Lynn's Box Joint Jig Size Errors

[biggles99]

Hello,

I have made a plywood version of the Lynn's jig following pretty closely to Lee Styron's plans. In the time since I have tried using it, I am getting inconsistent widths of the pins. For example, even if I go 16 half turns, I get a spacing of 1/4" pins to not be exactly 1/2" apart from each other. It seems to make the next pin a bit wider apart than 1/2". Also, the width of the slot is not always exactly consistent either.

I have tried a few different ways to use jam nuts to hold the whole sliding carriage in place, and it really seems like it does not have side to side play, but I figure it must.

My leading guesses of what could be going on?:

1. Because the handle is hard to turn, which I thought would be a good thing, because of the slight difference in the two t-nuts maybe not perfectly spaced, could the t-nuts be distorting the threads of the 3/8-16 threaded rod? And after I have made a few cuts and keep going back to the same spot to do the next one, am I distorting the same threads over and over? Maybe I need to get "GRADE 5 or 8" threaded rod (if they even make it...)

2. The whole jig is not rigid enough and is flexing somewhere and it creates variations

3. I still don't have the two ends of the rod held rigid enough to the base, so even though I think there is no play, there probably is?

4. Cheap 3/8-16 threaded rod isn't exactly 16 threads per inch exactly? (I don't think this is true, but I'll throw it out there

5. Cheap t-nuts are not good.

I am planning on bringing my dial calipers home from work to measure stuff more accurately as I turn, but I'm curious about other's opinions...

Thanks,

Tim

Also: thanks for the plans on your site, Lee...

[Stytooner]

It does sound like you have the carriage t-nuts too tight in relation to each other. I would just shim one out a hair or just loosen these up with a flat tip screw driver. I use candle wax on the threaded rod as a lube. It makes everything glide easily. The end of the threadded rod at the handle is the critical point. It needs to have no slop here. This is why this point has the jamb nuts. To get it precise.
I have a new version of the Lynn's jig. It is a lot beefier. It has a 1/4" ply sled under it. I am not sure when I can get up the plans, but you might consider putting such a sled under yours to help with the flimsyness.
Here are a couple of pics of some of its assembly.

[biggles99]

Thanks very much for the info. I will start by looking into the general flimsyness, I think. Thanks for the pictures, too! Much appreciated...

[Tony Caro]

Help!

I too have built this jig and am getting less than accurate results. It would be a lot easier if my saw could take dado blades but alas it can't so I have a blade which has a reputed 2.8 mm kerf but I believe it to be a bit more (approx 3 mm). The jig I built is very sturdy (using 3/4 inch plywood mainly) and I used regular nuts sunk into the timber rather than t-nuts but am yet to get any consistent results. There is no slop in the jig that I can discern, I can only think that the inconsistent saw blade (which I thought would cancel itself out) or my counting (I am mixing metric with imperial measurements) is throwing me out.
Has anyone come across a way to easily record fractional turns on this sort of jig?

[Black wallnut]

Tony, perhaps you should just start a new thread about box joint jigs. What do you have for a table saw and what blade are you using? Does your combination have problems that is causing you grief, such as a bent arbor or too much runout of the blade? Why does your saw not take a dado blade? Please do not bring back to life threads that are so old.

[Stytooner]

I just found this link today and put it on my LJ page as well.
Pretty detailed info. Should really help to clear up a lot of jig issues.
I admit I didn't do as good a job preserving and presenting the plans as I would like to have done. I didn't know how to use Cad back then. Do now and maybe one day I will get to it. May not need to though if this other gentleman comes through first.

http://worldofwood.proboards.com/ind...nt&thread=1509

[LCHIEN]

I do not have a Lynn's jig. However I can make these comments:
1. 16 tpi for threaded rod, even cheap ones is usually quite accurate, there will be 160 threads in a ten inch piece.
2. Each Turn moves the jig 1/16th inch or .06125 inches. This should be quite consistent given the above. Its important not to go backwards, just forwards or you will have "backlash" which is the slop of the threads relative to each other, might be .005". If you stay going in one direction then the threads will stay in contact.

3. You can get finer gradations of turns. One method I have seen used is to mark the corners of a hex nut on the shaft. If you note the starting handle position (as a time on a clock face) and come say two turns with the handle to the same clock position, then observe the nut position and go two more corners, you will have completed 2 and 2/6 turns. Each turn is .06125 and each 1/6th turn is .010" approx.
So in this simple way you can get .010 resolution. Remember only go one direction and do not overshoot and go back if at all possible. If you want more resolution then you can put a rotating disk on the shaft and mark it in tenths of a turn and use a fixed mark as an index. Really, I simply use 1/8th turns in my head for my planer and my table saw using the clock position method, I go N full turns and then I can easily do 1/4, 1/2, 3/4 turns as 3, 6, 9 hours, and also 1/8, 3/8, etc pretty easily. My saw is 12 turns per inch and my planer is 16 turns per inch. Just remember how much each turn is and if you use 6, 8, or ten (D) divisions that each fractional turn is 1/D x 1/tpi. Gotta do a little math but you only have to do it once.
4. an inexpensive digital caliper in SS are under $15 now, available at Rockler, Woodcraft, and Harbor freight. I have had pretty good luck with them - I have several since they are so cheap. They even have ones that do both fractions and decimals and metric but they cost a bit more. Clicking the mode switch will convert units. I recommend getting the stainless steel (SS) body calipers not the plastic ones.

[Tony Caro]

Thanks Loring, some good tips there.

[capncarl]

The version of Lynns jig that I made eat up a lot of wood before I was satisfied with the results. The slop in the threads is consistant with the mecanism on a worn out milling machine or metal lathe. You do not necessarily start from "0" on the crank, you start your measure when the slop is taken out of the threads, and do not go backwards. I took a good finger joint piece and cut it down and attached it on the stationary frame and installed a pointer on the moveable frame to use as an indicator. That way I know where I am at on the cut. It is quite easy to get confused and loose count and botch up the whole piece. I like the idea of using the dial caliper. It would help a lot.
A safety item I installed was a chunk of wood on the back of the jig as a blade guard as the jig passes over the blade. The guard has a stop that catches the table and prevents me from pushing the jig past the blade and leaving it exposed. It is tempting to push the jig on through the blade to look at your cut before returning.

After you get confident with your jig you can cut a large stack of wood in one set up, so build your jig deep enough to handle it. 4 set ups would be required for 1 job.
capncarl

[Stytooner]

To always start on zero, you can adjust your handle. My zero point was always at the 6 o'clock position on the handle.
There will be no backlash at all if you continuously turn in one direction. Loring stated a pitch accuracy of about .005" per foot. That seems about the norm for threaded rod. You will not see that in your material spread out over a foot length.
You can also eliminate some of the inherent backlash in the carriage if you want to by using two large nuts on one end of the carriage with a spring in between. it can be done to some extent without that by pressing the nuts toward each other during assembly. This way if you do screw up and have to back track, it doesn't throw you out as much.

There are two ways that I am aware of to be able to cut all fur sides at once. you have to cut the setup twice with either.

First you stack all 4 together. Then flip pieces and rotate 180 degrees when doing the other end.

Second way is to offset two sides or front and back the distance of a finger. Alternate that offset to the other two parts when doing the second pass.

There have also been gauges designed for these that can be either attached to the rear of the jig or your backer board. I will try to locate those and post a link.

[Stytooner]

Here are some printable scales that can be attached to the back of the jig. I couldn't locate the ones originally designed for this, so I drew these up.

http://leestyron.com/pdf/Box%20joint%20scales.pdf

[Tony Caro]

Thanks Capncarl and Lee again.
I've done some more test cuts and found that my counting has been a little off and that has accounted for much of the error. Remember I am not using a dado blade and I'm having to juggle saw kerfs as well as metric to imperial conversions as well as fractional revolutions of handle turning. Despite all this, I managed to make a box but I miscounted at one point resulting in a more narrow finger than was called for.
I have noticed no backlash on my jig but I desperately need some way of counting turns and using some sort of gauge to indicate progress during the cuts. With the loud saw whining it's easy to lose count. Anyone have any pictures of an easy way to gauge cutting progress?

[Stytooner]

The PDF I just posted can be printed out and used on the back of either style jig. You will need to print a new one each time, but should work well.

[Tony Caro]

The PDF I just posted can be printed out and used on the back of either style jig. You will need to print a new one each time, but should work well.

Lee, I may well take your suggestion up but I was hoping to build something into the jig that is more permanent.

[Stytooner]

I haven't actually used the scale myself, but if you applied each of them to identical hardboard backers, they could be swapped out when you cut different sized fingers. In that case, you could laminate them and they would be permanent on the backer. The backer could be screwed to the back of either style carriage. Keep it up high enough so that it isn't cut by the blade, but is close enough to use as a measuring tool. Might need to remove any jig stop temporarily so it would be able to be slide past and expose the blade for position check.
I like to use a piece of this in the carriage itself to help prevent chipping and tearout.