Are bone-dry biscuits necessary?

Questions about the dryness and size of biscuits. April 24, 2002

Maintaining bone dry biscuits seems to be an unnecessarily tedious process. I'd like to experiment and measure the blade/slot/dry biscuit/wet biscuit conditions. What do you think?

Forum Responses
While it's true that we may not need "bone dry biscuits", part of the advantage of the biscuit system is the clamping action achieved by the biscuit swelling, forcing contact with the sides of the slot. Add this to the fact that a bunch of compressed biscuits are a lot easier to insert than a bunch that have already swelled to working thickness.

From the original questioner:
I'm skeptical about the amount of swelling a biscuit exhibits from bone dry to humid. And I assume that a biscuit that has been exposed to humidity and swelled will not shrink back any significant amount if oven-dried at a later time.

All of this can be tested. If you have what you believe to be a dry biscuit (and not oven dried), measure its thickness. Then soak it in water for a couple of hours. Let it dry in open shade, then measure it again. I'd be surprised if it changes more than .002 in thickness. I'm still of the mind that a .005 oversize cutter would bring an end to all the screwing around we do with tight biscuits.

From contributor B:
Cutters for biscuit slots were developed in conjunction with the thickness of the biscuits. Original developers determined the optimum thickness for each. Later, manufacturers most likely confirmed these results prior to jumping on the bandwagon. To suggest a thicker cutter is to suggest the manufacturers were in error on their original theories or relative thicknesses.

From the original questioner:

You're correct. I am suggesting that the manufacturers are in error on the fit of biscuits. I've had my DeWalt joiner since they first came out and have gone through about 1000 biscuits in a variety of applications before reaching this conclusion. There's obviously a split in opinion on this issue, but I have to believe that most agree that they're a tad tighter than they should be.

From contributor B:
I've gone through 1000's upon 1000's of biscuits over the years in standard and hard to find size configurations. Again I'll say that with proper maintenance, keeping the biscuits in a closed container, they will fit over 95 percent of the time.

You may not need bone-dry pieces, but they will shrink a few 0.001s when dry. Then when wet with the glue, they will swell and provide a very close proximity between the wood and the biscuit, which means a strong glue joint.

Another reason for dry (thin) pieces is this - if the biscuit is too fat initially, you will scrape off all (or maybe a lot) of the glue when it is inserted.

Note: If a biscuit is exposed to high humidity before gluing, it will swell excessively and re-drying will not return it to its original size.

Gene Wengert, forum technical advisor

From contributor A:
You are right on target. My experience over the past 15 years has been that the biscuits swell in humid weather and do not return to their original thickness when dry. Keeping them in an airtight container helps. The problem is that distributors do not take such precautions. When you receive them, you may find some are already swollen. The biscuits are doing just what they are designed to do - swell when wet.

You might contact the original manufacturer and ask for the manufacturing tolerance (to determine minimum and maximum size allowed), the amount they expect the biscuit to expand when in contact with PVA and what thickness variation to expect when they are stored, unprotected, at 35% rh or any other rh applicable to your area.

Perhaps one of us should run an experiment to determine what the optimum thickness of the slot should be based on the manufacturing tolerance of the biscuit, the anticipated expansion during storage and the expansion when coated with glue. There are a lot of variables to be considered.

From contributor B:
It has always been my understanding that biscuits are "compressed" during manufacturing. As such, exposure to moisture not only swells the wood fibers as with normal wood expansion/contraction, but also releases the wafer from the compressed state as well.

So, if biscuits have been exposed to your local relative humidity you can bring them down in thickness a bit by "toasting" them in an oven or microwave. However, they won't get as thin as "factory spec" because oven drying only lowers the moisture content and doesn't recompress the biscuit.

From the original questioner:
Yesterday, I emailed with questions regarding this swelling issue but haven't heard back from them (I asked for hard numbers, which they might be reluctant to release). If they do respond, I'll post their response on this message board.

It was mentioned that biscuits are compressed during their manufacture, but so is plywood. I'm sure plywood expands too, and if the amount was very significant, we'd see a lot of cabinets and furniture coming apart. It's my understanding that furniture manufacturers bring their raw materials up to the ambient humidity of the intended market for their furniture prior to cutting and assembly. So why then would biscuits be sold bone dry when the bone dry spots on the planet are few and far between?

Another point that hasn't been raised is the effect on the outer surface of plywood if a biscuit expands significantly. A .005-.010 biscuit interference would probably translate as a bulge on the outside of the sheet material. Have you ever noticed a bulge due to an expanded biscuit? I'd still love to get my hands on a .165 (.007 oversize) cutter.

Have you ever seen a plywood countertop that has gotten wet and swollen and even delaminated? The pressure is fairly low for plywood, but this extra swelling (called springback) is well known in composite products. It is not uncommon for plywood to do this, but particleboard (and related boards) are even worse.

Gene Wengert, forum technical advisor

From contributor A:
I have noticed bulging from biscuits on 1\2" and 5\8" particleboard and MDF, but not on solid wood or plywood. This may be from the glue causing both the biscuit and the sheet goods to swell.

From contributor B:
Oddly enough I've heard of just the opposite about biscuits telescoping through the material surface. I've read articles that warned that the material surface over a biscuit actually "depresses", showing a recessed area in the surface the shape of the biscuit.

I believe this is due to some wood expansion at that biscuit area from the water in the glue, causing the surface above the biscuit to swell upwards. The surface is then heavily sanded before the wood fibers have had a chance to shrink back down. After the glue has dried the wood shrinks back down and a depression the shape of the biscuit appears.

Also, the idea behind "compressing" biscuits is that there will be increased expansion in the slot. If you only relied upon normal moisture increase swelling, then when the glue dried, the biscuit would shrink back down to its undersized state, leaving a loose biscuit and a weakened joint.

There are two problems with biscuits that I have seen. First, the swelling of the biscuit causes the shape to telegraph through to the surface, creating a bump that is quite obvious in glossy finished surfaces. The second is a depressed area (called a sunken joint when it happens to glue-lines, so maybe we should call this a sunken biscuit?) that contributor B explains very well.

Gene Wengert, forum technical advisor

I haven't experienced any problems with biscuits. I assume the manufacturers have looked at all angles before putting their product on the market, plus put in a lot of money in research and development. I find the dry biscuit fits the slot as it should and expands to fill the slot as it should, thus I keep them stored in the lidded container until used as recommended. If some do expand a little, I simply place them in my vice and recompress them. In a one-man shop meeting completion deadlines, I find it saves time and stress to just follow manufacturer recommendations.

From the original questioner:
This morning I received a detailed response to the email I sent to Lamello. Below is some information taken from this response.

High precision on the biscuit size is worthless if it is not accompanied by at least as high quality standards on the machine and the cutting tool side. Therefore, when discussing experiences with biscuits, never forget the "Systems Aspect".

Lamello plates are produced from beech wood, which is cut and then dried. Then, the lamello-form is cut and the platelet is stamped, leading to the known and typical surface of the Lamello biscuit. Original Lamello biscuits are 3.9 to 4.0 mm thick, measured at the two "bridges" on the surface which serve as gliding paths for the biscuit when inserted into groove and which give the plate a sound primary stability. The space next to the bridges is slightly thinner and is designed to carry the glue. When glue is applied, the biscuit swells by one to one and a half of a tenth of a millimeter, resulting in the known and tight joining.

As mentioned above, the accuracy of the size of the groove is as important as the precision of the biscuit itself. All Lamello plate joining machines come with a slip-on plate (in plastic materials) of 4 mm thickness. This plate can also be used to check the dimension of the groove. Remember that if the cutting tool has been sharpened several times this may have an impact on the size of the groove it produces.

All biscuits are oven-dried and it really depends on the area you keep them or where your shop is located. In Europe, we have high humidity like in the eastern US and biscuits are thicker than they would be in the dry western US.

Check the place where your cabinets will be mounted one day. In western states, don't keep the box open without plastic cover in fall. In the eastern states, keep biscuits in a plastic bag all the time.

Our experience: Buy a router bit (saw blade) for your Lamello machine that makes the cuts wide enough, so thick biscuits will fit inside easily. Mix yellow glue with some flour. When the water of the glue sucks into the flour, it will fill all the gaps by swelling.

Here are results from my own highly unscientific tests.

My Porter Cable blade thickness.

Plates stored in various ways.
open container 8 or 10 months
.1400 to .1580
new package opened for test
.1475 to .1590
plates kept in sealed container
.1485 to.1580

Plates from open container.
measurements taken ~3/8" from each end.
1. nuked for 30 sec. on high, door opened to evacuate moisture. 45 sec. high.

before .1550-.1570
after .1510-.1545
water soaked 20 min..1565-.1595

2. water for 20 min.
before .1525-.1555
after .1620-.1630

All plates measured from each container (30 in all) had thickness variations from end to end. Least was .001". Greatest was .007.

My shop is on the central Oregon coast. This place is just about as damp as you get. I've yet to have a plate joint fail. I do, however, have alignment problems from time to time, due to joiner blades being too thick.

Perhaps the overall thickness of the biscuits isn't the key. After my water tests, I noted that the impressions that are found on a dry biscuit (a nurled look with compressed edge) were all but gone. The impressions and the edges of the plates swelled more than the rest of the plate. So, possibly the swelling of the impressions is the locking power of the biscuit.

Us Texans take an 80 grit sanding block and just slightly sand the biscuit before we glue it and clamp it. We use lockweld yellow glue and we are not bashful about using "fresh" glue until it squeezes out everywhere. We achieve good alignment of panels in the clamp station. We do, in some applications, use biscuits just for panel alignment with no glue.

Biscuits will vary slightly in thickness from any manufacturer (I've used them all). Unless you are into high production, it is easy enough to test their fit before assembling a project. If you find a few that are a bit oversized, simply compress them a bit in a metal vise. Take note of the position of the vise handle when it produces the correct fit. Glue up your project ASAP after this and the biscuits will re-expand.

The comments below were added after this Forum discussion was archived as a Knowledge Base article (add your comment).

Comment from the original questioner:
After much input on this subject, my original thought of a .005 oversize cutter is still my preferred solution to the problem of too-tight bicuits. My current solution is to put each biscuit into my machinist's vise and give it a hard and fast squeeze. This is a bit time consuming, but it does create enough clearance to allow easy lateral alignment. Also, the swelling affect that has been so heavily discussed may kick in. Earlier in the discussion, it was said that the the biscuits swell a tenth to 1.5 tenths of a millimeter (.004-.006) when glued. I was never able to duplicate anywhere near that amount of swelling and frankly doubt that it occurs. My conclusion is that this superb fastener was just a tad over-engineered.

Comment from contributor T:
I have been using various plate joiners for years, at least 15 boxes of #0, #10 or #20 Lamello biscuits. The one common anomaly is the biscuits vary (humidity aside) and the technical reason for this is simple: the stamped pieces vary in thickness due to the natural variation in the wood they are made of. I tend to discard 6% after dry fitting (feeling the proper fit is key - too tight will swell and too loose will shrink. Simply sanding a bit off the tight one's with an orbital sander works for me.