PCB Encapsulants to Annoy Copy Pirates
D from BC
hell for copy cats..
So far I've found:
http://www.mgchemicals.com/products/832ht.html
'Extremely difficult to remove - grants incredible technology
protection '
Besides epoxy.. Are there alternatives?
ex: Very very hot melt? Low melting point glass?
D from BC
Amateur smps designer
British Columbia, Canada
Posted to sci.electronics.design
TTman
"D from BC" <myreal...@comic.com> wrote in message
news:dddue5hrjdthg692k...@4ax.com...
> I'd like to pot my smt pcb with the nastiest sh*t possible to make it
> hell for copy cats..
>
> So far I've found:
>
> http://www.mgchemicals.com/products/832ht.html
> 'Extremely difficult to remove - grants incredible technology
> protection '
>
> Besides epoxy.. Are there alternatives?
Araldaite 2014- grey epoxy. I use it, very, very good stuff !
D from BC
wrote:
Neato..
http://en.wikipedia.org/wiki/Araldite
'Highmark Manufacturing uses Araldite in the manufacture of advanced
ballistic protection body armor'
While on wiki. I came across this tidbit...
http://en.wikipedia.org/wiki/Epoxy_resin
'Vinegar is an effective and safe solvent to clean up tools, brushes,
skin, and most surfaces contaminated with epoxy resin or hardener.'
Huh..didn't know that..
Jon Slaughter
When you try something let us(me) know how it goes?
Archimedes' Lever
wrote:
>
Take a Dremel with a round rubber drum sanding attachment of about 1/2"
diameter.
Grind off all the ID marks on the chips.
Then, pot it in "Stycast".
No, epoxy is what you need. Most anything else can be removed one way
or another too easily.
There are high adhesion polyurethane potting compounds, but they are
far more costly.
An alternative to fully potting it, would be to do the chip grind thing,
and then 'paint' the epoxy over those chip-top locations. You could also
'paint' over the traces until the build up is high enough to hide them,
but they can be traced via continuity tests usually.
You could also build a small trace array that you selectively cut and
bridge to complete uniquely "keyed" circuits ala Dongle Key technology.
You could make each circuit unique, yet the same, yet harder to copy.
John Larkin
The disadvantages of potting generally far outweigh the likelihood
that someone will copy your design. And if they're determined, they
will anyhow.
John
Hammy
<myreal...@comic.com> wrote:
Patent it. Thats what everybody else does.
What did you invent a time machine ?
D from BC
It'll be awhile..
I just started the pcb design.
Might be potting in Jan 2010.
D from BC
<OneBi...@InfiniteSeries.Org> wrote:
>On Mon, 2 Nov 2009 22:28:03 -0000, "TTman" <someo...@ntlworld.com>
>wrote:
>
>>
>>"D from BC" <myreal...@comic.com> wrote in message
>>news:dddue5hrjdthg692k...@4ax.com...
>>> I'd like to pot my smt pcb with the nastiest sh*t possible to make it
>>> hell for copy cats..
>>>
>>> So far I've found:
>>>
>>> http://www.mgchemicals.com/products/832ht.html
>>> 'Extremely difficult to remove - grants incredible technology
>>> protection '
>>>
>>> Besides epoxy.. Are there alternatives?
>>
>>Araldaite 2014- grey epoxy. I use it, very, very good stuff !
>>
>
>
> Take a Dremel with a round rubber drum sanding attachment of about 1/2"
>diameter.
>
> Grind off all the ID marks on the chips.
I might try a solvent to smudge the ID on the IC.
>
> Then, pot it in "Stycast".
>
> No, epoxy is what you need. Most anything else can be removed one way
>or another too easily.
>
> There are high adhesion polyurethane potting compounds, but they are
>far more costly.
In addition to adhesion, the potting has to resist solvent/chemical
baths that pirates may use.
>
> An alternative to fully potting it, would be to do the chip grind thing,
>and then 'paint' the epoxy over those chip-top locations. You could also
>'paint' over the traces until the build up is high enough to hide them,
>but they can be traced via continuity tests usually.
>
> You could also build a small trace array that you selectively cut and
>bridge to complete uniquely "keyed" circuits ala Dongle Key technology.
>
> You could make each circuit unique, yet the same, yet harder to copy.
Maybe I could do some micro wire jumpers running through the epoxy.
If tools are used to remove the epoxy, the pirates might buzz through
the wires by accident.
D from BC
Potting will make me feel good? :)
This might be ironic.
It's possible the first people to tear into the epoxy potting will not
be Asian copy cats but instead US patent owners who suspect their
patent is being used without permission.
Joel Koltner
news:8b2ve5tmtarghq32l...@4ax.com...
> It's possible the first people to tear into the epoxy potting will not
> be Asian copy cats but instead US patent owners who suspect their
> patent is being used without permission.
Hmm... I wonder if you can be made to reveal your design
schematics/software/etc. if a patent holder is able to convince a judge that
there's a very good change you're infringing on their patent, yet you have
made it effectively impossible for an outside party to determine as much?
John Larkin
<myreal...@comic.com> wrote:
>On Mon, 02 Nov 2009 16:57:55 -0800, John Larkin
><jjla...@highNOTlandTHIStechnologyPART.com> wrote:
>
>>On Mon, 02 Nov 2009 13:16:28 -0800, D from BC
>><myreal...@comic.com> wrote:
>>
>>>I'd like to pot my smt pcb with the nastiest sh*t possible to make it
>>>hell for copy cats..
>>>
>>>So far I've found:
>>>
>>>http://www.mgchemicals.com/products/832ht.html
>>>'Extremely difficult to remove - grants incredible technology
>>>protection '
>>>
>>>Besides epoxy.. Are there alternatives?
>>>
>>>ex: Very very hot melt? Low melting point glass?
>>>
>>>D from BC
>>>Amateur smps designer
>>>British Columbia, Canada
>>>Posted to sci.electronics.design
>>
>>The disadvantages of potting generally far outweigh the likelihood
>>that someone will copy your design. And if they're determined, they
>>will anyhow.
>>
>>John
>
>Potting will make me feel good? :)
Not after you've actually done it a few times.
>
>This might be ironic.
>It's possible the first people to tear into the epoxy potting will not
>be Asian copy cats but instead US patent owners who suspect their
>patent is being used without permission.
Is it?
John
D from BC
A possibly unique smps controller.
MooseFET
> I'd like to pot my smt pcb with the nastiest sh*t possible to make it
> hell for copy cats..
>
> So far I've found:
>
> http://www.mgchemicals.com/products/832ht.html
> 'Extremely difficult to remove - grants incredible technology
> protection '
>
> Besides epoxy.. Are there alternatives?
>
> ex: Very very hot melt? Low melting point glass?
Make a nice aluminum housing for it and put it together with left
handed screws.
You can get chips marked with your company part number if you order
enough. Make your part number for a 74HC04 be LM324 and etc.
Make it all "firmware" in a FPGA and use a coin cell battery to keep
the FPGA always powered on. If they open it up poof goes the code.
mpm
wrote:
> On Mon, 2 Nov 2009 22:28:03 -0000, "TTman" <someone...@ntlworld.com>
> wrote:
>
>
>
>
>
>
>
> >"D from BC" <myrealaddr...@comic.com> wrote in message
>
> - Show quoted text -
Could you also throw 'em a curveball by re-printing bogus ID's on the
chips?
Something within the limits of credulity...
I'm not suggesting you paint "555" on the top of a TMS340 or anything
like that...
If you chose your part numbers carefully, you might send them on a
wild goose chase... :)
Archimedes' Lever
wrote:
>
>I might try a solvent to smudge the ID on the IC.
Many these days are laser burned.
The Dremel grind will not hurt them.
Archimedes' Lever
wrote:
>
>In addition to adhesion, the potting has to resist solvent/chemical
>baths that pirates may use.
>
The stuff I am thinking of is a tough rubbery thing that can stop a
knife point or a bullet, much less most solvents that you worry about
that would melt polycarbonate, but not this. It is used by NASA. It is
the ONLY encapsulant they have approved for HV space applications.
Archimedes' Lever
Of course. Why not? You would have to make the removed ID surface
appear acceptable or authentic before applying the new info.
>Something within the limits of credulity...
>I'm not suggesting you paint "555" on the top of a TMS340 or anything
>like that...
Anything different would at least cause lost time, and time is
everything to copy cats.
>
>If you chose your part numbers carefully, you might send them on a
>wild goose chase... :)
>
Well, a good diagnostician could do a pretty good job of determining
what something "must be" by basic examination and probing.
That would require it be visible, or accessible to probing.
Tim Williams
> Make it all "firmware" in a FPGA and use a coin cell battery to keep
> the FPGA always powered on. If they open it up poof goes the code.
Planned obsolescence, too. Good idea!
D: what about x-rays? Do they make lead oxide-filled epoxy? It's
pretty trivial to x-ray through potting otherwise. ;-)
Tim
D from BC
True..
I do try to dodge making plastic dust when I can.
I don't like cleaning.
I might even try melting the number with an old soldering tip.
Dremel is probably the low cancer solution.
Solvents make fumes and same goes for plastic melting.
D from BC
Nasa stuff can be pricey.
Maybe I can get an Asian pirated copy of that rubbery stuff. :)
D from BC
Example:
http://www.vidisco.com/images/full/app-ndt-Electronic-PCB-100.jpg
I can see the disc capacitors,resistors,jumpers and all traces.
But what an xray won't show chip functions and discrete part values.
imo ..encapsulation makes scope probing difficult.
Greegor
D from BC
It's:
Privacy by Encapsulation vs Design Disclosure by Patent Owner.
Who's the winner going to be?
Archimedes' Lever
<tmor...@gmail.com> wrote:
An x-ray does not mean a good analysis would result.
Use Aluminum traces and they are invisible to X-rays. Ha!
Have fun soldering or basically getting a good electrical bond on all
the nodes though.
You could can it in PCB media with 100% ground plane and ad a nice
thick layer of LEADED solder on it, OR little sheets of lead soldered
onto it. :-)
Archimedes' Lever
wrote:
>On Mon, 02 Nov 2009 20:30:23 -0800, Archimedes' Lever
><OneBi...@InfiniteSeries.Org> wrote:
>
>>On Mon, 02 Nov 2009 17:37:39 -0800, D from BC <myreal...@comic.com>
>>wrote:
>>
>>>
>>>I might try a solvent to smudge the ID on the IC.
>>
>>
>> Many these days are laser burned.
>>
>> The Dremel grind will not hurt them.
>
>True..
>I do try to dodge making plastic dust when I can.
>I don't like cleaning.
The high temp polymer dust is not the best stuff to be breathing
either. I would not recommend doing any high speed power sanding on
basically anything inside your house.
>
>I might even try melting the number with an old soldering tip.
It is not that kind of plastic. Use an engraving tool to gouge stripes
into it and that will obscure the numbering.
> Dremel is probably the low cancer solution.
If it is laser etched, it is one of the only solutions. You have to
move or remove media to remove etched info.
> Solvents make fumes and same goes for plastic melting.
There are not too many solvents that will attack the plastics most ICs
are packaged in. The print used is pretty hardy stuff too. Usually an
epoxy "ink".
Archimedes' Lever
wrote:
>On Mon, 02 Nov 2009 20:37:14 -0800, Archimedes' Lever
><OneBi...@InfiniteSeries.Org> wrote:
>
>>On Mon, 02 Nov 2009 17:37:39 -0800, D from BC <myreal...@comic.com>
>>wrote:
>>
>>>
>>>In addition to adhesion, the potting has to resist solvent/chemical
>>>baths that pirates may use.
>>>
>>
>>
>> The stuff I am thinking of is a tough rubbery thing that can stop a
>>knife point or a bullet, much less most solvents that you worry about
>>that would melt polycarbonate, but not this. It is used by NASA. It is
>>the ONLY encapsulant they have approved for HV space applications.
>
>Nasa stuff can be pricey.
>Maybe I can get an Asian pirated copy of that rubbery stuff. :)
It is called "Conap" and there are no substitutes, as far as NASA is
concerned, and it is VERY expensive, but I am sure there are plenty of
good 'turgid' other brand polyurethane encapsulants out there.
Archimedes' Lever
wrote:
>On Mon, 2 Nov 2009 20:51:05 -0800 (PST), Tim Williams
Looks like an old floppy controller card.
Archimedes' Lever
wrote:
>How do you like getting products made that way?
Unless YOU QUOTE who you are responding to, nobody will know who or
what the fuck you are talking about.
Not everyone reads or looks at there news the same way.
Relying on threaded views is retarded.
QUOTE what you are referring to. D'oh!
TTman
> Neato..
>
> http://en.wikipedia.org/wiki/Araldite
> 'Highmark Manufacturing uses Araldite in the manufacture of advanced
> ballistic protection body armor'
>
> While on wiki. I came across this tidbit...
>
> http://en.wikipedia.org/wiki/Epoxy_resin
> 'Vinegar is an effective and safe solvent to clean up tools, brushes,
> skin, and most surfaces contaminated with epoxy resin or hardener.'
>
> Huh..didn't know that..
>
It says vinegar will remove resin 'that is starting to harden'....
Dirk Bruere at NeoPax
> "D from BC" <myreal...@comic.com> wrote in message
>> I'd like to pot my smt pcb with the nastiest sh*t possible to make it
>> hell for copy cats..
>>
>> So far I've found:
>>
>> http://www.mgchemicals.com/products/832ht.html
>> 'Extremely difficult to remove - grants incredible technology
>> protection '
>>
>> Besides epoxy.. Are there alternatives?
>
> Araldaite 2014- grey epoxy. I use it, very, very good stuff !
>
>
--
Dirk
http://www.transcendence.me.uk/ - Transcendence UK
http://www.theconsensus.org/ - A UK political party
http://www.blogtalkradio.com/onetribe - Occult Talk Show
J.A. Legris
> On Mon, 2 Nov 2009 20:51:05 -0800 (PST), Tim Williams
>
I used to manufacture a product that was encapsulated and nobody ever
copied it. Then I stopped encapsulating, sales went up, and still
nobody copied it. Why? Because it used a micro. Last time I checked it
costs over $20K to read a locked micro. Also, the firmware was pretty
easy - no self-respecting programmer would pay $20K for someone else's
lousy object code, so they did it themselves and never got a faithful
copy.
If all you're protecting is a circuit design, don't count on epoxy
potting compound - it simply disintegrates under the heat and pressure
of an ordinary soldering iron, allowing you to drill down wherever you
need access. An x-ray would make the task almost routine.
More reasons for not encapsulating:
1) Honest dealing. Would you choose an encapsulated product over an
open one if there was no actual need for encapsulation? What are the
encapsulators hiding besides the functions?
2) A horrible mess. Epoxy is formulated to flow wherever it can, and
it does. Especially into DIP switches, connectors, and all over your
tools and work space.
3) Diagnostics and repair. How do you diagnose an encapsulated device
that fails. How do you repair it?
4) Expensive and time-consuming addtional steps to manufacture.
5) Environmentally unfriendly: for you, the customer and everyone
else.
--
Joe
MooseFET
Nearly any large atom works. The trick would be to make the masking
material very nonuniform. Here's an idea for doing it:
Dip coat the board twice and let each dip dry completely.
Dip the PCB a 3rd time and then roll in random bits of metal from a
machine shop.
Dip the PCB and let it dry again.
Dip the PCB and then cover with bits of scrap wire.
Pot the whole mess.
Spehro Pefhany
<jale...@sympatico.ca> wrote:
>On Nov 3, 12:44�am, D from BC <myrealaddr...@comic.com> wrote:
>> On Mon, 2 Nov 2009 20:51:05 -0800 (PST), Tim Williams
>>
>> <tmoran...@gmail.com> wrote:
>> >On Nov 2, 8:35�pm, MooseFET <kensm...@rahul.net> wrote:
>> >> Make it all "firmware" in a FPGA and use a coin cell battery to keep
>> >> the FPGA always powered on. �If they open it up poof goes the code.
>>
>> >Planned obsolescence, too. �Good idea!
>>
>> >D: what about x-rays? �Do they make lead oxide-filled epoxy? �It's
>> >pretty trivial to x-ray through potting otherwise. ;-)
>>
>> >Tim
>>
>> Example:http://www.vidisco.com/images/full/app-ndt-Electronic-PCB-100.jpg
>>
>> I can see the disc capacitors,resistors,jumpers and all traces.
>>
>> But what an xray won't show chip functions and discrete part values.
>>
>> imo ..encapsulation makes scope probing difficult.
>
>
>I used to manufacture a product that was encapsulated and nobody ever
>copied it. Then I stopped encapsulating, sales went up, and still
>nobody copied it. Why? Because it used a micro. Last time I checked it
>costs over $20K to read a locked micro.
You didn't check overseas.
>Also, the firmware was pretty
>easy - no self-respecting programmer would pay $20K for someone else's
>lousy object code, so they did it themselves and never got a faithful
>copy.
Hubris is a great protection mechanism, at least in the West.
>If all you're protecting is a circuit design, don't count on epoxy
>potting compound - it simply disintegrates under the heat and pressure
>of an ordinary soldering iron, allowing you to drill down wherever you
>need access. An x-ray would make the task almost routine.
>
>More reasons for not encapsulating:
>1) Honest dealing. Would you choose an encapsulated product over an
>open one if there was no actual need for encapsulation? What are the
>encapsulators hiding besides the functions?
>2) A horrible mess. Epoxy is formulated to flow wherever it can, and
>it does. Especially into DIP switches, connectors, and all over your
>tools and work space.
>3) Diagnostics and repair. How do you diagnose an encapsulated device
>that fails. How do you repair it?
>4) Expensive and time-consuming addtional steps to manufacture.
>5) Environmentally unfriendly: for you, the customer and everyone
>else.
Good points all.
baron
You could additionally use a multi layer board with ground planes top &
bottom ! That could get around the Xray machine.
--
Best Regards:
Baron.
legg
<myreal...@comic.com> wrote:
>On Mon, 2 Nov 2009 22:28:03 -0000, "TTman" <someo...@ntlworld.com>
>wrote:
>
>>
<snip>
>While on wiki. I came across this tidbit...
>
>http://en.wikipedia.org/wiki/Epoxy_resin
>'Vinegar is an effective and safe solvent to clean up tools, brushes,
>skin, and most surfaces contaminated with epoxy resin or hardener.'
>
>Huh..didn't know that..
Don't count on it.....
RL
D from BC
I smell a bot.
D from BC
wrote:
Sorry. I didn't mean to imply vinegar removes cured epoxy.
Archimedes' Lever
<jale...@sympatico.ca> wrote:
>3) Diagnostics and repair. How do you diagnose an encapsulated device
>that fails. How do you repair it?
Pitch and replace. Doh! Some products cost far more to service than
to simply replace, and that includes even many that are not ever
encapsulated.
Archimedes' Lever
<jale...@sympatico.ca> wrote:
>5) Environmentally unfriendly: for you, the customer and everyone
>else.
Cured potting compounds pose no environmental threat (nearly all), and
even when uncured they only pose a threat to the environment if you or
another of your dopes gets the shit all over the place.
Use some brains.,
J.A. Legris
wrote:
> On Tue, 3 Nov 2009 07:01:38 -0800 (PST), "J.A. Legris"
>
> >3) Diagnostics and repair. How do you diagnose an encapsulated device
> >that fails. How do you repair it?
>
> Pitch and replace. Doh! Some products cost far more to service than
> to simply replace, and that includes even many that are not ever
> encapsulated.
Some of us are interested in knowing why their products fail. In fact,
I attribute my success, in part, to just such an attitude.
--
Joe
J.A. Legris
wrote:
> On Tue, 3 Nov 2009 07:01:38 -0800 (PST), "J.A. Legris"
>
> >5) Environmentally unfriendly: for you, the customer and everyone
> >else.
>
> Cured potting compounds pose no environmental threat (nearly all), and
> even when uncured they only pose a threat to the environment if you or
> another of your dopes gets the shit all over the place.
>
> Use some brains.,
I did, which is why I no longer use potting compounds. The problem is
dopes like you who cannot reason beyond their next paycheck. The long-
term effects of most synthetic chemicals have never been
comprehensively assessed, so erring on the side of caution is the wise
thing to do.
--
Joe
D from BC
I'm wondering about thermal expansion.
My 1"x1" pcb that I want to encapsulate will be in an environment
where the temp may swing from 22C to 70C daily.
I have no idea if an epoxy encapsulant will flex with heat and shear
forces will push on the tiny smt parts..
I'll try to find specs.
Perhaps testing in an environmental chamber is needed.
But I'm cheap, so I'll be using a light bulb in a box controlled by a
thermostat.
Archimedes' Lever
<jale...@sympatico.ca> wrote:
>On Nov 3, 7:31�pm, Archimedes' Lever <OneBigLe...@InfiniteSeries.Org>
>wrote:
>> On Tue, 3 Nov 2009 07:01:38 -0800 (PST), "J.A. Legris"
>>
>> <jaleg...@sympatico.ca> wrote:
>> >3) Diagnostics and repair. How do you diagnose an encapsulated device
>> >that fails. How do you repair it?
>>
>> � Pitch and replace. �Doh! �Some products cost far more to service than
>> to simply replace, and that includes even many that are not ever
>> encapsulated.
>
>Some of us are interested in knowing why their products fail.
Some of us? You're a goddamned idiot for that snide remark. ALL of us
like just statistics.
> In fact,
>I attribute my success, in part, to just such an attitude.
Yes, but you apparently have no clue about hard potted devices.
If it is that small a device, you should already know all about every
failure mode there is for it, and that should be able to be determined
without examination in many cases. All you should need is your prior
knowledge of the circuit operation and the particulars of the failure.
If it is a power supply device and it is failing, you should not be
selling it as a product to begin with. A PROVEN design, will have all of
its failure modes documented. Any potted design should be a proven
design, especially if the potting is meant for IP security purposes.
Otherwise, a serviceable potting would be needed.
If you are selling hard potted products, you have already resolved to
full replacement upon failure.
Why a PS device would need IP security is the real question. Why this
dope thinks his supply circuit is so special that someone would be on his
heels to copy it is beyond me. If it is that revolutionary, he should
patent it, and SELL the design to a bigger player in the industry.
DUH.
Archimedes' Lever
<jale...@sympatico.ca> wrote:
>On Nov 3, 7:33�pm, Archimedes' Lever <OneBigLe...@InfiniteSeries.Org>
>wrote:
>> On Tue, 3 Nov 2009 07:01:38 -0800 (PST), "J.A. Legris"
>>
>> <jaleg...@sympatico.ca> wrote:
>> >5) Environmentally unfriendly: for you, the customer and everyone
>> >else.
>>
>> � Cured potting compounds pose no environmental threat (nearly all), and
>> even when uncured they only pose a threat to the environment if you or
>> another of your dopes gets the shit all over the place.
>>
>> � Use some brains.,
>
>I did, which is why I no longer use potting compounds. The problem is
>dopes like you who cannot reason beyond their next paycheck.
You're a goddamned retard.
> The long-
>term effects of most synthetic chemicals have never been
>comprehensively assessed,
Not with a dope like you at the wheel, but yes, there are long term
studies, and there is long term knowledge in the industry, you stupid
pussified twit.
> so erring on the side of caution is the wise
>thing to do.
You're a goddamned long term stye on the face of intelligent men of the
world. You meant to say "Being a PUSSY on the "side of caution" is a
stupid way to achieve 'progress'".
You are probably so retarded that you embrace RoHS, and you think that
metallic form lead is a problem for the environment.
If it were, the lead levels in water tables around lead mining and also
around land fills and especially around decades old police shooting
ranges would show higher numbers and they do NOT.
Idiots like you are pathetic.
Archimedes' Lever
wrote:
>On Tue, 3 Nov 2009 18:22:04 -0800 (PST), "J.A. Legris"
><jale...@sympatico.ca> wrote:
>
>>On Nov 3, 7:33�pm, Archimedes' Lever <OneBigLe...@InfiniteSeries.Org>
>>wrote:
>>> On Tue, 3 Nov 2009 07:01:38 -0800 (PST), "J.A. Legris"
>>>
>>> <jaleg...@sympatico.ca> wrote:
>>> >5) Environmentally unfriendly: for you, the customer and everyone
>>> >else.
>>>
>>> � Cured potting compounds pose no environmental threat (nearly all), and
>>> even when uncured they only pose a threat to the environment if you or
>>> another of your dopes gets the shit all over the place.
>>>
>>> � Use some brains.,
>>
>>I did, which is why I no longer use potting compounds. The problem is
>>dopes like you who cannot reason beyond their next paycheck. The long-
>>term effects of most synthetic chemicals have never been
>>comprehensively assessed, so erring on the side of caution is the wise
>>thing to do.
>
>I'm wondering about thermal expansion.
>
>My 1"x1" pcb that I want to encapsulate will be in an environment
>where the temp may swing from 22C to 70C daily.
The expansion numbers on potting are for what happens during curing.
>I have no idea if an epoxy encapsulant will flex with heat and shear
>forces will push on the tiny smt parts..
Then conformally coat it BEFORE you pot it.
>I'll try to find specs.
They are, of course, available. Stycast is a common potting media.
>Perhaps testing in an environmental chamber is needed.
Not likely. A simple chamber with known temp is fine.
>But I'm cheap, so I'll be using a light bulb in a box controlled by a
>thermostat.
OK.
MooseFET
> On Tue, 3 Nov 2009 18:22:04 -0800 (PST), "J.A. Legris"
>
>
>
> <jaleg...@sympatico.ca> wrote:
> >On Nov 3, 7:33 pm, Archimedes' Lever <OneBigLe...@InfiniteSeries.Org>
> >wrote:
> >> On Tue, 3 Nov 2009 07:01:38 -0800 (PST), "J.A. Legris"
>
> >> <jaleg...@sympatico.ca> wrote:
> >> >5) Environmentally unfriendly: for you, the customer and everyone
> >> >else.
>
> >> Cured potting compounds pose no environmental threat (nearly all), and
> >> even when uncured they only pose a threat to the environment if you or
> >> another of your dopes gets the shit all over the place.
>
> >> Use some brains.,
>
> >I did, which is why I no longer use potting compounds. The problem is
> >dopes like you who cannot reason beyond their next paycheck. The long-
> >term effects of most synthetic chemicals have never been
> >comprehensively assessed, so erring on the side of caution is the wise
> >thing to do.
>
> I'm wondering about thermal expansion.
You can find the thermal rate for most materials. There is likely to
be a temperature gradient but it will be smaller than the total swing.
> My 1"x1" pcb that I want to encapsulate will be in an environment
> where the temp may swing from 22C to 70C daily.
Smaller is better.
>
> I have no idea if an epoxy encapsulant will flex with heat and shear
> forces will push on the tiny smt parts..
> I'll try to find specs.
> Perhaps testing in an environmental chamber is needed.
> But I'm cheap, so I'll be using a light bulb in a box controlled by a
> thermostat.
If you put a very flexible layer over the parts it will help.
Unfortunately it also provides the nice parting line to let the
pirates pop it apart along the boundary.
I have used a "light bulb" based heat box. Mine used power
resistors. You can get thermostats that can switch the load. A
couple of jugs of water in the hot box helps to hold the temperature
steady.
Run you freezer down as cold as it can get. Put the item in the
freezer and let it soak till it is all the way down. Once it is fully
cold move it to the hot box. leave it in the hot box for a day and
then move it back to the freezer.
This will quickly show you that the potting was a bad idea.
D from BC
I've yet to find my design on the net nor in chip form.
I figure the lazy cheap 'just in case' action is to pot the
electronics in my product until it makes enough money for me to afford
a patent guy to do all the patent searching, applying,patent write
up,defending,offending and license granting...and maybe to run out to
get coffee sometimes.
J.A. Legris
> On Wed, 04 Nov 2009 04:01:10 -0800, Archimedes' Lever
>
>
>
IIRC you should apply for the patent before you release the product
(or even just keep it secret) or you may lose your rights. This
presumes patenting the thing is a good idea in the first place - which
is unlikely unless you have deep pockets.
http://en.wikipedia.org/wiki/Prior_art
--
Joe
D from BC
<jale...@sympatico.ca> wrote:
>IIRC you should apply for the patent before you release the product
>(or even just keep it secret) or you may lose your rights. This
>presumes patenting the thing is a good idea in the first place - which
>is unlikely unless you have deep pockets.
>
>http://en.wikipedia.org/wiki/Prior_art
It's not fair if I lose my patentability rights when I'd like to
release 200 units for ~field testing~ :)
to tweek and finalize the design.
Archimedes' Lever
wrote:
>On Wed, 4 Nov 2009 09:10:48 -0800 (PST), "J.A. Legris"
You can tweak the design at the bench by beating the hell out of a few
units, and being very good at recording and analysis of observed data.
D from BC
I'm just afraid my customers might do a better job at beating the hell
out my stuff more than I can.. :)
J.A. Legris
> On Wed, 4 Nov 2009 09:10:48 -0800 (PST), "J.A. Legris"
>
> >IIRC you should apply for the patent before you release the product
> >(or even just keep it secret) or you may lose your rights. This
> >presumes patenting the thing is a good idea in the first place - which
> >is unlikely unless you have deep pockets.
>
> >http://en.wikipedia.org/wiki/Prior_art
>
> It's not fair if I lose my patentability rights when I'd like to
> release 200 units for ~field testing~ :)
> to tweek and finalize the design.
The wikipedia article mentions that you can protect your rights by
obtaining non-disclosure agreements.
--
Joe
Ken S. Tucker
We spray paint our PCB's black, for a sealant and thermodynamic
(black-body) radiation, it also makes a direct copy difficult.
Ken
krw
Not sure why you think black paint increases thermal radiation, but
it's certainly not going to stop anyone from copying the design. It
may slow them down by about five minutes. Maybe.
MooseFET
It also makes the board likely to fail. The solvent in spray paint is
hard on parts. As a result the product will get a reputation for
being crap. Since nobody wants to make a copy of a crap product,
people will be less likely to copy it.
Ross Herbert
:I'd like to pot my smt pcb with the nastiest sh*t possible to make it
:hell for copy cats..
:
:So far I've found:
:
:http://www.mgchemicals.com/products/832ht.html
:'Extremely difficult to remove - grants incredible technology
:protection '
:
:Besides epoxy.. Are there alternatives?
:
:ex: Very very hot melt? Low melting point glass?
:
:D from BC
:Amateur smps designer
:British Columbia, Canada
:Posted to sci.electronics.design
One thing you need to remember.. if there are any heat generating components on
the board, encapsulation in a hard compound may cause subsequent fracturing of
components or solder joints because they can't expand and contract. This may not
be such a problem with smt but it has caused problems with thru-hole components
on some pcb's I have repaired.
Since Araldite 2014 (3 types btw)
http://www.kirkside.com.au/Uploads/Images/2014.pdf is a bonding adhesive it may
cure to be extremely hard and may not allow expansion/contraction of components.
If this is likely to be a problem I would recommend a thin coating of pourable
white silicone rubber (or similar) over the heat generating components before
applying the Araldite 2014.
Vladimir Vassilevsky
Ross Herbert wrote:
> One thing you need to remember.. if there are any heat generating components on
> the board, encapsulation in a hard compound may cause subsequent fracturing of
> components or solder joints because they can't expand and contract.
I know an engineer who spent many years mixing different compounds and
fillers trying to create solid substance with minimal expansion or
contraction while it is cured, and which will not rip the components off
the PCB because of thermal effects. This is no simple problem.
Vladimir Vassilevsky
DSP and Mixed Signal Design Consultant
http://www.abvolt.com
Ross Herbert
wrote:
:
:
I entirely agree with you Vladimir... However, depending upon the amount of heat
generated or the degree of expansion/contraction involved it may not be as
extreme as your engineer friend encountered.
The very first instance of this type of problem which I encountered was in
electronic ignition systems from Johnson outboard motors back in the 70's. After
dissolving the epoxy encapsulant (that took weeks using an expensive chemical
solvent supposed to be specific to this purpose) it was discovered that several
soldered joints had failed and pcb pads had actually been lifted from the board
by the expansion of vertically mounted resistors in particular. Incidentally,
the solvent removed the coatings on resistors and other capacitors so
determining values was impossible in many case. I never intended to repair the
units but I was interested in the failure mechanism of these units. I
communicated my findings to Johnson at the time but I never received any
acknowledgement.
D from BC
wrote:
Ha.. Copy protection by deliberate low quality.
Archimedes' Lever
<nos...@nowhere.com> wrote:
Quite simple. Conformal coat the PCB prior to potting.
Still getting breakage? Conformal coat... twice. Make a blanket.
Archimedes' Lever
wrote:
You did not write it, but I am lazy tonight...
>>It also makes the board likely to fail. The solvent in spray paint is
>>hard on parts.
So much horseshit.
Transformer encapsulant (varnish) from a spray can. Dolph's. brand, to
be specific.
We coated THOUSANDS of HIGH VOLTAGE power supplies with a volatile
vehicle spray can applied coating, and baked it dry, after a vacuuming of
the board.
We have even done vacuum impregnation of the PCB which draws the
coating up inside the unmasked PCB media.
And we used a volatile solvent type thinner to cut the main gallon
with.
MooseFET
> On Thu, 5 Nov 2009 18:42:45 -0800 (PST), MooseFET <kensm...@rahul.net>
[....]
>
> >It also makes the board likely to fail. The solvent in spray paint is
> >hard on parts. As a result the product will get a reputation for
> >being crap. Since nobody wants to make a copy of a crap product,
> >people will be less likely to copy it.
>
> Ha.. Copy protection by deliberate low quality.
This is why cars made today don't look like the Pinto.
A company in Redmond Washington was trying the idea on software. I
wonder what ever happened to them.
MooseFET
> Ross Herbert wrote:
> > One thing you need to remember.. if there are any heat generating components on
> > the board, encapsulation in a hard compound may cause subsequent fracturing of
> > components or solder joints because they can't expand and contract.
>
> I know an engineer who spent many years mixing different compounds and
> fillers trying to create solid substance with minimal expansion or
> contraction while it is cured, and which will not rip the components off
> the PCB because of thermal effects. This is no simple problem.
You also have to match the thermal expansion. This makes it extra
hard because you need to balance two things.
I went down the potting road in the past and discovered that an
aluminum housing was cheaper to do in production.
Ken S. Tucker
That's funny. I'll write a memo to QC and production right now!
"MooseFET" approves your spray paint.
Ken
Archimedes' Lever
wrote:
>On Nov 5, 9:46�pm, D from BC <myrealaddr...@comic.com> wrote:
You don't even rate the peanut gallery.
That was Retard Gallery Comment.
Right where you belong too.
MooseFET
> On Nov 5, 6:42 pm, MooseFET <kensm...@rahul.net> wrote:
>
> > It also makes the board likely to fail. The solvent in spray paint is
> > hard on parts. As a result the product will get a reputation for
> > being crap. Since nobody wants to make a copy of a crap product,
> > people will be less likely to copy it.
>
> That's funny. I'll write a memo to QC and production right now!
> "MooseFET" approves your spray paint.
> Ken
Quality control is like crowd control. You have riot control to make
sure crowds don't get out of hand and you have quality control to make
sure quality doesn't get out of hand. This way, people won't want to
copy your products.
Greegor
wrote:
> On Mon, 2 Nov 2009 22:03:41 -0800 (PST), Greegor <greego...@gmail.com>
> wrote:
>
> >How do you like getting products made that way?
>
> Unless YOU QUOTE who you are responding to, nobody will know who or
> what the fuck you are talking about.
>
> Not everyone reads or looks at there news the same way.
>
> Relying on threaded views is retarded.
>
> QUOTE what you are referring to. D'oh!
Read the subject title, Archy!
SUBJECT: "PCB Encapsulants to Annoy Copy Pirates"
G > How do you like getting products made that way?
Archimedes' Lever
wrote:
Folks respond to the BODY of a post, not the fucking topic title. Those
that do have their reward, and are pretty fucking stupid as well.
Also, why would some fucking total retard snip the entire body
containing the details of what the person wanted to know, just to make a
snide fucking peanut gallery comment?
Oh... wait... That was a Retard Gallery comment. It's all clear now.
Archimedes' Lever
You should get a good, thermally conductive epoxy paint to seal it up
in, and *then* coat it with real black body paint, which contains enough
Carbon to worry me as a direct spray on, but would be fine as an overcoat
on the thermal epoxy. Then, you have a conductively cooled device from
the source through the epoxy, and a radiant cooled device from the "heat
sink" POV. That paint is pretty good at "emitting" the energy 'behind
it'.
Ken S. Tucker
wrote:
Your thinking is appreciated.
Our standard was set using a 10M ohm resistor, measured accurately
then spraying it and the leads, and then measure the diff, you know.
Following that, we choose to protect PCB's that way to start.
Of course a good circuit design is where it begins.
Ken
Ross Herbert
<OneBi...@InfiniteSeries.Org> wrote:
:On Thu, 05 Nov 2009 22:37:53 -0600, Vladimir Vassilevsky
It is far simpler and more effective to simply dip the whole board in pourable
silicone rubber prior to encapsulation with the hard epoxy. That will leave
plenty of room for expansion.
Archimedes' Lever
wrote:
Conformal coating can be spray applied or used as a dip, and the
coating thickness can be altered by simply buying the right coating to
achieve the right thickness desired in a single dip.
Make it too thick, and the security features go away.
Ken S. Tucker
> On Fri, 06 Nov 2009 01:14:03 -0800, Archimedes' Lever
>
Well ok, as long as the silly-cone has the appropriate thermal
conduction. I'd order a thermistor measure on the x-PCB to
verify temp remains happy on power components, if it has any.
Ken
Archimedes' Lever
You find that out when you beat up ten finished assembly candidates
till they fail. Then find out where they failed and why.
Greegor
wrote:
> On Fri, 6 Nov 2009 17:56:40 -0800 (PST), Greegor <greego...@gmail.com>
> wrote:
>
>
>
>
>
> >On Nov 3, 1:15 am, Archimedes' Lever <OneBigLe...@InfiniteSeries.Org>
> >wrote:
> >> On Mon, 2 Nov 2009 22:03:41 -0800 (PST), Greegor <greego...@gmail.com>
> >> wrote:
>
> >> >How do you like getting products made that way?
>
> >> Unless YOU QUOTE who you are responding to, nobody will know who or
> >> what the fuck you are talking about.
>
> >> Not everyone reads or looks at there news the same way.
>
> >> Relying on threaded views is retarded.
>
> >> QUOTE what you are referring to. D'oh!
G > Read the subject title, Archy!
SUBJECT: "PCB Encapsulants to Annoy Copy Pirates"
G > How do you like getting products made that way?
AL > Folks respond to the BODY of a post, not
AL > the fucking topic title. Those that do have
AL > their reward, and are pretty fucking
AL > stupid as well.
AL >
AL > Also, why would some fucking total retard
AL > snip the entire body containing the details
AL > of what the person wanted to know, just
AL > to make a snide fucking peanut gallery
AL > comment?
AL >
AL > Oh... wait... That was a Retard Gallery
AL > comment. It's all clear now.
And here you are, Archy!
Don Lancaster
One rude surprise of some conformal coatings: Ferrite beads may no
longer work!
Hoo boy was a major Aerospace company surprised when they found this out
the hard way.
Apparently a ferrite bead has to magnetostrict to operate per spec.
Keep it from doing so and the rules change wildly.
--
Many thanks,
Don Lancaster voice phone: (928)428-4073
Synergetics 3860 West First Street Box 809 Thatcher, AZ 85552
rss: http://www.tinaja.com/whtnu.xml email: d...@tinaja.com
Please visit my GURU's LAIR web site at http://www.tinaja.com
D from BC
wrote:
>One rude surprise of some conformal coatings: Ferrite beads may no
>longer work!
>
>Hoo boy was a major Aerospace company surprised when they found this out
>the hard way.
>
>Apparently a ferrite bead has to magnetostrict to operate per spec.
>Keep it from doing so and the rules change wildly.
Neato....
The material has to move to act like the specs.
Ken S. Tucker
You can use vaseline to maintain the magnetostrictive constant,
constant.
Ken
Archimedes' Lever
wrote:
>And here you are, Archy!
You're an A-GREEG-ious IDIOT.
Archimedes' Lever
Silicone dry lube spray, wiped clean.
Similar to mold release.
MooseFET
Beads also make heat when they do it. They are often the most "point
source" like heaters in the who design.
JosephKK
<myreal...@comic.com> wrote:
>I'd like to pot my smt pcb with the nastiest sh*t possible to make it
>hell for copy cats..
>
>So far I've found:
>
>http://www.mgchemicals.com/products/832ht.html
>'Extremely difficult to remove - grants incredible technology
>protection '
>
>Besides epoxy.. Are there alternatives?
>
>ex: Very very hot melt? Low melting point glass?
>
>D from BC
>Amateur smps designer
>British Columbia, Canada
>Posted to sci.electronics.design
Regular opaque hobby casting resins?
JosephKK
wrote:
>On Nov 3, 7:56 pm, D from BC <myrealaddr...@comic.com> wrote:
>> On Tue, 3 Nov 2009 18:22:04 -0800 (PST), "J.A. Legris"
>>
>>
>>
>> <jaleg...@sympatico.ca> wrote:
>> >On Nov 3, 7:33 pm, Archimedes' Lever <OneBigLe...@InfiniteSeries.Org>
>> >wrote:
>> >> On Tue, 3 Nov 2009 07:01:38 -0800 (PST), "J.A. Legris"
>>
>> >> <jaleg...@sympatico.ca> wrote:
>> >> >5) Environmentally unfriendly: for you, the customer and everyone
>> >> >else.
>>
>> >> Cured potting compounds pose no environmental threat (nearly all), and
>> >> even when uncured they only pose a threat to the environment if you or
>> >> another of your dopes gets the shit all over the place.
>>
>> >> Use some brains.,
>>
>> >I did, which is why I no longer use potting compounds. The problem is
>> >dopes like you who cannot reason beyond their next paycheck. The long-
>> >term effects of most synthetic chemicals have never been
>> >comprehensively assessed, so erring on the side of caution is the wise
>> >thing to do.
>>
>> I'm wondering about thermal expansion.
>
>You can find the thermal rate for most materials. There is likely to
>be a temperature gradient but it will be smaller than the total swing.
>
>
>> My 1"x1" pcb that I want to encapsulate will be in an environment
>> where the temp may swing from 22C to 70C daily.
>
>Smaller is better.
>
>>
>> I have no idea if an epoxy encapsulant will flex with heat and shear
>> forces will push on the tiny smt parts..
>> I'll try to find specs.
>> Perhaps testing in an environmental chamber is needed.
>> But I'm cheap, so I'll be using a light bulb in a box controlled by a
>> thermostat.
>
>If you put a very flexible layer over the parts it will help.
>Unfortunately it also provides the nice parting line to let the
>pirates pop it apart along the boundary.
>
>I have used a "light bulb" based heat box. Mine used power
>resistors. You can get thermostats that can switch the load. A
>couple of jugs of water in the hot box helps to hold the temperature
>steady.
>
>Run you freezer down as cold as it can get. Put the item in the
>freezer and let it soak till it is all the way down. Once it is fully
>cold move it to the hot box. leave it in the hot box for a day and
>then move it back to the freezer.
>
>This will quickly show you that the potting was a bad idea.
>
Scotsman's thermal shock.
D from BC
Maybe...
The potting should be resistant to a hot tipped vacuum system.
The potting potting compound is removed by melting and burning in an
archaeological fashion with a hot vacuum nozzle.
afaik...the above is doable.
If I can make a potting removal tool, the techno duplicators at 'The
Happy Circuit Company' can make a potting removal tool.
JosephKK
<myreal...@comic.com> wrote:
Last i saw they were catalytic thermosetting plastics. Not quite to
the diallyphthalate class, but pretty good.