Major French Polishing Problem

David Agnew

Member
Messages
285
This is the top of a large desk I've recently built for myself. I've been fighting with it all week. What you see is 4 coats of Minwax oil polyurethane covered with two overly-thick wiped on coats of Zinnser amber shellac, followed by my first ever attempt at French Polishing.

There was a lot of sanding, dust, bugs, and problems. It's been a long week.

Anyway, I've got my own ideas that I'll share later, but first I'd appreciate the opinions of the learned folks here on:
A) What you think I did wrong and how to not do that anymore
B) How to recover from this

Thanks for the help.

Hopefully this photo uploads ok. This is a completely dry surface. Shellac sat for about 90 minutes, then I used Mineral Spirits to remove the oil, then waited until the MS evaporated away before I took the photo. You're seeing marks in the finish.
20161118_204916.jpg
 
Shellac and 'overly thick' do not go together. I'm only speaking generally but, a french polish finish would start by filling the pores of the material. Pumice sifted through a cotton cloth in a "pounce" bag tapped along the surface is one method, paste fillers also work. Whatever you fill with should react with the shellac to become more or less invisible. I assume the coats of poly were a recommended method to fill the irregularities but, I would have a hard time getting poly to a smoothness that would work well with polishing; was it sprayed? I'm not saying it is wrong; there is always more than one way to skin the cat.

Once you have a filled and super-smooth surface, thin shellac (a 1lb cut or thereabouts) is applied in very thin coats. I use cheese cloth wrapped in a cotton cloth; some folks prefer linen or a well worn piece of bed sheet. A figure 8 pattern is popular when applying the shellac. I tend to use overlapping circles. The thin coats dry very fast (about 30 minutes in my shop). However, I wait a few hours or even overnight after the last coat. This is because I am going to wet-sand and I want the shellac nearly brittle (probably a bad term) . Check your progress and flatten any irregularities (I use a clean cloth that is wet with DNA) before proceeding to the next coat. The thin coats avoid building too quickly and creating ridges or other marks. Any that do occur are also quite small and easily dealt with.

Once the shellac is completely dry I shine a light across the surface to see if I have picked up any irregularities from the continuing evaporation of the alcohol carrier used in the shellac. If required I pad again with the same 1lb cut to fill these irregularities, then face the hard truth that unless I want all my effort to go for not I will wait another 4 or 5 hours before sanding. Once I am satisfied with the surface I spritz with distilled water and sand with a block and 1500 grit or thereabouts. Dry sanding will accumulate shellac 'blobs' on your abrasive and mar your surface. I keep a pile of soft, clean, damp cotton cloths at hand and wipe away the spoil frequently refolding to a clean surface of the cloth for each wipe. I wipe the whole surface again along the grain with a damp cloth when I am done.

The burnishing step will finish things up. Here's where the elbow grease comes in. I skip the shellac entirely and just dampen the core of my new pad with DNA. The surface that contacts the material is practically dry to the touch. I use long, overlapping strokes that run with the grain. You are literally heating the shellac through friction and micro-polishing the finish itself. I skip a final top coat of wax as I find it just dulls the effect I have worked so hard for.
 
Last edited:
i havnt done french polish as the word implies but i have used both finishes a fair amount, so till DAVE comes in to tell the secret my thoughts are to wet sand the exist finish to allow another thinned coat of poly, i never put shellac on top of poly for a final finish.. the poly is tougher than shellac but shellac by itself is easier to repair. if it were mine i would have used the shellac first for color pop or to act as a sealer which would help to eliminate the roughness in finish to start with.. shellac is easy to sand and will fill in voids of grain to get smooth finish. poly will show witness lines if you sand to much that is why i would use a wipe on finish.. my personal finish regime now days is to use shellac then top coat of lacguer
 
Agreeing with what has been said, I think the first mistake was putting shellac over the top of urethane varnish, the second being the thick coats. Putting shellac on top of varnish is seldom an acceptable finish, and putting waxy shellac on top of a urethane formula may be part of the problem (urethane resin inhibits adhesion). That said, I'm not sure what the next step would be. Likely I would try to remove the shellac and then work on leveling the varnish and getting a good gloss on it.
 
i too rely o the info jeff jewitt has to offer as well i have one of his books and belong to his forum to get questions answered at times.. you can call his number and most times will be talking to him not a person filling in he is always willing to help others!!!
 
Thanks for the advice folks.

So, I self-diagnosed. I was attempting to use the Flexner method and I'm fairly certain I got too aggressive with the alcohol. I think the oil hid the damage I was doing as I was doing it.

I ended up putting a 320 pad on my random-orbit and scuffing the top. I didn't sand the shellac fully off, but the surface looked fairly uniformly dull with the damage showing up as ghosts.

I modified the Flexner method by skipping the alcohol. I simply added shellac to my pad without the added squirt of DNA. That worked pretty well. Here's the desktop after one polishing session of about 40 minutes. It's dried for about 90 minutes now. In about 30 minutes I'm going to run down and do another session and I figure it'll be good enough. It's made of construction lumber and a board of BC plywood, after all! :D

I'll wax and polish it in the morning, then let it sit until Monday, when I'll assemble it.

20161119_210853.jpg
 
I suspect your process may cause long term problems as well. The Poly is a softer finish than the shellac so I believe it likely that you will end up with the shellac cracking over time. Generally the rule of thumb is softer finish on top of harder to avoid that sort of problem. I would love to be wrong... and maybe you used a hard enough poly that it'll be ok.
 
If you google "Bob Flexner French Polish" I'm sure his method will come up. I based my technique on what he wrote in his book Understanding Wood Finishing.

That has nothing to do with shellac over poly. The shellac over poly was simply me being frustrated with the stupid project and also wanting to try something new. I didn't want to sand off the poly and take it down to bare wood as I'd have to stain it again too. As I mentioned, this isn't fine furniture, it's construction grade 2x8's and a slab of BC oak plywood from the BORG. I didn't run anything through a planer or jointer so it's really not all that flat to begin with. :) Sanding off the poly would've involved actually flattening the piece which would've taken a ton of time and also certainly worn through the face veneer of the plywood in a number of spots.

Flexner's French Polishing method is basically
1) Fill pores
2) Level surface
3) Load your pad using a combination of shellac and DNA
4) Lube the pad using mineral oil - this creates a long "vapor trail" as the DNA evaporates out from under the oil. Makes it very easy to see how/where you've worked and also keeps the pad moving easier as it dries
5) Eventually, he says to just add DNA... that the pad will have "enough" shellac that you just need to thin it and dribble it out.... at this point you're mostly burnishing and bringing up the gloss


Jeff Jewitt's method appears to be (based on that one webpage)
1) Fill pores
2) Load your pad with shellac
3) Apply a thin coat and wait for it to cure hard, then mechanically level
4) Repeat step 3 until happy (or crazy)
5) Mechanically polish and perfect the gloss


Basically, my mistake was in Flexner step 5. Perhaps I had too much oil so the pad was still moving well. I certainly had too much DNA. It seems that the pad was probably pulling up shellac instead of putting it down.

The second time around, I made three changes
  • I re-arranged my lighting to give me a better view on what was going on. It seems that proper lighting is VERY important for this process. You can't just have "lots" of light, you need to have it in the proper locations so you can easily see the reflection in your work.
  • I re-made my pad. The first one was a ball of t-shirt wrapped in more t-shirt. This time I used cotton balls wrapped in t-shirt. Next time, I'm definitely going for a large lump of cheese cloth for the core. The cotton balls just compressed and it seemed like the shellac didn't distribute evenly through them. Near the end, it seemed like more shellac was coming out via the edges/sides of the pad rather than the flat part touching the surface. Also, if I'm doing another 5'x3' desktop or table, I will make the pad bigger.
  • I did not add shellac and DNA, I put shellac only into the pad. This meant I was a little more aggressive with re-filling the pad as it became stickier faster. I'm assuming I was thinning too much earlier. By not adding additional DNA, it ensured I had a 2-ish lb cut of shellac at all times. The oil then helped keep everything moving and the vapor trail was a big plus.

My natural tendencies lean towards "Too much is just enough" but this process is certainly "a little goes a long way" or perhaps "less is more".

It was very nice to be able to achieve a "good enough" finish in about 6 hrs of work (2x 1hr applications + 2x 2hr cure times) which would've taken me probably 2 days using wipe-on polyurethane. Heck, considering the problems I had with the poly on this project (dust, bugs, bubbles, etc) I might not use it again until I can get a spray booth setup.

Took this photo this morning after about 9 hrs of curing. Oil's been cleaned off, this is perfectly dry. Certainly not perfect, I'm happy enough that I'm skipping the planned wax-n-polish.
20161119_231246.jpg
 
You have another major problem coming. If you french polish over oil the finish will begin to crack.
Don't use an oil grain filler.
There is a reason french polish is one of the most beautiful finishes, it takes work. Starting with rotten stone, shellac flakes, methyl alcohol, an a lot of elbow grease. I'll use a drop of pure tung oil to create the slide . Only 3 coats a day . Till I reach the finish and apply a couple of very thin coats of a fine wax.
 
Last edited:
Well, the adventure continues. It's been a 7 full days (168+ hours) and the finish can still be dented by my fingernail. So I guess that means the store bought shellac was old. :(

I found this very informative page describing how to decode the manufacture date from the "lot code" Zinsser prints on the cans:
http://www.popularwoodworking.com/w...log/user-unfriendly-zinsser-bulls-eye-shellac

UNFORTUNATELY, Zinsser uses an alcohol soluble ink to print this information!! Somehow, I got a couple drops of shellac on the lid, and it was -of course- right on the Lot code. So I only have the last two characters and have no idea how old the shellac is.

At this point, I'm just going to use the desk. The finish will get messed up. My computer and other things that don't move will get stuck to it. Oh well, who cares, it's tuba-fores anyway. :) I learned a ton about shellac, finishing, furniture design, and construction with this project. Plus this big desk really fits my 6' 2" frame well (I'd hope so, *I* designed it!) and it's very comfortable to work it. Resounding success, all around!
 
Well, the adventure continues. It's been a 7 full days (168+ hours) and the finish can still be dented by my fingernail. So I guess that means the store bought shellac was old. ...

Or it may mean the poly undercoat didn't cure.

Put some of the shellac on a piece of glass; leave it overnight, and see if it's hardened.

Too much shellac applied too fast will also take a long time to cure.

My bet is the poly undercoat. If it didn't cure for at least a week before you overcoated it with shellac, then it's not cured.

Strip it back to bare wood and start over - using ONLY THE SHELLAC. That's what French polish is supposed to be anyway.
 
Lets bring this one back from the dead again.

I keep learning fun and interesting things with this project. For example, Shellac has extremely poor heat resistance! The finish is starting to dull and pinhole near the exhaust port of my laptop. I haven't measured it, but I'd guess the air is around 110* or so. Seems other people (professional people, even!) have had this problem: http://www.woodweb.com/knowledge_base/Heat_Damage_to_a_Shellac_and_Polyurethane_Finish.html

Also, the shellac was definitely bad out of the can. It never got hard enough that I couldn't dent it with my fingernail.

I've got some time off around Christmas, so I'll put some drawers into this thing and make it a "real" desk soon. I'm sure I will learn a lot in that adventure also.
 
Top