LBSC Green Arrow
Not a lot to report this time. I decided to make the new valves next but found out that I didn't have any material of the right size (of course!) so had to order some and I've been waiting for that to arrive. I've also been having a tidy up in the workshop to try and make a bit more room and get some of the loose stuff put away on shelves and in drawers.
As I said above I didn't have any bits of bronze suitable for the valves so ordered some 1" diameter gunmetal bar which was just big enough in diameter to get the valves out of. That was set up in the lathe chuck and supported with the fixed steady as it would not have been rigid enough to part some slices off without it. One big disadvantage of the ML7 is that the bore of the spindle is small and will only take up to about 0.625" diameter material. I don't like cutting short bits off long bars if I can avoid it as you always seem to end up with short bits that you can't use.
I parted off three slices, one for each valve.
Then into the mill to machine the round blanks into rectangular ones.
The blanks for all three valves were rough milled to shape leaving the port control edges about 0.025" oversize for trimming later when the exhaust cavities had been machined.
I thought about cutting the slots for the valve spindle and the drive nuts in the outside valves next. The slot for the valve rod could just be flat bottomed but I prefer to make it round to leave a bit more metal on the bottom of the valve which will be a bit thin on these valves. The ideal tool of choice would have been a 4mm ball nosed endmill but, although I have a large quantity of endmills. I didn't have one. Instead, I drilled a 4mm hole in the valve block to form the bottom of the slot when it was milled, positioning the hole so that the valve would have a bit of vertical movement to allow the lift off the port face to release any condensate in the cylinders.
Despite getting the hole in the right place on the first valve I cocked it up by drilling it on the wrong face! Much bad language and back to square one to make another blank as the original was now unusable. Then, I nearly cocked it up again on the second valve, despite putting a cross on the face! (I put the cross on the wrong face!) Talk about having a bad day! Fortunately, I caught it just in time before I actually drilled the hole..
Giving the machining of the valves a bit more thought, I decided that it might be best to machine the exhaust cavity next whilst the valve was still solid. If I had machined the slots first the valve would be a bit flimsy to clamp securely with a lot of the metal removed. As it happens, it probably didn't make much difference as I still had to hold the valve to machine the slots anyway.
The cavities were machined by starting off plunging an 8mm endmill to the depth of the cavity and roughly machining the rectangular shape. It was then finished to size with a 3mm endmill, all done using the DRO.
The internal bore micrometer was ideal for checking the cavity for size. Whilst doing the machining it occurred to me to use some compressed air to continually blow the chips away. There's a tendency for the cavity to fill with chips which is not good for the cutter. I used my airbrush compressor and a long blunt hyperdermic needle taped to a indicator stand and this worked very well. Note the dimensions scribbled on the top of the vice to remind me of the DRO readings required!
All three cavities turned out well with no more cock ups.
Next job was milling the slots in the back of the valve which was straightforward using the DRO to get the correct sizes. Whilst set up like this I relieved the top of the sides of the valves just leaving the bottom part at the full width. This meant that when finally finishing the valve to the correct width I only had to machine a narrow strip instead of the full height of the valve.
I realised that by holding the valve this way I couldn't mill the slot for the drive nut deep enough so I milled them as deep as I could with this set up and then turned the valve through 90° to finish them.
That's the outside valves finished apart from machining the width to final size. You can see that there is not a lot of metal left on the bottom.
You may notice that the slot for the drive nut is quite wide. The drawings show it 0.25" wide with the nut secured to the valve rod by one 8BA setscrew. I didn't think that would be secure enough so I'm making the drive nut 0.375" wide and fitting two 2.5mm stainless grubscrews. I'll also be making the nuts deeper to give more thread for the screws.
I haven't quite made my mind up about how to drive the inside valve and have some sort of adjustment incorporated. I'll probably just thread the valve rod and the drive nut so the valve can be adjusted by turning the valve rod. It's going to be awkward to set it as you can't see the valve once the inside cylinder is fitted. I did think about fitting a couple of screwed plugs in the bottom of the valve chest that you could take out to see the valve but I don't think you would be able to see much through them.
When I was drawing up the 3D model I did find one mistake on the drawings with regard to the position of the valve rod slot on the valve but forgot all about it when I made the valves! The drawings show the centre of the slot to be 0.1875" from the top edge of the valve but it should be 0.1563" otherwise the exhaust cavity isn't vertically centred on the ports. It's offset by 0.0313". It's no big deal though and won't really affect anything. It's just annoying that I forgot about it.
Onto the drive nuts. These were milled from 0.375" square brass. First job was to drill and ream the hole for the valve rod.
Then two holes were drilled and tapped for two 2.5mm grubscrews to clamp the nut to the valve rod and allow for adjustment of the valve.
Finally, the top of the valve nut was milled to the same height as the valve leaving a boss a bit higher to give more thread depth for the grubscrews.
A trial assembly of the valve, drive nut and a valve rod showed everything to fit ok.
At the moment the grub screws stick up above the top of the valve nut and would foul the steamchest cover but I'm going to machine flats on the top of the valve rods which will solve that problem.
The inside valve was machined in a similar way to the outside ones but the height of the valve was reduced to make it as thin as practical. There is very little space in the inside valve chest to allow the steam to flow freely so reducing the height of the valve gives more room.
The valve nut was again machined from 0.375" square brass but threaded 5BA to take the valve rod which will also be threaded 5BA and allow the valve to be adjusted by rotating the valve rod.
I'm going to have to enlarge the cavity inside the valve chest as at the moment the valve touches the side of the cavity and won't slide freely.
A bit of a side note - when you are making the valves and the drive nuts it's important to get everything as good a fit as possible whilst still allowing the valve to float on the nut. I make the slot in the valve for the valve rod to also be a good fit on the rod as this helps to keep the valve square to the ports as it moves, as does making the nut in it's slot a good fit. If everything is a sloppy fit then the valve will twist as it moves and make a mess of the valve events.
Because I had changed the design of the valve for the middle cylinder the valve rod needed to have a reduced diameter on the end otherwise I would have had to thread it 5BA right to the end to screw the valve nut on and a thread would not run very well in the boss on the end of the valve chest. I had originally drilled and reamed this boss to .0125" diameter but it now needed to be 0.09375" diameter. To resize the hole in the boss I redrilled it and tapped it 1/4" x 40 to take a bronze bush. I also enlarged the cavity in the valve chest to give more clarance for the valve.
It all fitted together nicely.
I couldn't resist a trial assembly of the cylinders and valves so far.
I realised that I hadn't yet drilled the exhaust passage for the inside cylinder so thought I had better do that. I drilled it 4mm diameter but was very careful how deep I drilled as the end of the passage comes very close to the cylinder bore. I didn't want to break into the bore and ruin all the work that I had done! I drilled it short of reaching the actual exhaust port in the cylinder and then mounted the cylinder in the tilting vice so that I could drill from the exhaust port through to the drilled hole at an angle to avoid the bore. I actually used a 4mm endmill to drill that as it had to start in a corner and drills are not very good at doing that whereas an endmill has not problem.
I haven't really decided yet how the exhaust pipe assembly will connect to the inside cylinder block. If I use a bolt on flange then the plain drilled hole is alright as it is but if I use a stub that is a push fit in the passage then I will have to open it out to the diameter of the stub.
I had often considered the use of O rings for sealing cylinder end covers rather than use gaskets so decided to give it a try. I ordered some 1mm section Viton O rings in suitable diameters which arrived today. I reckoned that there was just enough room to squeeze a ring in between the outside edge of the end covers and the fixing bolts. I needed a suitable cutting tool to machine a groove for the ring and found one that was ideal in my stash of 'useful' bits of tool steel and was the right width at just over 1mm. I set the inside cylinder rear cover up in the lathe on the threaded stub that I had used before and machined the groove to 0.75mm deep.
There was only just enough room and the inside of the groove had to touch the edge of the bolt holes but it turned out ok.
Unfortunately, when I looked at the rear covers for the outside cylinders the recess in the edge to clear the boss on the valve chest would have cut across the O ring groove so I had to abandon the idea after all that. I could fit one on the front cover of the inside cylinder and may do that but the outside cylinders will have to use gaskets. I have actually ordered a bottle of Loctite 574 flange sealant so may try using that instead of gaskets. I've not used it before but everyone seems to swear by it.
I think the O ring idea is worth pursuing and I will try it again on later projects. I hate making gaskets!
The day after posting the above bit about the O ring seals I had an email from fellow N25GA member Nick who pointed out that my idea wouldn't work as the steam would just leak through the fixing bolt holes. He's quite right of course and I don't know why it didn't occur to me. I blame it on old age! The O rings would need to go on the inside of the bolt holes but I can't do that here as the steam passages would be in the way. Oh well, better luck next time! The Loctite 574 has arrived so I'll be trying that when I come to assemble the cylinders permanently.
The last few days have been spent on tidying up a few items. Firstly, I drilled and reamed the piston rods to take the taper pins that hold them to the crosshead. I fitted the rods to the crossheads and then held them in the mill to drill through the rod and then ream them with the taper reamer.
The crossheads already had the hole for the old pins so I just spotted through those. I had to ream the holes a bit bigger than I would have liked to clean up the old holes in the crossheads but I think they will be alright. I can always use a spot of Loctite to make sure.
The original crosshead pins were a right pain to fit as they are threaded into the crosshead from the back and there is no easy way to tighten them up. I think the previous builder had used a pair of pliers judging by the marks on the head of the pin. They were also threaded into the drop arms as well which made it even harder to fit them. The drop arms are supposed to have a clearance hole to fit over the thread of the pin and then secured with a nut.
I decided to make new longer pins and drill out the crosshead and the drop arm so that the pin could be easily pushed in from the back and then held with a nut. They were a simple turning job from 0.375" silver steel. The threads are 5/32" x 40 and the nut was made by retapping a 4BA nut.
It's now very easy to fit the pins and drop arms rather than a struggle. I did have to replace one of the little pins that stop the drop arm rotating about the crosshead pin as it was very loose. I drilled the crosshead and the drop arm to take a new larger pin made from 2mm silver steel.
It was time to start messing with the valve gear and see what sort of travel it was giving so I reassembled both sides with the driving axle. I noticed that the motion was now very stiff to turn since making the new crosshead pins, probably because I had made them a good fit in the little ends of the connecting rods. It turned out that the lefthand connecting rod was actually bent and binding up. A quick straighten using the mill as an arbor press soon sorted that out.
It was obvious that there was play in various bits of the valve gear. The trunnion bushes for the expansion links were quite sloppy so I turned up new ones from gunmetal bar. One of the pins that fasten the combination lever to the valve crosshead was quite loose but I just replaced it with a suitable drill whilst I was testing the valve gear! One of the die blocks is a loose fit in the expansion link and that will probably need replacing.
One of the first things that I wanted to check on the valve gear was how much travel the valves had when in mid gear. At this point the only travel given to the valve is from the combination lever. The calculated travel should be 0.117" which is interesting as that gives the valve a negative lead of -0.0039". I think that is actually a mistake as LBSC was a a big advocate of lead in his engines and I doubt he would have intentionally designed the valve gear to give negative lead. It can be corrected by altering the dimensions of the combination lever but it's much easier to shave a bit off the valves to reduce the lap. That's what I will do as I'm not making new combination levers!
A lot of people think now that lead is not necessary in our models and I tend to agree. I'll probably adjust the lap of the valves to give a lead of say 0.005" as that helps to compensate for wear in the valve gear over time. As the valve gear wears the valve travel decreases and the ports open later and later which gives the effect of negative lead.
Anyway, back to the job in hand and measuring the valve travel in mid gear. I faffed around for ages trying to measure it with a digital caliper but it was difficult to get consistent results. All of a sudden I thought 'you've got a drawer full of dial indicators' so use one of those. So I did. I put the indicator in a magnetic stand on the chassis and the shaft of the indicator on the end of the valve rods. Much easier!
I wasn't expecting the results to be good but they came out much better than I expected. The travel on the lefthand valve was pretty close at 0.118" and the righthand not bad at 0.113". I can live with that.
Next I measured the valve travel in full forward gear. The lefthand side gave a travel of 0.258" but the righthand side only gave 0.230". The difference may be due to the setting of the return cranks but some of the difference could be the loose fit of the die block on that side.. Doing a quick calculation on one of my spreadsheets that I set up to work this sort of stuff out, if I reduce the lap of the valve from 0.0625" to 0.054" that will give a lead of 0.0045" and will require a full gear valve travel of 0.22" to give a full gear cut off of 76%. That should be fine for a three cylinder loco. That valve travel will open the ports by 0.056" which is only just over half way. It doesn't seem a lot but if it's correct then we are somewhere in the ball park already.
Not much of an update tonight. I had a look at the loose die block tonight and it was the screw that holds the die block to the radius arm that was a loose fit in the die block. The fit of the die block itself in the expansion link was fine. That was good news as it meant that I didn't have to make a new die block, just a new screw. That was another simple turning job from silver steel. The thread on the end is 3/32" BSW but I had a die for that so no problem.
The slot in the head was just cut with a Junior hacksaw. I made the new screw longer than the original so that I can fit a locknut on it after screwing it into the radius arm. I'll probably replace the one on the otherside to make them the same.
After replacing the screw I checked the full gear valve travel again and both sides were now pretty much the same at 0.230" (That's a bit different from last nights results but the reverser is not a very rigid fit at the moment and has probably moved compared to last night) That's also good news because it means both return cranks are set the same and in the right position. It would have been a pain if I had to move one or both of them as they have been pinned to stop them moving and I would have possibly had to make new crankpins to reset them although I could have removed the crankpins and refitted them in a different position which would have been easier.
I still haven't checked the quartering of the crank axle yet to see what angle the centre crank is in relation to the outer ones. I think I will be very lucky if it turns out to be correct but stranger things have happened! I certainly hope that I don't have to reset it.
Whilst I was messing around with the valve gear I noticed that the radius rod on both sides were really tight against the outside edge of the slot in the motion brackets so I took the brackets off and milled a bit more off the edge of the slot to give some clearance.
The valve gear was a lot freer after that.
I've also been tying up a few loose ends on the cylinders and valve gear. I hadn't made the end covers for the outside steam chests yet so got those done, a straightforward turning job from brass bar. These only serve to retain the O rings that I'll be using to seal the valve rods. They have a spigot that fits into the valve chest to such a depth to leave the correct spacing for the O ring and a smaller diameter spigot on the outside that is really just for show.
The four holes for the 8BA fixing screws were drilled in the mill using the DRO. I made a jig from a bit of brass bar with a recess in it to take the spigot of the cover and tapped 4BA to take a bolt to hold the cover in place. This was then held in the vice using a Vee block and centred using the DRO.
The inside valve chest required a threaded 'gland' nut as you may recall that I had intended to fit a similar cover to the outside ones but the boss on the steam chest was too small to take any bolts. I had to tap it 9/32" x 40 instead. This was another straightforward turning job from hexagonal bronze bar. The threaded spigot was again sized to leave the correct gap for the O ring.
I thought it was about time to cut all the valve spindles to the finished length (they had all been left long) and thread the ends to take the forks that connect to the 2 to 1 levers so that was the next job. It was a case of trial and error, cutting a bit off and then assembling the fork and the flat links that connect to the 2 to 1 levers until the levers looked right when the valve rod was in mid position. I assumed that the levers should be straight across the frames at this point.
That looks pretty good to me. Note that the slotted screws holding the valve chest covers on are just temporary. It's very fiddly trying to fit the proper hex headed bolts so I'll leave those until final assembly!
I was a bit worried about how I would secure the 0.125" pin that connects the link for the inside valve rod to the end of the short lever. As it happens, it can't fall out under normal circumstances as it never travels outside of the edge of the long lever and so is held in place by the top and bottom fork of the lever. I'll probably just leave it floating.
Talking of the 0.125" silver steel pins that I used, I've always had a problem with the fit in the reamed holes that they go in. I normally use a 0.125" reamer to finish the holes and of course that is dead size for the pins, hence they are too tight. In the past I've lapped out the holes with some very fine aluminium oxide grit (left over from when I used to grind my own telescope mirrors) on a piece of 0.125" silver steel until the pins are a nice fit. A couple of weeks ago I won a large quantity of used small reamers and in amongst them was one of 0.1258" diameter. That's the perfect size for the 0.125" pins and gives just enough clearance. Problem solved!
I decided to look at the steam and exhaust pipes next.
I couldn't easily reuse the original exhaust as the elbows for the outside cylinders threaded into tapped holes in the original steam chests, also, the connection to the middle cylinder was just a stub that pushed into a hole in the old cylinder. I wanted to make this a proper flange fitting. I'm hoping that I can get away with using the old steam pipe though.
The new elbows for the outside cylinders were milled from a bit of brass flat. It was the only bit that I had in the size I wanted and it was just long enough to make the two (note to self - time to order some more brass sections!). The first job was to mill the two pieces of brass to shape and then drill the fixing holes in the flange.
Before I could go any further I needed to know the position of the cross pipe on the faces of the elbows. This would be determined by the thickness of the flange for the middle cylinder so that was made next. It started life as a piece of 0.375" square brass with a 0.25" hole drilled through it for the cross pipe. Most of it was then milled away to form the shape that I wanted.
I could now drill the holes in the faces of the outside elbows to take the cross pipe. The final job was to drill and mill the oval hole in the flanges to match up with the holes in the steam chests.
I found a nice piece of 0.25" diameter brass tubing for the cross pipe. I did have some copper pipe but it was very thick walled and would reduce the area for the steam to escape. The brass tube was from an assortment that I got from friend Dennis's workshop after he passed away. He was a mechanic in the RAF during the war and I suspect the tubing was probably oil pipes off a Lancaster bomber or similar!
To hold everything together whilst silver soldering the joints I thought it best to make a jig from a bit of steel bar that the elbows etc. could be bolted to.
The blast pipe is 0.3125" in diameter but I didn't have any suitable tubing so I machined it from solid brass bar. I prefered to do it this way anyway as the top of the blast pipe is threaded to take the brass nozzle so it needs to be thick walled. The end needed to be notched to fit onto the cross pipe and that was done with a 0.25" endmill in the mill.
To hold the blast pipe in position in the jig I bent up a piece of thin steel strip and drilled a hole in it to fit snugly over the blast pipe and another to bolt it to the jig. It's a little bit tricky as the blast pipe is not at 90° to the bolting faces of the flanges due to the 1.3° inclination of the steam chests. I just set the angle by eye using a protractor. The brass will be quite soft after soldering so it can always be 'tweaked' if the angle is not quite right.
I did the soldering in two stages. The first stage was to solder the blast pipe and the middle cylinder flange to the crosspipe. I had left drilling the holes through the flange and the bottom of the blast pipe into the crosspipe until they were soldered in position. Not fitting the elbows at this stage meant that I could remove any burrs from the drilled holes. If the elbows had been fitted as well than I wouldn't be able to easily do that.
Drilling the holes into the cross pipe:
The elbows were then soldered into position and the assembly cleaned up.
Next job - steam pipes.
The original steam pipes were a bit scruffy but nothing a bit of tidying up wouldn't cure. All they needed was a flange fitting to replace the stub pipe that originaly pushed into the inlet hole on the inside steam chest and reusing them would save a lot of work. You don't see them when the loco is assembled anyway.
I made a flange from a scrap of 0.125" brass flat and drilled a hole in it to fit over the existing stub and two holes for the fixing screws. The stub did need shortening as otherwise it would be too long. Once again, I made a jig to hold the flange in place whilst it was silver soldered onto the pipes. I just used the same piece of steel used for the exhaust pipes and drilled and tapped a couple more holes in it.
The soldering didn't go entirely to plan as I accidentally soldered the flange to the studs that I was using to align it! That wasn't too big a deal as I just cut them off close to the top of the flange and then drilled them out. Next I held the pipes in position on top of the steam chests and marked the position of the two outer holes in the flanges onto the outside steam chest covers. The covers were then taken off and the holes drilled and tapped 7BA. I could then fasten the steam pipes firmly in position whilst I spotted through all the other holes including those on the inside steam chest.
I now needed to drill the holes in the steam chest covers for the steam inlet and get them in the right position as they didn't line up with the fixing holes. I made a template by covering the flanges in black ink from a Sharpie and pressing them onto a piece of paper - instant template. Just line up the fixing holes with those in the covers and then centre pop the steam hole.
I had already decided to use O rings to seal the pipe connections onto the outer steam chest covers and the inside cylinder and steam chest so the steam inlet holes were counterbored to suit.
All holes drilled and counterbored:
It would have been nice to be able to use O rings on the outside exhaust connections but it would be difficult with the oval holes that I had to use. I suppose I could use a very thin O ring and put a recess around the oval holes to accommodate them? I might look into that. There may be room for a 1mm section ring? I have actually got some that I bought for sealing the glass on water gauges.
The boss on the front of the steam pipe assembly for the oil connection touches the top of the inside cylinder block so needs relieving to give a bit of clearance between that and the front cylinder cover. I could just file a bit off the top of the cylinder and the cover but I'd sooner take it off the bottom of the boss and leave the cylinder alone. I'm sure the boss doesn't need to be as long as it is but I'd better leave it alone incase it does! The thread in the end is to take a check valve for the oil feed.
We're getting close to trying the chassis on air at long last. I've still got to trim the valves to length and make the PTFE piston rings but not too much left now. The big 'if' now is whether the inside crank is at the right angle. What are the chances?