Cheap ckt. potting options

Mark:

First off let me comment that potting a PC board is not an effective duplication deterrent. One company that I worked for potted some of their products to seal the circuits for environmental protection. However whenever there was a field failure the potting would be taken off the PC board to see what had happened to the device !! So potting would be somewhat of a bother to the cloner of your product but would not prevent the copy process.

Secondly please be aware that there are some things to consider when potting a circuit board. Here is a list of some key areas of consideration.

1) Some cheap hardware store type epoxy products have fillers in them that may render the potting conductive. I am aware of both steel and aluminum filled types.

2) Epoxy products of some types and surface mounted components may not always be the best mix. The differences in thermal characteristics between FR4 PC boards and the potting can break small surface mount component solder connections.

3) There are plenty of epoxy products designed specifcally for electronics potting so you should not have too much problem finding a product to use. I would check out some electronics products retailers such as Newark or McMaster Carr.

4) Partitioning the area to get potted can be very difficult to implement effectively. In order to make potting practical and have the material be pourable means that the epoxy is somewhat viscous and it will run through the smallest hole or crack. If it is the battery area that you want to leave open then I would pot with a 9V battery harness protruding above the epoxy surface. In the case of batteries like AA or AAA then there are some nylon battery holders that you could use to hold the cells and isolate the area from epoxy fill-in.

5) Even with item 4 above said there can be problems getting complete fill of the areas under circuit boards if not enough space is left between the edges of the PC board and the case that it is in. I have seen the need to angle off the corner(s) of the PC board or provide a hole through the center of the PC board to allow the epoxy to flow past and under the board.

6) Two part epoxy can often generate some heat as a result of the curing process. In addition sometimes ovens are used to hasten the cure of the epoxy to speed up the manufacturing process. If your PC board has some sensitive low level analog type circuits it is possible that the curing heat can cause detrimental shifts in the circuit performance or destroy any pre-potting calibrations than may have been done.

7) There are some silicon based potting materials available for potting that have some advantages for surface mounted components.

It is finally my opinion that there are better ways of protecting a design than potting that are less bothersome than potting. One very good way is to capture a large part of the product design and intellectual property into the firmware of a microcontroller on the product PC board. Select a microcontroller that has a good locking protection system and you will be much farther toward protecting your ideas from prying eyes. Even the protections offered by "fuse" protected FLASH based microcontrollers can be sometimes overcome by persistent hackers if the product cloner is willing to spend enough money. So the best way for you to overcome that consideration is to stay ahead of the cloner by offering changes to the product every so often that make your product look more attractive to the customer than an older less capable clone.

Good luck

- mkaras

"Potting is such sweet sorrow" The bard knew in R+J. Harry

What does the circuit do? There are darn few "new" ideas that can't be re-engineered or reverse-engineered in a day or two.

Also a quick googling shows that almost anything can be dissolved, even epoxies.

Your best protection is offering a good product at a competitive price.

epoxy or polyester resin mixed with very fine sand - less flammable, stronger, cheaper and better heat conduction than epoxy alone

----== Posted via Newsfeeds.Com - Unlimited-Unrestricted-Secure Usenet News==----

The #1 Newsgroup Service in the World! 120,000+ Newsgroups

----= East and West-Coast Server Farms - Total Privacy via Encryption =----

Ooh, great line.

John

For runs in the hundreds and above all good advice, i think OP was talking one dozen or two.

--
 JosephKK
 Gegen dummheit kampfen die Gotter Selbst, vergebens.  
  --Schiller

In addition to the sound advice from mkaras I would add that using any hard epoxy to pot components which generate heat will definitely result in early failure of the device.

Way back in the dim dark ages when Evinrude/Johnson started making solid state ignition systems they used hard epoxy to encapsulate them. A service agent gave me a few dud ones to look at and see if I could find out their main failure mode. I had to import a gallon of exotic epoxy solvent from the US at great expense and after a couple of weeks I finally removed enough epoxy to get down to the board level. Inspection showed that components which ran hot would try to expand. Naturally, the epoxy resisted, being hard against both the underside of the pcb and the top of the component, and so the through-hole component leads expanded the other way and either ripped the solder pads off the board or produced dry solder joints.

If potting components which are going to expand I use a primary layer of soft silicon rubber prior to encapsulation in the hard epoxy potting compound. This gives some leeway for expansion.

$pend all you can; no matter what you do, it is impossible to prevent duplication.

News==----

Newsgroups

Adding sand only makes it slightly more difficult to "bust into". And sand is a good insulator; there is no such thing as "thermally conductive" epoxy *unless* a conductive additive like fine particles of metal (silver, copper, aluminum, etd) or carbon was used--and that makes it electrically conductive.

Though metals are good, cheap thermal conductors, alumina (aluminum oxide) is much better than sand, and the purer, the better.

Yes, but did he have to quote two hundred redundant lines to do it?

Jim

Expandable caulk from Home Depot. $2.99 the tube for about 1 m^3 potting.

Jim

Pretty much all of the low-cost potting epoxies will turn into a gell state when heated to between 80-150 degrees, then they can then be readily peeled off. Two-part araldite permanently changes composition at around 100 degrees.

As others have mentioned, it can be a pain in the arse to properly pot the device, and then you have to worry about thermal expansion issues.

A good technique for providing environmental protection and some degree of protection against hackers is to use a hard conformal coating. There are many varieties around, Dymax sell a very hard black conformal coating that is very difficult to remove. Advice: put some electrical tape around the edge of the board, put a plob of coating on top, let it settle nicely, shine a UV light at it (or buy a two-part agent), then peel off the tape. You can also buy hard spray on coatings, that work well (not silicone or acrylic).

In addition, or alone, you can also use a solvent that will dissolve the part numbers of your components. I haven't tried it so I can't tell you what would do the job. Its pretty hard to tell an opamp is an opamp, especially if its in a funny case like a PLCC.

/andrew

You could also hotstamp an obliterating label over the top of each part before assembly. But most competent techs can probe around and make a good guess at the functions.

You should read Don Lancaster on patents and protections. He makes valid points.

--
++++++++++++++++++++++++++++++++++++++++++++++++++++
+ Required crap appended to avoid restrictions imposed by brain   +
+ damaged idiots. 
+
+ Server Response: \'441 Posting Failed (Rejected by POST filter)\', +
+ Port: 119, Secure(SSL): No, Server Error: 441, 
+
+ Error Number: 0x800CCCA9 
+
++++++++++++++++++++++++++++++++++++++++++++++++++++

If your quantity is large enough, you can get the ships custom marked with your companies part number. This can save a bit of time.

If you really want to cause trouble for those copying the circuit, have the parts numbered with something that misleads. An op-amp with a logic gate's number is an example.

--
--
kensmith@rahul.net   forging knowledge

As with all the other comments a multi-stage approach is going to give you the best protection, but it still will never be perfect (more like disuading the average curious hacker and making the more determined reverse engineer's job more difficult). Shaving the top layer off of the chips with a file to remove the marking works well, along with using microchip protection like blowing internal fuses, software protection, etc). Also a two-part polyester potting compound with a shore A hardness of about 95 seems to be a good balance between being too rigid (encapsulation self-disassembles when you hit it with a hammer, no give for thermal expansion, etc) or too soft (can easily be removed with a craft knife). A dark coloured hard conformal coating may be the best idea (I have never tried this thou), especially if it is difficult to remove with heat or chemicals.

In the end thou, if it is a simple product with a potentialy large market, then you will not be able to protect it from the determined who see a buck to be made.

No wonder why i keep burning my fingers!

I imagine they don't recommend powdered beryllia. ;-)

Cheers! Rich

Probably not. Especially when the thin layers of low thermal conductivity epoxy between the beryllia particles makes their advantage pretty small, compared to the negatives. A good substitute would be aluminum granules or needles that had been oxidized to insulate them at the circuit's working voltage.