Recommend potting compound for my application?

Mar 8, 2005

Hello,

I have a small circuit board that is just several "high power" SMT
LED's which require reasonably large copper pads on the board to
dissipate the heat. This small LED board is about 1.5" long and 1"
wide and fits inside a similarly shaped recess in the back of a plastic
lens. It will be placed LED side down into the plastic lens and then I
need to expoxy/pot the PCB into the plastic lens housing. The epoxy
should be thick enough so that it doesn't run down the size of the PCB
and interfere with the LED's. It should also be clear so that any drops
do not show up as discolorations inside the clear lens. It also
(ideally) have good thermal properties and help to dissipate the heat
generated by the LED's. Of course it needs to be non-conductive and
should hold up well to outdoor temperatures and be unaffected by
contact with water.

Can anyone recommend such a beast? Or can anyone recommend a company I
can contact who has a line of expoxies/potting compounds and see which
one would work for me?

Thanks!
Mike

Mar 8, 2005

Mike,

I was in a similar situation with my product. Unfortunately I think
you're going to have a hard time finding something to suit all of your
criteria. In my experience potting compound is not very good at
thermal transfer. The stuff that has the lowest thermal resistance is
not clear. A lot of it is toxic and needs special handling precautions
(something to be concerned about if you go into production). That, and
try to perform failure analysis on a dead product through potting
material. You have a number of headaches in store for you.

Contact Dow Automotive and/or Loctite Corporation as a start. They
both sell potting materials. Maybe you'll get lucky and find something
right for you.

On a side note, you might want to try conformal coating the PCB and
sealing it off from the elements inside your housing with gaskets.
Better heat dissipation and cheaper process. Also, I am currently
looking into low temperature injection molding to replace our potting
process. Look into this too. It appears to be a reasonable
alternative.

Good luck,

Gerb

John Larkin · Mar 8, 2005

On 8 Mar 2005 14:56:01 -0800, fastturbovette@yahoo.com wrote:

Hello,

I have a small circuit board that is just several "high power" SMT
LED's which require reasonably large copper pads on the board to
dissipate the heat. This small LED board is about 1.5" long and 1"
wide and fits inside a similarly shaped recess in the back of a plastic
lens. It will be placed LED side down into the plastic lens and then I
need to expoxy/pot the PCB into the plastic lens housing. The epoxy
should be thick enough so that it doesn't run down the size of the PCB
and interfere with the LED's. It should also be clear so that any drops
do not show up as discolorations inside the clear lens. It also
(ideally) have good thermal properties and help to dissipate the heat
generated by the LED's. Of course it needs to be non-conductive and
should hold up well to outdoor temperatures and be unaffected by
contact with water.

Can anyone recommend such a beast? Or can anyone recommend a company I
can contact who has a line of expoxies/potting compounds and see which
one would work for me?

Thanks!
Mike

Tough requirement. Transparent and thermally conductive don't go
together. Consider gaskets or o-rings, maybe; much less messy.

Master Bond is maybe the best supplier for custom epoxies.

http://www.masterbond.com/

John

Spehro Pefhany · Mar 8, 2005

On Tue, 08 Mar 2005 15:43:34 -0800, the renowned John Larkin
<jjlarkin@highSNIPlandTHIStechPLEASEnology.com> wrote:

On 8 Mar 2005 14:56:01 -0800, fastturbovette@yahoo.com wrote:

Hello,

I have a small circuit board that is just several "high power" SMT
LED's which require reasonably large copper pads on the board to
dissipate the heat. This small LED board is about 1.5" long and 1"
wide and fits inside a similarly shaped recess in the back of a plastic
lens. It will be placed LED side down into the plastic lens and then I
need to expoxy/pot the PCB into the plastic lens housing. The epoxy
should be thick enough so that it doesn't run down the size of the PCB
and interfere with the LED's. It should also be clear so that any drops
do not show up as discolorations inside the clear lens. It also
(ideally) have good thermal properties and help to dissipate the heat
generated by the LED's. Of course it needs to be non-conductive and
should hold up well to outdoor temperatures and be unaffected by
contact with water.

Can anyone recommend such a beast? Or can anyone recommend a company I
can contact who has a line of expoxies/potting compounds and see which
one would work for me?

Thanks!
Mike

Tough requirement. Transparent and thermally conductive don't go
together.

Electrical conductivity and transparent don't go together, but for
thermal conductivity, diamond is VERY good (the best known, AFAIK) as
well as being transparent over a wide range of wavelengths.

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

Rich Grise · Mar 9, 2005

On Tue, 08 Mar 2005 14:56:01 -0800, fastturbovette wrote:

Hello,

I have a small circuit board that is just several "high power" SMT
LED's which require reasonably large copper pads on the board to
dissipate the heat. This small LED board is about 1.5" long and 1"
wide and fits inside a similarly shaped recess in the back of a plastic
lens. It will be placed LED side down into the plastic lens and then I
need to expoxy/pot the PCB into the plastic lens housing. The epoxy
should be thick enough so that it doesn't run down the size of the PCB
and interfere with the LED's. It should also be clear so that any drops
do not show up as discolorations inside the clear lens. It also
(ideally) have good thermal properties and help to dissipate the heat
generated by the LED's. Of course it needs to be non-conductive and
should hold up well to outdoor temperatures and be unaffected by
contact with water.

Can anyone recommend such a beast? Or can anyone recommend a company I
can contact who has a line of expoxies/potting compounds and see which
one would work for me?

These guys might have something:
http://www.dowcorning.com/applications/product_finder/ProductSearch.aspx?searchfor=potting%20compound&country=USA

Good Luck!
Rich

Rich Grise · Mar 9, 2005

On Tue, 08 Mar 2005 19:42:24 -0500, Spehro Pefhany wrote:

On Tue, 08 Mar 2005 19:33:37 -0500, Spehro Pefhany
speffSNIP@interlogDOTyou.knowwhat> wrote:

On Tue, 08 Mar 2005 15:43:34 -0800, the renowned John Larkin
jjlarkin@highSNIPlandTHIStechPLEASEnology.com> wrote:

On 8 Mar 2005 14:56:01 -0800, fastturbovette@yahoo.com wrote:

Hello,

I have a small circuit board that is just several "high power" SMT
LED's which require reasonably large copper pads on the board to
dissipate the heat. This small LED board is about 1.5" long and 1"
wide and fits inside a similarly shaped recess in the back of a plastic
lens. It will be placed LED side down into the plastic lens and then I
need to expoxy/pot the PCB into the plastic lens housing. The epoxy
should be thick enough so that it doesn't run down the size of the PCB
and interfere with the LED's. It should also be clear so that any drops
do not show up as discolorations inside the clear lens. It also
(ideally) have good thermal properties and help to dissipate the heat
generated by the LED's. Of course it needs to be non-conductive and
should hold up well to outdoor temperatures and be unaffected by
contact with water.

Can anyone recommend such a beast? Or can anyone recommend a company I
can contact who has a line of expoxies/potting compounds and see which
one would work for me?

Thanks!
Mike

Tough requirement. Transparent and thermally conductive don't go
together.

Electrical conductivity and transparent don't go together, but for
thermal conductivity, diamond is VERY good (the best known, AFAIK) as
well as being transparent over a wide range of wavelengths.

P.S., I don't think it's available as a potting compound.

Just fill your box with coal and have Superman squish it. ;-)

Cheers!
Rich

John Larkin · Mar 9, 2005

On Tue, 08 Mar 2005 19:33:37 -0500, Spehro Pefhany
<speffSNIP@interlogDOTyou.knowwhat> wrote:

On Tue, 08 Mar 2005 15:43:34 -0800, the renowned John Larkin
jjlarkin@highSNIPlandTHIStechPLEASEnology.com> wrote:

On 8 Mar 2005 14:56:01 -0800, fastturbovette@yahoo.com wrote:

Hello,

I have a small circuit board that is just several "high power" SMT
LED's which require reasonably large copper pads on the board to
dissipate the heat. This small LED board is about 1.5" long and 1"
wide and fits inside a similarly shaped recess in the back of a plastic
lens. It will be placed LED side down into the plastic lens and then I
need to expoxy/pot the PCB into the plastic lens housing. The epoxy
should be thick enough so that it doesn't run down the size of the PCB
and interfere with the LED's. It should also be clear so that any drops
do not show up as discolorations inside the clear lens. It also
(ideally) have good thermal properties and help to dissipate the heat
generated by the LED's. Of course it needs to be non-conductive and
should hold up well to outdoor temperatures and be unaffected by
contact with water.

Can anyone recommend such a beast? Or can anyone recommend a company I
can contact who has a line of expoxies/potting compounds and see which
one would work for me?

Thanks!
Mike

Tough requirement. Transparent and thermally conductive don't go
together.

Electrical conductivity and transparent don't go together, but for
thermal conductivity, diamond is VERY good (the best known, AFAIK) as
well as being transparent over a wide range of wavelengths.

Best regards,
Spehro Pefhany

Well, the crappy diamond is about 2x better than copper. But the
isotopically pure version is 50x.

John

John Larkin · Mar 9, 2005

On Tue, 08 Mar 2005 19:42:24 -0500, Spehro Pefhany
<speffSNIP@interlogDOTyou.knowwhat> wrote:

On Tue, 08 Mar 2005 19:33:37 -0500, Spehro Pefhany
speffSNIP@interlogDOTyou.knowwhat> wrote:

On Tue, 08 Mar 2005 15:43:34 -0800, the renowned John Larkin
jjlarkin@highSNIPlandTHIStechPLEASEnology.com> wrote:

On 8 Mar 2005 14:56:01 -0800, fastturbovette@yahoo.com wrote:

Hello,

I have a small circuit board that is just several "high power" SMT
LED's which require reasonably large copper pads on the board to
dissipate the heat. This small LED board is about 1.5" long and 1"
wide and fits inside a similarly shaped recess in the back of a plastic
lens. It will be placed LED side down into the plastic lens and then I
need to expoxy/pot the PCB into the plastic lens housing. The epoxy
should be thick enough so that it doesn't run down the size of the PCB
and interfere with the LED's. It should also be clear so that any drops
do not show up as discolorations inside the clear lens. It also
(ideally) have good thermal properties and help to dissipate the heat
generated by the LED's. Of course it needs to be non-conductive and
should hold up well to outdoor temperatures and be unaffected by
contact with water.

Can anyone recommend such a beast? Or can anyone recommend a company I
can contact who has a line of expoxies/potting compounds and see which
one would work for me?

Thanks!
Mike

Tough requirement. Transparent and thermally conductive don't go
together.

Electrical conductivity and transparent don't go together, but for
thermal conductivity, diamond is VERY good (the best known, AFAIK) as
well as being transparent over a wide range of wavelengths.

P.S., I don't think it's available as a potting compound.

You can buy diamond-filled epoxy. But the thermal conductivity is
still dominated by the epoxy between the grains, so it's not a whole
lot better than epoxy filled with AlN or something.

John

John Woodgate · Mar 9, 2005

I read in sci.electronics.design that fastturbovette@yahoo.com wrote (in
<1110322561.393141.116890@g14g2000cwa.googlegroups.com&gt

about
'Recommend potting compound for my application?', on Tue, 8 Mar 2005:

Can anyone recommend such a beast? Or can anyone recommend a company I
can contact who has a line of expoxies/potting compounds and see which
one would work for me?

I suggest a re-think. Once you get a whole series of special
requirements applied to a part or technique, it's probably the wrong way
to go.

If you just want to secure the board in place, consider one or more
spring clips.
--
Regards, John Woodgate, OOO - Own Opinions Only.
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

Al · Mar 9, 2005

In article <1110322561.393141.116890@g14g2000cwa.googlegroups.com>,
fastturbovette@yahoo.com wrote:

Hello,

I have a small circuit board that is just several "high power" SMT
LED's which require reasonably large copper pads on the board to
dissipate the heat. This small LED board is about 1.5" long and 1"
wide and fits inside a similarly shaped recess in the back of a plastic
lens. It will be placed LED side down into the plastic lens and then I
need to expoxy/pot the PCB into the plastic lens housing. The epoxy
should be thick enough so that it doesn't run down the size of the PCB
and interfere with the LED's. It should also be clear so that any drops
do not show up as discolorations inside the clear lens. It also
(ideally) have good thermal properties and help to dissipate the heat
generated by the LED's. Of course it needs to be non-conductive and
should hold up well to outdoor temperatures and be unaffected by
contact with water.

Can anyone recommend such a beast? Or can anyone recommend a company I
can contact who has a line of expoxies/potting compounds and see which
one would work for me?

Thanks!
Mike

First the rant. Packaging is an engineering discipline by itself. Don't
try this at home ;-)

One answer:http://www.epotek.com/

Al

Spehro Pefhany · Mar 9, 2005

On Wed, 9 Mar 2005 14:58:54 -0500, the renowned Keith Williams
<krw@att.bizzzz> wrote:

Oh, that must make a good high-security potting compound. We used some
carburundum (IIRC) filled epoxy on the crypto key storage unit I worked
on in the early '90s.

How would the diamond, carbide or whatever particles affect
decapsulation in the usual manner? I'm not sure you'd even notice the
fill.

It might be effective for filling a hollow bar for a lock though.
Apparently cheap cordless cutoff saws with a diamond wheel can deal
with hardened metal locks pretty effectively. Or maybe just fill it
with something that gums up a diamond saw.

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

John Larkin · Mar 9, 2005

On Wed, 09 Mar 2005 15:41:25 -0500, Spehro Pefhany
<speffSNIP@interlogDOTyou.knowwhat> wrote:

On Wed, 9 Mar 2005 14:58:54 -0500, the renowned Keith Williams
krw@att.bizzzz> wrote:

Oh, that must make a good high-security potting compound. We used some
carburundum (IIRC) filled epoxy on the crypto key storage unit I worked
on in the early '90s.

How would the diamond, carbide or whatever particles affect
decapsulation in the usual manner? I'm not sure you'd even notice the
fill.

It might be effective for filling a hollow bar for a lock though.
Apparently cheap cordless cutoff saws with a diamond wheel can deal
with hardened metal locks pretty effectively.

The lock on our gate froze up (lots of salt air here, eats locks) and
I cut off the hardened hasp with a Dremel and a fiber cutoff wheel in
under a minute. So much for security.

We once used some glass-filled epoxy for potting. If you tried to mill
open a brick, it would destroy a couple of end mills per block. I
can't imaging the tool life you'd get machining carburundum or
diamond-filled stuff. Sounds grim.

John

Jim Thompson · Mar 9, 2005

On Wed, 09 Mar 2005 13:16:11 -0800, John Larkin
<jjlarkin@highSNIPlandTHIStechPLEASEnology.com> wrote:

[snip]

The lock on our gate froze up (lots of salt air here, eats locks)

What's a "lock" ?

480)460-2350 | |
| E-mail Address at Website Fax

Spehro Pefhany · Mar 9, 2005

On Wed, 09 Mar 2005 13:16:11 -0800, the renowned John Larkin
<jjlarkin@highSNIPlandTHIStechPLEASEnology.com> wrote:

On Wed, 09 Mar 2005 15:41:25 -0500, Spehro Pefhany
speffSNIP@interlogDOTyou.knowwhat> wrote:

On Wed, 9 Mar 2005 14:58:54 -0500, the renowned Keith Williams
krw@att.bizzzz> wrote:

Oh, that must make a good high-security potting compound. We used some
carburundum (IIRC) filled epoxy on the crypto key storage unit I worked
on in the early '90s.

How would the diamond, carbide or whatever particles affect
decapsulation in the usual manner? I'm not sure you'd even notice the
fill.

It might be effective for filling a hollow bar for a lock though.
Apparently cheap cordless cutoff saws with a diamond wheel can deal
with hardened metal locks pretty effectively.

The lock on our gate froze up (lots of salt air here, eats locks) and
I cut off the hardened hasp with a Dremel and a fiber cutoff wheel in
under a minute. So much for security.

Heh. And that's a relatively wimpy tool.

We once used some glass-filled epoxy for potting. If you tried to mill
open a brick, it would destroy a couple of end mills per block. I
can't imaging the tool life you'd get machining carburundum or
diamond-filled stuff. Sounds grim.

John

Why even try? Just use the usual array of chemicals to dissolve the
binder away and the diamond particles get flushed away with everything
else.

Were the end mills solid carbide or some kind of HSS or HSS with TiN
coating? Of course PCB material eats HSS drills for just that reason,
so carbide is the way to go. The thin ones break if you look at them
sideways too hard, but you can't have everything.

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

John Larkin · Mar 9, 2005

On Wed, 09 Mar 2005 16:42:14 -0500, Spehro Pefhany
<speffSNIP@interlogDOTyou.knowwhat> wrote:

On Wed, 09 Mar 2005 13:16:11 -0800, the renowned John Larkin
jjlarkin@highSNIPlandTHIStechPLEASEnology.com> wrote:

On Wed, 09 Mar 2005 15:41:25 -0500, Spehro Pefhany
speffSNIP@interlogDOTyou.knowwhat> wrote:

On Wed, 9 Mar 2005 14:58:54 -0500, the renowned Keith Williams
krw@att.bizzzz> wrote:

Oh, that must make a good high-security potting compound. We used some
carburundum (IIRC) filled epoxy on the crypto key storage unit I worked
on in the early '90s.

How would the diamond, carbide or whatever particles affect
decapsulation in the usual manner? I'm not sure you'd even notice the
fill.

It might be effective for filling a hollow bar for a lock though.
Apparently cheap cordless cutoff saws with a diamond wheel can deal
with hardened metal locks pretty effectively.

The lock on our gate froze up (lots of salt air here, eats locks) and
I cut off the hardened hasp with a Dremel and a fiber cutoff wheel in
under a minute. So much for security.

Heh. And that's a relatively wimpy tool.

We once used some glass-filled epoxy for potting. If you tried to mill
open a brick, it would destroy a couple of end mills per block. I
can't imaging the tool life you'd get machining carburundum or
diamond-filled stuff. Sounds grim.

John

Why even try? Just use the usual array of chemicals to dissolve the
binder away and the diamond particles get flushed away with everything
else.

I guess, but that would keep going and eat the parts, too.

Were the end mills solid carbide or some kind of HSS or HSS with TiN
coating? Of course PCB material eats HSS drills for just that reason,
so carbide is the way to go. The thin ones break if you look at them
sideways too hard, but you can't have everything.

Just the regular steel ones, I think. And right, I've never had any
luck drilling a pcb with a carbide drill. Unless you have a 100K RPM
air-bearing excelon drill with the right entry/exit materials, they
snap off after a couple of holes. Better to use a steel drill and run
it until it gets dull. Better yet to buy the danged PCBs.

Potting always sounds cool, but in practise it's a messy nightmare.

John

Keith Williams · Mar 10, 2005

In article <besu215vt07406dj5hc6m9i08i1m20rijf@4ax.com>,
jjlarkin@highSNIPlandTHIStechPLEASEnology.com says...

On Wed, 09 Mar 2005 16:42:14 -0500, Spehro Pefhany
speffSNIP@interlogDOTyou.knowwhat> wrote:

On Wed, 09 Mar 2005 13:16:11 -0800, the renowned John Larkin
jjlarkin@highSNIPlandTHIStechPLEASEnology.com> wrote:

On Wed, 09 Mar 2005 15:41:25 -0500, Spehro Pefhany
speffSNIP@interlogDOTyou.knowwhat> wrote:

On Wed, 9 Mar 2005 14:58:54 -0500, the renowned Keith Williams
krw@att.bizzzz> wrote:

Oh, that must make a good high-security potting compound. We used some
carburundum (IIRC) filled epoxy on the crypto key storage unit I worked
on in the early '90s.

How would the diamond, carbide or whatever particles affect
decapsulation in the usual manner? I'm not sure you'd even notice the
fill.

It might be effective for filling a hollow bar for a lock though.
Apparently cheap cordless cutoff saws with a diamond wheel can deal
with hardened metal locks pretty effectively.

The lock on our gate froze up (lots of salt air here, eats locks) and
I cut off the hardened hasp with a Dremel and a fiber cutoff wheel in
under a minute. So much for security.

Heh. And that's a relatively wimpy tool.

We once used some glass-filled epoxy for potting. If you tried to mill
open a brick, it would destroy a couple of end mills per block. I
can't imaging the tool life you'd get machining carburundum or
diamond-filled stuff. Sounds grim.

John

Why even try? Just use the usual array of chemicals to dissolve the
binder away and the diamond particles get flushed away with everything
else.

I guess, but that would keep going and eat the parts, too.

Right. The crypto key storage unit had the PCB mounted in a plastic
(can't remember the particular plastic) case, then wound with four
insulated wires connected in a bridge configuration. Any two wires
shorting or any cut wire causes an alarm to the key storage circuitry
and the keys are zapped. The intention of the filler was to make a
mess of any tools trying to surgically remove the potting, thus greatly
increase the chance of cutting/shorting wire(s). The insulation on the
wire was designed to dissolve in any chemical that would attack the
potting, also causing an intrusion alarm.

Were the end mills solid carbide or some kind of HSS or HSS with TiN
coating? Of course PCB material eats HSS drills for just that reason,
so carbide is the way to go. The thin ones break if you look at them
sideways too hard, but you can't have everything.

Just the regular steel ones, I think. And right, I've never had any
luck drilling a pcb with a carbide drill. Unless you have a 100K RPM
air-bearing excelon drill with the right entry/exit materials, they
snap off after a couple of holes. Better to use a steel drill and run
it until it gets dull. Better yet to buy the danged PCBs.

Let someone else break their bits?

Potting always sounds cool, but in practise it's a messy nightmare.

It is cool, if it's someone else's messy nightmare. ;-) I was just
the (lead) design and physical security engineer and on the project.
We had the materials people pick the potting and yucky manufacturing
processes.

--
Keith

Recommend potting compound for my application?

Guest

Guest

John Larkin

Guest

Spehro Pefhany

Guest

Rich Grise

Guest

Rich Grise

Guest

John Larkin

Guest

John Larkin

Guest

John Woodgate

Guest

Al

Guest

Spehro Pefhany

Guest

John Larkin

Guest

Jim Thompson

Guest

Spehro Pefhany

Guest

John Larkin

Guest

Keith Williams

Guest

Welcome to EDABoard.com

Sponsor

Online statistics

Forum statistics

Recommend potting compound for my application?

Guest

Guest

John Larkin

Guest

Spehro Pefhany

Guest

Rich Grise

Guest

Rich Grise

Guest

John Larkin

Guest

John Larkin

Guest

John Woodgate

Guest

Al

Guest

Spehro Pefhany

Guest

John Larkin

Guest

Jim Thompson

Guest

Spehro Pefhany

Guest

John Larkin

Guest

Keith Williams

Guest

Log in

Welcome to EDABoard.com

Sponsor