Delrin repair

[Alex Eisenhut]

Anyone that knows about Delrin knows it's difficult to glue. I have a Tektronix potentiometer from a scope, it has a threaded hole in the Delrin for a set screw. That hole eventually cracks and fails.

So I drilled a hole and put a brass threaded insert, with JB Weld Plastic Bonder and made a big glop. The insert was a little bit squashed so that the set screw exerts pressure to hold it in place.

I let it cure 24hrs, and I can tighten the screw enough to hold on to the shaft... Thing is, anyone think this will hold or fail?

I'm pretty good at visualizing repairs like this, but don't have the experience to tell if they'll last. (My last improv repair was a weird switch with a home-made copper collet, it works but corrosion has set in now... green moss or verdigris is growing between the copper and the body of the switch. I repaired the switch but it looks like it'll eat itself in a few years.)

[Paul Moir]

I've had some very good success with loctite cyanoacrylate (superglue, I think 454) with their activator (770) with Acetron, which is another brand of Delrin which is also non-porous.   As with all glues, more surface area is better which can be achieved by scuffing.  The activator added about 3x in the shear direction, which is what I was interested at the time.
This was bonding a piece of stainless, and I had about 1 sq in bonding area.  You could knock it off with a hammer but it was some work.  It was stronger than any epoxy bond I could make.  The bond failed along the acetron, since supergluing stainless is easy.

EDIT:  the lesson on that was that for a strong bond, you need to activate the surface.  You don't need their activator to do it either:  a propane torch can do the same thing. 

[Gyro]

Any possibility of tapping what's left to take a larger screw?

Edit: Or fabricate a metal bush that fits around the coupling with a tapped hole to take a locking screw.

[justanothercanuck]

(My last improv repair was a weird switch with a home-made copper collet, it works but corrosion has set in now... green moss or verdigris is growing between the copper and the body of the switch. I repaired the switch but it looks like it'll eat itself in a few years.)

OT - verdigris isn't natural, however patina is.

[Alex Eisenhut]

Yeah I saw all the different ways to activate the surface, I went with mechanical roughening and making the blob wrap around the shape of the pot.
Where did you get 770? I guess I'll have to order the Loctite product, I couldn't find it at any store locally in Montreal, including hobby stores.

I'll re-assemble the 1A4 today and hope it'll last. If it fails again, I'll drill a tiny hole straight across the pot and the shaft and put in piano wire. This is absurd. Tek went a bit overboard with these mixed-material designs IMO. This threading right into the plastic makes no sense to me.

The front panel knob has a metal insert with plastic over it, this is the right way to secure a set screw IMO.

[Paul Moir]

Practically any industrial supply shop will have it, just check their line card for Loctite.  In most industrial parks you can't throw a brick without hitting six of them.  Unfortunately these types of shops tend to be dinosaurs so they have useless websites.  For your amusement here's the place I buy it from:  http://www.schoonerind.com/ You wouldn't think it was a fairly large and old supplier who will happily do a cash sale of anything they carry.  A trip to Nova Scotia's a little far though.
Or just buy it online.  The small glass bottle I have has lasted years.

[RobertBG]

This is the ONLY thing I've found over the years to work reasonably well on Delrin and other Acetal plastics like it http://www.loctiteproducts.com/p/epxy_plstc_s/overview/Loctite-Epoxy-Plastic-Bonder.htm

For a repair like you are trying I'd mask the front of the hole off and use a dremel to expand the rear a little along with some small groves in a star like pattern around the hole that will fill with epoxy and give it some bulk.If you have a little bevel on the front and rear along with the groves the epoxy will fill it and mechanically  be held in place so you can carefully redrill to size.The stuff adheres reasonably well considering the surface it is sticking too but a little mechanical help goes a long ways towards a seamless permanent repair.

[Alex Eisenhut]

I found the JB Weld plastic bonder, I hope it works well enough. I see what you mean re: bevel, but there's precious little material left after the 5/32 hole. I wonder if adding fine grit to the epoxy will cause the grit to serve as teeth to help the adhesion; this is part of the process to get copper to stick to exotic PCB material.

[glicos]

I think epoxy is the best in bonding this material..

[Alex Eisenhut]

I think epoxy is the best in bonding this material..

*Which* epoxy? Delrin has very low surface energy, nothing actually sticks to it, you have to prepare the surface with chromic acid.

https://en.wikipedia.org/wiki/Polyoxymethylene

[glicos]

I think epoxy is the best in bonding this material..

*Which* epoxy? Delrin has very low surface energy, nothing actually sticks to it, you have to prepare the surface with chromic acid.

https://en.wikipedia.org/wiki/Polyoxymethylene

Yes, i forgot to mention that it needs some preparation of the surface. Use sandpaper on the surface before using any adhesive or epoxy.

For the epoxy, i forgot the name of the epoxy we are using..

[Alex Eisenhut]

Well it was a miserable failure. The insert popped out. This epoxy is kind of soft and I guess it doesn't bind well to metal, hence the insert popping.

I did some reading about Delrin and found a chart of what solvents attack it, and I found a plastic pipe primer that has methyl ethyl ketone in it. That should soften it up and help the surface energy that regular epoxy needs.

They also say anise oil is very reactive towards Delrin, very odd. Got to try that!

[ChunkyPastaSauce]

Chromic acid will definitely etch the Delrin. Nasty stuff though. 3M recommends it to prep delrin for structural adhesives and lists the chemicals to make it (2) http://multimedia.3m.com/mws/media/933332O/surface-prep-pretreatment-for-structural-adhesive-techbulletin.pdf
Ben Krasnow made some and etched, looked like a good etch \

[ChunkyPastaSauce]

How about this? http://www.ebay.co.uk/itm/Tektronix-311-0283-00-Potentiometer-270-ohm-With-Shaft-NOS-/141654719569

[SeanB]

If it is a setscrew simply make a collar that fits around the whole thing, with a tapped hole for the setscrew, then use an activator on the roughened plastic to mechanically hold it there while you screw in the setscrew. The collar applies pressure all round the plastic part and the setscrew in the hole stops the collar falling off. Easy to make using some thin steel brazed to a squared off nut.

[Gyro]

That's what I said back in Feb. 

[Alex Eisenhut]

That's what I wanted to do as well, but how? The space is very limited due to the pot's position in the chassis. Then it needs to be tapped, which IME with so little material and hand-tapping always ends up a bit wobbly so it will fail to grip the shaft properly too.

Not just that, but you still have the Delrin gluing problem, since the pot not only rotates, but slides back and forth; the collar still needs to be glued to the body of the pot.

It's a nightmare. If the plugin had only one channel I'd have put it back in the bottom of the closet but since the other three channels work, my OCD must find a solution. Even though I rarely use variable volts/cm and never the invert... 

ChunkyPastaSauce: Thanks but I need a different pot. I wish I could just whip out some weak Canadian dollars and buy the replacement pot!

[Gyro]

Hard to tell how much space without seeing it in situ, I was thinking of a simple brass (or steel) collar with a tapped hole to take a screw of similar diameter to the original. It's the sort of thing that a friendly local with a small lathe would be able to turn up really quickly.

[ChunkyPastaSauce]

If you provide clear dimensions or CAD, I'll print you a PCABS or fiberglass filled PETG part. You pay shipping.

[Alex Eisenhut]

If you provide clear dimensions or CAD, I'll print you a PCABS or fiberglass filled PETG part. You pay shipping.

Quite generous, thanks. I'll keep it in mind, I want to try another glue approach. I've got some MEK here and lots of brass inserts.

It's amazing the mechanical complexity that Tektronix engaged in; a simple square hole in the pot and a square section in the control shaft would have been simpler, and still allow the shaft to slide back and forth, without needing a set screw threaded in plastic...

But you do get a very nice tactile feedback with this system, as well as positive engagement.

[Alex Eisenhut]

I think this part is really Nylon. After looking at some chemical compatibility charts, it makes sense since the part resisted CA glue, hot glue, MEK, dichloromethane, and anise oil.

I'm going to try some glass etching cream which is supposed to etch both plastics...

[ChunkyPastaSauce]

How old is the scope it came from? Nylon was invented in the 30's. Delrin was invented much later in the 50's, and probably not really available until the 60s-70s. Id guess you wouldn't see it in a scope designed in the 60s, maybe not 70s

[Alex Eisenhut]

So while I was waiting to receive all the parts, I decided to put the anise oil in the humidifer, it has a cup for scented oils... I figured I could make the place smell like black jelly beans.
Turns out warm anise oil eats up humidifier plastic quite nicely over a few days. 
The top of the humidifier is a fused mess now.

All this because I wanted to repair channel 3 on a plug-in. 

[KL27x]

I don't know how brittle is delrin. I use a bit of HDPE for odd parts. The bad is that it is not possible to glue. The good is that HDPE is fairly elastic and very tough. So my first thought is to make an interference fit. I know Delrin/acetal is not as elastic, and also it's hard to see how small, that part actually is, so take this with a grain of salt:

E.g., take the insert and spin it in a drill while cutting a series of extremely shallow rings into it with a dremel cutting disc. Almost just distress the surface. Drill the hole with the next smaller bit. File/dremel the hole and chamfer the end of the insert until you can just tap/press the insert in without forcing it too badly.

If you get it just right, the bond will be plenty strong. But if you get it wrong you'll crack the part, of course. If it were HDPE, anyway, you would possibly have a very hard time to get the insert back out if you wanted to. Even as slippery as it is.

For this purpose, perhaps I'd grind 2-3 deep and wide longitudinal channels on the insert, too. To prevent it from spinning, of course.

You can possibly file out the hole (to clean it out) and fill it in with a piece of compatible plastic and a soldering iron, and try again.

[Alex Eisenhut]

I don't know how brittle is delrin. I use a bit of HDPE for odd parts. The bad is that it is not possible to glue. The good is that HDPE is fairly elastic and very tough. So my first thought is to make an interference fit. I know Delrin/acetal is not as elastic, and also it's hard to see how small, that part actually is, so take this with a grain of salt:

E.g., take the insert and spin it in a drill while cutting a series of extremely shallow rings into it with a dremel cutting disc. Almost just distress the surface. Drill the hole with the next smaller bit. File/dremel the hole and chamfer the end of the insert until you can just tap/press the insert in without forcing it too badly.

If you get it just right, the bond will be plenty strong. But if you get it wrong you'll crack the part, of course. If it were HDPE, anyway, you would possibly have a very hard time to get the insert back out if you wanted to. Even as slippery as it is.

For this purpose, perhaps I'd grind 2-3 deep and wide longitudinal channels on the insert, too. To prevent it from spinning, of course.

You can possibly file out the hole (to clean it out) and fill it in with a piece of compatible plastic and a soldering iron, and try again.

That's what I did the second time, I used long set screws as a shaft, put the insert in my drill and used my XActo razor saw to put in some grooves.
It didn't help against the glue blob unsticking from the Delrin under the force of the set screw.
I got some cone-point set screws and this will be the next try. I'm going to clean the pot again, use regular JB Weld epoxy.
I think the cone-points will help.
Of course, the new set screws have slightly larger hex sockets and all my 0.050 hex keys are rounded out. So I ordered some new 0.050" and 1.3mm hex keys.

Unfortunately the new parts I ordered seem stuck in Canada Post limbo.