Richard Gawler
Western Thunderer
I bought one of those 5-sided tapered broaches for this sort of thing. Terribly expensive but you can remove barely a thou at a time until there is space for the oil.
Ian@StEnochs
Western Thunderer
Rather than buy an expensive reamer try just polishing the axle with metal polish. I put the axle in the drill or lathe chuck and polish the running surface with a bit of cloth and Brasso.
Ian
Ian
oldravendale
Western Thunderer
I bought about a dozen used hand reamers on the Bay of E for about £10 three or four years ago. Included were two 3/16" reamers, one of which is slightly bigger than the other (!) and is perfect for reaming out bearings.
Brian
Brian
Ian@StEnochs
Western Thunderer
A grand source of used, and sometimes unused, tools are auto jumble and traction engine shows. I have picked up a lot of useful things for not a lot of cash plus you get to inspect the goods AND haggle over the price.
Ian.
Ian.
michael080
Western Thunderer
Simon,
why not use toothpaste or cream cleaner between axle and bearing to polish the contact surface?
This would't need an expensive reamer and the process is slow, hence controllable.
Edit: Didn't see Ians proposal.
Michael
why not use toothpaste or cream cleaner between axle and bearing to polish the contact surface?
This would't need an expensive reamer and the process is slow, hence controllable.
Edit: Didn't see Ians proposal.
Michael
simond
Western Thunderer
Took a wee while, and to be honest, I’m still not sure what the problem was, but both pony trucks now behave themselves. I simply put the reamer in the lathe and ran the pony trucks on and off it repeatedly with a little “swashing” until the problem went away.
It may be that there simply isn’t enough load on the pony trucks, and I wonder if that might not be a good idea also to prevent the nodding dog effect - it’s not obvious when driving the loco sensibly, but a sudden stop does cause a little porpoising. I’ll ponder this.
I might get the brakes done tomorrow, that and shorten the front crankpins, as the following stage is to re-fit the cylinders, slidebars and motion brackets, and the sandboxes. Then a re-spray of the chassis.
I’ll also take the opportunity to 3D print a new speaker box, and hopefully it’ll all be back together at the weekend.
But I’m not holding my breath!
I haven’t clicked muting the sound when filming, so I apologise for the C4 news soundtrack in the background.
It may be that there simply isn’t enough load on the pony trucks, and I wonder if that might not be a good idea also to prevent the nodding dog effect - it’s not obvious when driving the loco sensibly, but a sudden stop does cause a little porpoising. I’ll ponder this.
I might get the brakes done tomorrow, that and shorten the front crankpins, as the following stage is to re-fit the cylinders, slidebars and motion brackets, and the sandboxes. Then a re-spray of the chassis.
I’ll also take the opportunity to 3D print a new speaker box, and hopefully it’ll all be back together at the weekend.
But I’m not holding my breath!
I haven’t clicked muting the sound when filming, so I apologise for the C4 news soundtrack in the background.
simond
Western Thunderer
Phil,
er, yes, I think so.
I think my skepticism was perhaps more related to seeing a method by which I could easily, effectively and enjoyably do the necessary manufacturing, and, I guess, have confidence that the outcome would work acceptably, thus justifying the effort. I was not (still am not!) wholly convinced by rows of handrail knobs…
It’s early days but the result so far appears to be marginally more effort than compensation, but much of that effort is in producing the sets of bits that are required, modified axleboxes, modified horns and the anchor brackets for the wires, and all these jobs are trivial repeat jobs on the milling machine. There’s a slightly tricky soldering job in fitting the columns to the axleboxes (I did it in pairs, and then cut them) but apart from that, it’s easy.
So I suppose it’s more a case of falling back on what you know, or trying something new.
Again, it’s early days, but the loco does certainly run smoothly. The pony trucks are not right yet, but that’s nothing to do with the main axles.
I don’t subscribe to @adrian point of view that compensation is the worst of all solutions, it is certainly better than a rigid chassis, but has two key deficiencies, you have to find room for the beams, which may or may not be easy, and you end up with a triangle of support, in which your centre of gravity wants to be somewhere near the fat end - if it isn’t, your loco might be a little prone to tipping over. And as Adrian said, it can mean that every bump in the track is immediately transmitted to the loco, if you use the traditional “one fixed axle” approach.
I'm going to attempt to repeat the success on Chris’ pannier chassis, and then, all being well, you can call me a convert. Thanks to you, Adrian, and everyone else who provided encouragement from the sidelines!
er, yes, I think so.
I think my skepticism was perhaps more related to seeing a method by which I could easily, effectively and enjoyably do the necessary manufacturing, and, I guess, have confidence that the outcome would work acceptably, thus justifying the effort. I was not (still am not!) wholly convinced by rows of handrail knobs…
It’s early days but the result so far appears to be marginally more effort than compensation, but much of that effort is in producing the sets of bits that are required, modified axleboxes, modified horns and the anchor brackets for the wires, and all these jobs are trivial repeat jobs on the milling machine. There’s a slightly tricky soldering job in fitting the columns to the axleboxes (I did it in pairs, and then cut them) but apart from that, it’s easy.
So I suppose it’s more a case of falling back on what you know, or trying something new.
Again, it’s early days, but the loco does certainly run smoothly. The pony trucks are not right yet, but that’s nothing to do with the main axles.
I don’t subscribe to @adrian point of view that compensation is the worst of all solutions, it is certainly better than a rigid chassis, but has two key deficiencies, you have to find room for the beams, which may or may not be easy, and you end up with a triangle of support, in which your centre of gravity wants to be somewhere near the fat end - if it isn’t, your loco might be a little prone to tipping over. And as Adrian said, it can mean that every bump in the track is immediately transmitted to the loco, if you use the traditional “one fixed axle” approach.
I'm going to attempt to repeat the success on Chris’ pannier chassis, and then, all being well, you can call me a convert. Thanks to you, Adrian, and everyone else who provided encouragement from the sidelines!
Richard Gawler
Western Thunderer
Simon, if you could post a photo of the insides of the frames showing the beams and their supports I expect I would find this educational. I have a packet of hornblocks for an unbuilt Connoisseur J79 and I know so little about building suspension everything I try beyond a rocking axle will be new to me.
Ian@StEnochs
Western Thunderer
Compensation is fine for locos that move slowly but springs give a superior ride to fast moving trains. It is a joy to watch the wheels rise and fall under an engine as it sails over a bit of rough track, the added bonus being a reduction in noise transmission.
My problem with CSB is one of weight distribution. With individual springs it is relatively simple, but time consuming, to individually adjust the weight carried by each axle. Just how that can be arranged with a single spring wire I haven’t figured out.
Ian.
My problem with CSB is one of weight distribution. With individual springs it is relatively simple, but time consuming, to individually adjust the weight carried by each axle. Just how that can be arranged with a single spring wire I haven’t figured out.
Ian.
Richard Gawler
Western Thunderer
Ian this must be the most telling thing I have ever read about suspension.
0-4-0 locos are usually shunters, so we don't find much written about applications of CSBs to them. Eight- and ten-coupled locos are working long trains on a main line, hopefully at some speed, and setting up four or five pairs of springs will be hard work. This leaves six-coupled chassis open to the greatest amount of debate, and British loco modellers build more six coupled chassis than the other arrangements.
Pencarrow
Western Thunderer
Looks right at home on your GWR shed at 1:12.
bambuko
Western Thunderer
The only way you could do it, is if you made fulcrum points adjustable horizontally.
"... Adjust values of a, b, c, d..." in the example shown above
(where I have changed spreadsheet by using different "Axle loads")
The values for a, b, c, d are the original ones where "Axle loads" assumed were all equal.
This is not easily done for the usual method of construction that solidly solders fulcrum points to the chassis.
Hence the importance of getting the calculations correct in the first place.
Alternatively fix fulcrum points such that they are removable? to allow adjustment...
Pencarrow
Western Thunderer
Useful if you're settled on solid compensation beams but doesn't cover springy beams as per the solution Simon is using.
simond
Western Thunderer
this seems rather complicated in comparison to the excel spreadsheet I used.
Continuous springy beams. I downloaded and worked through the Alan Turner model - I'd share it but it's at home and I'm having lunch in the office.
enter the relevant data in the yellow boxes, use the sliders to adjust a, c, d & f until the green diamond and red triangle are aligned, and confirm that you're happy with the weight distribution. I chose a deflection of 0.8mm and 0.48mm wire diameter. The loco body (whitemetal) is just shy of a kilo, and the wheelbases are per the drawing (not in the above screenshot, obviously)
I weighed the body (balance one end on the scales and the other on a bit of wood or whatever of roughly the same thickness, ditto repeato other end, to work out where the CoG is. (cog position from front buffer beam = length between buffer beams, multiplied by weight on rear buffer beam, divided by total weight)
This gives an actual deflection which is the same for all wheels. It's then a matter of arranging the supports so that the wire is straight, doesn't hit anything, and has room to flex. My fulcrum points are bits of brass angle soldered firmly to the frames. I arranged that they were all drilled the same, and the upper edge of the angle matches the upper edge of the frames.
Richard, I'll post some inner frame pictures later this evening. It's rather like the drawings a couple of pages back!
atb
Simon
Rob R
Western Thunderer
It seems all well and good if you have an existing model you want to upgrade but how do you determine the CofG of an unmade kit?
Or the total weight of a scratchbuilt loco before you start it?
It is almost as if you have to build it rigid to get the required data and then convert to CSB?
Am I missing something here? (The answer to that is probably yes but I just don't know what).
Or the total weight of a scratchbuilt loco before you start it?
It is almost as if you have to build it rigid to get the required data and then convert to CSB?
Am I missing something here? (The answer to that is probably yes but I just don't know what).
bambuko
Western Thunderer
@Rob R
You are correct about the problem of knowing total weight,
but it is not too much of an issue - you can try different dia spring beams when the model is finished...
Fulcrums remain the same it is just a spring rate you adjust to account for different model weight.
C of G is not much of a problem either - you can roughly guess it and correct it on finished model by adding weight in strategic places.
Of course it is easier with an existing and tried design, but it is more fun doing it yourself
You are correct about the problem of knowing total weight,
but it is not too much of an issue - you can try different dia spring beams when the model is finished...
Fulcrums remain the same it is just a spring rate you adjust to account for different model weight.
C of G is not much of a problem either - you can roughly guess it and correct it on finished model by adding weight in strategic places.
Of course it is easier with an existing and tried design, but it is more fun doing it yourself
Ian@StEnochs
Western Thunderer
When I started to model in S7, I moved up from P4, I decided to have all my locomotives fully, live sprung. In P4 I had used the ride on the stops with a spring to push wheel into dips method but let that the added mass of 7mm fully sprung would work. My layout featured colliery exchange sidings and had mainline locos and pugs. The pugs, short wheelbase 0-4-0s, rode beautifully and virtually silently over the indifferent, on purpose, pit lines with the wheels lifting and falling as the engine traversed the lumpy track. The bigger engines, 0-6-0, 0-4-2 and 4-4-0s did the same on the better mainline.
However the pugs nodded like the dogs heads seen on car parcel shelfs! Especially noticeable on starting and stopping despite being very careful and gentle, not realistic at all. Conversion to 3 point compensation, rear driven axle solid with frames, front axle on centre pivot, stopped the nodding but has not compromised running quality.
It seems all well and good if you have an existing model you want to upgrade but how do you determine the CofG of an unmade kit?
Or the total weight of a scratchbuilt loco before you start it?
It is almost as if you have to build it rigid to get the required data and then convert to CSB?
Am I missing something here? (The answer to that is probably yes but I just don't know what).
Most of my locos are built from scratch and start with the frames. I could make a rough estimate of final weight distribution based on previous models. I could also build in adjustable fulcrum points. However there are some models, 4-4-0s in particular, where it is impossible to shift the CoG enough by adding weight. Just where in an Edwardian 4-4-0 can you add weight to pull the CoG between the drivers?
Individual springs on each axlebox allow different loading on each axle. It can be fiddly, mostly by adding shim packers on the axlebox, but it works for me.
Ian.