Veneer Replacement Technique

[Don Stauffer]

I'm refinishing an end table and it's not going well. I guess there wasn't enough burled walnut veneer left on the top so I sanded through it in places. Well, I learned, so that's something. The veneer is framed by red oak boards. I'm considering replacing the veneer, maybe using a router to make room for the new piece. My fear there is I might end up with it too deep or not deep enough, or uneven, but it's the best idea I can think of. I'd be interested if you have a different way.

I contacted a veneer place and got this response:

"There is no way to do what you need to do. It's not something I've had success with either. The first challenge is getting the depth right and I think that could be nearly impossible. The second challenge is getting a veneer to fit exactly inside the frame which is very unlikely... partly because veneer expands and contracts bit when the glue is applied and has dried.

Your best bet may be to post your question on one of the many woodworking forums on the internet. I suggest the forum at Sawmill Creek (www.sawmillcreek.org (https://nam04.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.sawmillcreek.org%2F&data=04%7C01%7Cdon.stauffer%40okstate.edu%7Ce7de73 79cb2249da5f9908d8b71cbd5a%7C2a69c91de8494e34a230c df8b27e1964%7C0%7C0%7C637460681354779893%7CUnknown %7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzI iLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=pKMTDPK7RuSAa9B8XzWjKjD31ChvMcNRvvoc6X5G2B8% 3D&reserved=0)). I have no affiliation with this site but this particular forum is responsive and resourceful. Give that a try because those guys and gals know a lot more than I do."

Can anyone recommend a technique that has worked in a situation like this?

[Robert Engel]

I don't think the router idea is feasible.

A few ways I can think of:

1. Remove the central field, veneer a new panel and install it.

2. Make one veneer panel consisting of the central veneer surrounded by mitered strips to match the perimeter.

3. Take the panel apart & start over. If you can't reuse the oak, it would be fairly easy to remake that.

Of the 3, it would be a toss up between 2 and 3.

If there were an inlay strip it would be more complicated, that's what I'm dealing right now.

[Mike Henderson]

What you are trying to do is very difficult. What I would do is put new veneer over the whole surface.

But if you want a veneered center panel with real wood edging I'd do what Robert suggested above - rip the existing edging off, veneer the center and carefully glue edging slightly proud of the panel. Then comes the difficult part. You have to take the edging down to the level of the veneer without sanding through the veneer. It's easy to make a mistake but it can be done.

If you do make a mistake and damage the veneer at the junction of the edging, what I've done is to use a router and put in a decorative strip between the veneer and the edging. Of course, then you have to take that down to the level of the veneer.

If you can start with thicker veneer than the standard commercial stuff you'll save yourself some heartache. Maybe resaw some veneer yourself. I have done this with commercial veneer but you have to go very slow and work carefully.

In general, I avoid doing a panel with wide edging because it's so difficult. I'd just take the veneer out to the edge. If I wanted the look of a different edging, I'd lay the edging as veneer with real wood on the edge of the substrate so it looks the same and then put a profile that hid the joint between the veneer and the wood substrate.

I hope you can understand what I mean - it's hard to describe in words.

Mike

[Don Stauffer]

I considered covering the whole top with veneer. I hesitated to do that since that would expose the edge of the veneer, which would be visible and potentially fragile. I still may, but I also was hoping to keep to the original design as much as possible. Similarly, putting a panel on top would duplicate the top surface, but the side of the panel would be exposed, so I'd probably have to veneer that separately.

I'm not sure taking it apart is possible since it's pretty well glued. But I'll look at that a little. Maybe there's a technique to do that that I haven't learned yet?

[Mark e Kessler]

I used to work in a veneer dept of a very high end corporate furniture making business, we replaced sections of large conference tables some costing in excess of 100k with a cnc router removing the damaged section, the bigger challenge was finding a match

[Don Stauffer]

That's interesting - a CNC router would certainly make it possible to make a precisely sized space for the veneer. I think with some luck and care it might be possible to do that well enough with a hand-held router. The main advantage of a CNC would be the bit is positioned relative to the machine rather than a local surface of the work. The only CNC device I have access to can only handle things about an inch thick, I think, though.

How did you deal with veneer expansion and contraction? Presumably, make the veneer horizontal dimensions with a precise allowance just enough for any expected expansion, and maybe use an adhesive which wouldn't have that effect? I had thought of putting the veneer on a relatively thin MDF piece and routing deep enough for the combined depth of MDF and veneer, but that might be too much for an operation with a hand-held router. How much expansion would be expected from adhesive moisture? Presumably then it would shrink back to its original size when it dried. So it's a question of whether that gap would end up overly large. It occurred to me to do the veneer in 4 pieces to minimize the size of the gap. The original is actually in 4 pieces and does a good job of matching them for an "ink blot" effect.

[Don Stauffer]

I looked more closely at this. The frame seems to be very thin wood glued on top - thicker than veneer, but not very thick at all. So I think that's how they dealt with not having fragile veneer all the way out to the edge. I'm thinking they mounted the veneer on a fairly thin base, then maybe planed the oak boards down until they were the same thickness as base plus veneer, then glued it all together. Actually a reasonably smart way to make a veneered end table durable.

[Kevin Jenness]

I would do it on my cnc router if I had to retain the existing edgebands. You would have to work fast getting the new veneer in place and under pressure using a waterbased glue. You might be able to find walnut burl with a paper backing that would make the job easier. You don't want flexible glue or expansion room, an exact fit is what veneer excels at. Its expansion potential is held in check by its thinness and the glueline.

Easier would be to remove the banding, veneer the center panel and re-apply the bands. Easier still if the client agrees is to run the veneer over the edgeband, but it is a bit more fragile and liable to telegraph the joint below due to the banding width.

Go easy on the sanding next time. I heard about another guy that burned through once ;). If you see any color change while sanding stop at once and you might be able to save it.

[Steve Jenkins]

I would make a new top. Veneer on mdf then edgeband with oak. To flush the solid I use a hand held belt sander but I’ve been doing that for a lot of years.

[Don Stauffer]

Right now I'm thinking something similar to this: Veneer the frame as well. It seems on closer inspection the frame was hardly more than veneer anyway.

[Kevin Jenness]

"Veneer on mdf then edgeband with oak. To flush the solid I use a hand held belt sander but I’ve been doing that for a lot of years."

You're a braver man than I. ;)

[Mark e Kessler]

The Veneer room I worked in was tightly controlled as far as temperature and humidity, we had a system that would spray automized water in the room (would scare the crap out of me on occasion), didn’t have much of an issue that i can remember, if doing a complicated layup like a starburst then you would want to cut, tape and glue all in one day. I can’t remember the adhesive we used but would imagine we used a hard glue line, all of our tops were made with 1” MDF with the same species and thickness for the backs although a lower grade was allowed, they were glued in a joos hot press. Solid edges were sanded flush using a stroke sander.



That's interesting - a CNC router would certainly make it possible to make a precisely sized space for the veneer. I think with some luck and care it might be possible to do that well enough with a hand-held router. The main advantage of a CNC would be the bit is positioned relative to the machine rather than a local surface of the work. The only CNC device I have access to can only handle things about an inch thick, I think, though.

How did you deal with veneer expansion and contraction? Presumably, make the veneer horizontal dimensions with a precise allowance just enough for any expected expansion, and maybe use an adhesive which wouldn't have that effect? I had thought of putting the veneer on a relatively thin MDF piece and routing deep enough for the combined depth of MDF and veneer, but that might be too much for an operation with a hand-held router. How much expansion would be expected from adhesive moisture? Presumably then it would shrink back to its original size when it dried. So it's a question of whether that gap would end up overly large. It occurred to me to do the veneer in 4 pieces to minimize the size of the gap. The original is actually in 4 pieces and does a good job of matching them for an "ink blot" effect.

[Don Stauffer]

Fortunately, the client is me. But yeah, I learned that lesson. I think what I'm settling on is pretty much your second paragraph second sentence there. I'll get some red oak veneer and some burled oak veneer and just glue it on the top. The edges are a little rounded from sanding, so I may have to put some wood putty in there. I don't know. I'll see what happens, I guess.

[Don Stauffer]

This is fascinating. Since I'm considering veneer, I've been studying wood grains online for a couple days now, and I've reached a couple conclusion which surprise me.

The first thing is that despite the pink hue, I believe the frame is white oak, because of the length of the rays in the grain. That fits because the other way to tell white from red is whether the grain is plugged on the end, and the legs, which also have a pinkish color, do seem to have the typical tyloses blocking the pores that white oak has. So I don't believe there's red oak anywhere on this.

This is a Lane Furniture 920-02 End Table.

https://highendusedfurniture.com/product/lane-furniture-contemporary-bookmatched-walnut-22x27-end-lamp-table-920-02/

Everyone who identifies it by type of wood calls it "walnut", so that's what I've been assuming, but pretty much all the burled walnut veneer I can find is anywhere from a little to a lot darker. For example:

https://www.veneersupplies.com/products/Walnut-Burl-Veneer-Lot--95-x-1725-18-Sheets-Per-Lot.html

Then on impulse I looked at some burled white oak. The tone seemed the same, so I started looking at various grains, and ... I found this, which is far closer to the original than anything else I've found:

https://www.veneersupplies.com/products/Oak-White-Burl-Veneer-Lot--1125-x-28-7-Sheets-Per-Lot.html

I'm no expert, but I don't think it's walnut at all. I think it's burled white oak! That would make the entire piece white oak.

What do you think?

[Kevin Jenness]

" The edges are a little rounded from sanding, so I may have to put some wood putty in there."

Not a great idea. Flatten the panel before veneering if you want to get a reliable glueline. Do you have access to a thickness sander?.

[Don Stauffer]

What's a thickness sander? Am I likely to be able to get one locally (Tulsa)? Are they expensive?

[Kevin Jenness]

Like a thickness planer but using abrasives. Most commercial shops have a widebelt sander or drum sander and will run your panel for a fee. Call around.

Otherwise, use a long hand plane or a flat hard block with coarse abrasive to level your panel. Or start over. Veneering a panel with rounded edges is asking for trouble.

[Phil Mueller]

I’d agree it’s likely not walnut. There are a lot of light toned burled veneers, and whether it’s oak or not, the oak veneer you pictured would be a good match. veneersupplies is a great source. Joe’s a good guy to work with and the quality of the veneer is excellent.

You may already know this, but burl veneer often needs flattening first, so you might want to check out Joe’s veneer softener/conditioner. And it doesn’t work well with the hot iron press method...tends to leave lots of cracks. Cold press or veneer hammer would be my choice. Good luck, should be a fun project.

[Don Stauffer]

I got a good impression from VeneerSupplies. They're who suggested this forum.

Their veneer actually indicates whether they consider it to be flat or not. I was going to just go on that.

[Kevin Jenness]

"Their veneer actually indicates whether they consider it to be flat or not"

What does that mean?

[Phil Mueller]

Kevin, it means whether the veneer is flat enough to glue down as is or cut and edge tape it, or if it’s curly, wavy, etc and should be flattened before glueing down or cutting and edge taping.

For example, these veneers are relatively flat and can be easily worked or glued down:

449577 449578

This elm burl, however, is wavy and would be difficult to work with as is.

449579

You could probably get away with glueing it down as is to a flat surface, but it could buckle and create a crack. And it is certainly more difficult to cut and edge match without first flattening it with a softener/conditioner.

So, if Don is planing to do a four way match and his veneer comes like the elm burl, it would be best to flatten it before edge jointing and taping.

[Kevin Jenness]

I see. I was thinking about the panel's flatness. I agree, most burl veneer will need flattening. I have a piece of unidentified burl veneer in a sample pack with random bumps more than 3/8" high, probably best suited as a topographical relief map. I flattened a couple of dozen madrone burl leaves once and some of them had areas that folded over in the press rather than flattening out.

If I were laying up burl I would use powdered plastic resin glue for rigidity. Burl has a mind of its own.

[Don Stauffer]

Thing is there's a fancy solid edge and I don't want the top standing up above that. I wonder about PSA-backed veneer for the frame.

[Don Stauffer]

Some of their veneers have this note:

"Good news! This veneer has been pressed flat and should stay that way if stored (https://www.joewoodworker.com/veneering/storage.htm) and acclimated appropriately."

[Don Stauffer]

I appreciate all the advice. I'm currently digesting it all while reading up on veneering, and trying to decide exactly what to do. It seems like there's no easy way to deal with this.

[Mike Henderson]

Where are you located? Perhaps an experienced veneer person could help you.

Mike

[Don Stauffer]

Tulsa. That would be really great. I can handle the staining and finishing once the veneer is right.

[Don Stauffer]

One possibility would be for me to buy or build a CNC router. I'd have use for one since I make speakers as a hobby. I have lots of 3D printing experience and I've used the milling machine at FabLab Tulsa, so I know what I'd be getting into.

I had always assumed they would be too expensive, but I'm wondering now whether a good price point exists for a functional one, and what limitations there would be and for what price.

I've had a recommendation for the Shapeoko XXL with HDZ. It goes for about $2,500. I'm assuming it could be set up to hollow out the area for the burl veneer, except the corners I'd have to chisel or something.

[Bert McMahan]

I'm not sure how thick the top is, but if you're worried about the veneer expanding or contracting, you could attach it to a thin 1/4" MDF or plywood panel first. You'd then route out a bunch more material than the thickness of the veneer (and install the whole thing), but it would be easier to work with than a floppy veneer sheet. You could also do the routing with a hand router and jigs, getting it as square as possible, then tweak the veneer/MDF panel to get it to fit just right.

[Brad Shipton]

Before you start building a CNC take a straight edge and check the table for flatness. I doubt it is perfect, and a CNC has no way to measure imperfections within the parameters you are describing. It will mill off a plane perfectly flat based on whatever point you pick as a reference. I doubt that will work very well since most wood veneer is 1/42" thick and you do not have much to play with. You will still have to manually trim the corners since all tools have a radii, but that is not a big deal.

Flattening burls or crotch veneers is not complicated or difficult if you have a vac press, but it takes time. There are commercial solutions for that. Most building fine furniture using burls or crotch wood use a backer and the choice of adhesive is important.

[Don Stauffer]

The reason I'm thinking a CNC router is that I could only replace the burl veneer and leave the 2" frame at the outside. The edges and corners of the frame are where I noticed it's not flat.

The other thing is one could actually make the top flat with the router, using the lowest point as a reference and routing the rest to the same height. Then as a second operation, route the area for the veneer. Probably using 3 of the 4 corners to form the plane for the router would work since the corners are probably lower than anyplace else.

[Bert McMahan]

Before you start building a CNC take a straight edge and check the table for flatness. I doubt it is perfect, and a CNC has no way to measure imperfections within the parameters you are describing.

Actually, lots of CNC controllers can do a probe scan of a surface and adjust their profile based on actual surface height. It's mainly used for printed circuit boards, which need to be engraved only through a thin layer of copper. You will want to make sure the CNC machine you buy actually has this feature, but it's available on even very cheap models. The touch probe is as simple as a piece of conductive material (like a flat piece of copper or aluminum) to which you connect a wire and lay on your part. Another wire is connected to your frame, and the machine drops the tool down and detects when the tool touches the part. You do this with the spindle off, of course.

[Brad Shipton]

Bert, I am aware of that, but not in the price of the different machines he listed. The controller to collect that data is more expensive than the machines listed, and you need software that creates the 3D surface.

[Bert McMahan]

Bert, I am aware of that, but not in the price of the different machines he listed. The controller to collect that data is more expensive than the machines listed, and you need software that creates the 3D surface.

It's actually doable with the Shapeoko, one of the machines he mentioned. It's a little finnicky but not bad. The Carbide3D controller will accept a probe input, and since his part is wood you can make a probe with a piece of flat copper and a wire. You use a regular tool as the probe, which completes the circuit when it hits the target. The controller senses this and notes the position. Here's a fairly DIY way to do it:

https://hackaday.com/2014/12/12/mill-warped-pcb-blanks-on-an-uneven-bed/

Here's some info from Planet CNC, which makes a 4-axis controller for about $175:

https://planet-cnc.com/how-to-use-moveable-sensor/

And how to use their Warp feature to compensate:

https://planet-cnc.com/how-to-use-warp-feature/

Basically, instead of doing the warping in CAD, they do it to already-processed Gcode. It's also doable on many of the very cheap controllers that will run GRBL.

It won't be quite as full featured as doing it with a full-on CMM but it's definitely not expensive and you don't need anything more than some free software.

[Brad Shipton]

That is interesting Bert. I am familiar with the industrial versions and own software that can make these type of surfaces. I think one needs to temper their expectations in this machine class. I have a 4'x8' machine that has frustrated me at times, and I expect there is a lot of learning to get all these steps to work. I am not sure I would do it for this table unless it was a family heirloom.

[Bert McMahan]

I expect there is a lot of learning to get all these steps to work.

Now that, I can agree with :D

[Mike Henderson]

In my opinion you're making this a lot more complex than it is. What I'd do is use a drum or wide belt sander to sand the panel flat. Then I'd lay up veneer with a veneer edging of the same species and color as the existing edging. Use whatever veneer and pattern for the center panel. You could even do a radial match if you want to be fancy.

Lay that on the flat panel, preferably using a caul and a vacuum bag. When cured, put an edge around the outside to hide the joint between the wood of the substrate and the veneer. Done.

If you can't do it and don't want to learn veneering, shop the job out to someone who does veneering.

Mike

[Phil Mueller]

If it were my table, I’d create a simple jig similar to a slab flattening jig and with a router remove all the veneer to just a HAIR below the existing solid wood border. Square up the corners with a chisel. I’d glue in new veneer...whatever pattern you want. For that size table you could make a couple of plywood or MDF cauls and clamp down the glued veneer. Put a layer of felt on one side of the caul that goes against the veneer to make up for the slight recess of the veneer (wax paper in between the felt and veneer).

If you have any gaps between the new veneer and existing edge, you could fill it, or I would run a small stringing inlay.

Once it’s all down, carefully plane/card scrap the existing edging down to level with the veneer.

[Don Stauffer]

3D printers have a similar feature, inaccurately called "mesh bed leveling".

[Don Stauffer]

Phil, that is almost exactly what I had in mind when I first made the post. But I got responses discouraging that plan. Your "slab flattening jig" was not part of my original idea, though. But I did think of it later, and then I thought basically all that is is making the router have a reference plane with the correct orientation to the table, and that made me think a CNC router would do the same thing.

What I'm discovering about CNC routers, though, is they would require I remove the top. I could try dissolving the glue, but your way could be done with the top still attached.

I'd be interested how you'd fashion the jig.

[Phil Mueller]

I’d have to experiment a little, but was thinking just a 3/4” frame, a bit longer than the table width with an opening in the frame that would allow you to see what your doing, but still give good support to the router base. If for example the opening is two inches wide (really just depends on the size of your router base), clamp it down, do a 2 inch area, then move it, do two more inches, etc. Not the quickest, but would get the job done.

If you have a router guide, use the guide against the outside of the table edge and do the perimeter of the veneer first.

Set the router depth to the thickness of the new veneer, plus the frame thickness, plus a hair deeper.

[Don Stauffer]

I see! Doing it in small sections allows the jig to be narrow enough to support the router on both sides. But it does seem like the frame would have to be supported on both ends to the same height no matter where the 2" area is, so it would have to sit on pieces that went all the way across the table. Something like:

450086

[Phil Mueller]

That’s what I was thinking. And yes, the two frame pieces that sit on the hardwood edge of your table would need to be long enough to move completely from one side to the other. You could also make the router support pieces thicker if you want a little more support.