't
this be detected in testing, and up to them to fix?

2010\04\08@182550 by Alex Harford

face picon face
On Thu, Apr 8, 2010 at 10:05 AM, Vitaliy <spamBeGonepiclistspamBeGonespammaksimov.org> wrote:
> We have several projects here using the same PIC. We'd like to mark them
> with color dots to be able to easily tell the difference. Is there anything
> out there that will survive the RoHS reflow oven?

Ignoring my last email, what about an embossing ink like the last
product on this page?  I have no idea if it would stand up to reflow
temps, but they are made to be heated.

http://www.stampin.com/online/accs/embacc.htm

2010\04\08@183726 by PICdude

flavicon
face
I'll speak for my situation...


Quoting Olin Lathrop <TakeThisOuTolin_piclistEraseMEspamspam_OUTembedinc.com>:

> Something isn't making sense.  If you want the labels to survice reflow,
> then that means you are distinguishing the firmware in the PICs at assembly
> time.  That means you are programming the PICs before assembly or getting
> them programmed that way.

Correct.


> What about in-circuit programming?  ...
> ...  Some even load different firmware
> just for the testing, then load the operational firmware once the hardware
> has been verified.
> ...

One board of my product has multiple configurations depending on  
firmware.  I was pre-programming with the different firmware versions  
and then sending those to the CEM.  The firmware is setup to run a  
self-test on the first two times it's powered up -- the CEM would run  
one of those to verify operation, and we'd run the second.  I even  
setup the PCB panels to have common power and ground lines going to  
all boards, so they could power up and test a whole panel at once.  
(BTW, there are some minor issues with this.*).

But yes, as i was looking into using CEM's in China and other places,  
I'm now doing as you mention -- placing test-only firmware on the PICs  
before assembly, and then re-programming to specific configurations  
later.  But still, based on a couple hardware configurations, the test  
code is a bit different, so the PICs are pre-programmed with 2  
different test-code versions at the start.

Cheers,
-Neil.

* In case anyone else wants to do this, be aware that separating the  
PCB's from the panel (I'm using mousebites), occasionally would lead  
to the common traces getting pull a bit off the board, and would have  
to be cut off carefully.  However, with the proper technique this can  
be avoided -- snap the break-joint one way, then the other, then pull  
apart.




2010\04\08@183922 by PICdude

flavicon
face
Quoting Alex Harford <RemoveMEharfordspamTakeThisOuTgmail.com>:


> I don't follow the logic.  I assume you want to differentiate between
> PICs because they have different firmware preprogrammed.  So once it's
> on a board you know what PIC is on there, as you can inspect the PCB
> to determine the product.
>
> Is this so you can ensure the right PIC was installed onto the PCB?
> If the manufacturer installs the wrong parts on your boards, wouldn't
> this be detected in testing, and up to them to fix?
> --


I have different several configurations for the same PCB.

Cheers,
-Neil.


2010\04\08@184232 by Vitaliy

face
flavicon
face
"MCH" wrote:
> How about a plain ol' stamp? (the kind you use for a return address)

I can't get it to work properly on a regular envelope. :)

2010\04\08@191456 by Vitaliy

face
flavicon
face
Olin Lathrop wrote:
>> Hm, that's an interesting idea. I think using a sponge might work even
>> better? A tiny silkscreen fixture would be perfect.
>
> Something isn't making sense.  If you want the labels to survice reflow,
> then that means you are distinguishing the firmware in the PICs at
> assembly
> time.  That means you are programming the PICs before assembly or getting
> them programmed that way.

Correct, they are being programmed with a bootloader.


{Quote hidden}

It's about firmware security. We program the bootloader into the chips, and
the bootloader then decrypts and programs the actual firmware. We trust our
Chinese CM, but not enough to send them raw binaries.

The circuit has an ICSP connector, but we only use it for development or for
reprogramming the bootloader.

Vitaliy

2010\04\08@191833 by Vitaliy

face
flavicon
face
"PICdude" wrote:
> You'd need to get the ink to the sponge first though.  I would think
> this "paint" marker would work fine though.  BTW, these are the
> component stencils I have in mind, but just with my custom
> text/graphics on it.
>
> http://www.smtdispenser.com/index,5,Single-Component-Printing-System.html
> (see
> on the right side, just above the table at the bottom).

You would need to use stencil-style letters, and I'm sure you can only make
it so small before the Os become solid white circles.

Vitaliy

2010\04\08@191949 by Vitaliy

face
flavicon
face
"PICdude" wrote:
> Depends on whether the ink can handle the reflow temps though ... (OP
> requirement).  Not sure the paint/ink in these paint markers would be
> able to be poured into a stamp pad and not dry up within minutes though.

The dry time is specified as 25-30 minutes (oil based ink). So I'm sure it's
possible but the quality won't be that great.

Vitaliy

2010\04\08@192216 by Vitaliy

face
flavicon
face
"Alex Harford" wrote:
>> We have several projects here using the same PIC. We'd like to mark them
>> with color dots to be able to easily tell the difference. Is there
>> anything
>> out there that will survive the RoHS reflow oven?
>
> I don't follow the logic.  I assume you want to differentiate between
> PICs because they have different firmware preprogrammed.  So once it's
> on a board you know what PIC is on there, as you can inspect the PCB
> to determine the product.

I replied to Olin's question. Let me know if my explanation doesn't make
sense.


> Is this so you can ensure the right PIC was installed onto the PCB?

Yes.


> If the manufacturer installs the wrong parts on your boards, wouldn't
> this be detected in testing, and up to them to fix?

They already messed up once (not fatal, but if it happens in the future they
would have to replace the PICs). It's cheaper and easier to prevent this
from occuring. When they do it "at their expense", eventually we see our
prices go up.

Vitaliy

2010\04\08@193045 by Vitaliy

face
flavicon
face
Alex Harford wrote:
>> We have several projects here using the same PIC. We'd like to mark them
>> with color dots to be able to easily tell the difference. Is there
>> anything
>> out there that will survive the RoHS reflow oven?
>
> Ignoring my last email, what about an embossing ink like the last
> product on this page?  I have no idea if it would stand up to reflow
> temps, but they are made to be heated.
>
> http://www.stampin.com/online/accs/embacc.htm

Aren't they just liquid rubber? I would expect it to evaporate or burn at
RoHS reflow temperatures.

Vitaliy

2010\04\08@193657 by Xiaofan Chen

face picon face
On Fri, Apr 9, 2010 at 6:41 AM, Vitaliy <piclistEraseMEspam.....maksimov.org> wrote:
> "MCH" wrote:
>> How about a plain ol' stamp? (the kind you use for a return address)
>
> I can't get it to work properly on a regular envelope. :)

In my previous job, the pre-programmed chips have a sticky
label on it and I do not think they have any issues with
reflow process.

--
Xiaofan http://mcuee.blogspot.com

2010\04\08@194843 by Marcel Duchamp

picon face
On 4/8/2010 4:17 PM, Vitaliy wrote:
> "PICdude" wrote:
>> You'd need to get the ink to the sponge first though.  I would think
>> this "paint" marker would work fine though.  BTW, these are the
>> component stencils I have in mind, but just with my custom
>> text/graphics on it.
>>
>> www.smtdispenser.com/index,5,Single-Component-Printing-System.html
>> (see
>> on the right side, just above the table at the bottom).
>
> You would need to use stencil-style letters, and I'm sure you can only make
> it so small before the Os become solid white circles.
>
> Vitaliy
>

I wonder if you could give the different firmware versions a different
number each.  Then represent different numerals with a different color
ink.  For example three color stripes like brown, black, red might
indicate program number 102...
This might be worth getting a patent on!

2010\04\08@200752 by Vitaliy

face
flavicon
face
Xiaofan Chen wrote:
>>> How about a plain ol' stamp? (the kind you use for a return address)
>>
>> I can't get it to work properly on a regular envelope. :)
>
> In my previous job, the pre-programmed chips have a sticky
> label on it and I do not think they have any issues with
> reflow process.

We're looking into that for stand-alone chips but for production one ink dot
is so much cheaper.

Vitaliy

2010\04\08@200906 by Vitaliy

face
flavicon
face
Marcel Duchamp wrote:
> I wonder if you could give the different firmware versions a different
> number each.  Then represent different numerals with a different color
> ink.  For example three color stripes like brown, black, red might
> indicate program number 102...
> This might be worth getting a patent on!

I'm afraid it's not a new idea. ;)


2010\04\09@000659 by Djula Djarmati

flavicon
> We have several projects here using the same PIC. We'd like to mark them
> with color dots to be able to easily tell the difference. Is there anything
> out there that will survive the RoHS reflow oven?

We use an engraving pen for this, also for serial number engraving on
finished PCBs since it's indestructible. Paint is problematic since it
has to survive both the oven and the PCB washing. I have seen commercial
components with faded or erased markings after washing.

For paint, I would try dots of industrial heat resistant paint like this:
http://www.globalindustrial.com/g/maintenance/paint/aerosol/krylon-high-heat-paints

Djula

2010\04\09@005220 by William \Chops\ Westfield

face picon face

> We'd like to mark them with color dots ... that will survive the  
> RoHS reflow oven?

Well, since the epoxy packages survive the heat, I imagine that an  
epoxy paint/glue (at least one designed for high temperatures) would  
also survive.  You can probably use a clear-ish high temp epoxy glue  
and your choice of color of powdered glass or "mineral" cosmetics  
(titanium dioxide for white, assorted iron oxides for red and yellow,  
etc.)

Another possibility is the "powder coating" powders that are pretty  
readily available these days.  They seem to be designed to melt/cure  
at 400-600F or so (which covers the after-reflow; I guess you'd need  
some kind of binder or paint-like version to stick on the chips BEFORE  
assembly.)  (hmm.  Bought some of that at HF a while back.  Maybe I'll  
go make me some red chips!) (don't use the electrostatic-based powder-
coating equipment, though !  :-)

PVC Plastisol Inks (for screenprinting) seem to cure at less than  
350F; they're probably not good to use at the higher temperatures.  
IIRC you're not supposed to laser-cut PVC because of the caustic fumes  
produced (HCL)

An then there are all the inks and whatnot that they use for  
soldermask and silkscreen on the PCBs, though I'm not sure that these  
are convenient to buy or use in small quantities  (most are some  
variety of epoxy paint again, I think.)

I guess my main point is that reflow isn't THAT high a temperature, or  
you'd have lots of other problems!

BillW

2010\04\09@012337 by Ruben Jönsson

flavicon
face
> I wonder if you could give the different firmware versions a different
> number each.  Then represent different numerals with a different color
> ink.  For example three color stripes like brown, black, red might
> indicate program number 102...
> This might be worth getting a patent on!

Assuming that this was a new patentable idea to start with you have effectevly
just made it unpatentable (is that a word?) by making it publicly known.

In fact, there are companys putting adds containing the basics of the idea in
the personal area of low volume magazines just to make the idea publicly known
and then hindering other companies to take a patent. By having it in the
magazine you have proof of when it was invented.

/Ruben


==============================
Ruben Jönsson
AB Liros Electronic
Box 9124, 200 39 Malmö, Sweden
TEL INT +46 40142078
FAX INT +46 40947388
EraseMErubenspampp.sbbs.se
==============================

2010\04\09@014953 by Vitaliy

face
flavicon
face
AFAIK it's not true. I believe that under US law, you have one year to
patent your idea once you make it public.

{Original Message removed}

2010\04\09@015551 by Vitaliy

face
flavicon
face
www.uspto.gov/smallbusiness/patents/filing.html

"An invention cannot be patented if .. the invention was ... in public use
or on sale in this country more than one year prior to the application for
patent in the United States."

2010\04\09@015633 by PICdude

flavicon
face
Quoting Ruben Jönsson <RemoveMErubenEraseMEspamEraseMEpp.sbbs.se>:

> Assuming that this was a new patentable idea to start with you have  
> effectevly
> just made it unpatentable (is that a word?) by making it publicly known.
>

Is this something new (as of within the past few years)?  AFAIR, the  
USPTO says that you have one year *after* an idea is made public to  
patent it.  That is of course, assuming that no one has used the idea  
in that time.



2010\04\09@040530 by Ruben Jönsson

flavicon
face

> Quoting Ruben Jönsson <RemoveMErubenspam_OUTspamKILLspampp.sbbs.se>:
>
> > Assuming that this was a new patentable idea to start with you have  
> > effectevly
> > just made it unpatentable (is that a word?) by making it publicly known.
> >
>
> Is this something new (as of within the past few years)?  AFAIR, the  
> USPTO says that you have one year *after* an idea is made public to  
> patent it.  That is of course, assuming that no one has used the idea  
> in that time.
>
>

I heard this from a patent engineer here in Sweden when we were looking into
patent possibilities for a product we make. I thought patent rules where
international but apparently not so.

Here in Sweden you start the patent process by filing a patent application. The
invention can not be publicly known at at the date when the application is
filed. The patent office then makes a review of the application and this takes
12 months. Before that you don't have a patent but in that time it can be made
publicly known.

Here are the Swedish rules:

<www.prv.se/In-English/Patents/What-to-consider-before-filing/Conditions-
for-a-patent/>

and specifically:

<www.prv.se/In-English/Patents/What-to-consider-before-filing/Conditions-
for-a-patent/Keeping-your-invention-secret/>

In the link that Vitaliy posted:

<http://www.uspto.gov/smallbusiness/patents/filing.html>

It says:

An invention cannot be patented if "(a) the invention was known or used by
others in this country, or patented or described in a printed publication in
this or a foreign country, before the invention thereof by the applicant for
patent," or "(b) the invention was patented or described in a printed
publication in this or a foreign country or in public use or on sale in this
country more than one year prior to the application for patent in the United
States..."

Does this mean that (a) you can not file for a patent if someone else have
published the invention (made it publicly known).

And (b) that You can make your own invention publicly known and then you have
one year to file for a patent?

If so, then by writing about it here on the PICLIST, you have effectevly
stopped others from filing a patent for that invention and also stopped
yourself from filing a swedish patent.

/Ruben
==============================
Ruben Jönsson
AB Liros Electronic
Box 9124, 200 39 Malmö, Sweden
TEL INT +46 40142078
FAX INT +46 40947388
RemoveMErubenTakeThisOuTspamspampp.sbbs.se
==============================

2010\04\09@045543 by Alan B Pearce

face picon face
> In my previous job, the pre-programmed chips have a sticky
> label on it and I do not think they have any issues with
> reflow process.

Even if the label was burnt away in the reflow process, surely it has done
its job by identifying the chip for the assembly house? If you need to
identify the chip once the boards come back to you then you have 'real big'
(tm) problems.

2010\04\09@070501 by Peter

picon face
> Assuming that this was a new patentable idea to start with you have effectevly
> just made it unpatentable (is that a word?) by making it publicly known.

Ruben, while I understand where you are coming from, you cannot patent a color
code used for encoding any numbers or letters in 2010. Saying that copious
previous art exists would be an euphemism.

I recently had the "pleasure" to find no fewer than two magnetic amplifier
patents (controlling the inductance of an AC inductor using a second DC
inductor's magnetic field) issued in 1998 and later. One was not issued in the
US (surprise) and previous art is copious on the internet (found that too, US
reputable sources too, 1994 and earlier).

Back to the original poster, it is possible to mark chip bodies in many ways,
using just one color. One would be barcodes, which can also be applied with a
stamp, and inkjet printing. Epoxy inks exist which withstand oven temperatures
without trouble. Many other shapes can and have been used for this, including
simple geometric shapes.

The simplest method is probably to simply coat the entire chip's back with
brightly colored epoxy paint with a roller. Use a different color for different
chips.

I once designed and built a device which marked various electronic components
using a stack of abrasive disks on a spinning axle. The stack had spacers so the
flexible disks would leave a row of stripe like marks at defined intervals
resembling a barcode. It was used to mark chips and also other sorted parts,
manually or while fed through a jig. The first jig for chips was made from an
aluminum chip holding 'stick'. A similar system which I built  but saw little
use used a CNC router spindle to mark a pattern of 'dots' (non-penetrating or
penetrating abraded, drilled or milled holes) on a grid in PCBs and on chips.
Obviously the marking was done in software (generated G codes sequence). None of
these were new or patentable at the time I implemented them.

I personally consider people and organizations who patent very obvious methods
and procedures where copious previous art exists, to be pond scum and a public
danger of the 'unscrupulous lawyer' type. Anything goes to prevent the actions
of those, as, after all, they don't have any taboos either, and I am sure they
would understand that 'the goal excuses the means' more than anyone else.

I am very surprised that an engineer would bring up the issue of 'you just
destroyed a possibly patentable idea by bringing it up' on this list. At least
one other person on this list did register a patent on a method (not device)
which is very obvious, in the US, and the idea was discussed on this very
piclist while the patent was pending (!), no-one finding it obvious at the time.
I guess you know who you are, I won't bring it up again.

 Peter


2010\04\09@081216 by Olin Lathrop

face picon face
Vitaliy wrote:
> It's about firmware security. We program the bootloader into the
> chips, and the bootloader then decrypts and programs the actual
> firmware. We trust our Chinese CM, but not enough to send them raw
> binaries.
>
> The circuit has an ICSP connector, but we only use it for development
> or for reprogramming the bootloader.

That still doesn't make sense.  That would mean all the PICs going thru the
reflow oven would have the bootloader on them.  If so, what's the point of
marking them?  I can see you might want to later distinguish between
bootloader and fully programmed boards, but that would be after reflow.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\04\09@081559 by Olin Lathrop

face picon face
Vitaliy wrote:
>> If the manufacturer installs the wrong parts on your boards, wouldn't
>> this be detected in testing, and up to them to fix?
>
> They already messed up once (not fatal, but if it happens in the
> future they would have to replace the PICs). It's cheaper and easier
> to prevent this from occuring. When they do it "at their expense",
> eventually we see our prices go up.

But aren't you giving them a test jig?  Any wrong parts should be caught
there.  It's also usually easy to incorporate PIC programming into the test
procedure so that the manufacturer can just install bare PICs straight from
Microchip and they get programmed as part of the test process.  That also
helps in that a reworked unit can be run thru the test procedure regardless
of what firmware got loaded so far, since it will get overwritten anyway.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\04\09@085940 by Olin Lathrop

face picon face
Ruben Jönsson wrote:
> Assuming that this was a new patentable idea to start with you have
> effectevly just made it unpatentable (is that a word?) by making it
> publicly known.

Actually, in the US at least, he now has 1 year to file.  Of course the idea
(color bands to denote numbers) is clearly not a new concept and silly as a
patentable idea, so there is no hurry to file anything.

> In fact, there are companys putting adds containing the basics of the
> idea in the personal area of low volume magazines just to make the
> idea publicly known and then hindering other companies to take a
> patent. By having it in the magazine you have proof of when it was
> invented.

Mentioning it on the PIClist also works, since it's public and there are
various archival records.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\04\09@090226 by Olin Lathrop

face picon face
PICdude wrote:
> Is this something new (as of within the past few years)?  AFAIR, the
> USPTO says that you have one year *after* an idea is made public to
> patent it.  That is of course, assuming that no one has used the idea
> in that time.

No, the idea can be used during the 1 year.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\04\09@090907 by Olin Lathrop

face picon face
Ruben Jönsson wrote:
> Does this mean that (a) you can not file for a patent if someone else
> have published the invention (made it publicly known).

No.  First you can file anything you want, the question is whether there is
grounds for rejecting it.  In the US at least, a patent is granted to the
first to invent, not the first to file.  If someone describes a idea you had
for a while here on the PIClist, you can still get a patent on it if you
file within 1 year and you can show proof of having invented the idea before
the other guy.  This is where signed and dated lab notebooks get important.

> And (b) that You can make your own invention publicly known and then
> you have one year to file for a patent?

Yes, in the US at least.

> If so, then by writing about it here on the PICLIST, you have
> effectevly stopped others from filing a patent for that invention and
> also stopped yourself from filing a swedish patent.

Someone else could still be granted a US patent if they file within a year
and can prove they came up with the idea before the earliest date you can
prove.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\04\09@091614 by Ruben Jönsson

flavicon
face
> > Assuming that this was a new patentable idea to start with you have
> effectevly
> > just made it unpatentable (is that a word?) by making it publicly known.
>
> Ruben, while I understand where you are coming from, you cannot patent a
> color
> code used for encoding any numbers or letters in 2010. Saying that copious
> previous art exists would be an euphemism.
>

Peter, I have no intention to patent a color code (that was somebody elses
idea). I just pointed out that by making the idea/invention publicly known with
a post on the PICLIST, it may become unpatentable. At least, that is the case
here in Sweden, but not so in USA, I learned.

Regards / Ruben
==============================
Ruben Jönsson
AB Liros Electronic
Box 9124, 200 39 Malmö, Sweden
TEL INT +46 40142078
FAX INT +46 40947388
EraseMErubenspamspamspamBeGonepp.sbbs.se
==============================

2010\04\09@091743 by Olin Lathrop

face picon face
Peter wrote:
> At least one other person on this list did register a patent on
> a method (not device) which is very obvious, in the US, and the idea
> was discussed on this very piclist while the patent was pending (!),
> no-one finding it obvious at the time.

You are contradicting yourself.  You say the method was very obvious, but
noone found it obvious at the time.  Sure, lots of things are obvious once
they are explained to you.  In fact that's the mark of a really good idea,
one which didn't occur to you but was really obvious after someone else came
up with it.

I hope you're not referring to my and Dave Tweed's US patent 7,411,378,
which is on a way to do power factor correction without measuring the
current, and also a way of regulating the output tightly without defeating
the power factor correction.  It was discussed on the PIClist at the time,
but nobody mentioned any prior art, and the patent office couldn't find any
either.


********************************************************************
Embed Inc, Littleton Massachusetts, http://www.embedinc.com/products
(978) 742-9014.  Gold level PIC consultants since 2000.

2010\04\09@130112 by Peter

picon face
Olin Lathrop <olin_piclist <at> embedinc.com> writes:
> You are contradicting yourself.  You say the method was very obvious, but
> noone found it obvious at the time.  Sure, lots of things are obvious once

Thanks for pointing that out, no thanks for elaborating. The phrase I wrote
should have said 'no-one found it (the innovation) un-obvious or new at the time
when it was discussed'.

I don't want to go into further details. It does not surprise me at all that the
USPTO did not find any 'previous art' in general. They seem to have serious
trouble with that since, about, the late 1960s, approximately. It is none of my
business to tell them what to do.

-- Peter


2010\04\09@130756 by Peter

picon face
Oven paint is available in white, yellow, ochre and brown and can be used as is.
Drying time is long but an hour at 100 deg C in a toaster should work wonders.
One can write with oven paint using a stylus but I'd use a stamp.

-- Peter


2010\04\09@132654 by Russell McMahon

face picon face
> chips are actually a dark shade of gray.  Reminds me of my web
> development days -- I'd put debugging info (on beta versions) right on
> the screen in an inconspicuous area with white text on a white
> background.  The users wouldn't know it's there, but our team could
> click and drag in the area to highlight the text and make it visible.
> But I digress.

Joining your digression.
I do that now in some emails.
I add "tags" of the form _sometext and then convert to white on white.
This makes them searchable but not (initially) visible.
No doubt under some conditions of passage across the web they may
become visible, so no zero security is provided.

         Russell

2010\04\09@162049 by Vitaliy

face
flavicon
face
Olin Lathrop wrote:
>> It's about firmware security. We program the bootloader into the
>> chips, and the bootloader then decrypts and programs the actual
>> firmware. We trust our Chinese CM, but not enough to send them raw
>> binaries.
>>
>> The circuit has an ICSP connector, but we only use it for development
>> or for reprogramming the bootloader.
>
> That still doesn't make sense.  That would mean all the PICs going thru
> the
> reflow oven would have the bootloader on them.

Correct.


> If so, what's the point of
> marking them?  I can see you might want to later distinguish between
> bootloader and fully programmed boards, but that would be after reflow.

Different bootloader versions. Also, devices can be programmed with unique
family/device IDs and are therefore not interchangeable.

Vitaliy

2010\04\19@235226 by Joseph Bento

face
flavicon
face

On Apr 8, 2010, at 6:08 PM, Vitaliy wrote:

> Marcel Duchamp wrote:
>> I wonder if you could give the different firmware versions a different
>> number each.  Then represent different numerals with a different color
>> ink.  For example three color stripes like brown, black, red might
>> indicate program number 102...
>> This might be worth getting a patent on!
>
> I'm afraid it's not a new idea. ;)

I just thought of this catchy memory jogger dealing with bad boys and young girls!  I'll be famous!  :-)

Joe



2010\04\20@042333 by MCH

flavicon
face
Almost as famous as Violet is. Oh - what a girl...

Joe M.

Joseph Bento wrote:
{Quote hidden}

More... (looser matching)
- Last day of these posts
- In 2010 , 2011 only
- Today
- New search...