hardest plastic for joining axles?

[PGE_Modeller]

Max,

In case you do not already have it, here is a link to the 2mm Scale Association's "suggested chassis layout" for a split-frame 2mm scale locomotive:
http://www.2mm.org.uk/standards/locochassis.htm

Edit: I forgot to point out that, of course, all dimensions given are in mm.

The dimensions will need to be adjusted for 9 mm gauge (instead of 9.42 mm), but the design has been well proven - there are a lot of 2mm scale locomotives running that have been built to this basic design.  Most of the 2mm crowd seem to use Araldite ( a UK adhesive ) to hold the half axles in place in the muff.  Not shown in the drawing is a small hole that is generally drilled into the centre of the muff to allow excess glue to escape as the axles are pressed home.  Loctite would appear to be an appropriate North American substitute for Araldite.

Watching this ( and the subsequent build ) thread with interest!!

Cheers,

[mmagliaro]

I set up an experiment where I put a styrene tube over a steel axle, and then a brass tube over that.
I did this because I will probably build my axles that way (with a plastic tube between two half-axles,
and then a coupler of brass collars ^*1   or a whole brass tube over the plastic tube
to make sure it never cracks.

I secured them all over each other with that Loctite 680 Retaining Compound, and let it dry for 2 hours.
I with a firm grip, I was able to twist them apart.

Then I checked the data sheet and the stuff supposedly takes a big jump in bonding strength about
the 6 hours mark, so I cleaned the parts and tried again.  I'll wait overnight this time and see how it goes.

Results and photos when the next test is done.

^*1  Cracked axle tube brass collars pioneered by Victor Miranda   

[nscaleSPF2]

Uh, Max, I feel partially responsible here.  The Loctite 680 is not recommended for plastic (according to the Henkel website).  Loctite 648 (the one that I used and originally recommended) can be used for both metal and plastic.

[mmagliaro]

Not to worry, Jim.
In the overnight test, the results were much better.   The only way I could twist the steel
axle out of the plastic tube was with two pairs of pliers, and even then, it was
tough.   So I'd call that a success.

Henkel often says their products are "not recommended on plastic", but I have yet to see
any of the Loctite blues, reds, greens do anything untoward to plastic (except just not stick to it.).

I'll see if I can scare up some 648 for testing.
I need to test this on acetal/Delrin plastic anyway (not the styrene I used for this test).

[mmagliaro]

In a follow-up test, I put piece of brass tubing over a steel axle and secured it with 680.
(I am testing how that will hold the brass wheel onto the end of the axle).

As I expected, Loctite with metal-to-metal requires dynamite to get it off.
I could not get the axle out, or even budge it inside the brass tubing, even wrenching it with two
pairs of pliers.  I had to punch and destroy the brass tubing to get it off.
That's what I want to see.   When I put those wheels on the ends of the axles,
I don't want them to ever give way.

[narrowminded]

Loctite is fine on Delrin.  It is NOT fine on ABS.  It acts as a catalyst and embrittles it glass hard. Any impact at all or even the preload of a loctited bolt securing an ABS part and the ABS material in immediate contact as well as surrounding area will flake off in shards like chipped glass, including the shine.  The effect takes several hours to appear and there is no reversing the action.  Loctite has a chart somewhere that a search should find that advises which plastics are or aren't compatible.  It always bugged me that their warning reads to the effect, "may attack some plastics" which to me suggests it would dissolve it or soften it as solvents do.  In fact, those materials that are effected are actually embrittled and the effect takes some time, hours, to appear and even then is only realized with a load.  It should read "may embrittle some plastics" mainly because that's what it does.

[narrowminded]

And loctite will cure if at least one side is metal.  It can slow the curing time but it will eventually cure.  Searches should find more specific data on that. 

[mmagliaro]

Good info, guys.  I think I have only ever used it on styene and Delrin (or "acetal").

I ordered some Delrin rod that I am going to drill out to 1.5mm to make my tubes, so that should be fine.

Don't worry, I am going to conduct tests on it with a steel axle and a brass over-sleeve to make sure it
holds and doesn't get brittle, soft, or otherwise disintegrate from the Loctite.

[rrjim1]

680 and 648 are the same except the 680 is thick and for loose fitting part, (stronger). 648 is really thin and made for tight fitting parts, (weaker). On the Henkel website it states "LOCTITE® Retaining Compounds are designed for bonding cylindrical metal components. They are also used to assemble cylindrical metal and plastic component combinations. Plastic components need to be tested to ensure retaining compound or activator does not stress crack the component".

[mmagliaro]

680 and 648 are the same except the 680 is thick and for loose fitting part, (stronger). 648 is really thin and made for tight fitting parts, (weaker). On the Henkel website it states "LOCTITE® Retaining Compounds are designed for bonding cylindrical metal components. They are also used to assemble cylindrical metal and plastic component combinations. Plastic components need to be tested to ensure retaining compound or activator does not stress crack the component".

Correct.  I found their guide for choosing a retaining compound.  It also says that the products cure in the absence
of air and in the presence of metal ions.  If at least one of the two surfaces to be joined is metal, it will cure.
It is slower to cure if the metal doesn't oxidize well (like stainless steel).

I suspect that since Loctite works by expanding when it's in the joint, the expansion will crack the plastic if the plastic
can't take it.   In my case, since I will have a metal axle inside a plastic tube, surrounded by a metal sleeve,
there is really no danger of the plastic stress cracking.   (All the same, I'm not using ABS.  I'm using Delrin).

[narrowminded]

What about just a bored delrin sleeve that fits the axles well.  If keying seems important enough, put a small flat on each axle end where they tend to butt. Filed would be sufficient. Then at final assembly, a small amount of epoxy in the bore that flows around the flatted parts of each axle affording a glued but also effectively a cast in place, keyed engagement, axle to axle.  Done and no play.  Worst case and the axle needed disassembled someday the delrin could be cut away and the axles separated.  A cross drilled vent hole in the center of the sleeve would afford an escape for the excess epoxy or might be used as an injection hole to inject the epoxy into the sleeve cavity after assembly and alignment is completed. Better chance to assure assembly accuracy before finalizing.  An approach something like that seems to me like a better chance at success with basic tools without risking the parts with huge numbers of hours in them.  Just a thought.

[mmagliaro]

Filing a flat on the steel axles won't help unless the inside of the slippery Delrin tube also has a flat
in it.  The epoxy will stick fast to the metal axle, but would still happily slip right around inside the plastic
tube.   

I am going to bore out a Delrin rod, and the fit should be snug, but I still want to put brass collars around
the Delrin because of what we have seen happen with other plastic axle tubes that crack under the stress
of the metal axle.  If the tube is the right thickness, and the pressure of insertion is right, it all works fine.
But who knows what those exact specs are?  And I'd rather not take the chance.  Wrapping it in
a brass tube makes sure that the plastic can never crack, and never stretch or expand over time.

[peteski]

Filing a flat on the steel axles won't help unless the inside of the slippery Delrin tube also has a flat
in it.  The epoxy will stick fast to the metal axle, but would still happily slip right around inside the plastic
tube.   

I think the epoxy is supposed to form a solid plug filling all the gaps inside the Delrin tube. So the epoxy plug will fill the voids where the flats are filed in the half-axles and also the gap between the axles.  While it is an interesting idea I'm not sure how much shear strength an epoxy plug would have (since the only areas which would positively lock the axles would be the small pieces which fill the flat axle areas).  Those seem like they could easily be sheared by twisting the axles.

[Lemosteam]

Max, I am concerned about one thing:

Accurately centering a hole in a solid delrin rod seems near impossible without some excessive runout.  I would be surprised if there is not extruded tubing with a 1.5mm or 0.625" ID.

Peteski, while dlerin is slippery, it is still porous.  The entire surface area of the bond inside the tube will be adhered to because it will enter those pores.  The only load that willl ever be applied is the force of the worm against the worm gear.   The chances of that load overcoming the bond is small even with the driver in a locked position.    On the track it would be impossible, there simply is not enough friction at the tread.

Max, will you be soldering a worm gear onto the outer brass tube?

[narrowminded]

To elaborate just a little, the injection method might be a little more fool proof.  Less chance for trapped air bubbles, even with a centered vent hole with its inherent hit or miss opportunities for air entrapment where the flat transitions to the full round.  The injection method would afford a flow through, self bleeding action which assures a greater chance of success first time, every time.  What that might look like is two staggered holes, one each drilled from opposite sides of the sleeve, through to center axle hole, not all of the way through the part, and with  the access holes aimed directly at the axle flats when the axles are installed, positioned with the axle flats 180 degrees opposite, flats directly facing each access hole located directly above the location where the axle flat transitions to the axle diameter.  This would afford a bleeding effect as the epoxy flows in at one flat, through the cavity including the center gap, and then escaping out the other hole positioned above the second axle flat.  This minimizes the possibility of trapping air in the cavity as it is really flow through and self bleeding with just a small amount of excess material pushed through to assure this effect.  The injection tools might be made with the micro tubing that is available as standard stuff.  Very small tube sizes that could be inserted in each hole, almost like needles, and at a length that affords working room and controls the excess epoxy in a position away from the part so you aren't left with a solidly glued up glob of mess. Those sleeves might be solvent cleaned out after each use or just treated as sacrificial, one use, disposable tools. Much cleaner method I think.

As far as strength, the ultimate cast in place shape with tabs and such would be stoutly girdled by the sleeve it is inserted in and in a sense, even though the epoxy will adhere to the sleeve at least minimally, it is not required to do so as the epoxy becomes a one piece, zero clearance, drive dog engaging axle to axle.  While I THINK the strength would be more than adequate, it would be easy to make a few test pieces with just the axle material and some delrin to make some test pieces to prove the process and ultimately for destructive testing.  I think it will hold long after some very nice wheels are destroyed.   And to add one more option if test pieces were tried, I would try just CA with accelerator.  It's more brittle but again, being girdled, might be more than enough and might be easier to use.  The tests would be the final proof.