Fish-Eye Problems in a Table Refinishing Job

Here's a detailed and comprehensive thread about the causes of fisheye and the options for fixing it, from simple to advanced.June 18, 2013

I'm thinking this piece is a lost cause, but here's the story. As a favor to one of my cabinetmaker customers, I accepted a job to refinish his 50 year old walnut coffee table that had water damage from plant watering.

I stripped it, rubbed with a Scotchbrite and xylene followed by a Scotchbrite and lacquer thinner wipe, let it sit overnight, then washed it with oxalic acid and good rinse of clear water. Let it sit for a couple of days, then sanded it with the orbital with 120, 150, 180 and a hand sand of 180, then stained it to match the base. Then (this is where I probably messed up) I shot Valspar 60% self-seal pre-cat on with no sealer while I was doing an office desk and book case set that I sealed the day before. I didn't look at it... sprayed it and placed it on the drying rack. When it came time to sand, I noticed a few small fish eyes and hand sanded like hell and sprayed on another coat and it now has fish eyes in the hundreds if not over a thousand, on a 2' x 5' table. What do I do?

P.S. I finish new cabinets for a living and from what I've read, smoothie is not going into any of my spray equipment.

Forum Responses
(Finishing Forum)
From contributor C:
Well, lots of mistakes for sure. You will have to remove the present finish and start from scratch. Depending where you stripped and sanded, you may already have contaminated the shop by introducing sanding particles of siliconized wood into the environment. Only time will tell if it's a problem. If the sanding was not done in the area where you're finishing, probably not, but no guarantees.

When re-stripping, after the new finish is removed, you really need to use oxalic acid followed by TSP to neutralize the acid and then a few washes with clean water. Before that, you need to remove the silicone contamination with 50/50 xylol and toluol using fresh rags each time and one wipe only per section, otherwise you're just going to spread it around instead of getting rid of it. Several times or more is usually necessary for this.

Another way I hate (but that is affective) is to get a tub big enough to hold more than half the top and scrub it with ammonia, non-sudsing type. This will act as an emulsifier on the silicone oils, so they can then be washed away by running water, much like soaps gather oil off your hands and hold them in suspension until you run them under clean water when washing.

After you have removed as much of the oil as possible, apply shellac, which will prevent the oil from bleeding through to your preferred top coat. It has to be applied properly in order for it to work as well as it can.

I use a 1/4 lb cut for the first coat applied in a mist coat that dries in 30 minutes or less. When dry, I do not sand it. The upper surface of the lac is what is needed not to be disturbed. In order to lock the oil down, a second coat of 1/2 lb cut is applied, and when dry, a 1 lb cut. None of these coats are sanded and applied thinly.

After a few hours or more, a final coat of 2 lb lac is applied. Then sand with fine paper (400/600) till smooth and ready the surface for the coatings to follow.

Most likely at 50 years old, the finish was nitrocellulose lacquer. I would use the same material, only using a bar top lacquer to better protect it from further water and/or alcohol damage. It will also knit perfectly to the shellac barrier coating. It should be applied in a dry coat for the first couple of coats. Wait till dry to also form a solid barrier of lacquer that will help the coating from possibly fish eyeing, which can still happen if you were careless with the previous shellac steps outlined. How would I know that?:)

At this point, there should be no evidence of even the most minute circles/holes which would indicate there is still an oil problem. If this is the case, unfortunately, you will have no choice but to add smoothie to the lacquer and go forward.

Not much will be needed, though, if the problems are tiny in size, maybe several drops. Make sure to do this away from your other work, outside if possible, and that includes all prep work. I would recommend a cheap gun like Astros to dedicate to siliconized finishes so that if ever this arises again you will have it on hand.

If you have to spray in your booth, I recommend spraying the top as close to the filters as possible and changing the filters after the coatings have dried hard. I would then advise you to put a fresh coat of booth-coat over the entire booth to seal in any possible contamination. Overkill? I'll let your future non siliconed work answer that question for you - it is always truthful.

From contributor M:
The prep job you did was outstanding. I do not see how the silicone could have come from the original finish. That leads to another question. If the fish eye is not showing up on everything you have sprayed since then, the problem was with the Valspar and whatever you thinned it down with. I am sure you figured this out already. Getting rid of it is another problem.

While I respect contributor C's assessment, I would never go to all that time and trouble. I could not afford to be down that long. I would replace the hoses if I had to. Hell, I would make a new top as well. That's just me. Time is money. You know you have silicone contamination. You just have to find what works best for you.

From contributor R:
Contributor C's answer is no doubt more technically correct than mine. But for whatever it's worth, we have had success with spraying very dry coats over the stain - 2 or 3 where it's just misted on. The droplets should not flow out/flow together. Let each coat dry thoroughly. The theory is that the finish dries before anything has a chance to migrate. It will feel a bit pebbly. No sanding between coats. Then after the surface is sealed, one flow-coat that barely flows out, but does, so you can start to level the finish. Maybe, maybe, at this point you can do a very, very light sand to somewhat level the finish. Remember, you're trapping the contaminants in. Don't do anything to break through the barriers you've sprayed. And, again, this coat needs to be very dry before spraying again. Perhaps one more coat like this if you need it. Otherwise, one good flow coat and you're done.

From contributor E:
Then again you could always add Smoothie to get rid of the fish eye.

From contributor L:
Spray vinyl sealer on it and then finish as usual.

From contributor R:
I've had little to no success with the silicon additives. I've had some success with vinyl sealer, but not nearly as much as with the method I described. But it may be the nut behind the trigger.

From contributor I:
I understood years ago, and it is said to be a folktale by some, that fish eye eliminator is just more silicone. Use it and all your equipment is contaminated. Maybe true, maybe not. Just use caution and talk to your supplier.

From contributor J:
That's a lot of prep. I'd do the heavy sanding (which you now shouldn't need) before doing the rinses. Then rinse and only lightly sand at whatever final grit to knock down the grain raise so as to not expose the deeper non-purged wood. In place of ammonia, I'd use vegetable oil soap (potassium oleate) as a surfactant - won't darken the wood nor mess with the PH. SealCoat (dewaxed shellac) over the stain, 2 even coats. Vinyl sealer or precat sealer piss coat over unsanded shellac (important to not get solvent burn through the shellac), then fuller coat, then sand, etc. If fisheyes still show up in some areas, the sand plus multiple dry/mist method mentioned earlier does work to blend them in (think 5 or 6 coats with the fluid so choked that you barely see the solvent flare on the surface - and let it dry hard between coats so the wet coat doesn't burn through).

From contributor T:
I'd never second guess contributor C! The procedures described by contributors R and O have been successful in my work 99% of the time. I treat silicone additives like hazardous materials in my shop and only use them as a last resort. Contamination risk from overspray is real, as is risk from handling the product and anything it comes in contact with, including your hands. Before I strip any table top and many other items, they are scrubbed with detergent, rinsed, wiped and dried with mineral spirits, maybe alcohol too. Any silicone contamination (or other unwanted residues) that are on the finish when remover is applied will amalgamate with the dissolved finish/stripper slurry and transfer to the wood surface during the stripping rinsing process. The pre-strip cleaning has proven effective for me in greatly limiting the fish-eye incidence once the coloring and finishing steps are begun.

From Paul Snyder, forum technical advisor:
This is a fairly easy fix. Fish eyes are caused by a contaminant with low surface tension, often silicone oil found in furniture polish. To prevent the fisheyes, you just have to lower the surface tension of the finish you're using. There are a few companies that make additives that perform this function; I use one called Smoothie and it works great. Just add it to your finish and spray per usual. It usually takes 1-3 squirts from the bottle to get the finish to flow out without the fish eyes.

In this case, you already have fish eyes in the surface of the finish. The way to fix the problem is to sand the finish back as far as you can without cutting through and getting into the stain. Then add some Smoothie to your finish and spray additional coat(s) over the leveled finish. You may have to perform this sanding/leveling 2-3 times if you sprayed the finish heavy and have deep craters.

Fish eye eliminator like Smoothie is silicone oil. I've been using it for years and have never had a problem with contamination of my sanding area, spray booth, or spray equipment. I do throw away the sandpaper I use between coats of finish and clean the spray equipment well with lacquer thinner or acetone after use (like always).

From contributor I:
Silicone is a standard paint additive in many coatings already. They use it in metallic paints, for instance, to make the flakes stand out more. I have used Smoothie many times. In the old days of lacquer we used to squirt it in everything! I went through the panic with everyone else for a while, but have never really seen any contamination after using it.

I like using a 2k isolante coat myself to seal off the offending contamination and then if that doesn't work, it gets a shot of smoothie through my gravity cup. (I use my Sata and just clean it well after use.)

From contributor U:
If you're asked to change the color of a kitchen and top coat with conversion varnish, would it be best to start with smoothie? And after a lacquer cleaning/sanding process, add it to both dye stain and CV?

From contributor I:
I would never use it unless I had a contamination problem and then only if a 2k isolante didn't take care of it. I assume you are adding a binder to your stain if going over an existing finish, so you would add smoothie in that case. Also once you use smoothie, every coat after that requires using the same amount.

From contributor A:
The modern silicones do not resemble the older ones. Smoothie has been the professional's choice since 1951. It is not the same stuff.

Paul, do you have a Smoothie gun? I would be afraid of future contamination. Sometimes regular finishes can be difficult to clean.

I would think the newer Sealcoat shellac would be ideal as a barrier coat for these old mystery pieces covered with lard and/or Pledge. A few unsanded coats should bury most contaminants.

From Paul Snyder, forum technical advisor:
One of the main points I was hoping to make is that the fear of contamination is exaggerated/unfounded. I have been using Smoothie for years and have never experienced any sign of a contamination problem that it caused. I've used it in the spray gun, pressure pot, and pump. A normal clean up using lacquer thinner eliminates the residue.

When refinishing furniture and cabinetry, I think it's wise to use Smoothie every time. It doesn't cause any harm but it does prevent fisheyes which are a pain to fix once you have them. I have tried various contamination removal processes including a flow over bath of wax and grease remover solvents with limited success. I've had the same results using shellac (limited success).

If you're using 2K poly, the isolante for oily woods will seal the contamination. I have used it only once on a refinish job, and it worked well.

From contributor U:
Paul, thank you. What are your steps for refinishing old lacquer kitchen cabinets with conversion varnish? Just looking for the most efficient path.

From contributor L:
You would have to strip the cabinets. You cannot put a conversion varnish over lacquer. The rule of thumb is you cannot put a harder finish over a softer one. CV is harder than lacquer so you would need to remove the lacquer before you sprayed the CV.

From Paul Snyder, forum technical advisor:
Remove the existing finish as described at the link below:

Related article:Stripping Furniture

Give the stripped items a good wash down with generous amounts of lacquer thinner and clean rags. Let the stripped items sit for a day.

Apply the new finish using the steps appropriate to get the color and effect you want. When using pre-cat and CV on bath and kitchen cabinetry, I prefer to use vinyl sealer over the stain and glaze coat(s). With CV, the vinyl sealer needs to be catalyzed to avoid wrinkling.

Add Smoothie to the vinyl sealer and CV topcoats.

If you get spots that don't want to dry (sticky), don't panic. It's caused by the wax from the stripper migrating to the surface of the finish. Give the spots a really light wipe with a clean cloth and naphtha and they will dry pretty quickly. Otherwise, it will take days for them to dry on their own.

From contributor G:
I just got done refinishing a 15x20 butcher block center island and it also had two matching tables on the side. The side tables when sprayed would fish eye. I had done everything that contributor C mentioned. I tried a trick that I use in stone restoration to remove stains (this is called a poultice). All I did was take some blue shop towels and poured some lacquer thinner on it and sure enough, within a couple of minutes, I could see the contamination bleeding up into the towels. Re-shot it (for the 4th time) and it was $$$.