Frames and frame stretchers

[neathabbeyironworks]

I am seeking general advice in regards to assembling plate frames. I am fully aware that accuracy is very important when assembling frames. The prototype I am modelling is the 0-4-0 0-4-0 Garratt, William Francis. The frame stretchers on the original are steel plate bent at right angles. There is also a horizontal plate between the frames secured with angle iron. The front an rear buffer beams also secured with angle iron. I am going to have to depart from the original, as the bent plate will not be accurate in my view, and 13mm angle iron is the smallest on offer. So my questions boil down to how to fabricate rigid accurately dimensioned frame stretchers, and how to join plates at right angles to each other? Thamks in advance for your expertise. 

[steamer5]

Hi Huw,
I had the same dilemma with K1.
First plan was to cut the plates to shape, less the thickness of plate return on the edges & have strips TIG welded on. Worked great & looked fine, only issue was dimensionally not accurate! Two things could have helped here, make the returns thicker so there was machining allowance… hindsight is great, 2nd a jig to hold the parts tightly together for welding, we only used magnets.
Second plan have them folded as a trough…. You need to work out what stretchers are 3 sided return & what are only two & combine into as many as you require, once again getting the dimensions you require & the returns truely at right angles is a challenge, unless you are setup to do it at home
Third plan, get plate slightly over the required thickness machine to out side dimension & thickness, then start producing swarf by machining out the center! I went down this route for K1’s stretchers. I tried it on the smallest stretchers to see how it would go, it takes time but you get there…. eventually! By the time I had finished I had in excess of 30 litres of swarf…. collected it in 10 litre tubs! I used a face mill to carve out the majority of the material, then smaller end mills to tidy up the inside of the returns & to give a small internal radi in the corners. I left any holes & shaping until the end, it’s kinder on the face mill!
A fourth option would be the slot & tab method, having got laser or water cut plates, but I was unsure on how this would go for strength given that the returns would end up with an open side, I’m sure somebody can tell us

To give you a bit of an idea, the largest horizontal plates for K1 only have returns on the sides against the frames. These started out once machined to size at 3.4 kgs, once the extra thickness was removed, holes & cut outs added they are down to 0.75 kgs!

If you like I can send you some pictures

Cheers Kerrin

PS you are lucky to beable to get 13 mm angle, best I can do is 20mm, luckily it’s 2.5mm thick & scale is 2.35. I set up my bandsaw with a couple of fences & trimmed the majority of it off the milled the last bit, which was fine for the engine frames. The boiler frames have the same size angle holding the running boards on, 1 pierce each side approx 1400 long so approx 6 meters to trim off, I’m going to sort the fences on the bandsaw so I only have less than a millimeter to clean up!

[firebird]

Hi

We recently completed a pair of electric loco's shown here on this forum. One problem I had was the steel angle was not a true 90° so don't trust that it is. When it came to bolting the frames together the out of true angle started to pull the frames out of square. I ended up skimming every piece of angle to square it up. These frames are 1/4'' thick and 30'' long but still pulled out of square. I imagine your frames will be thinner so will twist more easily. I have recently completed a part built CHUB (an eBay) I found the same problem on that, the frames were out of square. I had to remake all the angles to true it up.





Hope this helps

Cheers

Rich

[uuu]

I had to mill all the angles on Jessie to be square, too.

It helps a lot if you can get all the holes in the plates and angles accurately placed. You mention accuracy, too, so I'm not really saying anything new. Then when you loosely assemble the frames, there's nothing pulling things off-square; nothing fighting against you.

If you can make up some rods dead on the width between the frames, threaded in both ends, you can bolt these into any spare holes (bracket holes, brake hanger holes, etc) to help hold things true as you tighten up.

Wilf

[racinjason]

I drew mine in fusion 360 sheet metal to produce a DXF file for the laser cutters with the flanges, then with a 90 degree carbide counter sink I V'ed the part where the bend needed to be leaving about .2mm then the folds were very accurate. My Z17 build on March 24 there is a post that shows How I did it.
 Cheers Jason.

[brucevoelkerding]

I agree with Rich and Wilf - plan on milling the Angles square. Another Item I have found handy on my 3-1/2" Gauge Frame Builds is a piece of 1/4" dia Stock which is just short of the Length of the Frames. It is tapped 5-40 on both Ends (any 1/8" dia OD Thread would be OK). I bolt this tight thru the front Draw Hook hole and then thru the rear Drag Beam slot as a Clamp. I use this to hold the Frames together while aligning everything and match marking. It is much less obtrusive as compared with using Clamps and much lighter.

[David]

By all means mill the angles square - certainly worth doing. But I'm wondering if complete and absolute squareness is a requirement. Consider how much movement and stress there will be when the loco is together and moving, heating up and cooling down, etc.

The loco I made has 3mm water jet cut frames and 10mm water jet cut horncheeks bolted to the frames - no machining done at all on the running faces. The frame stretchers are 3mm steel with small angles made from hollow square section steel cut up and riveted to them, then spotted through the frames for bolt clearance holes and put on with nuts and bolts. There's no way any squareness found its way into that setup, although I did my best.

I had to use an eccentric bush somewhere in the coupling rods, but really the thing went together fine and seems to work. It runs on less than 10 psi of air and goes well with steam on a ground level track.

I'd say the build up of my machining inaccuracies in all the many bits that go into a loco were a much bigger problem than the frames being a bit skewed or wonky. And none of it seems to have mattered much.

It's not like it's a solid unsprung chassis for an OO scale loco with top hat bearings soldered in place.

Of course do as well as you can, but my guess is most locos are a bit wonky.

[neathabbeyironworks]

Thanks for these replies.The suggestion to use a V cutter is a good one along with making the flanges slightly oversize and milling them back to dimension. Does the bend at the V have to be welded up? Would MIG welding minimise distortion? 13mm angle was obtained from Rapid metals in Redditch. Managed to bring them back with me from UK to Oz, in a plastic pipe, 8 lengths a metre each. This angle seems to be 90degrees but will check each time I use in future.

[Neale]

For folded plates like this, at least in thinner steel, I CNC machine to shape (Solid Edge sheet metal design) and machine a light V-groove on the fold line. This is intended mainly as a marker rather than a "weak point" to define the fold position. I then use my little vice-mounted folder - one of the two-piece 100mm designs available from a number of suppliers - and use the engraved line to define the folder die position. The edge can be "felt" into the groove. Then just crank up the vice jaws to give a 90deg fold. The process isn't super-accurate but it is fairly repeatable so with a couple of test pieces and tweaking the bend allowance in Solid Edge you can get pretty good results. But I also aim to be very slightly oversize to allow trimming on the mill.

Above 2mm or so steel, though, I tend to redesign with a tab-and-slot variation and machine separate flat pieces to silver-solder together.

[racinjason]

I V’ed mine almost right through then folded the bend and silver soldered the joint, this produced a very accurate assembly. I had all the rivet holes laser cut into the plates and they all lined up perfectly when assembled.
Cheers Jason

[steamer5]

Hey Jason,
What thickness of material please? Did you make any allowance for the bend, if so how much, for the rivet holes?

Cheers Kerrin

[racinjason]

Kerrin,
I have done the same on 3, 2 and 1.5mm sheet, I try to leave .2mm of the sheet thickness when I V it out. For say a 2mm sheet with a flange 8mm high on the inside I put the centre of the V 8.9mm from the edge. I did do a couple of trial cut to test the lengths and test corners I hope that makes sense.
cheers Jason.