Another Fisheye Story

Fisheye rears its ugly head, setting off another round of troubleshooting. February 8, 2008

I've been spraying my own work for approximately 3 years now using primarily Enduro poly and acrylic from Compliant Spray Systems and now General Finishes. Recently (8 months) I added Target Ultraseal to my mix. My typical regime is 2 coats of Ultraseal shellac, hand sand with 320 grit, one coat of Enduro, sand again, topcoat Enduro.

I've been accustomed to decent results and I'm always experimenting with coat thickness. My current week is throwing me for a loop, though. Fisheye trouble. First on a khaya project we built. Craters of .125 - .25" were visible when I laid down the last coat. The whole surface looked like the finish didn't want to lie down, particularly around the open grain.

I was reading other posts here and at Target coatings and thinking surface or product contamination. So I switched from Klingspor silicon carbide stearated 320 grit to 3M 216u 320. I bought two new parker filters - a coalescing filter and a particulate and absorber filter. I've flushed my hoses, guns, and started with fresh product. Today I'm spraying cherry, and getting better but still bad results. The craters are small - 1/16" - but on one piece in particular, everywhere. I notice on the cherry (relatively closed grain) that the freshly sprayed surface doesn't really level. If I look to catch a reflection, I see the surface looks kind of scaly, for lack of a better descriptor. It stays that way, too, rather than leveling out. Of course, this shrinks enough that you can't really see it once it dries. What does it all mean? How much could temperature be a factor? It's been 55 - 65 in my shop this past week.

I'm also spraying glass which I haven't done before. When I spray the topcoat on a clean piece of glass, I see specks of something and the finish is thinned out around the speck, kind of like the 1/16" craters I was seeing on the cherry. So that is contamination, right? Where could it be coming from? By the way, I'm cleaning the glass with alcohol and water. Does this seem definitive?

Forum Responses
(Finishing Forum)
From contributor A:
Several thoughts... Are you thinning the finish with water? Adding water to WB finishes can inhibit their ability to wet a surface and flow out due to water's high surface tension.
How thick of a wet film are you spraying? Most WB finishes don't like to be applied too thickly. Keep each coat to 2 - 3 mils wet film.

How many coats are you spraying in a day? Many WB manufacturers recommend no more than 3 coats in a day. If it's cold, spray fewer coats. The cool temperature may be affecting your results. Check with your supplier, but I think at 55 - 65 F, you're at the lower limit (particularly 55 degrees) where you can expect the finish to dry properly in a reasonable amount of time. I don't like to spray WB at temps less than 65 F.

Another unrelated consideration... Shellac is a hard, brittle finish, and if applied too thickly under other topcoats, can eventually cause them to fail. I don't like to use more than one coat of a 1 lb cut under WB and solvent-based topcoats.

From contributor R:
Check the date of the product. Some waterbornes don't have a very long shelf life. I had some go bad, and all it wanted to do was fisheye.

From contributor D:
The fact that the finish is cratering around the pores makes me think that the finish is just not flowing into them. I often get this with mahogany, and I just coat until it's filled. Is this wood oily at all? You could wipe the bare wood with acetone before you seal. The specks in the finish are weird. Did you ask the manufacturer about them?

From contributor X:
Causes of fisheyes in WB coatings include:
Old finish
Aluminum contamination
Other metal oxide contamination
Silicon sterated sandpaper
Incompatible coating
Wax, oil or silicon contamination

If you only notice it when spraying the finish coat, then try spraying 2-3 coats of just the finish on a scrap of wood. If it fisheyes on the third coat, it's the finish and you can send it off to the recycle center.

When spraying onto glass, you should not notice any contamination if you properly filter your material. The materials should always be filtered before going into the gun.

Never dump old finish back into the main container of finish. Always pour old finish (the stuff that's in the gun, pot and line) into a separate container or you could ruin your finish if there's contamination somewhere.

If the Enduro is the culprit, then you have some old finish on your hands and should send it back to where you bought it if it's recently purchased.

From contributor G:
The specks on the glass sound like the flattening isn't fully mixed in. Try filtering with a cloth filter, which will be finer than the standard paper paint filter. Also try pouring some finish on the glass to eliminate the gun as the problem.

From contributor W:
Contact General Finishes to see what they recommend about the water reduction and percentages of water to use with their products. Contrary to what contributor A states, WB's can be reduced with rather high percentages of water and will still form a decent film without fisheyes. You might see spidering and edge crawl, but fisheye is a sign of surface contamination or a problem with the finish formulation itself.

Secondly, as contributor D correctly states, the effect on the end-grain is called pin-holing and WB's will do this on mahogany, oak and walnut if not pre-sealed with fresh shellac. I see that you are using the Target UltraSeal; try 2 coats at 2-3mils wet and sand back with 320 aluminum oxide paper, then apply your WB topcoat.

Lastly, your temperature of 55-65F is slightly on the low side but not enough to cause severe film formation problems. If you stated 40-45F, then I'd agree. Again, check with General Finishes to see what they recommend for this specific formula. I've sprayed certain WB finishes at temps as low as 40F, maintaining a film thickness of 2 mils, and only see a slow dry cycle, without crazing or checking. Some formulas will vary on minimum temp.

P.S. Contributor A, didn't mean to single you out on the water reduction statement. Just wanted to clarify that some finishes are more/less sensitive to water reduction than others.

From contributor A:
No offense taken. I mentioned my concern about thinning WB finishes because some folks think they can be thinned just like solvent-based finishes, and are confused when the thinned finish doesn't flow out better than unthinned finish (unlike most solvent-based finishes), or worse yet, creeps, etc. The chemistry of WB and WB-hybrid finishes is more complex (emulsion/suspension vs. solution) than their solvent-based counterparts (as you likely already know), and many don't respond well to indiscriminate thinning. Temperature also seems to have a more profound effect on both unthinned and thinned WB finishes regarding their ability to wet surfaces and flow out well. I generally don't like to thin WB finishes more than 5 - 10% - unless the manufacturer's instructions say otherwise.

I thought it odd that, according to the original poster, the fisheyes didn't show up until the final coat, and that something other than contamination might be the culprit.

From contributor P:
When I've had fisheye problems, they occurred when spraying the second or third coats. Check the archives. I think you'll find this is the typical experience. I agree that it's probably not contamination. I've long suspected that is has something to do with shelf life or the recoat window. Like you said, WB finishes are complex. Maybe components that are held in suspension are released as the product cures (or ages in the can), and cause problems with surface tension? Or it could be that fisheyes are just more likely to form on the slick surface of a topcoat versus bare wood or primer.

It would be nice if someone from one of the WB product manufacturers would weigh in on this issue. I've written them directly, and haven't received an answer that didn't sound like it came from the marketing department. (Have not written General Finishes, though, because I have not experienced fisheyes with their products.)

From contributor A:
I wasn't implying that fisheyes are expected to turn up in the first coat. I've only experienced one episode of fisheye that revealed itself in the first coat of finish - and that was using pre-cat lacquer. When I've experienced fisheye problems, they have most often raised their ugly heads (eyes?) shortly after applying the second coat of finish, and are just one or two occurrences on a surface, or are randomly spaced over a surface. I usually apply quite thin first coats (1 - 2 mils), unless it's CV.

Incidentally, I think WD-40, otherwise a good product, has been responsible for some cases of fisheyes in finishes I've applied. I try not to use it around any machinery where it can come into contact with wood to be finished.

I agree with other posters that this could possibly be a case of old finish, or finish that's been stored in conditions that are either too cold or too hot, and as a result, may be partly polymerizing or beginning to separate out in the can. The WB products I've used, with one exception, don't seem to have shelf lives as long as solvent-based finishes, and to me, are more sensitive to their storage conditions.

To me, WB finishes are still somewhat of a "black box," and I don't understand causes and effects with them like I do solvent-based finishes (I use them about 25% of the time).

From contributor O:
I have experienced exactly the same problem on a few occasions with different manufacturer's waterbased satin clear lacquers. Looks like a bunch of zits over the surface: it definitely wasn't the classic fisheye. Close inspection revealed a core of very fine powdery material, presumably flatting compound. Even got them when spraying on glass, and a couple when pouring the offending finish on glass.

Things I tried were really cleaning out my Kremlin AAA system as I thought the flatting agent may be left over from the Durovar I spray a lot of. As none of the cans were freshly opened, I took them to my local dealer for a twirl on the shaker rather than to try to get everything into suspension by stirring. To get the viscosity lower, rather than add water or reducer/flow enhancer, I immersed the cans in hot water for a while before stirring and spraying; finally, I double filtered before applying each coat.

My conclusions were that while each action seemed to help reduce the problem, nothing actually totally cured it. I still got the odd zit on horizontal surfaces, while I didn't notice anything on the verticals on a desk that I was doing. In total frustration, I combined all the steps and sprayed the desktop vertical. This took a lot of time, and isn't something that I expect to have to do every time I want to spray something. It was okay, but there were still signs that the finish was trying to crater around a central bump in a few places (client was getting a glass top, so it wasn't critical for this project).

Contacting the respective manufacturers was not productive, so I resolved to go back to solvent based, as this was too much of a pain and took way too much time. I still have no idea what caused this problem, but I suspect that it had something to do with letting the can sit too long and not being able to get the flatting agents back into the proper suspended state. I always used to turn my cans of WB finishes every couple weeks or so, but again, that's something extra that I don't think I should have to do.

I realize this doesn't give a solution, but I wanted to let you know you're not alone in your frustration.

From contributor M:
I have had the exact same experience with the Enduro, and I stopped using it altogether. It looks like little specs of sand in the finish. I switched to Campbell's tinted lacquers and have had much better success. As much as I liked the finish that the Enduro was, it was just too many problems for me. I hope you will find out the problem soon!