A Ming-inspired Cabinet (86)

It’s been more than a month since my last post. The reason for the absence is simple: I’ve been sick as a dog.  In truth, I can’t really say how sick dogs might get or why we even have that expression, but I’ve been sick with some kind of virus for a full month. This is highly unusual for me, as I almost never catch a cold, however this illness has been widespread in the Northeast it would appear with many folks getting sick. My wife and son have also had it bad, but I have had it the worst. I spent a solid week in bed, then another week staggering around and seeing stars, losing weight, and filling snot rags with all kinds of interesting things. In the third week there was a glimmer of a feeling of getting better, and I even made it to the shop for a few hours here and there, but by the weekend I slipped back and came down with a brutal fever of 103.6˚F (nearly 40˚C) for a whole day, which was a day spent more or less shivering, or alternately sweating, in my bed. Quality time with family, as they say.

I’ve been to the doctor twice during this time, but no antibiotics, fortunately, have been prescribed. As I wind my way towards the end of week 4, I can say that I am definitely on the mend, maybe back to the 50% mark as of today.

Anyway, when I did make it into the shop a couple of weeks back I got to work on milling and drilling the ends of the bifold door hinge stiles for their mounting pins.

As per usual, the mill was the tool of choice, initially just for marking out the centerline of the drill holes, where the ease of fixing the stick to the flat table was a help:

 

 

The crossed lines were then dimpled at their intersection with a Starrett centering punch:

Result:

 

The process was repeated on the other end, and then again on the other three hinge stiles.

The holes which get drilled into the ends of these stiles have to be well aligned with the axis of the hinge rod. The better the alignment, the more perfectly the hinge will work.

Then the vise was mounted on the table, trammed into alignment, and then the right angle attachment also mounted, and aligned to the vise. I used the Starrett centering punch in the drilling chuck to align the spindle to the mark, and then fitted a drill and got to work:

A closer look:

Three steps were required to get to the required size of hole, starting with the brad point bit above.

Then the same alignment to spindle allowed for the self-tapping threaded insert to be fitted using a bolt with its head cut off, and nut and washer placed, as the driving tool:

The mill’s chuck should mean that the insert gets fitted in correct alignment from the get-go. I slightly loosen the chuck to allow the bolt to spin, and then use a wrench, or spanner as the Brits like to call them, to get the insert on its way:

Once the insert fully seated, the mounting tool is easily removed:

That takes care of the upper end of the stile. On the lower end, a different arrangement is realized, and a counterbored hole required. I drilled the outer one first with a 0.75″ Forstner:

Then I use an end mill to bore the main hole (a different stick is shown below than in the previous picture):

Here’s the brass mounting pin being fitted up:

Strangely, the flanged portion would not fit in the counterbore:

I checked the flange diameter, and it was spot on. That left the Forstner:

 

I hate it when a tool like that is not correctly sized. I have had that Forstner set for many years, so it is possible that when i had it resharpened at some point the sharpening shop decided to dress the outside of the cutter, for some reason, and that left it undersize (?). Or maybe it was always undersize, right from new?

In any case, I went over to a neighboring shop and borrowed a Forstner which was the correct size:

The counterbore was cleaned up with the correctly-sized bit, and then the brass hinge pin was then able to be fitted all the way:

These pins are not an off-the-shelf item, but were rather designed by me and fabricated at a local machine shop to my drawings.

After the brass pin is slipped into a place, a countersinking drill is run into the end of the stile, with the pivot pin as a guide, and then an assembly screw with Square Drive head was placed:

Assembly screws are coated with a lubricant, and unlike a decking screw do not have ridges under the screw head to aid in seating into wood, so they will fit cleanly to the brass.

The completed assembly:

 

However, while the party just seemed to be getting started, save for the mild sidetrack with the undersize Forstner bit, I started to notice a problem with the 3rd one of the brass inserts. Look at the picture above, and you’ll perhaps spot the issue: the brass part is not centered on the mass of the wood. It had been marked out on centerlines to start, but something had happened during the various cut out steps.

Somehow the hole had gotten drilled off center, and that was vexing. This was not the worst of it though. I investigated and discovered that the right angle attachment on the mill was no longer aligned with the jaws of the vise. So, while I could center the spindle with the centering point to the tick to start, when I chucked in the tools to follow, which were shorter, they drilled off the mark, and worse, they drilled holes which were not axially aligned with the stick. And what’s the most important thing with these holes? That’s right, that they be axially aligned with the stick.

I was a tad cranky to find that this was the outcome despite many precautions and careful work. I took some time to realign the right angle attachment and cinch the allen head bolts down about as tight as I could. I started telling myself that perhaps I had not tightened them sufficiently the first time around….

I plugged the long drill holes in the lower ends of the stiles, which was a pain as I do not have bubinga dowel stock -not something widely available – and besides, most commercial dowel stock isn’t cylindrical, regardless of species, at least not by the time you find it at the store. So, I made my own, and without a lathe, this is a tedious task. At last though, the three holes were plugged and I re-drilled.

But, to my considerable chagrin, by the time I was again working on the third one I noticed, during a double-check, that after I removed the centering point and chucked in the end mill, that the end mill was clearly not meeting the end of the stick centered on the same point. I had that feeling like I was starting to lose my mind – how could this be?! I also checked the metal threaded inserts with their associated hinge pins, and also found they were not properly aligned.

Everything was going sideways and pear-shaped, as they say. I’m not sure what the expression exactly refers to, but I was in a world of misery.

Re-checking, I found that the right angle attachment was again out of alignment. Grrr…. I checked the mounting bolts and they were still plenty tight. Hmm…. What is going on here?

I loosened the bolts and re-aligned the attachment, and then after cinching down again, did a test, turning the spindle on and off a few times. I discovered the from the first start up the angle attachment went out of alignment by a slight amount, and with each start up, the effect compounded. Basically, the torque of starting, when transmitted through the right angle attachment, caused it to go out of alignment. It was a bug not a feature, and certainly not the type of bug that could be lived with or be optimistically called a feature.

I realized that having modern soft start on that motor would be really nice, as I’m sure that would solve the problem, but I don’t have that feature at this time. I called it a day and slinked off home to lick my wounds.

————–

It’s a drag that the right angle attachment on this mill doesn’t work seamlessly. It is a 46-year old machine, so I guess I can cut it some slack. It might have worked great at some point and now simply be worn to the point that it no longer will hold position. I’m speculating. It’s a pity the attachment does not index to the quill somehow, or could be more rigidly locked in place.

I thought it over, considering the carnage, and decided that I needed to remake the hinge stiles from new, from raw stock again, despite the considerable number of hours I had into them. This was not an easy decision in some respects, mostly due to existing bubinga supply issues in my area. The place in Connecticut where I had bought sticks from in recent months seemed to have material a bit prone to more movement than I would like, which I think relates to how the wood is dried, and their stock was low anyhow. The local hardwood dealer in Greenfield hasn’t had much bubinga for months. Berkshire Products, a 2.5 hour drive away, specializes in wide slab lumber and doesn’t have much in the way of 8/4 S4S stock.

I did have one last stick on the shelf, a 96″ long, 4″x8″ beam that I have been hoarding for a few years. Though I was loathe to cut into it, I felt it was my only option really. Ugh! It’s not completely irreplaceable, but getting more 4″ thick stock in bubinga likely means going to Germany for it. So, the term ‘irreplaceable’ isn’t so far off really.

A few days on now, I have prepared 4 new hinge stiles, just barely squeezing them out of the portion of the beam I crosscut. Here’s three of the four:

They still need a couple of things done to them, but they are at the point where they can be drilled for the hinge hardware, so that’s the next step.

A couple of days ago I did a test on the mill, dropping the speed down to 2800rpm in the thought that this would mean less start up torque, which hopefully would preclude the angle attachment from further movement. No such luck. After 10 start/stop cycles, the centering point had shifted over 1/16″. And I couldn’t tighten the mounting bolts much more without risk of thread galling or bolt snapping. It just didn’t seem like the mill was going to work for me in this way.

The rest of that afternoon was spent exploring various alternate solutions to the challenge of drilling a straight hole, such as working the stick vertically off the end of the table (however the stick is too tall and would actually require a hole be cut in the floor, or the 5500lb. mill raised in the air, to make room under the drilling chuck), tilting the head of the mill over (spindle ends up too high relative to the work table and there is not enough room left on the table to support the stile), seeing if a router jig could be made to work (however not all the cuts are done with router tooling). It was a head-scratcher, and after several hours futzing I had accomplished nothing.

A slot mortiser would be handy to have for this job, but I don’t have one. I’ll put it on the Christmas list I guess.

A phone call to a friend later that evening was productive however. He suggested finding some way to rigidly fix the right angle attachment to the column of the mill. I had considered this solution previously, but had discounted it, largely because I was thinking that the right angle attachment had to be free to move in other directions – a conceptual error on my part. In this work, the table moves, and the angle attachment stays in a fixed relationship to the machine column.

I had a glimmer of hope.

So, yesterday I headed into the shop to explore this solution. First off, I stopped at the hardware store to pick up some longer bolts with which to tighten the angle attachment sleeve clamp, along with a 10mmx1.5 tap to clean out the threads. After the thread holes were done, I wiped the inside of the sleeve and the outside of the quill with solvent to remove any lingering lubricants. Then the angle attachment was set aside.

Re-tramming the mill head was next, since I had previously explored the idea of tilting the head over to horizontal. The tramming is a tedious task and took about an hour to complete. Then I mounted the Kurt vise to the side of the work table and aligned it to the x-axis. Finally, I mounted the angle attachment and lightly clamped it. Using a transfer punch as a test bar, I aligned it to the back jaw of the vise using a feeler gage and bump-and-check method.

Then I acquired a piece of plywood off-cut from the guys upstairs – a piece of what is called HDF, a plywood with MDF faces – and formed it into the shape I wanted. I found that a piece of 8/4 Canary wood off-cut could be clamped directly to the tops of the column ways, and this then is the basic configuration I am setting up:

Another view:

The plywood will be clamped to the angle attachment soon enough.

A few things to be done on that fixture yet, but I took a side step to set up the indicator and check how close to true to the x-axis was the angle attachment alignment, previously roughed in with using a feeler gage to the back jaw of the vise:

Sliding over, the discrepancy appears to be about 0.0016″, which seems decent:

 

The feeler gage gave a pretty close result it would appear.

I then thought to give the chuck a spin to see how round the transfer punch was:

 

Not so much.

Checking the other end:

Rotate:

 

So, hardly what you might call a proper ‘test bar’. Not something I could rely upon for anything other than transferring marks. When you try to be precise, the number of things that can trip you up or affect the result definitely multiply.

A proper tool for my mills tool-holding system, which uses DIN 2080 (ISO 40 taper) is a test arbor, which looks like this, by the way:

 

That however is something to obtain in the future – maybe when the time comes to add the soft start to the main motor. For the time being I just need to find something more accurate than the transfer punch rod. I think a piece of 1/2″ (12.7mm) or 5/8″ (15.875mm) drill rod should suffice, and on Monday I’ll traipse over to the local machine shop and see if he has something suitable kicking around. I’m sure he’ll have something which will do the job.

All for this round. Should be posting more regularly now I’m feeling semi-human again. Post 87 follows.