NSW 'B' class mogul

[Deleted]

I love how you made your brackets David...I require a lot of brackets myself and I think your method will make life much easier...

Cheers

Pete

[Jim]

That's a very neat way for pressing the retaining bracket David.

Jim

[David]

Thanks guys, I was pleased how well it worked and how simple it was.

I had a couple of days off after the disaster that was Saturday but had an idea yesterday about how to do the lubricator tank with the CNC machine so tried that out today.

The idea is to cut 45deg V grooves at the corners as fold lines. I tried it on a scrap of 1.6mm brass of the correct length, about half height. It worked very well. The sides were about .020" over because I left a small amount at the bottom of the V groove.

I'm cutting the real part now including cutting the profile from a larger piece of brass. It's sounding a bit dodgy because I didn't consider the fact that it just plunges the end mill at the end of each tab and it's a 4 flute tool. So that's not great. It makes the 2deg entry taper a bit pointless!

Here is the test piece. You can see how close the clamps were; I did a lot of checking with the tool in my hand to see it would clear them.

Profile cut: 4 flute 3.5mm end mill, 1500rpm, ~0.4mm depth, 90mm/min.
V grooves: 2 flute 45deg chamfer mill, 1500rpm, 0.25 depth, 60mm/min.



Cleaned up the Vs and just drawing a file over the long edges.



The folds were really easy to make. I'm not clever enough to make the 4th side meet up cleanly so that needs some fitting.



The plan is to silver solder the corners on the real one at the same time I solder the bottom on.

EDIT: Of course that plunging broke a tooth off the tool. I figured it would as soon as I realised what it was doing!

[David]

The test went better than the real thing but I think it will still come out better than if I'd tried to bend it myself.

The problem was the usual culprit of the stock lifting. Luckily I'd only specified the V grooves as 1.2 deep, in 1.6mm material. I did mean to only leave .2mm but as it is the cutter went right through on one groove, and mostly through on another.

I'll bind it up with some wire when I solder it and see how it goes.

It's a damned nuisance the way this thin stock behaves and I still haven't found a reliable way of holding it down. I tried double-sided tape over the weekend on a different experiment and it was useless. But it wasn't good quality, I haven't found any good stuff yet.

Hindsight shows I could/should have done it like the test piece, manually cutting the stock to size and just grooving on the CNC. Wouldn't have broken a tool that way either. For that matter, I could have done the grooves on the manual mill!

[Roger]

Nov 20, 2018 8:06:34 GMT David said:

Thanks guys, I was pleased how well it worked and how simple it was.

I had a couple of days off after the disaster that was Saturday but had an idea yesterday about how to do the lubricator tank with the CNC machine so tried that out today.

The idea is to cut 45deg V grooves at the corners as fold lines. I tried it on a scrap of 1.6mm brass of the correct length, about half height. It worked very well. The sides were about .020" over because I left a small amount at the bottom of the V groove.

I'm cutting the real part now including cutting the profile from a larger piece of brass. It's sounding a bit dodgy because I didn't consider the fact that it just plunges the end mill at the end of each tab and it's a 4 flute tool. So that's not great. It makes the 2deg entry taper a bit pointless!

Here is the test piece. You can see how close the clamps were; I did a lot of checking with the tool in my hand to see it would clear them.

Profile cut: 4 flute 3.5mm end mill, 1500rpm, ~0.4mm depth, 90mm/min.
V grooves: 2 flute 45deg chamfer mill, 1500rpm, 0.25 depth, 60mm/min.



Cleaned up the Vs and just drawing a file over the long edges.



The folds were really easy to make. I'm not clever enough to make the 4th side meet up cleanly so that needs some fitting.



The plan is to silver solder the corners on the real one at the same time I solder the bottom on.

EDIT: Of course that plunging broke a tooth off the tool. I figured it would as soon as I realised what it was doing!

Hi David,
That's a really neat way of doing it. Why not do exactly the same, but use a piece of Brass that's say 10mm longer than you need and machine five folds. When it's machined, just break off the short end ones and you will know that all of the sides are the same. Then you won't have to file it. If you machine complete Vees at the ends, you know that the tool won't deflect, changing the length on those ones.

[Roger]

Nov 20, 2018 8:50:36 GMT David said:

The test went better than the real thing but I think it will still come out better than if I'd tried to bend it myself.

The problem was the usual culprit of the stock lifting. Luckily I'd only specified the V grooves as 1.2 deep, in 1.6mm material. I did mean to only leave .2mm but as it is the cutter went right through on one groove, and mostly through on another.

I'll bind it up with some wire when I solder it and see how it goes.

It's a damned nuisance the way this thin stock behaves and I still haven't found a reliable way of holding it down. I tried double-sided tape over the weekend on a different experiment and it was useless. But it wasn't good quality, I haven't found any good stuff yet.

Hindsight shows I could/should have done it like the test piece, manually cutting the stock to size and just grooving on the CNC. Wouldn't have broken a tool that way either. For that matter, I could have done the grooves on the manual mill!

Hi David,
A couple more thoughts... for holding down, I sometimes use a length of flat bar along the edges and clamp that down. Brown packing tape works quite well if you have a vertical edge to tape it to so the tape is being pulled sideways rather than being peeled off, if you see what I mean.

Since it's not a big part, you could clamp the back with a bar like I've mentioned above. It looks like you're clamping onto some sort of fibre board, but I may be mistaken. If it's really important to keep it flat, you need to clamp onto something more rigid and dimensionally accurate.

It also looks like the plunges at the ends of the nibs are vertical. Those would benefit from being sloping at 2-3 degrees. Plunging like that will surely lift the material. When you machine to the exact depth of the stock, you will almost always get places where it doesn't go right through, and those will lift the stock. If you don't want that to happen, you would have to go deeper.


I'm sure you've done the Vees first, but if not, they should be.

[David]

Hi Roger,

Thanks for the ideas. I am going to try strips of metal along edges at some point. But I really want to find a good double-sided tape or superglue to hold the middle down.

I like the one about an extra V and just snap it off, that is a much better way to do it!

The MDF is pretty flat for a few goes. Certainly more flat than the unevenness due to lifting. I hadn't intended to cut through, it was meant to leave about .1mm all around, as I thought it would help hold things steady. It's only cut through due to the lifting which must have been in the order of 0.35 - 0.4mm.

I cut the Vs second with the idea that the chamfer mill could then get a clean entry from the side and wouldn't require any plunging moves.

If I do it again I'll use your idea of the extra length to give 4 full sides and start with pre-profiled stock so I can clamp it down like I did the test piece.

[Roger]

Hi David,
This is the sort of faced board that I use, it's pretty cheap and being thin it's not so easy to compress. It's also resistant to soaking up oil.

I also like my skinny clamps which can be turned round to reach a long way from the edge.

Double sided tape will certainly hold it down, but whether you'll get it off again is the question. Maybe you would if you used a hot air gun on it. That's what we used to use to get tape off cardboard boxes without damaging the cardboard. The glue lets go pretty quickly when it gets hot. I also have some label remover which also does a good job on tape, but getting it where you need it might be the problem.

[jon38r80]

Lidl sometimes have 3M double sided tape which has pretty good grab. I use it to hold pieces of wood down to a wooden base to allow edges to be routed. Did the same for a long piece of flat bar on the mill. The metal didnt move but the wood it was stuck to arched up from the mill bed. Think gettitting it hot was the cause. Hard plastic sheet seems to be the favourite substrate of others. I'm just too mean to buy it.

[stevep]

If you use good quality double sided tape, the attachment is very good. As Roger said, getting it off might be the problem.

Here is how I faced the steam chest covers for the three Pansys I am building.

[David]

A week and a day after I started, the lubricator is about 60% finished. Still need to do the scotch crank and pump ram, the operating crank and arm, o-ring spacers, etc. Plenty of machining, filing, and threading to go. This component really is trying my patience. A 6BA tapping size drill and a 3.5mm endmill are the casualties so far. And a lot of propane!

I was making a part I thought had a 5/16 blind thread in it yesterday so I ground the tap down and it still cut a thread. Then I realised that part was not threaded but should be bored for a press fit onto a clutch bearing! I'm not sure what OD the bearings have as I don't have any, but hopefully it is more than 5/16.

With that 'success' I decided to grind about 5mm off the 3/8 tap yesterday to see if I could make the nut capturing the delivery pipe a bit thinner. You'll see in a photo how little room there is and I didn't know it at the time but the old nut would not have fit. So a new nut was made and I got it to something close to the drawing dimensions.

This required the pump body to have the 3/8 threaded part shortened by about 5mm, with a corresponding reduction in length of the delivery pipe flange that goes into it and forms the chamber for the o-rings. I now had to 'undo' the calculations I'd done last weekend when I lengthened all of that to try and arrive at the same gap as the drawing.



I've been soldering the tank together over a few days. I tried making a new one by replicating the trial - cut the stock to size and clamp as close to square to the table as I could. The test piece was about 1/16 or so longer to a side and I decided to cut the new blank to the drawing dimensions. This made it small enough that there was now not enough clearance for the cutter! So that was a waste of time. The one I did have had fallen apart at the fold lines because they were too deep. I just did my best and resigned myself to a shoddy looking tank.



Then there was the fun of drilling and transferring the mounting holes in the frames with slidebars and wheels in the way. Of course the transfer ended up wonky so I had to use various die grinder tools, the broken 3.5mm endmill, and files to move the 2nd hole to somewhere it allowed the tank to look straight. Luckily this is up under the running boards so won't been seen from normal angles. But freehanding never goes well.

You can see in this photo that I had to move the pump body as close to the tank wall as possible for the clearance. It's 1/16 over from where the drawing puts it. I don't know how my friend got the clearance!

I can't wait to have this job finished. I've even given up swearing at it - there's only so many times you can curse in a day. "They're pretty easy to make" says my friend. I didn't dare reply.

[jimnswgr]

You're making excellent progress David. It is a very difficult spot for the lubricator to go and to try and install it with your chassis assembled. When I built my "I" class, not too dissimilar to the "B" I had my lubricator in a very similar spot, when it failed it was an absolute nightmare. A few months ago I moved it from that horrible spot underneath the boiler to right behind the front buffer beam and it has made filling it much easier. 

[David]

Thanks Jim. I think (hope? Not measured it yet) the lubricator can be unscrewed, the delivery pipe lifted over the valve rod, than the whole thing dropped down the middle when necessary. This just seems to be the place they always go in our club. I think filling will be OK as there should be a decent gap under the boiler there. How did yours fail?

I cut the scotch crank and crank web on the CNC last night. This is the second crank web as I tapped the holes crooked in the first one. I was eyeing the tap straight but I think the part was tilted in the jaws of my bench vise! I figured it's easy to cut another one so here it is. The finish off the mill on the first one was essentially perfect with a finishing cut but I marred it 'cleaning' it up so didn't bother with these ones.

Next step is to flip the stock over and mill the supporting material away until the I can break the parts off by hand. I do that on the manual mill. I'll tap the holes first. Then a hole for the ram has to be drilled in the bottom of the scotch crank - that's going to be the tricky part.

The profiles were cut with a 3.5mm 4 flute endmill, 1500rpm, 45mm/min feed, 0.5mm doc.
The slot in the scotch crank is 1/8" so that was done with a 2.5mm 4 flute endmill, same figures as above.

I put the parts at the ends of the stock so I could bring the tool down to the height of the pass off the end of the stock and then just cut in from there to avoid ramping. This worked well.

[David]

In an alternate reality the parts all dropped in and worked. In this one it took all afternoon and evening fettling, modifying parts, junking parts, and it still doesn't work smoothly all the time. There is a misalignment somewhere and insufficient clearance in the scotch crank to handle it. I've been trying to open it up to no avail. No amount filing to make the slot longer is improving the situation.

However it does work after a fashion, if you have the closed end of the scotch crank towards the 'front'. Turn it around and it binds unless the brass part carrying the driving spindle is loose.

I'm going to show it to the club guru and if he can either tweak the spindle alignment or doesn't think it will bind in service and cause damage to some part of the loco I actually like, I'm moving onto the o-rings and spacers.

I think this particular variant of the Ewins design design has two pitfalls, one leading to the other. The ram is right at the front because otherwise it will foul the forked end of the valve spindle. This limits the size of the scotch crank meaning it can only have one closed end which unless you thought about it first and allowed a lot of meat means you have nowhere to go if it binds up. I actually did foresee this and drew the closed end about 1mm wider than the design to allow for filing out to alleviate binding, plus I'd left about 1mm clearance to begin with on that end. I'm not saying it's a bad design - it works fine for the designer, but I seem to need a lot more leeway than it provides.

0/10 - should have bought one, however my pride didn't allow it and demands I make this work. It will be good to know how it was made and works so I can fix it again when it breaks. 9 days so far - that's an expensive lubricator!



Also, how would you all make the tank? It seems a simple enough job but without a tiny bending brake I'm at a loss as to a good way to do it.

[jimnswgr]

I don't understand why OSME lubricators always go there. My lubricator on the 26 class was a steam fittings one, inevitably a bit of rubbish got in there, but I did have a roller clutch fail too which was disappointing. From experience if anything will give trouble, it will be the lubricator.

James

[Roger]

Nov 26, 2018 22:48:59 GMT David said:

Thanks Jim. I think (hope? Not measured it yet) the lubricator can be unscrewed, the delivery pipe lifted over the valve rod, than the whole thing dropped down the middle when necessary. This just seems to be the place they always go in our club. I think filling will be OK as there should be a decent gap under the boiler there. How did yours fail?

I cut the scotch crank and crank web on the CNC last night. This is the second crank web as I tapped the holes crooked in the first one. I was eyeing the tap straight but I think the part was tilted in the jaws of my bench vise! I figured it's easy to cut another one so here it is. The finish off the mill on the first one was essentially perfect with a finishing cut but I marred it 'cleaning' it up so didn't bother with these ones.

Next step is to flip the stock over and mill the supporting material away until the I can break the parts off by hand. I do that on the manual mill. I'll tap the holes first. Then a hole for the ram has to be drilled in the bottom of the scotch crank - that's going to be the tricky part.

The profiles were cut with a 3.5mm 4 flute endmill, 1500rpm, 45mm/min feed, 0.5mm doc.
The slot in the scotch crank is 1/8" so that was done with a 2.5mm 4 flute endmill, same figures as above.

I put the parts at the ends of the stock so I could bring the tool down to the height of the pass off the end of the stock and then just cut in from there to avoid ramping. This worked well.

Hi David,
Just a quick question... why don't you use the CNC machine to position over each tapped hole so you can get them exactly vertical? I made a sprung point that engages with the back of the tap wrench so I can tap everything on the machine while I can find the exact hole centres.

[Roger]

Nov 27, 2018 8:58:57 GMT David said:

Also, how would you all make the tank? It seems a simple enough job but without a tiny bending brake I'm at a loss as to a good way to do it.

I'm impressed that you're determined to make these things yourself, it's satisfying even if the road to getting there can be rocky.

Using thicker material for the box reduces the amount it will bend when it's Silver Soldered. I opted to use the CNC to machine slots in the sides and bottom to locate them all. The intention was to then file away the excess material on the outside, but I liked the look of it when that was done and decided to leave it as it was.

Doing it this way allows you to machine all of the holes you need while it's still flat.

20170604_220509 by Roger Froud, on Flickr

20170605_213059 by Roger Froud, on Flickr

[David]

Thanks guys.

James, I guess it is inertia. My father-in-law fitted the lubricator to his latest loco in this position without taking account of the boiler and couldn't take the lid off it... I think he put a bit of a dent in the cladding to make clearance! On the plus side I guess it is mid-way between where the drive comes from and where the oil goes, and if you get it right it is a pretty convenient place for filling. Some people put little filler pipes in the lid to avoid having to remove it.

I get the impression roller clutches wear out and need replacement eventually. The question is how long can you make it before this happens and keeping them clean might be a big factor.

Roger, what you say is exactly how I'd do it on the manual mill but I'm a bit shy of drilling on the CNC because I'm lazy - on the Tormach you can't set the tool length while a job is running so you need to have your required drills in chucks with lengths set in the tool table before starting. I'd also need to know the hole co-ords from the g-code because the tapping would have to be done after the job is finished by positioning the table & spindle manually - you can't jog them while the job is paused. But it is possible, for sure, if I put the time into it. These are the only things in PathPilot I have found lacking so far. It would be really useful for a beginner to be able to pause a job, fiddle about, and then restart. But I can see that's odd practice in CNC where you should press a button and walk away.

It might be quicker to make the whole part on the manual mill.

As for the tanks, wow! The hex captures are a good use of CNC and the slotting looks like a great idea. That could work with 1.6mm and I think soft solder would be ok as it is not under any sort of pressure. I'm going through your thread from page 1 so I don't miss anything but haven't got to the bit shown.

[Roger]

Hi David,
I'm also not keen to drill on the CNC, using the Z-axis. It's ok for a few holes, but I wouldn't walk away and let it drill many unsupervised. I've got caught out a few times, and for most jobs I just use the hole centres in the CNC and the quill manually. That's a big advantage of a powered knee CNC conversion that's not always appreciated.

Doing things on the fly isn't really possible with Mach either, I single step to do the drilling.

I agree, the tanks with slots would work with 1.6mm material, they don't have to be as deep as mine. I did it that way with massively strong walls because I wanted to make sure the alignment would be good enough for the drive shaft going through the whole assembly.

You've got your work cut out reading the whole thread. Hopefully the pictures won't get pulled by Flickr before you've read it all.

[David]

A short update, no pictures necessary. I've been left with my real job, the children, the housework, and what is hopefully only hayfever which has all left me a bit unmotivated to work on my nemesis.

But I took 45 mins in the workshop this morning to take a guess at how much to grind down the lubricator ram and then draw some chicken scratchings and make the o-ring spacer - and it pumps oil! About one drop per full turn of the spindle which might be a bit much but I'm going to run it before I muck about.

I won't know if the ball seal is good enough until there is pressure against the outlet, obviously.

Now I need some roller clutches but I can make the lid and a few other bits. But not today, I'm quitting while I'm ahead.