After years of using Bessey K-Body clamps...

[Todd Burch]

I'm a bit disenfranchised.

Needless to say, Bessey K-Body clamp faces are NOT parallel, and as a result, as I have observed too many times, when glueing up cabinet doors constructed with cope & stick stub tenons, because of the out-of-parallelness of the clamping jaws, the stiles are pushed out of plane with the rails.

Here's an example of what I mean:

257343

This is plan view. The bottom drawing of the top of a door shows a 12" long rail and two 2" wide stiles. The upper drawing, (in exaggeration), is what a door will look like after being clamped with a Bessey K-Body without cauls.

I've seen this before, but the cause and effect has never really sunk in.

I've considered a couple solutions.


I could cut cauls to counteract the incline of the jaws, and place a caul between each stile and the clamp, requiring 2 cauls per door. Last night, I glued up 6 doors, so I would have needed to cut 12 cauls, and the assembly would be more complicated, with the fumbling of extra pieces. Do I store the 12 cauls when I'm done? These doors would have used 37" cauls. Next time, if I need 52" cauls, I would have to cut more. This solution doesn't scale well.
I could not use the Besseys like they were intended (perpendicular to the glue joint), but I could use them parallel with the glue joint, as in the (rough) picture below. The foreground clamp is the traditional use that leads to the above warped door. The background clamp illustrates my thought on alternate use.


257345

This, again, would be an adventure in fumbling with the stock and clamps, and storage of the clamped up assembly "standing up" would be pretty unstable, compared to how sturdy Besseys are when used as designed. Plus, on softer woods like pine, cherry and walnut, there is an opportunity for clamp marks to be left on the outside edge of the stiles.

So the questions are, have you observed this too, and what do you do to circumvent this?

Thanks, Todd

[Dave Zellers]

I get what you're saying but I would also think this would be a measure of the tightness and accuracy of the cope and stick fitting.

If the fit is sloppy and the clamps are not parallel, this could result. If the fit is tight, not so much.

When I used to encounter this with pipe clamps, I would immediately switch the clamp on the offending joint to the opposite side reversing the effect. Sometimes a clamp on each side was required, adjusted to the correct pressure to achieve a flat joint.

I didn't get into parallel clamps until the Jets came along and since buying and using them I never experience this. But I don't think cauls are needed here. Just apply the right amount of force to the opposite side to create a flat joint tested by a straight edge.

The length of the tenon is also a factor. A 1/4" tenon won't align the rail and stile as well as a 1/2" tenon. Again, assuming a tight fit.

[Ole Anderson]

I have seen this in my panels and traced it down to the fence on my jointer not being exactly 90 degrees or the end of the rails not being cut 90 degrees.. Had nothing to due with my Bessy K-Bodies.

[ian maybury]

It's probably a matter of what's going on Todd.

All bar clamps with offset jaws by definition bow when done up - plus there has to be some potential for movement given clearance fits and wear etc too. The problem by definition gets worse with length.

It's hard to see how any clamp maker can eliminate the possibility - while they can play the trade offs between stiffness, weight and material cost, all will in the end broadly be subject to the same constraints. A badly designed or manufactured or low spec clamp will be more prone to problems of this sort, but there's no escaping the problem. My guess is that there's a lot of moderate out of flatness in doors being removed in wide belt sanders and the like.

Clamping technique has to be a big part of it. The bowing will be minimal in the case of short clamps, so it's necessary to taking account of this and the nature of the work call when it's likely to become a problem for you - and take steps to compensate. It's common to see parallel clamps flipped and used alternately on both sides of a panel so that any movement is cancelled. You likewise see self aligning clamping pads with curved backs used to overcome the issue. (think i remember a piece by Davis Charlesworth years ago on the topic)

Accurate/square joints that don't require heavy clamping forces to pull them tight are a fundamental requirement too - bending/deflection will increase with any increase in the clamping force.

Even a good quality parallel clamp like a Bessey can't avoid the physics - it's just a matter of degree. A question of whether there may or may not be a better compromise out there...

That expensive and specialist panel/door clamping systems exist as a result of the need to maintain flatness and speed assembly (they use various means to avoid the offset/bow inducing loads you get with parallel clamps) suggests that it's a stock problem with no cheap answer. There's stuff like this Plano system: http://toolmonger.com/category/85/adhesives/page/2/ - or much more complex and sophisticated examples: http://www.jamesltaylor.com/stile_and_rail_squaring

ian

[Mark Bolton]

If you make doors regularly why are you not clamping them up on some sort of purchased or shop made face frame table? It keeps all the doors flat and dead square with much less fumbling and no balancing act. Plus it's fast. You can assemble 20 doors in no time assembly line style.

[Steve Baumgartner]

This is an example of "buckling", which is the way a long object fails under compression if everything isn't perfectly aligned. It can occur using any kind of clamps that don't also hold the pieces flat, not just with Bessey or other "parallel jaw" clamps. The very short mortise and tenon on a cope and stick joint is especially vulnerable, but buckling can also happen when you glue up wider boards. You should always check with a straightedge and take corrective action before the glue sets up. While you are at it, you should also check for twist.

[Jeff Duncan]

My thoughts having glued up more doors than I can possibly remember with Bessey's....1) as the others said, double check all your setups to make sure your parts are being milled correctly. 2) very important...don't use excessive force with the clamps! If your milling is done correctly you just need enough force to pull the parts together nice and tight. If your trying to tighten the clamps beyond that your just putting undo stress on the assembly which is going to do exactly what you've shown. 3) lastly check every glue up after you have the clamps snugged down. I'll find doors that have not stayed flat on occasion and a quick tap with the dead blow will get them to lie flat again.

If you have these things in order you should not have any need for cauls or any other complicated setup.

good luck,
JeffD

[Phil Thien]

257345


My recollection was that Norm always used the clamps as pictured in your foreground example. He'd then set the entire assembly on the floor, using the clamps to provide stability to hold the clamped assembly upright.

You don't really need to worry about whether it is the way the clamps were intended to be used, or not. You just need to know that you're getting sufficient clamping pressure, and that the assembly is keeping flat.

So I'd try that method if I were you.

[Todd Burch]

Obviously, I take responsibility for a glued-up not-flat door.

I've considered all the comments and suggestions.

@Dave: Although I have made loose joints before, these copes and sticks were not loose. I think I took about 6 stabs to get the shims just right on these. It's amazing how two woods like poplar and cherry will require the use of different thicknesses of shims to get a good, snug fit. (Last time I used this cope/stick set, I was making poplar doors.)

@Ole: I don't always joint edges before profiling them. If the blade I'm ripping with is giving satisfactory results, I'll skip the edge jointing prior to profiling. In this case, I might have jointed 1/4 of the edges. I know my tablesaw blade is true (as true needs to be). However, your post did get me to thinking about my router table where I cut the cope and stick. I have an old CMT (Orange) router table / cabinet, and there are two concentric aluminum rings that work as guards around the bit. The smaller ring doesn't get used that much, but I do use the large ring, and I was using it the other night. AND... I was having to EVER SO SLIGHTLY, not really "raise" my stock to slide past a poorly adjusted set screw that caused the ring to be a bit proud of the table, but I did have to ease up on the downward pressure to allow the rails and stiles to slide past. It may be this ever-so-slight proud ring that caused the effect that Dave referred to above. With the rails and stiles being cut face down, if they were higher at the bit because the ring was higher than the table, then, indeed, when the two pieces met, they would have a tendency to create the effect I got. This may have been the issue that allowed the clamps to push the pieces out of plane.

@Ian: probably not a bad clamp, or bad clamps, as I was getting the same effect on all six doors. Some worse than others. The actual out of plane results I got was not noticeable when I was clamping (but I also admit I was not looking for it). However, it was noticeable when I started sanding the fronts of the doors. I always draw pencil lines across joints before sanding, so that when I use my ROS, I know when to stop. I stop when the pencil lines are gone. When I noticed that the pencil lines weren't sanding off (the door fronts), I realized that they weren't sanding off because that part of the joint was low. Then, I eyeballed the joint and realized that yes, indeed, the stiles were bowed up just ever so slightly.

@Mark: While my issue, now more thought out after considering these replies, is probably a user error that started at the router table, I will probably still invest in something like the plano system linked to, if it's something that will support large batches of doors without having to single-thread the clamping system.

@Steve: yes. The bow was so slight, I did not notice it while clamping.

@Jeff: I usually don't put undo pressure on joints, but I know what you are saying. I've actually gotten to where, when I'm gluing up a drawer box with finger joints, for example, that once I get all fingers seated via the clamps, I immediately remove the clamps and set the box aside to allow the glue to cure. (I do this partly, also, because when making a batch of drawer boxes, I don't have enough clamps to leave each box "clamped-up" through the curing process.

@Phil: Now that I think I know my problem, I'm thinking that alternating the clamp as I proposed might not be the proper solution for this issue, this time.


Thanks everyone for making me re-analyze the whole process.

[Mark Bolton]

@Mark: While my issue, now more thought out after considering these replies, is probably a user error that started at the router table, I will probably still invest in something like the plano system linked to, if it's something that will support large batches of doors without having to single-thread the clamping system.

Just a suggestion but you may start out trying a simple shop made face frame jig out of ply. We assemble face frames and doors on one. A 4x8 sheet, cut dead square, hanging on the wall leaning back slightly, 2" strips on on bottom edge and both sides, destaco clamps, and parallel clamps. Its a compromise compared to something like a ritter but its very fast and accurate compared to assembling on a bench, checking for square with a tape, balancing parts on clamps, and so on. After breaking down the parts one man can assemble, clamp, and pin, dozens of doors in no time.

Could of course make a small one just for doors if space is limited.

Kinda been mentioned but, I personally don't think there is any clamp setup, parallel or otherwise, that can clamp a 3/4" cope and stick dead flat on its own. Even with tight fit and a 1/2" tenon the joint is naturally going to want to roll to the beaded edge simply because there is a wider flat bearing surface on the back side of the joint. The front face, while it has the bead, only has a very small reveal giving less bearing than the back. Once the glue softens the wood fibers that front face is going to crush ever so slightly if you have to use clamping to close the joint. Shaker doors, for us, never want to roll.

We use very light clamping pressure and pin through tennons from the back. The door is in the clamps for a very short time, and never a problem.

[Todd Burch]

Mark, can you show a few pictures of your setup?

Thanks.

[ian maybury]

It requires knowing that your clamping jobs will come within a particular size range - but one way to keeping stuff truly flat while clamping is probably to use a strong perimeter frame, and to mount either pneumatic cylinders, veneer press screws, or some sort of ratcheting/jacking clamp through it's centreline. So that because there is no offset in the line the forces act through from the frame the clamping force will have no twisting or bowing effect. It'd need a flat surface (or at least a system of cross members) set up to be adjustable for height which would fit inside the perimeter frame - to provide a flat surface on which to assemble the parts before clamping...

[Mark Bolton]

Its not much to see, about as basic as it gets, but this is one table that's less than 8'. https://plus.google.com/photos/104773695898951990946/albums/5857079685308875537?authkey=CI_pyv_XgZidqQE

I just put new strips on this one so there are only a couple destaco's on there now. From moving the destaco's around they get full of holes over time. Its a pain and I have continually thought of mounting them on a T-track and all kinds of improvements but there is always more pressing things to be done and even as basic as this is it works really well and is super fast for assembling virtually anything that will go on it. I think on my last small batch of cabs for a laundry I put together around 15 complete in about an hour or so but again, they are only clamped for a short time. I was looking for a picture of that stack of doors but couldnt find it. Stacked like a deck of cards but as you say, a lot is in the machining.

The berries would be to put together a quick rolling vertical beam like on a ritter. It would eliminate the need to flip face frames during assembly and would allow for clamping a stubborn door flat to the table. If you imagine with this setup and pocket screwing face frames, your only able to clamp the joints on the bottom and right edges (against the fences) so you have to roll the face frame once during assembly, but its still very fast.

Without the beam you can still occasionally get a slight bit of roll if the joint is not perfect but its never anything thats been a problem. When they go through the sander its all good.

I just stuck a couple medicine cab doors (will get mirror) on there for example.

We dont do a ton of doors so investing a couple thousand or more into a nice setup, or even continually modifying this setup, is really not cost effective for me.

[john bateman]

If you put a dowel the same thickness as your door frame, on either side of the frame you're clamping, between the clamp jaws and the stock, then the clamp can only apply pressure at the very center of the stock. If there's still any out-of-squareness it would be caused by something other than the clamps.

[Mark Bolton]

Wouldnt that require that the clamp be fixed dead center in the clamp face as well? If the clamp face is 1.5" wide, the door is .75" wide, you would technically have to use a 1.5" dowel with the door sitting on .375" spacers to get the dowel centered in both the clamp face AND the door center? Beyond that you would have to allow for some quantity of slop/flexure in the clamp head, bar, and so on? We are delving into the realm of NASA's 250K crescent wrench, but interesting none the less.

I personally have never been satisfied with any of the parallel clamps. Spending nearly 40 hard earned smacks for a 24" Bessey is painful when they are so clumsy to use. I prefer the Jet's but even at that I would simply never spend the kind of money necessary to populate my shop with such clamps. The two Jets I have were free-bee's for a tool purchase. The two Bessey's I have (4 total) I purchased retail and regret it to this day. I use the Besseys only because I feel obligated as I wasted a perfectly good meal out at a fairly nice restaurant with a lovely lady when I forked over the cash.

[Todd Burch]

...I use the Besseys only because I feel obligated as I wasted a perfectly good meal out at a fairly nice restaurant with a lovely lady when I forked over the cash.

Well, did the Lady ever get over it? ;)

[Phil Thien]

I personally have never been satisfied with any of the parallel clamps. Spending nearly 40 hard earned smacks for a 24" Bessey is painful when they are so clumsy to use. I prefer the Jet's but even at that I would simply never spend the kind of money necessary to populate my shop with such clamps. The two Jets I have were free-bee's for a tool purchase. The two Bessey's I have (4 total) I purchased retail and regret it to this day. I use the Besseys only because I feel obligated as I wasted a perfectly good meal out at a fairly nice restaurant with a lovely lady when I forked over the cash.

I love reading stuff like this because I simply cannot justify the expense of parallel clamps.

The first time I saw them used was on New Yankee Workshop. Nahm used them for everything, from panel glue-ups to frame/panel construction to what not. Each time I saw him glue up a panel, and sit it on the floor with the clamp heads acting as a stand, I thought they must be pretty great.

I have an assortment of inexpensive (cheap) aluminum bar clamps that were sourced at Harbor Freight. They are okay, they certainly aren't parallel. They really don't have enough "oomph" to overcome a poorly-fitting joint. But that is probably good in the long run, as I'm forced to make sure all my joints fit nicely. My crappy clamps have forced me to be a better woodworker.

[Chris Fournier]

I've used Bessey K-Bodies for over 20 years now and I'll use them for another 20 years. They are fantastic clamps for most all applications especially casework, they will not however overcome crap workmanship or a poorly executed glue up (both scenarios for which I have been responsible). For all but the most basic glue ups I do a dry run and it is not uncommon to find a poorly placed or overtightened clamp that pulls things out of square. My bad not Bessey's! Bessys are not expensive clamps they are however excellent value.

[Chris Padilla]

I love my Bessey K-Bodies...have a wall full of them. I think most problems arise from overtightening as that is usually my problem. It is very easy to pull panels out of coplanarity cranking on the clamps too hard. Also, one must be semi-careful centering the clamps appropriately both on the wood and within the clamp itself so that the pressure is applied uniformly.

[Mark Bolton]

Well, did the Lady ever get over it? ;)
Well of course, I might as well have l bought the bessey's twice because Im sure I went out to dinner anyway. Hah

[Andrew Joiner]

I love reading stuff like this because I simply cannot justify the expense of parallel clamps.

But that is probably good in the long run, as I'm forced to make sure all my joints fit nicely. My crappy clamps have forced me to be a better woodworker.
Yup. So we'll change the saying to " you can never have enough crappy clamps"

I find it hard to pay way more for a plastic covered tool that does the same basic job,squeezing.

[Bryan Cowing]

I have the same problem, traced it to the cope & stick set up. I was not taking a deep enough cut when making the stile, so when clamped the stile tilted into the rail cut when clamped. When setting the shaper fence for the stile cut, make sure the cutter is removing some stock each side of the groove. Heres an example, the piece on the left is slightly flat on top. Your cut should be full like the one on the right. Another thing I learned the hard way,, the cut at the end of the rail has to be 90*. I had a shop made hold down sled and had a fixed 90* when made or so I thought. Out slightly, my doors came out un square. Once I made the sled ajustable I could set it to 90*, I didn't have to trim the doors to be square:D I now use a Shop Fox hold down sled.
257602

[Jeff Duncan]

Only thing I can add is that clamps in general, even the plastic coated Bessey's, are cheap! I've used mine hundreds of times and will use them hundreds more times, and years down the road when I'm all done, will still be able to sell them for probably close to what I paid for them....if not more! Not many other things in a woodshop that you can say the same for. It's like that famous quote "everyone knows the price of everything, and the value of nothing":D

JeffD

[Kevin Jenness]

I bring the joints together by hand and set the assembly on supports on the bench the proper height to center the clamp screws in the thickness of the door with the clamps sitting on the bench. I use blocks several inches long and as thick as the door between the bar clamp heads and the stiles. I check across, the joints with a straightedge and if the stiles are rolled I move the blocks up or down to compensate. Counting on the clamp heads to square up the joints on their own doesn't work for me. For thick doors I will often double up clamps, one up and one down for balanced pressure.