Wiping Varnish - Gloss vs. Satin

[Robert McFarland]

Hello everyone. I recently purchased a case of satin P&L 38 to apply to a Cherry dining table I am building.

I bought satin because that's the sheen I want as a final result. Then I came across a couple threads stating that gloss should be used on all coats up to the last.

You can imagine my frustration at that discovery after dropping a wad of cash on a case of satin. My question is how much "muddier" will my finish be if I use satin for all coats? Will using gloss add enough clarity to warrant the purchase of another case of 38?

Thanks for any input!

[Scott Holmes]

In MY experience there is not that much difference. Also you could let one can set for several days and then pour the varnish into another container leaving the flattening agent in the bottom of the original can. Make sure you don't shake or stir the can before decanting the gloss varnish off the top.

[Robert McFarland]

Thanks for the input Scott. I never thought to let the flattening solids settle. I am in the process of finishing a test piece using straight satin. If I don't like the end result I'll try your suggestion to let settle and decant.

Thanks again!

[Bob Wingard]

This is one time I would have to agree 100% with Scott ... the ONLY difference between the two is the flatting agent ... that's why they tell you to stir constantly while in use ... the stuff settles out pretty rapidly ... give it a few days on the shelf ... pour off the top stuff ... if you should get a little of the goop in the finish, it's not a big deal ... just try to omit as much of it as possible. Filtering it is probably also a possibility, but I don't have a clue what the particle size of the flatting agent would be.

[Robert McFarland]

Thanks for the feedback Bob.

I went ahead and finished a cut-off piece of cherry I had left over from the table top. The flattening solids were stirred thoroughly to get an indication of what 10 coats of satin would look like. Here's the end result after sanding lightly with 600 grit:

268586 268587

I liked how the sheen and color turned out. I was a little unsure after the first few coats (it looked super flat; no reflection whatsoever), but overall I am pleased. One thing I noticed was the grain was a little blah. So, off to HD I went to purchase some BLO. I flipped the board over, wiped on and off a coat of BLO, let it dry for two days and applied 9 coats of thoroughly stirred satin. Again, I finished by lightly sanding with 600 grit. Here's the result of this experiment:

268588 268589

Now I'm happy with the sheen and grain intensity. However, I'm not a fan of the yellowish/amber color. I went with P&L 38 in part because of the light color. At some point I might have to make a concession. Right now I'm leaning toward the second option in favor of grain pop over color.

P&L uses soya oil to produce it's mild color (from what I understand). My question is can soya oil be used as an alternative to BLO to produce grain pop? Is it even available for sale as a stand-alone product?

Thanks for taking interest in my situation.

Robby

[Howard Acheson]

Here is some info relevant to this discussion.

A few years ago I wrote up the following for woodworking club newsletter. The most recent edition of Flexner's book has a section on the same subject arriving at the same conclusions.



The additive used by most finish manufacturers is a silica (glass) that is optically clear. Compare a well stirred satin finish to a gloss finish and you will see little or no difference in clarity.

The way the flatters work is that as a finish dries, it shrinks slightly. This pulls the top of the film tighter causing the granular sand-like silica items to make the surface un-smooth. It's the non-smooth surface that refracts and scatters the light rays giving the appearance of a non-gloss surface. The whole effect of the non-gloss is only the surface of the finish. If you apply a second coat, the hills and valleys of the underneath coat are filled in and the drying of the new coat again creates new hills and valley at the surface. If you apply a gloss finish (which has no flatters) it too will fill in the hills and valleys in the prior coat but no new hills and valleys will be created. You are left now with a gloss finish even though the underneath coat was a non-gloss.

Think about scuff sanding a surface as preparation for another coat. The scuff sanded surface will have a non-gloss surface appearence because of the scratches (hills and valleys) that the abrasive material creates. But, what happens when you apply another coat of gloss to your scratched surface? The resulting new surface is gloss because your gloss finish has filled in the scratches just like it does to the hills and valleys resulting from a non-gloss finish.

So, what should be taken away from this overly long dissertation is that for almost all finishes used in woodworking, it makes no difference what the gloss of the underneath coats were, the final gloss will be dependent on the gloss of the final coat.


>>>> The silicates do refract light to some degree, don't they?

Yes they do, but only at the surface where the shrinkage has created the non-smooth surface. As I said, there is no difference in the scattering of the light created by flatting agents and the scattering of the light created by sanding or otherwise roughening of the surface. If a gloss finish is applied over either roughened surface, the end result of a gloss finish is the same.

Well, you have forced me to get out my copy of Flexner's Understanding Wood Finishing. On pages 110-111 in his latest edition he covers this subject. The article also contains diagrams and photos illustrating the points I made.

Given that the silica flatting agents are optically clear, within a reasonable and normal film thickness, there will be no affect on clarity.

All that said, I do question the need to apply as many coats of finish whether satin or gloss is used.

[John TenEyck]

Flexner's explanation my be correct IF the index of refraction of both the resin and glass flatters are close to being the same. However, if they differ then there will be light scattering below the surface, in addition to and regardless of what happens at the surface. The flatters may be optically transparent, but they will still scatter light based on their size and how different their refractive index is from the polymer resin. Here's a link for those who would like further reading: http://en.wikipedia.org/wiki/Refractive_index You have only to look at a glass jar filled with satin finish and another with gloss to see how much light the satin one scatters. The finish contains thinners which have a lower index of refraction than the polymer, but it still proves the point that the surface has nothing to do with it.

John

[dan sherman]

Robert,

This is the finish I used on my current Jatoba project. Blo followed by 8 coats of satin P&L 38 cut with 50% MS.

The finishing schedule i used was as follows.
1. sand all the parts to 220-240 grit
2. apply the blo and wet sand it in by hand with 600 grit.
3. wipe off all the excess and let the blo dry for 24 hrs.
4 apply 4 coats of the P&L with at-least 5 hrs between coats.
5. let the first 4 coats cure for 24 hours.
6. get rid of the dust nibs with 1500 grit Mirlon. The coats are so thin the dust nibs come right out.
7. apply the next 4 coats.
8. let the last 4 coats cure for 24 hours.
9. get rid of the dust nibs with 1500 grit Mirlon.
10. let the finish fully cure for several days, and then wax.

This is a shelf panel after the above, it looks a lot better in person as my photography skills are crap.
268609

[Robert McFarland]

All that said, I do question the need to apply as many coats of finish whether satin or gloss is used.

Thank you for the information Howard. That supports my decision to use satin for all coats. In regards to your quote above, I'm not using the PL38 full strength. I cut it 50/50 with MS to make a wiping varnish. So the 12 coats I mentioned translates to 4ish full strength coats. Hopefully this clarifies things (pun intended).

John, thanks for your input. Your argument makes sense. I'm inclined to agree with you based on your example of merely looking at satin vs. gloss. Regardless of which viewpoint I agree with, the most important thing is that I like what it looks like in the end. I am happy with the way the test piece turned out on both sides in regards to clarity.

Dan, thanks for sharing your schedule. The one I used on my test piece is very similar, with the exception of number of coats in a set and dry time. I laid on three coats at a time 45 min - 1 hour apart, let dry 24 hours before applying the next set of three. I used 600 grit to sand the ribs off, but I might try a finer grit like you specified. Your project is looking good!

I appreciate all the helpful information, but I have yet to get my question answered about a BLO alternative. I do not like the yellow color BLO imparts on the cherry (especially the sap wood). And knowing that BLO amber's more over time furthers my dislike. What are some lighter-colored oil alternatives? I would like to use soya oil, since that's in the ingredient list of PL38, but I'm not sure if that can be bought separately and/or if the drying times are manageable. My ultimate goal is to find a light colored oil that pops the grain and can be top coated within a few days.

[John TenEyck]

I think you'll find no difference in the how much the grain pops with or w/o BLO or any other oil under your P&L 38. What you are seeing is a color change which make the grain more pronounced. Water does the same thing but it goes away when it dries. But if you want to do something I would look at dying the piece rather than looking for an oil, which may or may not ever cure. Transtint dyes do amazing things to highlight the grain and you can get nearly any color you want. And your P&L 38 will go over the Transtint with no problems.

John

[Rich Enders]

On the issue of finish clarity my experience puts me in the same camp as John TenEyck. We formulate polymer systems for industrial use, and clear added to clear does not necessarily result in clear. Unless you use ingredients having the same refractive index (RI) you will have some level of opacity. Also since the RI can change with conditions (eg. temperature; humidity), clarity of compound formulas can be somewhat transient.

I am no expert on woodworking finishes, but the use of flatters to modify gloss in a finish may lead to clarity variability over time. There is nothing like the look of a hand rubbed finish.