Knob casting experiments

[kellys_eye]

When using the jig to hold the original knob in the alginate for casting, could you not then re-insert a slightly wider shaft in the jig and place it into the mould centre and, after pouring the resin, remove it to leave a 'ready-centred' hole for the bushing?

As much as I love your execution of the method, having to resort to a lathe to complete the project must make it a lot more complicated for many potential knob-makers, although there's no denying the professional finish.

Lovely work. Thanks for sharing.

[Aerodyne]

Great thread, David. It echoes much of the method I used some years ago - I too built a 'jig' similar but not as well-made as yours - to hold the knob. As Kelly's Eye suggests above, I did indeed replace the knob spindle with (in my case, a quarter-inch) shaft - I had no access to a lathe for ferrule making - but the problem comes when attempting to withdraw the shaft from the casting. Despite coating it with release compounds of various kinds, this proved to be a sticky (literally!) problem. A lathe certainly would have been a boon.
Simple knobs are not too difficult to reproduce but older types with large flanges tend to damage the mould when removal of the original is attempted.
Alginate is easy to use. There are other stronger compounds available but these tend to be expensive in use.

Tony

[David G4EBT]

Originally Posted by kellys_eye

When using the jig to hold the original knob in the alginate for casting, could you not then re-insert a slightly wider shaft in the jig and place it into the mould centre and, after pouring the resin, remove it to leave a 'ready-centred' hole for the bushing?

As much as I love your execution of the method, having to resort to a lathe to complete the project must make it a lot more complicated for many potential knob-makers, although there's no denying the professional finish.

Lovely work. Thanks for sharing.

Yes, that's a possibility.

Initially, when I made the jig, what I had in mind was to first make the Alginate mould, then remove the cup, fit a brass ferrule on the shaft in place of the knob, fill the hollow in the alginate with the wood-filler, place the cup back into the jig and lower the brass ferrule on the shaft. By the time I'd done that, the wood-filler was already setting. Hence, I resorted to simply filling the cup with wood-filler, allowing it to set, then when set, putting it on the lathe to face-off the rear of the knob and mill out the hole for the ferrule.

After the ferrule has been glued in, the knob has to be held in a vice on the pillar drill table, drilled through the side and tapped 4BA for the grub screw.

I accept of course that not everyone has a lathe of any sort, let alone a woodturning and a metalworking lathe. Hence, much of what I've written in this thread will be of limited use, (as was the case with my recent thread on turning replica knobs in wood, which this thread follows on from).

I'd also say that there are clearly better materials from which to make the mould and cast the knobs. Silicone would be a better medium for making the mould as once set, it's stable, where alginate is a mixture of powder and water, and within maybe two hours, starts to shrink as it dries out. Hence, if say a set of three or four knobs is required, they'd have to be cast within perhaps half an hour or they'd be of slightly different sizes.

Furthermore, though the wood-filler and epoxy 'Plastic Metal' have made perfectly acceptable knobs, that is not their intended use, and I've no doubt that there are more suitable resin-based casting materials available. For the most part, the sort of radios I tend to restore are of low value, with no headroom for spending much in the way of materials, so I just used what was to hand, for a low-cost option. I had all of the bits and pieces to make the jig - the only thing I had to buy was the alginate.

Mounting the knob on the lathe to mill out the 12mm hole for the brass ferrule does have the merit of ensuring that the 1/4" hole for the control shaft to which the knob is to be fitted, will be perfectly concentric and perpendicular. Just 1mm out will give a swing of 2mm when the knob is on the shaft and rotated. Hence, if a lathe isn't available, whatever means of drilling the hole and mounting the shaft is chosen, it must meet that requirement. Next to impossible to do that on say a pillar drill.

I have to admit that it all sounds a frightful and labour intensive rigmarole, but if a knob or knobs is called for, which are unlikely to be found by any other means, it's one solution for anyone who has a lathe, the time, inclination and skill. That is until we have simple to use 3-D printers which accepts voice commands such as 'ere - make another just like this one' and dutifully complies. For now, I just know that were I rash enough to buy a 3-D printer, it would swallow me whole for breakfast, along with a book entitled '3-D printers for Dummies' if there is such a book. In my 79th year, still full of enthusiasm, I find myself positioned on a spectrum between 'You can't teach and old dog new tricks' and 'you're never too old to learn'. I strive towards the later, but drift towards the former!

Thanks for the interest shown by those who have contributed to the thread, and for their helpful input and kind comments.

[dseymo1]

Having encountered a similar need to drill-out knobs (I have a quantity of excellent knobs with non-standard small bores), I simply secured a rod in the existing hole, similar to that used in David's casting technique, and chucked that in the press drill before securing the knob in the machine vice. I then removed the rod from both the knob and the chuck, and replaced it with a drill. The results, while not perfect, were sufficiently concentric for practical use.

[nebogipfel]

As always your posts are fascinating David. Such attention to detail.

From what I've seen I doubt if 3D printed knobs would be any better than your moulded ones. To my eye 3D printed objects, certainly on hobby printers have a certain look and are not finished items. Having never done it, the thought of having to design objects first in 3D software puts me off the whole process.

[David G4EBT]

I managed to finish off two knobs today that I've cast as replicas for a Portadyne 'Princess' TRF, from an original knob. I had some brown aerosol paint which was a good match for the original Bakelite knob. I've attached a pic below, with the replicas either side of the original that was used for making the mould. They look a respectable match, even at close quarters.

As I said earlier, this was a low budget operation to make a few knobs for sets of little value, using materials and workshop equipment that were to hand.

If I were to make any more, I'd probably investigate using silicon rubber to create the moulds rather than alginate, which begins to shrink within an hour or two, so the mould can't be retained for further use. Also, though the wood-filler and 'Super Steel' two-part epoxy made perfectly acceptable replicas, I've no doubt that proper casting resins would be superior - after all, they're designed for that purpose.

As to 3-D printing, really, setting aside cost considerations, there are two aspects to consider. Firstly, the texture of the finished knob compared to the original, and secondly, the colour match. I've not actually seen any knobs made using that technique - I've only seen other objects and even when a fine nozzle has been used on professional equipment costing thousands, the surface of the objects has invariably not been smooth or shiny. Also, while there is a fairly wide range of filament colours available, I think it would be difficult to get an exact colour match to original knobs - brown Bakelite for example.

I guess the acid test would be whether it's possible to replicate something basic like a Lego brick so that the replica - in terms of texture and colour - is all but indiscernible from the original? I suspect not. It might perform like a Lego brick, dimensionally identical, but it won't quite look like one, and neither do I think would a 3-D printed DAC90A knob for example, (if ever there were such a thing look like an original, quite apart from any cost considerations). There might well be a whole range of knobs which - because of their texture and colour - do indeed lend themselves to 3-D printing - I'm not well enough informed to know one way or another, and 'driving' 3-D printing seems so involved that it becomes a hobby in its own right, and with a steep learning curve at that.

Maybe a few forum members have asked Santa one, and if so, perhaps they can update us in the New Year!

[McMurdo]

I've been watching this progress! If anyone can master it, you can.

There's a company called ABL Stevens who supply DIY casting materials for reproducing more or less anything. They have a comprehensive 'how to' section which might make for interesting reading.

I have some end-caps for an aluminium extrusion that I may need to reproduce in the future (in the day job) so I'm doubly interested.

http://www.resin-supplies.co.uk/information_pages.htm

[David G4EBT]

Some interesting and informative info there Kevin.

I notice that when using a non-porous specimen (such as a radio knob) to create a silicon rubber mould, you don't immerse the object in the liquid silicon rubber as you would do a porous object (in their example, a wooden door knob). Instead, you paint thin coats of silicon rubber onto the object, let the rubber air dry, then paint on successive coats till the mould is robust enough to use, then you peel it off the specimen. It's stated that shrinkage is about 10 percent. That may not matter for creating some craft items, but it would in cases where accuracy is called for.

As to the casting resin, I see that they sell filler powders, so maybe the resin can be coloured to be a close match to the original. I wonder if anyone on the forum has cast knobs which look like authentic replicas of Bakelite ones?

[McMurdo]

I have to say I've never moulded anything in my life, but that silicone putty looks interesting. You mix it, then mould it around the sample and let it go off. You can then 'pop' the master out and use the resulting mould for casting.

Somewhere there's a demo of someone casting a precision plastic nut for a plumbing fitting.

[David G4EBT]

That sounds promising Kevin!

[McMurdo]

Found the link, here

http://www.resin-supplies.co.uk/help...tic%20part.doc