Boring Problems

An Eccentric Anomaly: Ed Davies's Blog

I've been making some progress on the house but it's all been painfully slow. The main problem has been with boring holes through the largish bits of timber with sufficient accuracy to match up with the existing steel work for reasons I've discussed previously.

The summer having been abnormally wet, with only a handful of days with no rain at all, somewhat chilly and quite windy hasn't been exactly helpful, either. On the other hand, there have only been a few days where it's rained all day meaning nothing could be done. At least the wind has kept the midges down most of the time.

This drilling is the sort of thing which is pretty easy to do in a workshop but a bit more tricky using battery powered tools on a Scottish hilltop. I can fairly easily get the drill within 2° of perpendicular to the surface using various simple guides but that's a lot more error than is acceptable. 1° is 1 in 60 so 2° is 1 in 30 which results in a possible error of just over 2 mm when drilling half way through a 145 mm log and nearly 5 mm going the whole way through.

The first lot of holes I've been doing are through the bottoms of the 18 stub (464 mm, to go at the sides of the A-frame) and 18 main posts (3066 mm, to go at the corners of the rooms) for 16 mm bolts to go through the 18 mm holes in the tongues of the base brackets. They really need to be more accurate than 1 mm though if they're out by 2 mm left or right it's not too terrible, so long as they're consistently off by the same amounts in the same direction.

A stub post:

…and 8 of the 9 on the north side of the house with a main post (the one on your left as you walk from the kitchen into the living room):

I've got most of the posts done just about satisfactorily now. I messed up a few (5 stub posts and 3 main posts) but have chopped up one of the messed-up main posts to make good replacements for the missing stub posts so just have a few more main posts to do.

However, the purlins will need holes going the whole way through to the same level of linear accuracy so will need twice as good angular precision. Also, each will have four holes rather than two, increasing the need for consistency and opportunity for error. For this reason I've been spending my time on the last few of the post bases trying to improve things.

Here are some experiments with offcuts from the purlins I did in the middle of July.

First, with just one bit of timber it seemed to go quite well. The “purlin” is on end here with the gable bracket on the left and the bolts though the timber lining up with the holes in the backing plate quite nicely.

Then I tried with two offcuts to “simulate” two purlins meeting at the top of a main post. The gable bracket sits happily on one side with the bolts going through…

but when we look at the backing plate on the other side the bolts don't line up:

The bottom pair are offset sideways relative to the top pair:

Not by a huge distance but plenty to prevent proper alignment.

I tried this sort of set up a few times with different methods of aligning the drill. The best method seemed to be to drill through a backing plate using a try-square as a guide, aligning the drill with the blade of the try-square in two axes by eye. Putting the try-square on a backing plate rather than directly on the timber seemed to work best but still I couldn't be sure of being accurate enough always. With the purlin timbers costing £90 (+VAT) each it'd be nice to not destroy too many.

I seriously considered just flipping the back plates over and drilling (or getting drilled) new holes to align with the actual locations that the bolts come out. I'd have painted the inside of the new holes with cold-galv spray but I'm much less paranoid about the galvanizing on this metalwork than on the post-base brackets as they'll be inside the insulated envelope of the house and well clear of potential damp in the roof. The structural engineer didn't specify any particular coating for them though I'd assume something would be required.

That's still an option if one or two don't go right but I'd rather avoid that if I can.

As well as guiding the drill by eye with reference to a try-square I also tried various field-expedient drill guides. This ply- and softwood sandwich worked reasonably well in itself but needed long bits which could wander more than I'd hoped.

This little bracket with angle aluminium evostuck on also worked reasonably well, was more convenient than the big big one above and allowed shorter bits to be used but still couldn't get consistently good results and the aluminium wore out quite quickly so I finished up making a few before giving up on the idea completely. (Here I'm just starting on the second hole in the first of the main posts I did.)

Still not really improving on the possibility of 2° errors though so I got a Wolfcraft drill stand, seen here set up to drill the second hole in the first of the reworked stub posts being made out of one of the messed-up main posts.

That needs a drill with a 43 mm collar to fit. They don't make battery drills like that any more (DeWalt used to make NiCd ones, I believe). My old Black and Decker mains drill doesn't have one, either, so I had to get a new one and a 1500 watt pure-sine-wave inverter to run it off the van (really not wanting to mess around with a generator).

Even that wasn't terribly consistent but yesterday afternoon I ground the little “nibs” off the collar on the drill so it sits more firmly on the stand which I think should get things under control. I'll give it a try when it stops raining - which is beyond the reliable reach of current forecasts, it seems.