Horizontal boring for doweling

[Matanuska]

I use horizontal boring fairly frequently for edge joining boards with dowels. Up until today I've struggled to consistently obtain a flush joint. It didn't happen every time and it wasn't off by much, maybe 1/16", but I knew I should be able to do better. I had tried all the obvious things: Using a brad point bit to minimize bit wandering, making sure the boards aren't warped, tying the main table to the end table using the steel tubes, making sure the table height adjustment lock is solid so the table doesn't creep down, holding the board firmly to the table when drilling, and using the telescoping legs to minimize table flex. I use dowel cup centers to locate the mating hole locations and figured out early on that using the rip fence as a brace when I tapped the boards together to mark the holes wasn't a good idea.

When the problem happened again today I thought through all those thing again and had an "Aha!" moment. Here's a picture of the configuration that worked:

image0.jpeg (447.23 KiB) Viewed 28927 times

The issue was apparently table flex related to an inconsistent amount of pressure I applied when pressing the board the main table. Up until today I had installed the telescoping legs on the far end of the machine next to the end table. I realized today that since I only have two telescoping legs, they should really be located on the main table end so when I applied downward pressure it was supported by the legs. The wood I've been joining is old and hard and without firm downward pressure the bit pressure tends to rotate the main table. After I made this change I got consistent flush joints. Finally!

Anyone else have a similar experience or a different technique?

[RFGuy]

Anyone else have a similar experience or a different technique?

The observations you stated and more are why I don't like horizontal boring on Shopsmith equipment. Yes, I know, I am a heretic on this forum for saying this. I am sure if I had some kind of industrial machine for horizontal boring instead, I would love it. One of my biggest complaints is alignment, vertical and horizontal for this operation on a Mark V. You have to kinda lean over the table to sight the horizontal and there is parallax which sucks. On the vertical you have to kneel down to see it, or also parallax. Then even if you get it perfectly aligned there is table and quill flex, not to mention bit walking unless you do as you describe. It is a tedious and time consuming operation IMHO. Don't get me wrong, you can make beautiful creations using the horizontal boring on a Shopsmith if that is all you have for joining. I much prefer my Festool Domino and if I didn't have it, I would continue to use my Jessem doweling jig or my biscuit joiner. Both superior, in my opinion, to horizontal boring on Shopsmith equipment. Only have to align one axis with both (because of the reference plate), you can bring the tool to the work and sight right down on top of the workpiece so no parallax and easy to see when you are on the mark.

Have you tried telescoping legs both fore and aft of the fence, if you happen to have four of these lying around? Wondering if there is still table flex even with them on the fore side. Just a thought of something to try.

[Matanuska]

Two more telescoping legs would do the trick but I don’t have them. Might have to consider a purchase.

[larryhrockisland]

The telescoping legs are on sale, I got the email today.

[JPG]

HOKAY, just gotta flap about this.

IF(And I think tis so) the "table flex" is actually mounting tube rocking in the carriage bores(and/or) carriage rock on the way tubes) then I think attaching the telescoping legs at the far end of the MAIN table is the better location. That is where the downward movement needs to be minimized.

Now re dimpling with the dowel centers, Squeezing them together will be a more gentle method.(look out for summer ring growth!)

Now with the dimple, aligning the bit center becomes a matter of 'dropping' the bit into the dimple while making H/V alignment adjustments.

Now quill play can maybe be minimized by snugging the quill lock.

IF one is more attuned to locating by measurement, a router bit makes an excellent flat bottom 'bit'.( does not wander IF workpiece IS held secure ). That can lead to misalignment due to many more potential causes however. Reliable jigging is a must.

[Matanuska]

IF(And I think tis so) the "table flex" is actually mounting tube rocking in the carriage bores(and/or) carriage rock on the way tubes) then I think attaching the telescoping legs at the far end of the MAIN table is the better location. That is where the downward movement needs to be minimized.

Now re dimpling with the dowel centers, Squeezing them together will be a more gentle method.(look out for summer ring growth!)

Now with the dimple, aligning the bit center becomes a matter of 'dropping' the bit into the dimple while making H/V alignment adjustments.

Now quill play can maybe be minimized by snugging the quill lock.

Thanks for this - all valid points.

In my case I think the table movement is likely to be flexing of the entire table support system since the table is only supported by two tubes whose orientation doesn't resist side to side motion well. I believe the reason I was able to remedy the problem by placing the telescoping legs on the near end of the main table is because I've been applying significant downward pressure on that end of the table to make sure the wood is seated squarely against the table and fence. Without this downward pressure I agree that the bit force would tend to push the far end of the main table down. I'll have to do some tests with the legs on the far end of the main table and only slight downward pressure on the wood and see if this works as well.

I have noticed that the brad point now centers consistently on the dimple using my new setup. This is an easy check I didn't pay enough attention to previously.

Also agree squeezing to set the dimple is preferable if possible. Unfortunately this wood is so hard it's difficult to get a good dimple that way. I've found it easier to move the wood to my workbench where I can brace the boards up against its back rail and tap gently with my dead blow hammer. I've also observed the significant hardness variation between growth rings that you mentioned, enough to skid over the twist drill bits I was using initially.

From what I can tell quill play is not an issue. Any tighter on the set screw and the quill wouldn't move freely.

[JPG]

From what I can tell quill play is not an issue. Any tighter on the set screw and the quill wouldn't move freely.

NOT the set screw, the quill lock.

[RFGuy]

Without some kind of preload on the table system, I do think we might be asking too much of the telescoping legs and table system here. Those legs aren't some kind of rigid structural member. Even if you get them perfectly square (orthogonal to table in x-y) to the table, the rubber feet will squish a bit on the floor. I think there are too many variables with the connecting rods movement, variation in the telescoping legs connection to the connecting rod, possible slippage of the telescoping part, etc. IMHO, all of these factors can be enough to throw off this operation resulting in unexpected movement while boring. IF you like sanding and/or don't mind vertical misalignment in horizontal boring for joining two boards, then it is fine. In my opinion, some type of adjustable support between the upper waytubes and the bottom of the main table would be preferred if you want to investigate and/or fix this deficiency with horizontal boring on the Mark V. Just a thought...

[BuckeyeDennis]

When I need to maximize table rigidity, I slide the main table immediately adjacent to the extension table, and tie them together with the connector tubes. That by and large takes connector-tube flex out of the equation.

The table system now resembles a parallelogram, with the tables and connector tubes at the top, the way tubes at the bottom, the main table posts on the left, and the aux table posts on the right. Now, even assuming that the “corners” of the parallelogram all have some angular flex, the top and bottom remain parallel, and have negligible change in separation distance for small angular deflections.

In this configuration, the vertical position of bored holes should be pretty repeatable. But the depth could still be affected by longitudinal flex of the table system. But even the longitudinal stiffness is greatly improved, as compared to using the main table by itself.

I’ve honestly never found my 520 table-system rigidity to be insufficient in this configuration. But I will admit to grumbling to myself almost everytime I need to set it up, because of the extra time required.

[RFGuy]

Dennis,

I agree with you that it helps quite a bit, in my experience with my Mark V when I have done horizontal boring. Two main points I made earlier are that other systems, e.g. Domino, biscuit joiner, good quality dowel jigs make setting the vertical alignment bulletproof because there is a fixed, usually rigid, reference plate. Instead of aligning in 2 axes, you are now only aligning in one (horizontal) with them compared to Shopsmith horizontal boring. Reducing alignment from 2 axes to only 1 means less chance of human error, tool slipping, etc. Also, heck of a lot less fuss and less setup time with those in comparison. IF you don't mind the extra setup time and aligning two axes simultaneously instead of one then by all means go with horizontal boring for joining. I know there are a few applications where horizontal boring is a real benefit, but if all one is doing is joining two boards, e.g. making panels for table tops, etc. then it seems like a lot of hassle, to me, compared to conventional woodworking techniques/tools that I mentioned above. In short, I think the horizontal boring feature has been oversold by Shopsmith for years. Again, I know I am in the extreme minority on this forum regarding this.