Troubleshooting a Peeling Waterborne Finish Job

Other Versions
Glaze and water-based topcoat peel off an old urethaned tabletop. Pros weigh in on the diagnosis. July 8, 2005

I've got a problem. I posted earlier about changing color on 50 year old urethane. I called Target and asked again. I must use waterbourne products because of fume issues for the customer. I told Target that I would scuff the urethane and shade with a mixture of em9300 and dyes from Sherwin Williams. After drying I applied a water based glaze, dry brushing it on. After drying (2 days) I applied a coat of em9300 and after 4 hours scuffed and removed dust with vacuum and applied another coat of em9300. It really looked wonderful and I even rubbed out another section of the project with even more wonderful results.

The problem began when I was removing the masking from the wall and pulled a piece of the topcoats with it. I went ahead and removed the whole top section because I knew I would have to start over. You can tell from the picture that it didn't adhere to the glaze but that the glaze, for the most part, adhered to the urethane. It is possible that the glaze is too thick, but I'm not certain of that. I removed the masking two days after spraying. How can I improve my process with waterbourne?

Click here for full size image

Forum Responses
(Finishing Forum)
From contributor N:
Did you use a seal coat between the original finish and the first step in your process? When doing that many things on top of an old finish, I would use a barrier coat of Zinsser seal coat unless you can do some testing for adhesion on a similar piece first, which is usually not possible.

From Paul Snyder, forum technical advisor:
That's way too much glaze. You can't put that much color between coats of finish because the binder in the glaze isn't strong enough and the finish will peel (just like your picture). From the picture, the glaze looks opaque. Are you going for a color change or a painted finish?

If you want a color change, and that much color, you'll get much better results by using a very thin glaze and then toner. Or skip the glaze altogether if you're not using it for highlighting (pores or accenting recesses/corners). To make the toner, you can use dyes and/or pigments. The pigments will give you more of a semi-transparent look while the dye will be completely transparent. Mixed in with the finish, the dyes and/or pigment will add the color you want while retaining the adhesion and durability properties of the finish.

From the original questioner:
Thanks for responding. I agree that the glaze doesn't have great binding properties. The purpose of the glaze was to add another dimension to the color. I shaded darker because of all the mars and discolored areas. I believe I could have thinned the glaze and used less. This may help, however I think that maybe the finish is not quite cured. It's only three days old. The em9300 is designed to be its own sealer. Would Zinsers be better or do you think it is a glazing problem?

From contributor D:
The problem may be twofold. First, you have the problem with adhesion. Second, you left a water based product sit for two days on top of masking materials. Water based finishes, be they opaque or clear, have a bridging characteristic. They will bridge a gap between the surface being finished and masking materials. If you pulled the masking immediately after spraying the finish, it would not have pulled. For some reason other finishes do not do this. When you let water based finished cure, as you did, you must break the bridging. Scoring with a utility knife at the seam usually does the trick. This would not have solved your adhesion issues, but it may have allowed the finish to cure and without the initial pull from the masking, you may not have created the problem.

From the original questioner:
It was actually a small child that pulled the tape off. Wow! Did I want to cry! Do you think that I need to worry about the rest of the project not adhering? I have more to do and am planning now to use less glaze. Would a good schedule be: Em9300 toner (em9300 and dye), two hours later apply thin glaze (water based), 2 hours later apply the first coat of em9300 clear, four hours later apply the second clear coat, then pull tape before someone else does?

From contributor S:
If you got your glaze from Target Coatings, then it is a product. Get on the phone with GoldenPaint and ask to speak to their tech support Mike Townsend. He can walk you through the process of using their glaze medium in a furniture application setting.

As for recommending sealing with shellac, since their is an odor/chemical sensitivity for the homeowner, how is the use of this possible? It is not. If you are using the product to create a barrier coat, then it must be sprayed. If you brush on a barrier coat in hopes that you are trying to seal in underlying contaminants, then you are not going to be effective because brushing the sealer will bring the contaminants to the surface of your barrier coat. Then you would be back to square one. And spraying is not an option because of the chemical odor (once again). The chemical odor issue is why waterbase is being used in the first place.

From contributor N:
I thought his exact words were "fume issues." I didn't read deep enough to pull out the chemical odor sensitivity. As far as pulling out contaminants by brushing, I didn't see where there were contaminants, but there were adhesion problems. My suggestion was that the shellac may stick to the existing finish better than the glaze did. Since the finish is fully cured, I saw no indication that the finish might migrate through the shellac. I may be wrong.

From contributor M:
Are you sure it's urethane on the original piece? Otto Bayer created urethane in the late '30's, and it was used primarily for industrial purposes until the mid '50s, when it became more widely used for retail end use protective coatings.

In any case, it's not compatible with your water base applications, so must be sealed in. That's the base of your problem, although both Paul and contributor D are right on target. Since you can't use shellac, this is a case for only one thing - you need a tie coat.

From the original questioner:
I feel confident that it is urethane and I am spraying everything with an HVLP cap spray. I didn't get the glaze from Target - I bought it from another shop and don't remember what brand. After reading your comments and reviewing the way it peeled up, I believe that it was the glaze. I was able to do 4 test shelves and they all are different but have all adhered quite well (not with as much glaze). As I mentioned earlier, I used a dry brush method to add depth and help even out the inconsistent colors in the cabinet of which the shelves didn't have. I may have a poor glaze product or have applied too much or have not applied the first coat in the window time of the toner. Isn't this a wonderful business?

Do you think I need to be concerned with the rest of the cabinet's finish or do you think time will cure it? I did rub out another area of the cabinet and was premature in doing so, but it rubbed out beautifully.

I understand the value of vinyl sealers and wished I could use them in this situation, but can't because of the odor problem. I have already thought of sending my customers away for awhile so I can use the vinyl sealer, but I want to make this waterbourne work. This wont be the last job I'll do under these same circumstances. Thank you for your experiences and knowledge.

From contributor W:
Have you looked into Target's waterborne dewaxed shellac? Could be a solution to the fume constraints.

From contributor S:
If the urethane is cured, then there are no compatibility issues with a waterborne topcoat. That is because the waterborne topcoat does not have enough chemical bite to disturb what is underneath. The waterborne topcoat relies on adhesion only by mechanical means, and that means scuffing the underlying coating.

I mentioned brushing because if someone were intent on using shellac and there is an odor issue, they may think that by brushing it on there is far less odor, and that is true. However, if there are surface contaminants and if they do need to be sealed in, then brushing is not the right way to apply sealer, in this case shellac.

Vinyl sealer is out because of the odor issue. The reason that vinyl sealer works so well as a washcoat/sealer prior to using fillers and glaze is that vinyl sealer wipes cleanly. It has a slickness to it that stearated sanding sealer does not have.

Regarding adhesion, shellac has great adhesion qualities because of the way that the shellac molecule can grip itself to the walls of the scuff ridges. This gripping is not chemical; it is electromagnetic and it occurs on a molecular level. It is the same stuff that causes static electricity (and the reason that we ground our metal cans of finish to prevent sparks).

That's how scuffing works - the molecules of finish lodge themselves in the sand scratches and they stick to the walls of the sand scratches with tenacity. That is also why powdered finish dust is difficult to remove from crevices - it has a magnetic grip you have to find a way of disturbing.

Was the original glaze solvent? Or was it waterbase?

I am no chemist, and I do not know chemistry. So if what I wrote needs correcting or expanding, go ahead. Write on.

From contributor J:
Split your toning and/or glazing up into two separate steps, thus reducing the amount of pigmented layer that will separate a good bonding process. Second, apply wet on wet coats and I'd switch over to a toner with the topcoat being its base. Sorry if I missed the reason you went with the glaze, but it didn't look as though you were trying to highlight the areas seen on the picture.

From the original questioner:
I want to thank all of you for your support, concern, and knowledge. It is good to have you here, bringing your expertise.

From contributor M:
Have to respectfully disagree with the compatibility issue, and the glaze not adhering is a bit of evidence. If vinyl is too much odor to bear, ML Campbell makes a waterbourne sealer called Utlrastar. Other companies do as well. There are also water bourne shellacs out now, although I haven't tried any.

From contributor M:
I hope I'm not overdoing this, but since this is one of the biggest problems those of us who do refinishing face, and it's immensely costly when it goes wrong, it's worth a few more words.

I spoke with a regular customer today who's a successful organic chemist, who is in our line of work sort of, and after a primer on molecules, etc., here's the essence of what he said.

First, he doubts the cabinet has urethane on it if it's 50 years old. But that is not really the issue.

Second, he says there is not a severe compatibility issue if it's urethane or lacquer with water base on top, but it is a compatibility issue if, like most cabinetry, it was maintained with wax polishes, and the water based glaze will not adhere if there is any residue on the piece.

Third, the other compatibility issue is the type of bond the urethane and water base undergo, and that is the mechanical one. To create a good mechanical bond (not magnetic bond) a thorough sanding needs to be done first.

Fourth, the mechanical bond is not very strong, not as strong as the magnetic bond which occurs between the water based glaze and water based top coat.

Fifth, because of the 4th reason, and the thickness of the glaze and its long cure time, it's quite possible when the top coat was added over the glaze that the glaze was not fully cured and a magnetic bond was formed between the glaze and topcoat. When that dried and the vehicle evaporated, it shrank, actually pulling the weaker glaze/urethane mechanical bond apart.

Still with me? Potential solutions:
1) Prewash the piece down better to remove potential incompatible products.
2) Sand better to create a better mechanical bond with the glaze.
3) Let the glaze cure longer before applying top coat.
4) Or apply the glaze and top coat one after the other with no cure time at all.

From contributor N:
That is about the most coherent thing I have heard in a long time. That picture looks like my spraying table in my booth. I can spray about 50 layers of wb urethanes and stains over each other repeatedly and after they build up to a certain point, I just peel the whole works off in one piece down to the phenolic top. The layers all adhere to each other without separation, just like in the picture, but not to the phenolic top. I especially like the part about the possibility of wax or polish on the top preventing the bond between the original finish and the applied coatings.

From the original questioner:
I'm impressed! Everyone is correct, for the most part. The photo shows the glaze coming up with the top coat in that area, an area where many hands have rested for many years. In the rest of the areas, the finish came up without the glaze (not shown in the photo). I could have improved on abrading and cleaning the area better, especially where the hands rested the most. I still feel that the finish didn't adhere as well as I thought it should've, but the rest has no sign of problems at this time. Today the finish has hardened more and I feel more confident of its success. It may have been the timing of it all. This is the first time that I have used waterbourne products. It's frustrating to learn how they behave, but this product shows great qualities. In some ways, it's the future of some finishes.