MIMF

I am thinking about a guitar in which I use turquoise and silver in the rosette and purfling.
I don't think there would be any new issues in cutting a groove and fitting the parts.
I do think I would need to use epoxy to get a bond with the metal, but that's OK.

The thing that worries me is bringing the metal down level with the soundboard and
then polishing it afterwards. Silver is soft enough that I can probably still use a scraper to
get it down to level, though I will need to try that to be sure. But polishing?
I am worried about getting metal dust embedded in the soundboard when I polish
the silver to a shine. Ordinarily, I use little abrasive wheels on my dremel, but I'm
worried they would throw metal dust onto the soundboard.

Any tricks of the trade for me, folks?

-Doug Shaker

Aluminum looks like silver but it is softer so it is easier to work and level. Also it won't tarnish.

Be careful of getting the metal hot during sanding level because the glue will loose grip and piece will pop out. DAMHIKT

Actually, aluminum does oxidize, really rapidly, but the oxide is very hard and mostly transparent. It is a very thin, tough, layer that seals the aluminium from further tarnish. The aluminum then takes that matte look that you see in lots of bicycle parts. But that's a real option - thanks.

Rio Grande makes an alloy of silver that they call Argentium silver. It contains germanium in sufficient quantity to inhibit most of the silver oxidation. It isn't perfect, though, and eventually it will tarnish some. I was planning on sealing it under a thin layer of epoxy, but maybe that's just delaying the inevitable.

If you use 'fine' silver it will resist tarnishing pretty well. I have used sterling and recommend you not use it but the fine (pure) silver is much better in that regard. As stated above, aluminum looks like silver under finish.

I wouldn't recommend scraping metal in the purfling as it can catch and pull out. To avoid a lot of damage I would recommend block sanding the metal to level it, but work it down slowly so it doesn't get too hot, as previously mentioned. You can scrape clean the surrounding wood surfaces afterward but use a sharp scraper, did I mention use a sharp scraper? Practice, practice, practice!

Thanks, I appreciate the advice.

I use "dead-soft" half-round silver wire for vines on fretboards with great success. But, as said earlier, silver conducts heat so well that it pops out of the epoxy when sanding--a bit of CA and all is well (if I can find the piece).

One thing I forgot to mention is the surrounding wood will sand away much easier than the stone, metal, glue composite. That is a really good reason to use a very nice flat block for doing this. Also I suggest getting a roll of self adhesive abrasive and stick that to your block. This is because if you use ordinary sand paper it can flex and move, and sand deeper than the surface of the block. The adhesive stuff stays stuck to the block and the abrasive surface stays flat, which leaves you with a flatter surface instead of a rippled one. You can put a piece of tape around one end of the block so that end of the abrasive is held above the surface of the area you are sanding, that way you get good and level sanding from one end of the block but none from the other end. A little practice will show you what you need to do.

Good suggestions! I appreciate the advice. Thanks.

I've used copper and second all of the recommendations already presented.

Update:

Barry Daniels wrote: Be careful of getting the metal hot during sanding level because the glue will loose grip and piece will pop out. DAMHIKT

Barry, you are so right. I was thickness sanding my rosette yesterday and had forgotten your advice. Sure enough, the metal got so hot the epoxy melted, the metal expanded out of the channel and was sanded to death before I figured it out. Sigh. If I had figured it out one pass earlier, I would have been able to let it cool and CA'd it back in place.

Now I am trying to figure out if this metal purling idea is worth the effort or not. That is, I am trying to decide if I am acquiring valuable experience in an interesting technique or wasting time on a bad idea.

I gave up on it years ago except for small dots and circles.

I used brass filing to fill the gaps between irregular pieces of shell. Result was really nice, used super glue nd light belt sanding, no problems.

If the design you are contemplating would allow it, at least with a rosette, you can build the rosette in a trough cut out in polyethylene 9think white plastic cutting board). As long as you cut the trough in the top at the same time as in the top, the two will be identical.

You can then flood the rosette with CA, pop it out, sand it down thinner than the trough cut in the top, and fit it into the top's trough perfectly. If you then flood this with thin CA until it mounds higher than the surface of the top, you can sand the CA down level. This will leave the metal submerged below the surface of the CA and you will have sealed it away from the air and corrosion, and also not have to sand two greatly different hardnesses when you sand the top. If you sand down all the grits with Micromesh, it leaves a glass clear surface.

If this isn't making sense to you, I can post a picture of an example

Randy, please read your first paragraph and if it makes sense to you then explain it to me. I get the idea of making the rosette in a poly trough, but especially the last sentence in the first paragraph needs some clarification or restatement. Thanks.

Michael,
Sorry it's unclear. We got hit with golfball size hail (car in driveway) and I didn't get back in until edit expired.
Set circle cutter for inside radius and cut both the poly and the top. Set cutter for outside radius, and cut top and poly. Hog out between inner and outer, in poly and top.

By making the cuts in both materials before you change the set up of your circle cutter, each trough is identical, including depth. Sanding down the constructed rosette while it's in the poly, or popped out but before gluing into the top, will leave the rosette thinner than the trough in the top, leaving space above it to fill with the thin CA

Thanks Randy. I can understand that.

Randy's idea is the way to go, at least the part about the channel being a little deep, so that when you're doing the final leveling it's all CA that you're sanding.
I've been doing all my inlay that way for a while, after being forced into doing it by the constraints of a project. It was a 5 string bass with block position markers, so the inlay had to be thicker in the middle due to the fretboard radius. The pearl I had was not thick enough, so I inlaid the pearl a little deeper and covered it with CA. Leveling was easier, the color/pattern of the shell didn't change from being sanded/scraped, and it ended up looking better. There is a difference in the way shell looks when it's covered by CA, so if some of the shell is covered by CA, all of it should be. I'm not saying it looks better one way or the other, but some one way and some the other looks like a mistake.

Best to set any inlays a few thou below the top and cover with CA in my opinon. Won't sand through the top layer of shell if it's ablam or destroy the markings of the abalone if they are used as engraved lines and protect metal if used. Although it takes more percision when setting the cavity but if it was easy anyone could do it.