dealing with drips and fish-eye with WB finishes

[Chuck Morrison]

I've been experimenting with water borne finishes for several decades. After KTM9 tragically disappeared I tried several alternatives and have for now settled on the crystalac brite tone finish, although I find their pore filler to be largely useless. Over the years I've tried Miniwax, Delft, General finishes, Varithane, Bullseye and I don't know how many others. They all seem to have had the same general working issues.

One thing that all of these finishes seem to have in common is that they are a bit thick initially. Since I no longer have space for a spray booth, nor can I afford to heat/humidify/dehumidify a steady stream of outside air, I brush it on with foam brushes. As guitars are 3 dimensional objects, there tends to be sagging, if not outright dripping somewhere that has to be dealt with. There are several factors influencing this issue. Temperature is the first. When they say 70 degrees F as a minimum temperature for application, they really mean it. Lower temps mean higher relative humidity and slower evaporation allowing more opportunity for the finish to flow downhill. Since viscosity is generally pretty thick, the film being applied is also thick. A thinner film will go on thinner if you load the brush sparingly, dry faster and run less, so I usually thin with about 30% by volume of distilled water. The thinner mixture flows better initially and keeps overlap ridging lower.

But the primary factor is gravity itself working on a wet flowing liquid. The only way I know of to mitigate gravity is to move the object so that gravity is always changing direction, mitigating the sagging/dripping. A number of years ago I came up with a jig that allows me to do this without having to manually hold/turn a wet instrument. I bought a cheap little rotisserie motor, put it on a mount I could clamp to a bench and made a few adapters that would let me attach necks and bodies. For the last 10 years or so I've built mostly with bolt on necks, so this fit well with my building style.

The little adapter jig fits onto a 1/4" square rod that fits into the shaft end of the motor. It has to be glued or physically attached or it will just slide out and stop turning. It's not a huge feat of engineering. I used pipe fittings, but a more resourceful person could come up with a more elegant solution.

Tangentially related is fish-eye, which is sort of like sagging. It's really a shame not to be able to use stearated sandpaper, which works so nicely with WB finishes, between coats. I can't seem to help myself and I'm always trying to sneak in a sanding with the wrong paper and paying the price. Since stearate is a water based soap (at least that's what I read) we should be able to wash it off before re-coating and avoid fish-eye. I have done this after the fact (sanded the fish-eye down and rinsed it) and had some success in reducing the effect dramatically. Since using stearated abrasives is such a time saver, it might be worth rinsing the whole thing down before recoating and just using the better paper. I also find that after rinsing, I get a lot less "dust" showing in the next coat, showing that all the dust isn't coming from the air.

Has anyone else tried this and what results did you get ?

[Bob Gramann]

I have been using Target EM 6000 waterbase for many years. Usually, I spray, but in the winter, when I don’t want to set up the spray booth indoors (vented out), I brush. I use a 1” wide sable brush and apply the finish as thin as I possibly can, thus no runs. It takes about as many coats as it does to spray it, 8-12. When it starts to get uneven, I level between coats. My results still are not quite as nice as a sprayed job.

I have tried really hard to keep silicone out of my shop. In the past, when I’ve had fisheyes appear, I have washed the area with alcohol and put on a shellac coat. Usually that removes the fisheye tendency. I have limited myself to the yellow sandpapers—either 3M or Mirka. If any of these are stearated, they haven’t caused me trouble. I have a tight nap carpet on my work surface to keep from scratching the guitar. I find that rubbing the sandpaper on the carpet removes the loading. I vacuum the guitar and the carpet often when I’m sanding finish. I use no water in the process. Water makes it harder to see the progress of my sanding.

[Steve Sawyer]

I did a test piece a while back using epoxy (System Three Silvertip) as a filler then 8 coats of Brite Tone with a really good brush. I applied all 8 coats then wet sanded and got beautiful results. I've seen folks here and elsewhere say that WB finishes should never be wet-sanded, so I'm not sure why I had good results, although I did let the finish cure for a long time (like a month) before sanding. Interested to know what your experiences have been as I'll be using this schedule on the current build within the next few weeks. I'm taking note of your techniques as my test piece was flat, and I didn't have to deal with gravity messing with the finish.

When trying to keep sandpaper clean, I'll brush it briskly with a brass brush (one of those that look like a toothbrush that you can buy in any hardware store or Harbor Freight), then a swipe with the brush attachment on my vacuum.

As to the fish-eye problem, Dan Erlewine's finishing book identifies grease, oil and wax as troublesome contaminants in addition to silicone. His recommendations for removal of contaminants are 1) Naphtha to remove wax, 2) mineral spirits to dissolve silicone to allow it to be mopped off and 3) ammonia and water to remove any silicone residue. He covers this in discussion of and treating of raw wood, but I've interpreted this to suggest that any contamination from non-loading sandpaper is to wipe down with naphtha first, followed by mineral spirits. I would then do a final wipe with naphtha to ensure that no residue from the mineral spirits remain, though that might be overkill.

I too won't have anything containing silicone in my shop, other than one small tube of silicone grease. So far, so good - haven't had a fisheye problem yet.

[Marshall Dixon]

I've used KTM and Target Coatings products. I like the EM6000. But I've always sprayed it. I've sprayed this stuff through a Badger air brush doing touch-up.I like the way it sprays right out of the can.They make a thinner that I've used, but only because the can was half full and been sitting for a long time. That worked fine too.

I get some drips and runs even when spraying. I've discovered these the next day at times and sliced them off with a bare safety razor, let dry another day and scrape them smooth with the razor/scotch tape method.

As far as sanding, I use appropriate grit wet/dry automotive paper with water as a lube. Keeping it really clean by having several pieces of the same grit soaking in a tub to swap out often.

[David King]

I'd have thought a 30% cut with water would be excessive but I've never tried the brite tone. Most WB's seem to suggest 10% or less. If Brite Tone have a retarder available that should help with the issues but the rotisserie is brilliant in the meanwhile.

[Steve Sawyer]

but the rotisserie is brilliant in the meanwhile.

+1

[Chuck Morrison]

Thanks for the comments, everyone.

Yes, I've seen the cautions about the amount of water to add, but in my experience it didn't really change the viscosity as much as I was hoping. I still have an issue with stiff surface tension, and adding water won't change that. At times this can result in nasty little bubbles being trapped. Brite Tone is one of the better ones for less bubbles, which is why I've been using it. I think temperature is a factor. I tend to keep the shop right around 70, which is at the low end of the recommended range. It seems to do better if the finish is a little bit warmer than that. Maybe I just need to work warmer.

Another option I've tried a few times is padding the finish. It avoids bubbles and sagging, but it requires a different approach than french polish, unfortunately.

[Brian Evans]

One thing to remember is that Brite-tone is a catalyzed finish, which means in cures in a couple of days, while EM6000 is non-catalyzed which means it dries but does not "cure". After Brite-tone is cured, it seemed to wet sand just fine. I read that EM6000 shouldn't be wet sanded so I used mineral oil thinned with naptha as a lubricant. I found that with Brite-tone you need to put subsequent coats on no more than a day apart to get the burn-in of the fresh coat.

[Steve Sawyer]

One thing to remember is that Brite-tone is a catalyzed finish, which means in cures in a couple of days, while EM6000 is non-catalyzed which means it dries but does not "cure". After Brite-tone is cured, it seemed to wet sand just fine. I read that EM6000 shouldn't be wet sanded so I used mineral oil thinned with naptha as a lubricant. I found that with Brite-tone you need to put subsequent coats on no more than a day apart to get the burn-in of the fresh coat.

So no more than one day between coats, no wet-sanding with less than a two-day cure time. I'm going to do one more test piece before starting on the guitar currently on the bench, and I'll try that schedule.

Sounds like you're up a creek if you get a sand-through then, which I did in one small spot on my test piece. That only had 8 coats so I was planning on doing 12 on the current build just to provide a little insurance. Never sanded through on my last (my first) build, but I'm chalking that up to beginner's luck!

[Chuck Morrison]

My experience with witness lines is that they tend to buff out to nothing when I get to a full on gloss. I haven't tried this with Brite tone yet, but I'll get to test that on this most recent build since I finished the top before I put the back on (including no binding yet on the top) so I could put the bridge on and mess with the insides while playing the instrument with the back clamped/taped on. I don't like removing the bridge for finishing and I hate buffing around the bridge. I seem to enjoy making life harder than it needs to be, but what's life without a challenge ?

The top was sanded and buffed out about 5 days ago after just a day and a half cure time, and it buffed out very nicely. It was already quite hard. I was going on an assumption of being able to work in a 5 day window. So yesterday I glued on the binding and cleaned it up today with minimal scratching to the top and I brushed on the first coats of sanding sealer on the back and sides. The plan is to brush the top coat full strength on just the binding and purfling of the top with minimal overlap, just enough to fill in some scratches from the final sanding. A little super fine sanding around the outside and a rebuff of the top should tell the tale. In theory I shouldn't get any witness lines.

[Chuck Morrison]

I have tried really hard to keep silicone out of my shop. In the past, when I’ve had fisheyes appear, I have washed the area with alcohol and put on a shellac coat. Usually that removes the fisheye tendency. I have limited myself to the yellow sandpapers—either 3M or Mirka. If any of these are stearated, they haven’t caused me trouble. I have a tight nap carpet on my work surface to keep from scratching the guitar. I find that rubbing the sandpaper on the carpet removes the loading. I vacuum the guitar and the carpet often when I’m sanding finish. I use no water in the process. Water makes it harder to see the progress of my sanding.

I've been doing a lot of checking over the last few days and it seems that all the Gold/yellow sandpapers are stearated. Mirka, 3M and Norton all are. MIrka claims a calcium stearate, which may be the same for the others I don't know. I know there are zinc and magnesium stearates.

In my experience fisheyes don't seem to be as much of a problem if these papers are used on the finish between coats. Having said that, I just brushed on a coat having used 3x on it a few minutes before and vacuumed off the dust prior to brushing with full strength Brite Tone. I see a weak fisheye forming, but it isn't all that bad. I think that brushing may show it better than spraying.

The 3m silicon carbide papers will guarantee severe fisheyes for me. Different type of stearate I guess. This was never a problem for me when I was spraying Nitro.

[Bob Gramann]

I just finished spraying 12 coats of em 6000 on two guitars. The only sandpapers I used from the shellac preparation coat through the end were the Mirka P220 and P 320 yellow “finishing” paper. I had absolutely no fisheye issues. Last winter, when I brushed one and sanded between coats with these papers, I had no problems. If these papers have stearates, then these stearates are not the problem, at least with Target EM 6000. Is it possible you have some contamination in your shop?

I’ve been using Mirka red hook-and-loop paper (from on my orbital sander before I get to finishing (and before the shellac undercoat and pumice pore fill). Again, no problems. I hope never to see it again, but I haven’t seen fisheye in my shop for a long time.

I will use a yellow 3M 216U P800 paper to level my finish before I polish in a couple of weeks. When I’ve had to recoat after using this paper, I’ve had no problems. After sanding between coats, I usually wipe down the surface with an alcohol-damped rag after vacuuming the dust and before spraying.

If I were experiencing your problems, I would thoroughly clean my shop and wipe down all of the surfaces with solvent damped clean rags, changing rag surfaces often so as not to spread around any contamination. All you have to do is touch something with some silicone in the dust and then touch your surface to be finished and then you have fisheye. It doesn’t take much.

If you solve the problem, please let us know how you did it.

[Chuck Morrison]

All good advice, and I have been cleaning up the shop as much as I can before brushing anything. However, I don't think that is the story. In this case I have one surface, a EI rosewood back, that is giving me fits. The Cedar top is no problem and the grenadillo sides are coating very nicely. The only thing I can think of is that I had put a few coats of sealer on the back a week or so ago and then today sanded it (with Norton 3x 220) before coating it with fresh Brite Tone. Aside from the fish eye issues, it has whitish splotches over much of it. So it's looking like either the rosewood is at issue or there is a problem with putting fresh Brite Tone over it's own cured sanding sealer. Since I used the same sandpaper to sand the sides, which have no issue (and were sealed last night) I'm guessing the issue here is timing. Don't let the sealer cure before putting on the the finish coats. I'll sand it all down and start over on the back since that blotchy color scheme isn't going to fly.

[Barry Daniels]

Just let it be known that the silicone carbide in the sandpaper is not the same chemical as silicone oil and it should not be the cause of fish eyes.

[Chuck Morrison]

True, but the grey silicon carbide papers tend to be coated with some form of stearate, which apparently can cause fish eye in varying degrees. It is used as a lubricant, keeping sanding dust from adhering to the sandpaper particles. The "silicon" in silicon carbide abrasives is silica, the mineral or element Si. From what I can see, it's virtually impossible to get open coat aluminum oxide (gold) or silicon carbide (grey) sandpaper that isn't coated with some form of stearate.

Silicone oil should probably be referred to as siloxane. It can cause fish eye and worse in Nitro and probably water borne finishes too. While some polishes have this as an ingredient (it's a lubricant too, among other things) I've noticed that some brands (like turtle wax) now specify that their polishing compounds do not contain it. I don't have any of this in the shop from long habit with nitro.

[Bob Gramann]

You might try dewaxed shellac instead of the sanding sealer and see if that makes a difference. The splotches would worry me.

[Chuck Morrison]

OK, I just put the last coat on the recalcitrant rosewood back. Since I had a previous coat of several weeks old sanding sealer and a few coats of top coat that had gone so poorly, I sanded all that off down to the wood and started over again. I used the crystalac sanding sealer again, used some LMI microbead filler this time. I did a light cleanup (of the filler) sanding with 400 grit wet/dry (done dry) then six coats of Brite tone. The results are dazzling. I will flatten it in a few days and buff it out in a week or two.

After I sanded it down to the bare wood, I put on an initial coat of sanding sealer, The first thing I noticed was that the few spots where I hadn't got all of the previous sanding sealer off were opaque white. So there's the first clue, Don't let the sanding sealer cure before putting anything else on it, more sealer or top coat. Especially when sanded, it doesn't really seem to adhere well and you can see the witness lines as blotches. Easy enough to find those spots and sand them out, then another coat of sealer. After that 3 full coats of top coat, sanded flat with 600 grit Abranet and another 3 coats. I think I'm done. I gave each coat 2 hours before re-coating except once overnight.

I didn't see any fish eye or blotching at all this time around and bubbles were minimal and leveled out without my help for the most part. The only other thing I didn't mention is that the last 3 coats of top coat were done with a new clean foam brush and diluted finish, but since the previous 3 coats looked great I don't think the old brush was a factor. I believe the problems stemmed from NOT doing everything in a couple days. Time is of the essence with this kind of finish, do it all quickly. In this case I only sanded once in the whole process, and that just because I had some brush ridge lines I wanted to flatten. The contamination I was fighting was the sanding sealer itself, fully cured. One of those "don't do that" moments.

Oh, about diluting it with water. It turns out that the viscosity of the finish is pretty much set by the water itself, so adding more water doesn't make it much thinner viscosity wise. The FAQ on the crystalac site says that it would take a huge amount of water to actually make it thinner. They do sell a prepared thinner that will reduce the viscosity for those who need it. Happily it's much cheaper than the finish itself.

Oh and one last point about the rotisserie. The only things I've found that it helps with are the sides and the neck, as some part of those are going to be vertical while drying. The top and back, if done separately, are best left horizontal to flow out. Since my tops and backs are fairly highly arched, that means the outer edges get a bit thicker build up than the centers. That's fine since that's where I'd usually sand through anyway.

[Bob Gramann]

Congratulations.