Page 14 of 17

Re: Constructing the Martin Lewis Little Engines Northern Te

Posted: Wed Jun 08, 2016 7:46 pm
by Carrdo
To push on and to finish these couplers, the knuckle assembly is the last major component, a lot of which I have described previously.

However, they were made using more or less the original template. See the first photo.

For these couplers to lock and unlock positively in any scale, the the coupler knuckle head has to have a very specific profile and the coupler draft gear assembly has to to provide sufficient swing and clearances to allow both horizontal and vertical movement between cars, as well as twisting, bumping, shock damping etc.

Examination of both full size and other model live steam couplers made in the past and present indicated that a slightly altered knuckle head profile would aid in this goal. The second rather poor photo shows the slightly revised profile.

Re: Constructing the Martin Lewis Little Engines Northern Te

Posted: Fri Jun 17, 2016 3:32 pm
by Carrdo
Machining new coupler cover plate patterns.

Re: Constructing the Martin Lewis Little Engines Northern Te

Posted: Mon Jun 20, 2016 10:17 am
by Carrdo
Richard is also producing some of the same couplers except he is using his CNC mill whenever possible.

Here are some photos of the CNC machining of the knuckle head profile from various blank sticks. The cutter is a 1/8" dia. ball end mill and the blank is indexed every 5 degrees in the rotary head. To finish the coupler head profile from the cutter ridges, one uses a smooth file.

It is a lot of preparation to produce the CAD drawing and the computer program, but is it ever nice to see the machine do all of the work and not make a single mistake.

Also, once the program is written, it is so easy to make any adjustments to the coupler head profile if found necessary to do so.

Re: Constructing the Martin Lewis Little Engines Northern Te

Posted: Mon Jun 20, 2016 10:29 am
by Carrdo
A final photo showing the coupler head production machining setup.

Re: Constructing the Martin Lewis Little Engines Northern Te

Posted: Wed Jun 22, 2016 3:01 pm
by Carrdo
Just some further photos on machining the new coupler cover plate patterns.

As the four exterior sides of the cover plate patterns are finished 3 degree tapered (for casting) as well as the interior opening, the last two exterior sides have to be machined on a sub plate as shown in the first photo as the pieces cannot now be held directly between the vise jaws as the first two exterior sides have been machined tapered. One has to think about and plan how these pieces are to held right from the beginning as only the top and bottom surfaces of the pattern remain flat and parallel in the end.

Ensure that everything is square before machining. One can easily be deceived when using a square against a tapered edge as it can appear that one is out of square when one isn't (or the opposite) if the blade of the square is running up or down on the taper even slightly.

Re: Constructing the Martin Lewis Little Engines Northern Te

Posted: Wed Jun 29, 2016 6:19 pm
by Carrdo
The final operation on the knuckle/tailpiece assembly was to machine the lead in and out cam on the knuckle tailpiece.

The original builder partly machined and partly filed this cam but I came up with a special eccentric holding fixture which could be used with my rotary table.

I was a little apprehensive if the fixture would work as designed but it did (even much better). See the first three photos.

Time to assembly all the parts together to see how they worked.

These couplers still had a few tricks up their sleeves!

The knuckle/tailpiece assembly pin hole now wouldn't quite align with the machined coupler body pin holes due to the casting draft around the knuckle pin bosses fouling the ends of the knuckle. Time to get out the miniature drum sander recently purchased just for a job like this. But you don't want to remove any more metal than is absolutely necessary here as it affects the strength of the coupler and I wanted to leave a little more metal in this area than the original builder did.

Then, I found that the knuckle assembly, once fitted, would not rotate the full amount needed to fully open and close the knuckle due to :

1. the tailpiece was scraping the coupler half casting internal pocket and

2. the knuckle head base circle where the tailpiece was silver soldered to it was fouling the internal half coupler pockets due to a small change which I made in the internal pocket configuration to add more corner wall thickness in an area which I noticed was an issue on some of the original couplers.

The first issue was easy to spot but the second one wasn't.

A little belt sanding of the knuckle tailpiece solved the first issue but the internal pocket configuration needed to be changed slightly for the second issue to finally be solved.

A further complication arose when I noticed that the coupler heads were a little short (needed to have more engagement when the couplers were closed) based on the original knuckle head template so a new longer knuckle head template has been produced and all future knuckle heads will be made longer.

The last photo shows the first fully functional new coupler but it has the short head which will be replaced. At least I know now that the couplers will work as intended with the modifications made to date.

This battle isn't over yet as I have yet to finish a matched pair of couplers. This will involve opening up the mouth of each coupler as they have to closely fit inside each other and yet have sufficient clearance for the knuckles to be able move back and forth slightly, for the couplers to be able to rotate/swivel substantially and for the coupler bodies/knuckles to be able to twist slightly all at the same time.

One cannot produce a simple casting and expect it to work properly. There is a lot more to it than that.

Re: Constructing the Martin Lewis Little Engines Northern Te

Posted: Thu Jun 30, 2016 12:22 am
by Harold_V
Carrdo wrote: One cannot produce a simple casting and expect it to work properly. There is a lot more to it than that.
I expect that that wouldn't be true if one were to use the investment/shell molding process, where fine detail isn't an issue, nor is tight tolerance. I've seen non-ferrous castings made that held .005" tolerance. A true work of art.


Re: Constructing the Martin Lewis Little Engines Northern Te

Posted: Thu Jun 30, 2016 3:31 pm
by Carrdo
Hi Harold,

Maybe I should have clarified that I was talking about basic sand casting.

Yes, I agree that investment (lost wax) or centrifugal casting will produce extremely fine detail but have you priced what the minimum orders are for the number of pieces we as live steam modellers use?

Back in the 1970's I was quoted a minimum order of $800. for a small lost wax caster to even look at the project and another foundry recently said, when I enquired, that for any type of bronze casting, they needed a minimum order of 700 lbs. of bronze and this was not a large foundry.

Today, I am extremely happy if I can find anyone who will cast anything for the number of pieces involved.

My caster is another model engineering club member's home foundry and I am extremely grateful for the care and quality of work he does for the number of pieces involved. Really unheard of. So as challenging as the technical problems are, the economic ones are even worse. I would love to have the couplers cast in cast steel, but just forget it, it won't happen.

Bronze is bad enough and don't start asking about the alloy or the quality of the finished pieces; you take what you are given or go away empty handed.

Anyway, I had perhaps hoped to produce a few finished couplers and market them here - they will not be cheap or as an alternative, I could supply castings only with all of the instructions and technical support on how to produce a finished coupler. One would need machining and silver soldering skills as well as be able to produce the basic fixtures, tooling and templates involved. But not until I have all of the issues resolved as one can see from the posts made to date, this is not a 1/2 hour project.

Re: Constructing the Martin Lewis Little Engines Northern Te

Posted: Fri Jul 01, 2016 3:32 am
by Harold_V
Wow! I had no idea of the price, nor the required volume, commercial shops would demand.
Don, I have a long history of working with the jewelry industry. They use investment casting routinely, often casting silicon bronze.
If you are of a mind, it might pay you to investigate the equipment used by benchmen. There are small casting machines on the market that would allow for making the parts you are now having sand cast. The mold for the waxes could be made of rubber,and there are ways to ensure reasonable quality with rubber molds, although metallic molds for the waxes are preferred. We have one such mold maker on this board, Doug Edwards.

I make mention of this only because you are making gestures of producing these items beyond your own needs. If you got seriously involved in their manufacture, I expect you'd save considerable time, with no reduction in quality. All in all, a win/win situation.

In regards to cast steel, you'd probably be well served to just cast in ductile iron, as the problems are fewer, and the tensile strength of ductile rivals that of mild steel. It's a very nice alloy, which machines with the ease of gray iron, and pours easily. It does require the use of an induction furnace (a cupola can be used, but the resulting metal must be desulfurized before converting to ductile), however, so it most likely would be beyond the scope of your project. In this case, I'm just thinking out loud.


Re: Constructing the Martin Lewis Little Engines Northern Te

Posted: Mon Jul 11, 2016 2:16 pm
by Carrdo
While I am still struggling to finish the couplers (I still need to use Richard's digitally equipped mill to make the revised drilling fixture for the new coupler knuckle heads), I decided it was time to start on the tender tank itself.

The first question is what material to use in its construction?

While there are a number of possibilities, which have been thoroughly discussed here on Chaski, I again decided to employ structural or hot rolled steel on the tender tank.

My reasons for using this material again are (and you can always disagree):

- it's the devil I know,

- it is relatively easy to bend, form, machine, silver solder or weld,

- it was more readily available (for me) than other materials as I had some of the hot rolled sheet steel on hand already,

- it is reasonably priced (although nothing really is these days),

- it was used extensively by the big three American locomotive builders to build their locomotive tenders so if it was good enough for them, it is good enough for me,

- no matter what tender material is employed, it will end up with all sorts of holes in it (for water lines, rivets, the internal tank frames and bulkheads, etc.) which will have to be sealed and made absolutely watertight),

- there are many sealing products and systems now available to protect steel surfaces from rust and water leaks which, if properly applied to a properly prepared steel surface, make that surface virtually watertight for many decades.

The tender floor will be constructed from a piece of 1/8" thick plate and the tender tank sides, internal bulkheads, etc. from 1/16" thick material. The internal tank frame will be made from structural 1/2" by 1/8" angle iron.

In the original construction articles Martin Lewis used a 1/16" thick formed copper sheet flanged with a 1/2" lip all around for the tender bottom made much like a copper boiler tubeplate.

Re: Constructing the Martin Lewis Little Engines Northern Te

Posted: Fri Jul 15, 2016 9:40 pm
by Carrdo
Don't ask what I am doing here.

A less than common facing method when one is out of options as I have no equipment which will swallow this size of bar. This facing method is painfully slow but it works. I am using the finest automatic cross feed on the lathe. Both ends of the CR steel bar will be faced, lathe centers put in and then the bar will be turned between centers. The bar is 1-3/4" square and will be 12" long (finished length).

It does have something to do with the tender tank construction but is about two degrees of separation from it.

Gary Gillam likely knows what this is the start of.

Re: Constructing the Martin Lewis Little Engines Northern Te

Posted: Tue Jul 19, 2016 7:17 pm
by Carrdo
Moving right along with all of this insanity.

To put in centers exactly in each end of the bar, see the attached photos.

I do have an 18" long parallel test bar made for me commercially many years ago for my lathes so this was used to exactly locate the horizontal center on the bar end. Two surface ground spacers also had to be made to set the correct height for the bar and the correct offset horizontally from the parallel test bar (as seen in the first two photos). The bar end also had to be set dead square to the faceplate which was initially used to face each end of the bar.

I used a series of center drills starting with the smallest to gradually increase the size of the center hole which ended up being a shade over 5/16" diameter at the top of the cone. I didn't make the center hole any larger as with the increasing size of the center drill, the lathe speed had to be reduced drastically and even at the reduced speed, the largest center drill would start to vibrate and chatter a bit when the size of the hole was a bit over 5/16" dia. and I did not want to use a lot of force to continue the cut and throw something out of alignment. I also had to sharpen all of the center drills used which can be a fun exercise by itself.

The center drills initially were held in a drill chuck but I soon switched to holding each size center drill in a collet for the center drilling operation. I also marked which bar side was "top" and which side was "rear" (the one which was set against the angle plate) so that when I switched the bar end for end, the marked sides were in the same relative position. Of course all nicks and burrs, any raised edges etc., must first be removed from the bar before doing any of this.

I probably achieved putting in the bar centers to the degree of accuracy the bar was rolled to but this normally 5 minute operation took a week to perfect.

I can now turn the bar between centers, roughing it out in my big lathe and finishing it in the 9" SB lathe which is the more accurate of the two. My 13" SB toolroom lathe is WWII vintage (made in June, 1942) and I need to shim the tailstock to get it to turn parallel. Another job I have been avoiding for a long time but now will have to undertake.

It will soon become apparent what all of this is going to be used for.