First problem: To make headpins with perfect balled ends, I used sterling silver and later fine silver. I kept the wire perpendicular to the flame, either in the hot flame tip, or at the end of the large flame. The result with the sterling was balls full of craters and marks, even after trying to do it slowly, in the lower-temperature flame area, or in the hot tip of the flame. With fine silver I managed to get rid of most of craters, but the balls are not perfectly round. If I add flux over fine silver, I get extremely ugly and deformed balls. My flux is made of 50/50 borax acid and salt (plus some water).
The second problem (the bigger) with balls: I use a firebrick for soldering. The thing is that when I want to make a silver ball, the ball sticks to the firebrick and takes its shape. I used both sterling and fine silver and all balls get the flat part underneath, right where they sat on the firebrick. How can I get rid of that flat part and make silver balls perfectly round?
The trick is to get the the metal to cool steadily from the inside out by moving it very slowly out of the flame and keeping it steady.
Flux coating sterling stops the copper from oxidising and helps the silver form balls from the surface tension in the molten silver (the stiff copper oxides that form without flux spoil the ball). I don't know what you mean by "salt" in your flux mix. I never use table salt (sodium chloride) in flux - it reacts with the silver and the copper at high temperatures and can result in pitting. You can use commercial fluxes, borax or boric acid.
You do not need flux on pure silver when balling but is does cool quicker than sterling so you have to be extra slow when pulling the flame away.
When making balls that are not connected to a wire, much depends on the size of the ball. Above about 3mm the surface tension in the molten silver is insufficient to counter gravity and it will slump into a flattened oval.
Very small balls (1mm or less) are easy to form in little pits in a carbon block but most of the balls I form are in small numbers and most easily formed in a scorifier or crucible with plenty of borax - the balls tend to float a little in the molten borax and have no flat. It is essential to remove the balls AFTER the silver has solidified but BEFORE the borax has solidified.
Forming the balls on a soldering brick is quite possible, with a generous coating of flux, but you tend to end up with a ball glued to a bit of brick by flux. This comes off easily in the pickle but make a bit of a mess!
All that said... I keep a stock of argentium wire specifically for forming balls - it cools at a rate which avoids most pitting and doesn't need flux.
I tried to move the wire/ball slowly from the flame, but there was no difference. I tried that several times, with larger or smaller flame, etc.
I was using flux made from boric acid (those crystals) and water, but yesterday I added table salt because I found this recipe in several places on the internet and wanted to see the difference.
Unfortunately, I'm a beginner and I only have a firebrick and no charcoal block or others yet.
When you say the ball glues to a bit of brick by flux, do you mean it will still be perfectly round, without a flat area? I will try to make smaller balls of fine silver and see if it'll work.
I love Argentium; I read about it and my target is to make jewelry from Argentium, before going on gold, but now I need to practice with sterling, because it's cheaper and easier to find where I live.
I sympathise with the frustration in changing the speed and getting the same results - it really is a very delicate thing and on a bad day I struggle to get it to work on sterling. If you (very carefully) get a close look at the silver as it solidifies in the flame you can just see when the pit forms and see the transition that need to be slowed down (usually the bit where it passes just out of the pale blue cone).
The flux that is most traditional for silver working is very cheap and easy to get hold - household borax. This is sold in hardware stores for laundry as a powder. You can mix it to a paste with water and paint it on, or sprinkle a pinch of dry powder over the silver on your block before balling.
Boric acid isn't very soluble in cool water and I tend only to use it in methylated spirit as a firestain preventer. It will work as a flux though it tends to be very viscous. I have never heard of its use with table salt. The chlorine in table salt reacts with both copper and silver and used to be used to purify gold. I would avoid this.
The thing about balling sterling is not only to get a good cover of flux, but to get enough flux under the ball to form a cushion and this does seem to help stop flats forming on small balls. Maybe take a small pinch of flux, say 5 times the volume of the silver you are melting, and melt it onto your block in a corner - it will form a glassy area. Then add your silver, sprinkle a little more flux on top and try balling it - the ball will float slightly in the molten borax.
When it cools, the flux will be a hard, glassy material and the ball will be firmly adhered to the soldering block by it. You can then warm the flux with your torch until only the flux melts and pick the ball out of the sticky flux with tweezers or a soldering pick. For me, it always comes away with a bit of the block.
The part of your block with the flux on will then be a bit of a nuisance for soldering later on - tending to stick-down other pieces you are working on, so choose a corner you can reserve for this purpose.
None of this will help for balls larger than about 3mm - they will ALWAYS have a flat bottom; but sometimes that is just what is needed to attach it to a piece.
I threw my flux away and prepared the mix with water only (boric acid and water). I tried with 0.5 mm, 1 mm and 1.5 mm balls and still have the same result. I will try with adding more flux under the ball and form the "cushion" before adding silver.
It didn't work with the "cushion" either. I placed the deformed ball (about 2 mm) on the white "cushion" formed by the flux after heated, and the mix turned into that sticky paste and protected the ball without allowing me to melt it, to form the ball. The ball just got even more deformed.
I wonder how they make those perfect balls on jewelry items sold at jewelry shops...
Ah, that's a hint perhaps... I suppose your torch may not be powerful enough to heat the flux and silver to the melting point. I did just nip down to my bench and found my little "crème brûlée" torch could only just about manage 2mm and took a minute or two to get there.
It's the Max Flame Butane Torch used for PMC. I can melt silver with it on the firebrick within seconds but not when I have tones of flux on the metal. That's the scope of the borax flux, to cover the piece so you can't melt it while the solder flows.
No, the flux is there to coat the metal so the copper in the sterling won't react with the air. The solder won't stick to the copper oxides so is needed for soldering; likewise, the ball surface won't be smooth if covered in oxides (basically rust!).
Yes, you will have difficulty getting both a dollop of flux and silver up to the melting point with that (otherwise excellent) little torch. It's a matter of you having to get more material up to temperature.
However, nipping down to my bench again, I can get about 1.5mm balls reasonably round with just a small painting of flux using the smaller torch similar to the one you have. There is still a small flat area, but that is usually quite handy when soldering balls onto plate.
Don't forget that commercially produced jewellery will use ready-made granules produced using complex machines that drop the silver into a melting tube. These are hard to source in small quantities.
There is a method I've heard of producing random sized balls by pouring molten silver onto a cool surface (http://ganoksin.com/blog/lillianjones/2011/07/06/making-granules-the-sandkuhler-splash/), but have no experience in it.
Otherwise, do consider getting hold of a small amount of argentium which is a breeze to use.