The Home Machinist!

I always fear the last 10% of the construction of anything as this is usually where I make serious mistakes. It has happened many times before.

Anyway, to press on.

The first photo shows the quarter rounding operation on the bending head plate bottom edge as this edge will be the one in contact with the 1/16" thick steel tank sheet as it is being bent around the 1.5" diameter wrapper bar to form the 3/4" corner radii on the tender tank. You want to have a radiused - not a sharp edge here. And mark which side of the bending head bar is the top or bottom as it is easy to get it reversed.

Now for the tricky bit.

The bending head plate has to be precisely located/drilled/bolted between the bending head end plates to closely follow the wrapper bar to make the 90 degree corner radius. Actually, it has to be able to rotate slightly more than 90 degrees in order to compensate for any spring back in the sheet metal. That is why the jig base plate has a center cutout to allow for this. Since it is the bottom edge doing the bending on the bending head plate, it is this edge which has to be on the center line of the 1.5'' wrapper bar centers and not the center line of the bending head plate. In addition, the bending head plate has to be raised the 1/16" thickness of the sheet metal to be bent (as well as + 0.002" to give some minimum clearance).

The bending head plate also has to have a cutout at each end of the plate as shown in the first photo so it will clear the wrapper bar bearing end plates. In addition, the bending head plate has to be square to the feed table surface and the 1.5" diameter wrapper bar itself. As well, the gap between the wrapper bar and the bending head plate has to be uniform along its entire length and not be tilted in any way.

That is a lot of things one has to get right.

The light weight, made up threaded rod clamps are the right ones need for this operation. Initially, I had the end plates clamped to the bending head center plate with a large furniture makers clamp (as seen in the previous post) which totally unbalanced the whole setup as it was so heavy and cantilevered out on one side. Of course disaster ensued, everything tipped over and the result was the broken file seen in the second photo.

One of the last operations and one of the most complicated to setup. I had to use every trick in the book to make this one work.

Firstly, I did not have enough vertical travel on the mill to install the drill chuck, a drill and to set the work on the table of the mill. Had to hang the work off the side of the mill's table and support it on the machined ways of the knee.

Secondly, 4 separate parts had to be clamped in place to make the setup work. The forming head assembly wants to rotate all over the place now as it was designed to do this. But it mustn't for the drilling and tapping operation.

Thirdly, the setup must leave room for all of the parts to be aligned, checked for square, parallelism and have a uniform and on size gap. So one cannot just clamp things anywhere without considering this aspect of the setup.

It was "interesting"

Finally finished but not without incident which I won't go into.

To drill and tap the opposite end plate for the two 3/8-16 NC cap bolts, the entire complicated setup had to be reversed! See the first two photos.

In addition, I found that I needed another extra large toolmakers clamp, which I didn't have, to keep the end plate exactly where I needed it to be for the drilling and tapping operation. So, spend $$$$ as well.

The finished basic unit is shown in the last photo. The bender works as designed AND IT SHOULD after all of the misery it has put me through.

Will add my own touches such as handles on the bending head center plate to give added hand control while bending the tender tank sheet metal and to drag the beast about as it is heavy.

That seams like a lot of work for just four corners. I did that by using a vee block made from angle iron. Carefully lined up and pressed in the arbor press. Works well In fact I made four tenders with rounded corners using that method.

Hi Dave,

Yes, it involves a lot of work to make this bender but the advantage with it is that it is very difficult to make a mistake in the bending operation with it.

Perhaps you could explain with a few additional pictures your method?

I understand the principle behind it but I would like to know exactly how you locate and set the bend location and how you prevent the flat sheet from moving or sliding out of position when bending it in the arbor press?

Setting something up " carefully" as you describe it could involve many little tricks that I and others have yet to learn.

Perhaps there will be another video on this?

Back with some additional work on the 3/4" scale couplers. These things are fighting me to the bitter end but I have been in this situation many times before so there is nothing new here.

My old camera died recently and I am just learning the basics on the new one so my pictures may not be the best.

After silver soldering some two piece knuckle assemblies together, nothing would fit or move in a coupler body as I added too much solder as I wanted to ensure a strong joint.

Had to employ the setup shown in the second photo to get rid of the excess solder while still leaving a 1/32" root fillet for strength.

The first photo is of the knuckle head - tailpiece joint with the excess solder and the last photo is after the excess has been removed.

This should not happen if one uses the right grade of solder for the job and if one takes care about what one is doing but sometimes things don't work out as planned.

Looking back on my posts, I don't see any description or photo of how the two piece knuckle/tailpiece is aligned and held for silver soldering these parts together.

Well, just design and build another fixture!

And then there is the lead in cam on the tailpiece to allow for a smooth locking/unlocking of the tailpiece in the lifting pin slot. On the original design, this was filed but...

Well, just design and build another fixture! (which goes on the rotary table)

The metal bender setup and ready for action (which it won't see for some time) as now I have to profile the tender tank which will be made from a single piece of sheet steel to give as few seams as possible (as per the original M. Lewis construction notes). The single sheet will be 7-9/16" wide by 62" long by 1/16" thick. This is going to be an interesting job to profile such a large thin sheet and Martin Lewis doesn't give any instructions on how he did it (the original was made out of copper).

Frustration!!

The last time I posted regarding coupler construction, I thought that construction of the couplers was nearly finished. Not so!

When I tried to assemble a matched pair of couplers, I found that the couplers would not lock or mesh properly. I found that the overall length of the knuckle head was about 1/32" too short (or about 1/16" overall per pair). In addition, I noticed that the nose profile of the knuckle head could be altered slightly to give a more positive lock while at the same time allowing for some slack when the couplers were coupled together as per the originals.

Next, it was discovered that the coupler knuckle tail piece profile needed to be changed slightly as small changes here made a huge difference in how these couplers function (lock and unlock). This only became apparent with a matched pair of couplers were assembled.

All of the above meant that weeks of work now had to be repeated as new, slightly altered, knuckle head templates and tail pieces had to be produced along with all new knuckle heads and tailpieces. Existing coupler knuckle pieces just became so much expensive scrap. See the first photo. As a side benefit, the fixture used to hold the knuckle head and tailpiece in the correct relation one to the other for silver soldering together needed to be modified. This wasn't a particularly easy fixture to make to begin with as much trial and error was involved. Fortunately, I had original coupler parts which gave the necessary alignment.

Then, to compound the problem, I had the misfortune to jam and break a drill on the first new modified knuckle head produced (on the very last drilling operation) which absolutely refused to move without destroying the knuckle head. At that point, I quit in disgust.

It is not often that a New Year's present comes along but today it did. By drilling through from the opposite side of the knuckle head nose (the last operation repeated again), I was able to "push out" the broken drill pieces (by pure luck) and salvage the part. See the second photo - the broken drill is that minuscule bit on the vise closest to the viewer. The new revised tailpiece is the one on the left -it looks almost the same as the two unmodified tailpieces to it's right but...

So now we are back to about where we were four months ago. Better luck this time ??

Success at last(!!!) but these coupler parts fought me to the bitter end and with some of my own mistakes thrown in for added frustration.

The first, somewhat blurry, photo shows three of the new silvered soldered complete knuckles. It was at this point that I discovered that one of the tail pieces had been silver soldered in backwards - very easy to do but I knew about what could happen and had checked the unsoldered assembly twice to avoid this mistake....

Then, the next two tail pieces would not silver solder to the knuckle heads!

Did I use some leaded steel by mistake- I was certain that I hadn't and after a very bad few hours, I discovered what was wrong. As I had not been doing any silver soldering recently, I did not remember my own advice (oft repeated here in the past), the parts were not hot enough.

To repeat what I have found out before, with any silver soldering I do:

I only use structural (hot rolled) steel as it silver solders very well and gives a very strong structural joint. I do not trust any cold rolled steel as many grades have a high lead content and I have never been able to successfully silver solder any leaded steel (especially if one needs great structural strength). Others disagree but I will stick to my guns...

I have found one needs to heat, with hot rolled steel parts, (I use a propane flame only) to a much higher parts temperature than red hot. For me, I have to heat the parts to a very bright orange heat with a touch of yellow heat which is about 300 degrees F higher than a red heat. Especially when using a non cadmium silver solder and

One has to heat more than just the local joint area. I find that I have to heat the entire parts to have enough general reservoir heat to ensure that the silver solder will flow freely through the joint.

Black or white flux - no difference.

There it is, one has to sock the heat aggressively to the parts until the above is achieved.

The second photo shows three new couplers (the bottom two are a new matched pair and locked while the upper two are not a matched pair and are unlocked - the one on the upper left is an original coupler and the upper one on the right is new as I wanted to check that the new couplers are interchangeable with the originals - and they are.

Still to do is machine the dummy hollows on the side of the new coupler body castings as per the originals and to silver solder the two couplers casting halves together to make a solid coupler body.

The last photo is of all of the bits and pieces needed. Some of the more important bits and pieces have labels as shown in the photo, otherwise I tend to forget what they are for. Not getting any younger.

The three photos referred to in my previous post.

Whew! That coupler adventure was a fight for you!!!! Thanks for sharing your trials and tribulations.
I now have renewed patience for the machining challenges I discover on my projects.
Happy New Year!
~RN