gluing pcb to wood

Ok, so you need it ALIGNED to something when attached to a piece of wood... Difficult. I think you are now in the range of screws and spacers,with maybe a frame around your board to hold it. You said conformal coat, which means the board has been dipped into some sort of rubbery solution that has hardened. That is going to make any type of glue difficult, as the rubber is pretty smooth. The double stick tape is still the best idea, maybe with some sort of spacers to hold your alignment...

Charlie

afaik smoothness is not a big problem with epoxy or methacrylate. If so then sand or prime or both on surfaces first. On smooth surfaces the bond strength depends more on the electron interactions than a mechanical interlock from surface roughness.

A rubbery conformal coating is not the only type.

If alignment is important than take care. Many adhesives shrink during cure. Ex: silicone, contact cement, white glue, gorilla glue

Point 1: Gluing anything that "can't move even a few mils" to wood, which moves all the time with every change in humidity, is a recipe for disaster from the get-go...and if it's large, the differential expansion between wood and circuit board may well peel it off with any glue (or distort/warp it with some super-duper glue that holds it tight.

Point 2: Hot Melt glue, used poorly, works poorly. Used well, you can do some nice work with the stuff. I've rebuild many laptop power adapter connectors and replaced the molded connector end with a construction of hot-melt glue and popsicle-stick that have usually lasted longer than the original molded connectors did.

Point 2a: Hot glues vary. The general purpose stuff I use the most of is milky when cold, clear when hot. It is solid but not brittle when cold. Others range from quite brittle when cold to rubbery.

--
Cats, coffee, chocolate...vices to live by
Please don't feed the trolls. Killfile and ignore them so they will go away.

Yeah. I removed a picture stuck to the wall with that stuff. It removed the surface of the wall.

E-6000 is a similar product - it's a solvent-based clear adhesive gel/paste. Strongy, slightly flexible when dried, and cuttable-apart.

Using a bead of this stuff at each corner, thick enough to allow a box-cutter blade to be pushed into it (if necessary to remove at a later date) might do the job for you. It'd be strong enough to hold the PCB, but repairable.

I agree with the concerns about gluing to wood, though. Differential expansion across the grain of natural wood, as humidity changes, can run to around 1/8" per foot of width. If you need really precise placement, this expansion would be enough to mess up your alignment, even if it isn't enough to cause anything to bend or split (which is entirely possible).

If you must have some sort of glue-down to a rigid surface, you'll need to have something with more dimensional stability than wood.

--
Dave Platt                                    AE6EO
Friends of Jade Warrior home page:  http://www.radagast.org/jade-warrior
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

Ha! I've done the same. It works great for reconstructing hacked up power plugs. Glue gun is now standard equipment on the bench.

Magnets. Drill a hole in the wood, glue a magnet in the hole flush with the surface of the wood, lay the PCB on top and put a magnet on the PCB. Uses two small super magnets at each mounting location. The super magnet on the top of the PCB does not need to be glued to it, but you can if you want. I just tried it with a double sided PCB and it works great.

Ed

Generally, it's about how well the coating sticks to the wood, and how well the PCB sticks to the coating.

What's the application? If it's for sale, don't do it, period.

Good Luck! Rich

E-6000 is a similar product - it's a solvent-based clear adhesive gel/paste. Strongy, slightly flexible when dried, and cuttable-apart.

Using a bead of this stuff at each corner, thick enough to allow a box-cutter blade to be pushed into it (if necessary to remove at a later date) might do the job for you. It'd be strong enough to hold the PCB, but repairable.

I agree with the concerns about gluing to wood, though. Differential expansion across the grain of natural wood, as humidity changes, can run to around 1/8" per foot of width. If you need really precise placement, this expansion would be enough to mess up your alignment, even if it isn't enough to cause anything to bend or split (which is entirely possible).

If you must have some sort of glue-down to a rigid surface, you'll need to have something with more dimensional stability than wood.

-----

The alignment issues are not dynamic issues as these can be calibrated out. But one can't have the pcb move around to any degree during user interaction. The alignment discussion came about when talking about using tape... which is entirely different from glue. Usually double sided tape is thick and elastic and therefore will have a lot of give possibly creating alignment errors for any reason(such as the relative orientation due to gravity).

My point is that the wood, itself, creates dynamic issues, due to its tendency to shrink or expand as the relative humidity changes.

If you use wood, and plan to fasten the PCB to the wood, you've got a not-good choice to make:

- Use a very rigid fastening method (e.g. epoxy). The fastener itself won't have elasticity which could create alignment errors as (e.g.) the position of the device changes. However, the shrink/expand behavior of the wood *will* create such errors, and will also place stress on the wood and PCB which may bend or split them.

- Use a less-rigid fastening method (e.g. an adhesive with some amount of elasticity). This will reduce the bending/splitting stress, but may increase the tendency to create gravity-related alignment errors... and you'll still have the humidity-related alignment changes to deal with. If you want to glue it down, it sounds as if the combination of a truly rigid substrate (i.e. not wood), and a relative stiff adhesive.

--
Dave Platt                                    AE6EO
Friends of Jade Warrior home page:  http://www.radagast.org/jade-warrior
  I do _not_ wish to receive unsolicited commercial email, and I will
     boycott any company which has the gall to send me such ads!

g

Have you seen the price of permanent magnets these days! Until more mines come online outside China, I expect the price to go nowhere but up.

Otherwise magnets sound good.

George H.

Yep. $6.80 for 8 - 2 per corner - of the small neo's I tested: CAT# MAG-76 at Allelectronics. Shipping will probably bring the cost to double that or a bit more.

I don't know if it will ever go down. I paid ~ $15.00 as I recall for a large washer shaped 52 magnet 1 1/4 diameter and 3/16 thick with the center hole. While I was looking for that magnet, I found a number of really large magnets north of $100!

Yes, they work well through a double sided PCB - until the OP throws another wrinkle at us. I just wasted ~3 hours today because the client spec'd 2400 C when they should have spec'd

1000 C. When I asked what they were going to use for a sensor they proposed first type S, then type R and finally type K thermocouples. Something was clearly rotten in Denmark. None of those are rated for 2400 C. That confusion highlighted the incorrect temperature spec. It's all sorted now - what they *really* need is a max 1000 C and the type K works fine with that - but it took multiple conversations with 4 different people and about 3 hours to get it cleared up. Now there's another issue with the PID parameters, which is still to be resolved.

Ed

1/2" x 1/16" Nd disks - $11.00 shipped for 50 from sleazebay (vendor in Florida, as best I recall - I prefer to get things in country, they show up faster, usually.) Don't know the precise strength (so likely on the low end for Nd), but they stick pretty good, even through a sheet of 1/16 FR4

--
Cats, coffee, chocolate...vices to live by
Please don't feed the trolls. Killfile and ignore them so they will go away.

That's a good price. As a comparison I looked at the K&J Magnetics site. They have 1/2" x 1/16" grade 42 neo discs for $.72 each, 3 cents less each at qty 50.

Ed

So have I. One day I'll scrape it all off the floor ;-)

--
"For a successful technology, reality must take precedence 
over public relations, for nature cannot be fooled."
                                       (Richard Feynman)