Help with Rubbing through to bare wood in random areas

[Tegmark]

I sanded all 3 first coats with a small hard sanding block that fits in the palm of my hand sing 400grit paper with no major rub throughs.


But starting with my 4th coat I wanted to try my new large sanding block with 400grit paper. I made sure the sanding was dead flat & perfect,
but I start to cut through the clear & stain in certain areas to bare wood, the photo below is after sanded the sixth coat.

https://s32.postimg.org/688iripv5/wood2.jpg (https://postimg.org/image/688iripv5/)


Looking on a low angle the wood has slight warps & isn't dead flat. But I not sure why, as I did sand it thoroughly with 180grit before stain & clear. Does this mean I should have sanded the bare wood aggressively with 80 grit right at the start?




In 2nd pic you can see the layers of the clear exposed where I rubbed through, why do the layers show up like this? Wouldn't the coats merge seamless?

https://s32.postimg.org/nho5ykou9/wood3.jpg (https://postimg.org/image/nho5ykou9/)




When seeking a mirror flat finish, should I always use the large block where ever possible? Or is it ok to use the small one to take some off the top coat off then finish with the large?

thanks

[Xanthorrhoeas]

I don't know about your finish (I almost always use shellac) but the reason for the irregularities in the timber is likely to be because timber varies in how it reacts to moisture and temperature and will move in strange ways. The higher the gloss level the more it will show. There is really no way to stop it moving that I know of - perhaps encapsulating it resin would do it - but then it would not be timber!

Hope this helps.

David

[Arron]

Tegmark I'd like to answer your questions but it's hard to work out what you are asking. You should always start these questions by telling us what finish you have used. I'm going to assume it's polyurethane (ie standard hardware store 'varnish').

It's often useful to tell us the timber species and application method as well.

I also don't understand what those semicircular, concentric rings are on your workpiece. They look like machining marks, if that is so then the piece is a long way from being ready to have finish applied. You need to get the marks out and the surface flat before you start finishing.

It's true that you are often better off doing the final sanding after using a sanding sealer or first coat of poly (it stiffens up the little tufts and makes them amendable to sanding), but the surface must still be smooth and flat and free of machine marks before you apply any finish.

As a rule of thumb, if you can see ANY machining marks on your workpiece before applying finish then they will be a LOT more visible after applying finish. You can get a preview of how well your piece will stand up to finishing by wiping it down with white spirits, any glue marks will become visible as pale patches and any marks will become visible as dark scratches.

You probably do need to go through 80, 120, 180 and 240 grits. 180 is a reasonable grit to finish with for high build industrial lacquers etc and thus is the industry standard, but I think 240 or 320 is needed for polyurethane.

The sanding you do between coats should be pretty minimal. It's really just a light scuffing off of the gritty-feeling surface. If too much sanding is needed between coats then you are trying to remove problems you should have removed before starting the finishing. 400 grit is good for this, it doesn't give you surprises because it doesn't cut deep.

Also, if you are using stain then you need to be doubly cautious. If you are doing multiple coats then delay the sanding till you have built up the body with several coats.

Also, it's pretty unusual to need to sand between each coat. Sanding after coat one should be all that's needed. Maybe another sand before the final coat, maybe.

The first photo shows you sanded too hard, end of story. Well, nearly the end, it also shows you biased the sanding towards the edges, which is hard to avoid but possible with experience.

Also, those witness marks you see in the second photo are standard issue with polyurethane. With polyurethane, each coat cures independently of the coat below, so when you sand through the top coat you get a squiggly white line because the coats do not meld into each other. Applying another coat on top will minimise the lines but not remove them. Contrast this with a lacquer like nitrocellulose where each coat partially dissolves the coat below and thus melds into it, and thus all coats become one and there are no witness lines should you sand through any single coat.

So ultimately, it's a matter of preparation, and understanding that hardware-store polyurethane varnishes are not designed to give you a real high quality piano-finish. The average buyer of polyurethane knows how to bog-sand, slap on two coats and call it finished. The makers of these products know this so they optimise the products for such users.

You can try and cheat the system by using wipe-on poly (like Minwax) or maybe additives like Penetrol but my experience is you get a slight improvement in initial presentation but only for a fairly short period of time.

Helps I hope
Cheers
Arron

[Amrjon]

It looks like they might be marks from original milling of the timber. In which case they would need sanding out before re-finishing if you want to completely eliminate them.


Sent from my SM-G900I using Tapatalk

[Tegmark]

Tegmark I'd like to answer your questions but it's hard to work out what you are asking. You should always start these questions by telling us what finish you have used. I'm going to assume it's polyurethane (ie standard hardware store 'varnish').

It's often useful to tell us the timber species and application method as well.

I also don't understand what those semicircular, concentric rings are on your workpiece. They look like machining marks, if that is so then the piece is a long way from being ready to have finish applied. You need to get the marks out and the surface flat before you start finishing.

It's true that you are often better off doing the final sanding after using a sanding sealer or first coat of poly (it stiffens up the little tufts and makes them amendable to sanding), but the surface must still be smooth and flat and free of machine marks before you apply any finish.

As a rule of thumb, if you can see ANY machining marks on your workpiece before applying finish then they will be a LOT more visible after applying finish. You can get a preview of how well your piece will stand up to finishing by wiping it down with white spirits, any glue marks will become visible as pale patches and any marks will become visible as dark scratches.

You probably do need to go through 80, 120, 180 and 240 grits. 180 is a reasonable grit to finish with for high build industrial lacquers etc and thus is the industry standard, but I think 240 or 320 is needed for polyurethane.

The sanding you do between coats should be pretty minimal. It's really just a light scuffing off of the gritty-feeling surface. If too much sanding is needed between coats then you are trying to remove problems you should have removed before starting the finishing. 400 grit is good for this, it doesn't give you surprises because it doesn't cut deep.

Also, if you are using stain then you need to be doubly cautious. If you are doing multiple coats then delay the sanding till you have built up the body with several coats.

Also, it's pretty unusual to need to sand between each coat. Sanding after coat one should be all that's needed. Maybe another sand before the final coat, maybe.

The first photo shows you sanded too hard, end of story. Well, nearly the end, it also shows you biased the sanding towards the edges, which is hard to avoid but possible with experience.

Also, those witness marks you see in the second photo are standard issue with polyurethane. With polyurethane, each coat cures independently of the coat below, so when you sand through the top coat you get a squiggly white line because the coats do not meld into each other. Applying another coat on top will minimise the lines but not remove them. Contrast this with a lacquer like nitrocellulose where each coat partially dissolves the coat below and thus melds into it, and thus all coats become one and there are no witness lines should you sand through any single coat.

So ultimately, it's a matter of preparation, and understanding that hardware-store polyurethane varnishes are not designed to give you a real high quality piano-finish. The average buyer of polyurethane knows how to bog-sand, slap on two coats and call it finished. The makers of these products know this so they optimise the products for such users.

You can try and cheat the system by using wipe-on poly (like Minwax) or maybe additives like Penetrol but my experience is you get a slight improvement in initial presentation but only for a fairly short period of time.

Helps I hope
Cheers
Arron

Hi Arron, thanks for taking the time for the long post & sorry I didn't mention more details. I was using 2pack polyU brushed on & the wood piece is just some sample pine wood.

Those circular marks I assume are machine marks, but they can't be felt with the finger at all so I'm not sure how they are there. I did sand the bare wood for a while with 120, then 180grit, just to get the wood somewhat in the realm of flat.

I disagree about the sanding of clear, I used a large block & kept it flat, those areas that I rubbed through are the high on the wood. Maybe the wood warped a bit or maybe it was never flat from the first sanding I did.

Thanks for the rest of the advice & info on PolyU witness lines. Just another con of this varnish.

[ian]

Tegmark, I think you are wasting your time and 2 pack.

Machining marks can only rarely be felt, but as advised they show up really well under a finish.

If you are experimenting with the 2 pack finish, you really should be using a piece of timber that matches your project in both species and surface preparation.

are you sure that your sample is entirely dry? Could it still be a little above optimum moisture and the "high" spots the result of movement?