Troubleshooting Adhesion Failure in Conversion Varnish Over Lacquer
Conversion varnish needs a mechanical bond. Here, pros analyze a case where it's not sticking. May 17, 2005
Question
I have been doing wood finishing for a couple of years. I recently stained a maple staircase with Valspar lacquer stain in a dark cherry with several coats of stain. I let it dry overnight, then applied three coats of Valspar conversion varnish. The homeowner called me a week later and told me the finish was coming off. I went to the jobsite to take a look. 99% of the finish was bonding fine. The treads, risers, skirting, and newel post were fine. But on the handrail along the grooves, the finish, when scratched with a fingernail, was scratching down to bare wood. You really had to dig at it, though. What could I have done wrong? What are the guidelines for inspection of proper adhesion of a coating? I've been reading a lot regarding staining maple. I cannot figure out how to do a maple staircase with full treads, risers, skirting, balusters, and rail. I do just fine with oak.
Forum Responses
(Finishing Forum)
From contributor D:
Did you sand before applying your CV? If so, with what grit paper? CV requires a mechanical bond to adhere. It's possible that the open grain of the oak allowed for enough grip, but the closed grain of the maple did not allow the same grip and the failure was in the application process and not the finish itself.
From the original questioner:
Yes, I used 220 3M non-loading grit paper. The areas affected are all in the same area of the handrail and trim piece above the skirt. I tried to fix the spots once, but scratched off in the same areas. The Valspar rep told me I put too much stain on and the surface plasticized, causing poor adhesion, and that I should have done a sample first. I did. How can I fix the problem, and how do you do maple in a dark color with difficult angles?
From contributor B:
I think that rep is right. Too much stain. No adhesion bonding between stain and CV. The way to correct it, I don't think you want to hear.
From the original questioner:
What bothers me is I bought the Valspar material at Vista Paints. I told their lacquer expert what I was doing and how much stain I was putting on. He told me as long as I could sand the stain, the surface was okay to apply the CV. Well, I'm still learning, but at a high price.
From contributor D:
Did you apply the CV directly over the lacquer without an isolating coat of vinyl sealer? This may also be an issue. You should not apply a catalyzed finish over a non-catalyzed finish. Be thankful that it's only happening on the rail and not the rest of the surfaces.
You may be faced with having to strip the rails. Seeing as the finish is peeling anyway, may not be too tough a job. I would do a sample of your original schedule and materials, only apply a coat of vinyl sealer between your shading lacquer and CV. After allowing the sample to dry overnight, scratch and see if there is a better result.
From contributor J:
I think that even though you gave plenty of time for the stain to dry, it wasn't in these recesses, thus building a surface that CV could not hold onto. That is why I cringe every time I do a glazed kitchen or hear of one done without a cat vinyl barrier coat. Another thing that could happen is that the stain is wicking out of wood.
From the original questioner:
Thank you. The Valspar expert did not tell me to use a vinyl sealer between the lacquer stain and CV. I think the issue might be the stain quality, because I did the samples with the lacquer stain and CV topcoat and one with gel stain and CV topcoat. It checked fine, but now after the problem with the staircase, I came back to my shop and scratched hard at the same areas of the samples and sure enough, the lacquer stain came off in the same spots, but the gel stain sample did not. Could you tell me a little more about wicking of the wood? I've never heard that before.
What is the exact procedure for determining proper coating adhesion? Is there an industry standard? I had a gut feeling about this staircase. I have only used CV three other times and it was all oak. I had my concerns about lacquer stain. I should have stayed away. But how do you learn if you don't try?
I have a small residential company in San Diego. I've been painting for 20 years and have been doing wood finishing for two years. I'm not going to give up on wood finishing.
From contributor D:
I just re-read your posts. You mentioned speaking to the lacquer expert. I'm assuming this was someone in the store you purchased the products from. We have a lot of those experts (HA!) here also. I have found that people selling the products, claiming to be experts, are not. You need to use the products in real world scenarios, be confronted with situations and solve problems with chemicals with limitations, and get it right 99.9% of the time. That makes an expert, not reading a label or MSDS sheet or sitting in on a sales seminar.
From contributor M:
As for your question about coloring maple, I use dyes.
And a bit of advice: Get hooked up with a company with good tech support. Don't be afraid to question their reasoning. It doesn't take much to know more than some of these reps. They may be capable of fielding homeowner questions, but don't cut it when you have a technical question. In my experience, many of these guys have some experience in finishing and base their answers on their experience, not verifiable, technical information. I asked a rep about using shellac over one of their products. He said, "Yeah, sure, no problem." Then I said that I would be spraying several pieces and I did not want to be responsible for compatibility issues. Then he got serious and started making some calls. This "expert rep" has sold paint for 5-6 years and then watched someone spray wood finishes. In short, check out the background of your rep/tech.
I wouldn't have put CV over coats of lacquer, even with a vinyl sealer. Hard over soft, in my opinion, is not a good combination. I would have either shot the toner with clear, being careful not to exceed mil thickness; or shot dye, vinyl sealer, CV, or tinted some CV.
Also, I have had mixed results with Valspar CV. In my opinion, their viscosity is not consistent. While they do have some good products, their tech support stinks. Try calling corporate. It may take calls to 5 people to find out where the product is made, much less answer questions about it.
From the original questioner:
I'm now replacing the handrail, newels, and small trim above the skirt, and I'm considering replacing the whole thing. I can tell the owner feels the whole staircase is failing. It's not, but it's hard to argue with him while I'm standing there scraping finish off his handrail with my fingernail. If I keep going with this staircase, I'm losing money there and now I'm losing money on other jobs. I should just buy my way out. I did learn not to spray CV over lacquer stain. Won't do that one again. I would like to try a different method but I don't even know what dye stain is. I do know that most stains have pigment and dye. Where do I get dye stain, and how do I do the maple with confidence? The way I've been doing them, even when it was just regular nitro lacquer and lacquer stain, the maple staircases are just too difficult. I would be on my last coat of clear and bump the rail going all the way back to white maple. I thought I was going to burst a blood vessel. I think I will stay away from the maple, for now. Oak, cherry, mahogany, no problems. And I'm doing my first alder cabinets next month.
Maybe I'm using the wrong product to produce my color? Regardless of the binder to adhere it. I'm not very good with UTC tints. So I have two choices: oil stain or lacquer stain. As far as adding the color to the clear, I don't understand… How would you do a new maple staircase? The owner looked at a Minwax stain chart and picked red mahogany.
From contributor R:
I think you have 2 problems.
1. I agree with the rep - too much stain. Use one coat of stain only, wipe off real well and seal with a washcoat or full coat of vinyl sealer. Then you may apply a glaze or shading coat or both to achieve your final color.
2. 220 grit sandpaper is too fine for maple. Switch to 150 and you'll have better luck. You can do an adhesion test on a sample by cutting 5 or 6 parallel lines with a razor blade in a dried and cured sample and then 5 or 6 more at right angles to the first ones. Make sure you cat all the way into the substrate. Then stick on some really sticky tape, like packing tape, burnish it down good and then rip it off. Look at the grid you made. If there are several squares missing, you don't have good adhesion. If most or all of the squares are still there, you have a winner. They make a really cool gadget that does all the cutting for you at Paul Gardner Scientific, but this will work for now.
From the original questioner:
I've learned some things from a business perspective. It makes sense to pay to install a new staircase. If I have new handrail installed, I still have to do the finish work, and match the rest of the staircase. And I still don't have a new game plan. Why would they sell me lacquer stain and a CV topcoat if the CV won't bond?
Ok, so I should do the replacement handrail with the same technique, but after I spray the lacquer stain, I should spray a coat of vinyl sealer and then shoot the CV? What if I make a mistake again? I used the adhesion method mentioned earlier. The tape removed very little of the stain.
If I can sand down the areas that are failing, can I seal the rail with vinyl sealer and then stain with the lacquer stain, followed by another seal of vinyl and then CV? I did the sample in nitrocellulose lacquer over the lacquer stain and after, I looked at the owner's staircase handrail. I decided to check the sample I had in my shop. Oddly enough, it came off with a hard fingernail in the same spots as the CV. I don't understand that. And wouldn't the finish be failing all over the staircase, instead of exact spots on the handrail?
From contributor J:
I think I finally figured out what is going on here. Tell me you're not applying the stain as coats?! You had mentioned in your first entry that you had given the wood several coats - several being the key word. Any time you get up a layer of stain, that prevents the topcoat from biting into either the wood or a roughed up seal coat. You will have delamination sooner or later. Also, since you're putting the stain on heavy, then it would take more than 8-10 hours to dry completely, especially in crevices or details of the woodwork. How close am I?
You are right in thinking that it would be better to replace the handrail, newels and spindles if you can and then strip down the treads, risers and stringers. This would make the job a whole lot easier. I'd suggest you use a pre-catalyzed lacquer. This is so much easier than a CV is, plus it will melt into each coat, making the bond so much less work since, if you do miss scuffing all areas, it won't be a big deal. See if you can't make a sample using a dye stain first, then go over that with the lacquer based stain, only spraying it on and wiping it off. If you can find a spray, non-wiping dye stain, that is the way I'd go - don't worry if it looks a little blotchy, because the lacquer stain will even things out. You'll be fine coming in the next day to stain and you don't have to use a vinyl sealer, you can go with the self-seal route. Unfortunately, skipping not stripping some areas might keep bringing up this problem, so it's best to bite the bullet and just do it in one shot.
Just explain to the customer that you got bum info on the stain and that will at least relieve you of the pressure of a worrying customer always looking over your shoulder. I've been there too many times.
From the original questioner:
Several coats of lacquer stain, some not by design. I spoke to the owner this morning and scheduled one more attempt to fix the handrail. Do I have to continue using the same process, and just make sure I use 150 grit paper on the rail?
From contributor J:
Yeah, you might as well because switching now will only create more problems. Strip all effected parts off real good and sand with 150 and then a quick pass over with 180. Just be aware that your finish failure was due to the stain being too thick and/or not dry enough for the CV coat. That means you might have to back off on the color and make it up with a real light toner. Roughing up the green wood will only telegraph sanding and roughness through the finish. If you think you've got the guts, just remove the areas that are coming off, feather the CV existing finish into the raw wood, and hand apply the stain, feathering out the transition from new stain color onto the existing finish (you can cheat and blend more to not be noticeable). Seal, sand, wipe on more color if needed, then topcoat when dry. It would be advisable to add a little stain controller to the stripped areas if wiping on the stain.
From the original questioner:
How can I apply a lacquer stain by hand? Is that what you meant, or did you mean a different kind of stain? I figure it's not going to hurt to try one more time. I'll never make this mistake again. Got some alder cabinets coming up. I've never done alder before. Owner wants kinda like walnut distressed. Any ideas? I went to Sherwin Williams and they carry dye stain. The next maple staircase I get, I will put dye stain followed by cat vinyl toner, sealer and top it with the CV. What do you think?
From contributor S:
They real key to finding out what went wrong is to examine where the break in adhesion is taking place. If the break goes as far down as the bare wood, then the stain is not sticking to the wood's surface. A few things can cause that:
1. Sanding is too fine.
2. There was too much stain left on the surface.
3. The stain was not dry before the clear went on.
4. There was oil present on the wood's surface prior to staining.
From the original questioner:
I'm going to try to fix it tomorrow. Gotta give one more chance before I pay to have a new rail installed. I won't make that mistake again. Do you think I got a chance at a fix, or am I just prolonging the fact that I have an adhesion issue? I tried real hard to scrape that finish off.
From contributor J:
Any time you use a light colored wood and go straight to toner for evenness of tone, a dye stain has to be used to darken the wood so that nicks and devil's claw marks don't show up so bad! You said the coating was real hard scratching off, so that tells me the finish did its job the best it could. Another reason that the finish started to chip off was maybe that you sanded the wood too finely and the coating had nothing to adhere to - same thing goes with not scuffing in between dried coats of the CV. I also believe that you did not apply enough coats of CV to protect the color coats. Since the finish is intact on the rougher ogee part where your forearms usually rest, it's due to having something to grab a hold of to adhere.
Since you're this far, you might as well give touch-up a shot! Sand or feather out as much of the hard edges as you can without getting into the color coat. Once you have the feathering complete and cleaned from dust, seal the damage, repeatedly sanding in between to level surface or better yet just a hair below grade. Add colored toner with a touch-up brush until you think it could use one more coat to match (that last coat always will put you over the mark). Light sand and apply a couple topcoats to whole finished edge (scuff sanding first). Better yet, strip it and save yourself the frustration and time.
From the original questioner:
What sealer should I use, and what do you mean by colored toner?
From Paul Snyder, forum technical advisor:
It looks like all the color is in the topcoats and that's why the wood is bare when you scratch the finish off. Did you stain the bare wood and then topcoat with CV? If so, there's a problem with the stain; it shouldn't come off down to bare wood like that.
How did you prep sand? How did you apply the stain? Did you use a stain controller of some type? What do you mean by "lacquer stain?"
When you say "lacquer toner," it is my understanding that you are using a toner with nitrocellulose in the material. Nitrocellulose under a CV is a certain path to failure. I would have had you use a dye stain to achieve as much of the color as possible and then a wiping stain to get the rest. I think you should bite the bullet on this one and redo the job the right way for your sake. This will not go away! Even if you clean up the existing problems, it will keep failing as time goes on and your reputation is all you have.
Strip the pieces, sand them with 150 grit paper. Apply a red ngr stain, then apply a wiping stain to get the final color you want. Apply a self-seal CV to get the build you want and be done with it.
From contributor X:
I don't believe that to be a failure of the conversion varnish. If it was, the lacquer stain would have remained on the wood and just the top coat bubbled and chipped. What it looks like is the lacquer stain has failed, which makes more sense to me. The quick drying lq stains do not penetrate the fibers of the wood, as a penetrating stain does. Any decent sales rep will tell you that.
To get good adhesion, the wood should only be sanded to about 120 grit max. My guess is you sanded the heck out of it before lq stain application with 150 or 220, which made the wood too smooth to accept the product.
Lq stains are weak, in my opinion. Fortunately, because the lq stain doesn't penetrate the wood, stripping should be much easier.
From the original questioner:
Here's what I did. We sanded with 180 to 220 to remove fingerprints, pencil marks and any oil residue. Then dust vacuumed. I used Valpsar lacquer stain color red cherry. A couple of coats of that, then dry overnight. Then three coats of conversion varnish. I decided to replace the handrail and trim pieces above skirting - the only places that are failing. I think it was too much stain. What dye stain can I use on the new rail that will match Valspar's red cherry? I don't think lacquer stain is a very good product. Fine for shading, but not for the actual color.
Lacquer stain and conversion varnish? Why not? If I wanted to do a crackle finish with a conversion varnish topcoat, it's done with a lacquer system for the crackle. Then it's sealed with vinyl sealer (no lacquer content) and topcoated with conversion varnish.
What I am reading is that you sanded way too fine and the lacquer stain is letting go in places. 150 grit, garnet paper or open coat production on white wood. Then you apply the stain. Scuff 220 grit or even a 320 grit stearated sandpaper (3M's FreCut is a stearated sandpaper). That's enough finish tooth for the vinyl sealer to grip to (but it is too fine a tooth if the sealer needs to adhere to wood). Apply vinyl sealer, no more than 1/2 dry mil. Scuff 220 or 280 or 320 grit stearated sandpaper. Then shoot your conversion varnish.
Adhesion tests are best performed with a cross-hatch tester. The DIY method of carving the tic-tac-to criss-cross pattern is not accurate enough to take to the bank. Comparing the real crosshatch test to the DIY method confirms beyond doubt that the DIY is unreliable.
The comments below were added after this Forum discussion was archived as a Knowledge Base article (add your comment).
Comment from contributor C:
I do rails and floors. Sometimes if I have a non absorbing wood like maple I water pop the wood. I take my moisture meter and read the moisture in untouched wood. Then I wipe the wood with a water filled sponge and it will raise the grain. You need to wait until the moisture content returns to your original number before applying the stain. That will allow the wood to take in more stain in the initial staining, reducing the multiple coats. Then continue with your finish as the others have suggested.
