Maker Pro
Maker Pro

Epoxying over chip numbers?

J

John Muchow

Jan 1, 1970
0
acetone

Isn't there a chance that the acetone could affect the bond between
the IC legs and the case material, i.e., some of the acetone could
work its way inside and cause damage? Or is the chance of that no
greater than the possibility of damaging the chip with our Dremel?


John Muchow
-- remove SPAMMENOT for e-mail responses --
 
J

John Muchow

Jan 1, 1970
0
Abrasion does the trick: sanding, milling, laser.
However, anyone can still find out what chips you used, so why bother?
No other company i know of does something as infantile as hiding part
numbers.

Infantile...interesting choice of words. I'm finding out that this is
a surprisingly emotional topic for a lot of people. Frank has brought
up some great reasons for not hiding chip numbers, but there's still
the problem of micros, etc.

Because of the simple design of these products (it's the unique
packaging and combination of features that separates them from the
competition), we'd like to slow down anyone interested in copying the
design just a bit...until they come to market. Then, we can use
market penetration, great pricing, great tech support, etc. to make it
not profitable (we hope) for others to use our design.

It will inevitably happen, but if a few seconds of sanding (or
epoxying) of our prototypes can slow this process down until the
production models appear, that sounds like a damn good investment in
time and money to me.

How is hiding part numbers infantile? How is using a micro without
posting the source code not infantile then? I'm serious.


John Muchow
-- remove SPAMMENOT for e-mail responses --
 
W

Wouter van Ooijen

Jan 1, 1970
0
You do realise there are simple designs that will identify most logic
chips automatically?


Wouter van Ooijen

-- ------------------------------------
http://www.voti.nl
PICmicro chips, programmers, consulting
 
B

Brane2

Jan 1, 1970
0
John Muchow wrote:

How is hiding part numbers infantile? How is using a micro without
posting the source code not infantile then? I'm serious.


John Muchow
-- remove SPAMMENOT for e-mail responses --

Hiding part numbers is infantile because it is major pain in the ass for
anyone trying to service the device and because it doesn't stop anyone
trying to duplicate it.

It can buy you a few days at the most with the copycats and a frontrow
seat in hell, reserved by anyone who have tried to service your product
without full documentation and portfolio of spare parts...

With code inside micro, it's much different thing. It is still awkward
for servicing, but at least it is efficient against hobbyistic copiers,
so it achieves its basic goal- reasonable IP protection.

regards,


Branko
 
J

Jim Thompson

Jan 1, 1970
0
John Muchow wrote:



Hiding part numbers is infantile because it is major pain in the ass for
anyone trying to service the device and because it doesn't stop anyone
trying to duplicate it.

It can buy you a few days at the most with the copycats and a frontrow
seat in hell, reserved by anyone who have tried to service your product
without full documentation and portfolio of spare parts...

With code inside micro, it's much different thing. It is still awkward
for servicing, but at least it is efficient against hobbyistic copiers,
so it achieves its basic goal- reasonable IP protection.

regards,


Branko

I'm astonished at the number of people on this group who are incapable
of reading.

The OP said "prototypes".

...Jim Thompson
 
B

Brane2

Jan 1, 1970
0
Jim Thompson wrote:

I'm astonished at the number of people on this group who are incapable
of reading.

The OP said "prototypes".

...Jim Thompson

Same here. I was merely responding about the difference between relying
on epoxy and using codeprotected components...

Branko
 
Infantile...interesting choice of words. I'm finding out that this is
a surprisingly emotional topic for a lot of people. Frank has brought
up some great reasons for not hiding chip numbers, but there's still
the problem of micros, etc.

Because of the simple design of these products (it's the unique
packaging and combination of features that separates them from the
competition), we'd like to slow down anyone interested in copying the
design just a bit...until they come to market. Then, we can use
market penetration, great pricing, great tech support, etc. to make it
not profitable (we hope) for others to use our design.

It will inevitably happen, but if a few seconds of sanding (or
epoxying) of our prototypes can slow this process down until the
production models appear, that sounds like a damn good investment in
time and money to me.

How is hiding part numbers infantile? How is using a micro without
posting the source code not infantile then? I'm serious.


I feel the usual reason major manufacturers obliterate or house label
parts is so that they can't be repaired, not to prevent duplication.
Take a TV or VCR as an example. The company doesn't want them
repaired as each repaired unit is a lost sale on a new, dispoable
unit.

I have no issue with hiding numbers on a prototype. I find it very
annoying though, trying to repair something with unlabeled parts.

What kind of product are we discussing anyhow? Are these particular
parts even likely to fail in the long term?

-Chris
 
R

Rich Grise

Jan 1, 1970
0
Isn't there a chance that the acetone could affect the bond between
the IC legs and the case material, i.e., some of the acetone could
work its way inside and cause damage? Or is the chance of that no
greater than the possibility of damaging the chip with our Dremel?

I saw a couple of guys cleaning graffiti off a sign once with laquer
thinner, which is a lot like acetone. The sign had epoxy paint, and
it shrugged off the laquer thinner like rainwater. I think the epoxy
that chips are in wouldn't even notice acetone.

Fuming nitric acid, on the other hand...
 
J

John Muchow

Jan 1, 1970
0
With code inside micro, it's much different thing. It is still awkward
for servicing, but at least it is efficient against hobbyistic copiers,
so it achieves its basic goal- reasonable IP protection.

So, both are a PITA but because hiding chip numbers isn't as effective
a method of IP protection, it's infantile?

Perhaps I'm just overreacting to the word itself.

John Muchow
-- remove SPAMMENOT for e-mail responses --
 
J

John Muchow

Jan 1, 1970
0
What kind of product are we discussing anyhow? Are these particular
parts even likely to fail in the long term?

We've never had a failure of any of our electronic components in 16
years so I don't have much data to offer...we've never had to repair
anything. Someone did rip off a battery connector a few years ago,
but he was able to repair that one himself. :)

We've always been ready with loaners to ship overnight and postpaid
(both ways) repair service if needed though.

John Muchow
-- remove SPAMMENOT for e-mail responses --
 
J

John Muchow

Jan 1, 1970
0
You do realise there are simple designs that will identify most logic
chips automatically?

Yea, Mark reminded about those things. :)
We've got some chips that aren't 74/54 (and similar series) logic
chips (they're analog).

If I'm not mistaken, can they use the testers in-circuit? We'd surely
notice if anything was removed and soldered back in again.

We are only looking for a bit more time here, not long-term protection
from copying. We figured that a couple of minutes with the prototypes
is worth it just to make any copycats with the right equipment really
have to earn their keep.

Any day I can aggravate someone who wants to steal our designs for
commercial use is a good day.

John Muchow
-- remove SPAMMENOT for e-mail responses --
 
J

John Muchow

Jan 1, 1970
0
OK, we've determined that anyone who covers chip numbers will burn in
hell, that sanding and electric erasers work great, that Dremels work
great (but be careful), but no one had any recommendations for an
epoxy? :)

Guess we'll have to keep using the Dremel for the rest of the
prototypes. Thanks for your suggestions and advice!!.

....hey, is it getting hot in here?

John Muchow
-- remove SPAMMENOT for e-mail responses --
 
R

Robert Baer

Jan 1, 1970
0
John said:
Isn't there a chance that the acetone could affect the bond between
the IC legs and the case material, i.e., some of the acetone could
work its way inside and cause damage? Or is the chance of that no
greater than the possibility of damaging the chip with our Dremel?

John Muchow
-- remove SPAMMENOT for e-mail responses --

Acetone has about the same effect as water; the dremel tool can do a
lot of damage if one wants...
 
R

Robert Baer

Jan 1, 1970
0
John said:
Infantile...interesting choice of words. I'm finding out that this is
a surprisingly emotional topic for a lot of people. Frank has brought
up some great reasons for not hiding chip numbers, but there's still
the problem of micros, etc.

Because of the simple design of these products (it's the unique
packaging and combination of features that separates them from the
competition), we'd like to slow down anyone interested in copying the
design just a bit...until they come to market. Then, we can use
market penetration, great pricing, great tech support, etc. to make it
not profitable (we hope) for others to use our design.

It will inevitably happen, but if a few seconds of sanding (or
epoxying) of our prototypes can slow this process down until the
production models appear, that sounds like a damn good investment in
time and money to me.

How is hiding part numbers infantile? How is using a micro without
posting the source code not infantile then? I'm serious.

John Muchow
-- remove SPAMMENOT for e-mail responses --

Epoxy will be of no use.
Sanding, grinding or laser abrasion are the only effective ways of
"hiding" (removing) part numbers.
Use of FPGAs, PALs or micros is standard industry practice, and coding
is usually kept in house (also standard industry practice.
If you want to confuse the troops, add logic and linear that does no
useful purpose, but looks like that it is being used.
Also, put some critical traces under ICs with other traces that look
like they are needed both under ICs and visible, that do no useful
function.
Try added layers with used and extra traces; blind vias, etc.

Housesomeever, it all can be eventually decoded by someone with the
right tools and time...

I think that removing the legends off ICs will only pique someone's
interest, and therby hasten the eventual "decoding".
 
R

Robert Baer

Jan 1, 1970
0
Jim said:
I'm astonished at the number of people on this group who are incapable
of reading.

The OP said "prototypes".

...Jim Thompson

FPGAs, PALs and micros all have been used in prototypes...
 
M

Mark (UK)

Jan 1, 1970
0
Ha!

It's more like - what the hell WAS in the case in Ronin :)

Yours, Mark.
 
K

Keith Wootten

Jan 1, 1970
0
John Muchow said:
OK, we've determined that anyone who covers chip numbers will burn in
hell, that sanding and electric erasers work great, that Dremels work
great (but be careful), but no one had any recommendations for an
epoxy? :)

<Snip>

If they're though-hole chips, fold the legs over and mount the chip on
the reverse side of the board (or re-lay the PCB). Or, instead of a
Dremel, use a sheet of emery paper flat on the desk and erase the
markings before soldering.

Cheers
 
J

Julie

Jan 1, 1970
0
John said:
We have several prototypes being built over the next few weeks and
wanted to hide the chip numbers of some of the glue logic. We've
sanded them off up to now, but that's an awful and time consuming way
to do it. A search of the sci.electronics groups came up
empty....amazingly.

Does anyone have any recommendations for covering up the numbers on
plastic and ceramic DIP packages? Loctite's black Prism 410 adhesive
looks good, but we're not sure if it will stick well enough or if it
can be pried off like a "pancake".

Or is sanding still the best option for doing this?

Thanks!

John Muchow
-- remove SPAMMENOT for e-mail responses --

How many protos are you talking about? 10s, 100s, more?

Can you just pot the entire board? Or do you need to have subsequent access to
it?

Several have mentioned laser etching -- I'm presuming that you don't have
immediate access to one of those, nor want to spend $25k for one at the
moment... You may be able to achieve this in a cost-effective way by visiting
your local sign/engraving shop. Most have (on-site) access to a laser engraver
which should perform adequately well in a situation as this. You could
probably get the cost down pretty low per piece as well.
 
R

Rich Grise

Jan 1, 1970
0
OK, we've determined that anyone who covers chip numbers will burn in
hell, that sanding and electric erasers work great, that Dremels work
great (but be careful), but no one had any recommendations for an
epoxy? :)

Guess we'll have to keep using the Dremel for the rest of the
prototypes. Thanks for your suggestions and advice!!.

...hey, is it getting hot in here?

To use epoxy and make it stick, you'd have to sand the chips anyway.
But you can get a more aggressive wheel than an eraser! I work in a
weld shop, and they have little (air-operated) angle grinders, just
like a Dremel, but with a right angle at the business end, so you
can use a sanding disk, or whatever it's called. :)
 
Top