Wadkin Resurrection: Back Miter Fence

The previous post in this series dealt with the first stages of milling the rough casting of the protractor head. In this post, we look at the back miter fence, the rough casting for which was obtained recently from a dedicated Wadkin fan out on the west coast.

As I mentioned in that previous post, the design of the quadrant, that is, the pairing of independent front and back miter fences, seemed to lend itself to the processing of miter cuts for various polygons, while the later Wadkin one-piece cast mitering fence set at a 90˚angular relationship between front and rear fence faces was only going to be especially useful for 4-sided polygons (a tetragon, to use the technical term). Now, the 45˚ miter is the type of miter cut that will be executed by 99% of woodworkers in 99% of cases involving miter cuts, so as far as that goes having a one-piece miter fence is no great disadvantage at all.

Still, I do process a fair amount of other polygon miter cuts, and wanted the functionality promised with split front/rear miter fences to achieve the capacity to readily cut polygons from 3-sided (trigon) to 12-sided (dodecagon).

Ah, such hopeful imaginings. I took the rough casting for the back miter and decided to stick it in place, mocked up as it were, to see what was what. That’s when I discovered, to paraphrase a line from a Bruce Springsteen song (“Racing in the Street”) that, “all my pretty dreams had been torn”.

The way the pair of castings are configured does not allow for nearly the swing range I had imagined. The swept arm which allows the fence to be clamped at a given setting can only swing so far inwards before the end of that circular arm runs into the back face of the front miter casting. In order to get in much further than a 90˚ relationship between the two fence faces, the length of that swept arm needs to be on the shorter side. Anything beyond that, the base portions of the castings will run into one another. I found the tightest placement it appeared to be capable of would likely make it possible to cut the miter from a trigon, at 60˚ between fence faces.

Swinging back outward, the short swept arm now presents a constraint as to how far the back miter fence may be swung open while still being capable of being bolted tight to the front miter fence. This is as far out in swing as it looked like it would go:


Keep in mind that this is not the normal relationship of the paired fences for miter cutting. In this case I have the front miter fence in its the normal 90˚ relationship for cross cutting and have simply placed the back miter fence against it to check the relationship between the two fences. If I were to fix the fences together as you see in the above picture, obviously I would risk slamming the blade into the back miter fence, which would not be among the list of desired activities in my shop.

I reckon that the widest setting between fence faces I could obtain would be on the order of 108˚, which would allow pentagon miters to be cut. The back miter fence can open plenty wider, but then it cannot be clamped to the front fence, and given the lack of capacity on the sliding table for affixing hold downs, it would involve some head-scratching, and maybe an extra locking bolt hole in the table, some sort of pad upon which it could rest, and provide a means of fixing it all in place.

It’s an advantage to have the two fences capable of being bolted to one another as a unit, as that makes for greater convenience for those cases in which one is cutting unequal miters due to the framing members being mitered having different dimensions from one another. For example, you might have a 2″ wide frame to joint to a 2.5″ wide frame. With a front/back miter fence paired and connected to one another, then all one needs to do is leave the pair bolted in their same orientation and swing the assembled pair of fences out of the regular cutting angle to the blade to a position which produces the unequal miters. If the front fence and back fence are to be instead bolted independently to the table, then resetting for unequal miters becomes much more of a PITA.

An alternative idea would be to make a second version of back miter fence which would have a special longer swept arm and which could be placed in position when you want to cut with the fences more obtusely related than 108˚. If I went that route, I’m not sure I would go the way of wooden pattern and sand cast and then hours of milling. It’s not efficient, especially when there are good alternatives these days.

So, as a stock factory option, the front/rear miter fence pairing gives a functionality of cutting polygons in a range from 3-sided to 5-sided. That’s better than being able to cut only 4-sided polygon miters, but only very slightly better.

You know, as mentioned above, a vast preponderance of miter cuts are going to be for 4-sided right angled frames. Now, it seems to me that the next two miter angles one would expect on the basis of ‘likely to be done’ would be for 6-sided (hexagonal) or 8-sided (octagonal) frames. And the stock miter quadrant set up does not provide that functionality. As it only provides the capacity beyond the 45˚ miter to cut 3-sided and 5 sided frame corners, which are rather unlikely choices it seems to me, it might as well really only be a fixed unit giving the standard tetragon’s miter. Better to have a fixed 90˚ miter fence where you can be sure it is set at the right angle, no pun intended, than to have a device with fences which can move independently thus opening up a greater possibility that an error will occur in setting up which will lead to undesirable results, not to mention the inherent inefficiency of having to perform a secondary step to set the angular relationship between fences after the fence halve have been set onto the table. Sigh!

Still…what I want is a miter cutting device which allows me to quickly establish the angles required for any polygon up to 12 sides. The factory design and execution of quadrant does not allow for that. So, if I want that functionality, then I’ll have to find another way beside the stock set up.

I’m working on that other way presently, and have some percolating ideas which I have to exercise tremendous self restraint not to just spill the beans about. I’ll save that story for when concrete steps get taken towards that end. Right now I’ve settled upon a design approach and have executed a sketch which is most of the way to completion now.

In the meantime, I have this rough fence casting which I paid for, and is of no use until it is machined to final form. I’ve decided that I will machine it to produce an accurate replica of a factory back miter fence, which I will then have as a very desirable accessory for the saw, and the connected pair of fences will be perfectly usable for regular miters, which are going to be the most frequently tackled. So, it will still be useful. A rough casting is frankly not going to be useful to a lot of folks, unlike a machined one, which is at least going to be useful for cutting the standard miter.

So, on to some more cast iron machining. This back fence can be clamped in my Kurt vise, which is most helpful. What remains unhelpful is the shape of the piece, from the inside of which I must obtain the perfect back miter fence. There’s nothing to reference from to start, so it is unwise to mill too much off of any surface until you get your bearings and have a plan, so to speak.

The back miter fence millwork is slightly complicated by the fact that, in order to fit well to the front miter fence, three separate surfaces, at three separate heights, all have to be co-planar with the front fence, and all meeting at the same time. If one surface meets ahead of the others, then tightening the fence down with the lock bolt will introduce distortion to the casting which may throw it out of square or cause other unforeseen headaches. So, it seems I have one chance to get it right, though it seems like even if I get some of it wrong there are probably remedies involving shims or special washers that could address problems. I’m hoping though to get it right the first time without having to take such ‘plan b’s’.

I went initially with an orientation for the part which was easy to locate and clamp in the vise, and then took a couple of skims off the lower edge of the fence:


With the same clamping position, I also skimmed the underside of the pivot:

Another view:

Not shown was the taking of a skim off of the swept arm portion. Once that was done, I flipped the fence over, placing it on the cleaned and flat-milled surface, atop a pair of parallels, and clamped it using a brass block in the middle of the vise jaw:

A pass or two also off of the top of the pivot:

At this stage there is plenty of extra material on the pivot boss for adjustment, and the fence section remains about 0.06″ fat of the mark.

My next step will be to deck the clamping pad on the front miter fence, and obtain a height which I can match with later millwork on the back miter fence:

That’s where things stand for now. More to come. Thanks for tuning in!

2 thoughts on “Wadkin Resurrection: Back Miter Fence

  1. I am really enjoying this. Plan B is for people not committed to plan A. The book of five rings. Miyamoto Mushashi Your kind of book.

Leave a Reply