CSBs on tender

[David Thorpe]

While waiting the arrival of the Loctite 243 I've ordered to complete my current loco, I thought I'd get on with a GER S23 Holden tender that I'm building. I've completed the body and am about to get on with the chassis for which I had intended using the conventional method of a fixed rear axle with the front two compensated on a single centre beam. However, I then wondered if this would be a suitable project to dip my toe into the mysterious world of CSBs (I only say mysterious because of the maths that are involved). So I would be very grateful for some help with the fulcrum positioning.

The tender frames are 73mm long and, starting from the front, the axles are at 11mm, 37mm, and 59mm. I have got a High Level CSB jig to help me mark out the positions. And Am I right in assuming that Markits coarse handrail knobs are still recommended as fulcrums? I have got quite a few spare Gibson handrail knobs and although they are finer I wondered if they would do as I'd assume that, this being a tender, a thinner wire would be required for the springy beam than would be necessary on a loco.

DT

P.S. I don't know if it's relevant, but the weight of the tender, frames, proposed hornblocks and wheels is approx. 55g. To that I can of course add plenty of lead.

[David Knight]

Davey,

The Markits knobs have the advantage of a nice big hole that makes it easier to thread the guitar string through, a considerable advantage in restricted conditions. FWIW have you had a look at some of the downloads on Bill B's site? They may give some useful ideas for setup plus the ever useful CLAG info.

http://www.mousa.biz/downloads.html

http://www.clag.org.uk/csb-gallery.html

HTH

David

[David Thorpe]

Thanks David - I'd already had a look at the CLAG site and had tried to make head or tail of the spreadsheet, but without any success at all. That was not perhaps surprising as I've never had to use a spreadsheet in my life and really had no idea what it was all about. I hadn't previously seen the Gallery, and noted with interest that Dave Franks is producing CSB kits for some tenders. I also noted the remark that "Tender frame lengths are often a problem area with CSBs because there isn't much room fore and aft of the front and rear axles respectively". That would tend to suggest that for my tender I'd be looking at CSBs without outer fulcrums, so I was further disheartened to read in Russ Elliott's comments on the positioning of the fulcrums in such a case that even when you've managed to work out where the fulcrums should be "the positioning of these fulcrum points is therefore exceptionally critical and arguably far too sensitive, given the practical constraints that real fulcrum points have finite widths, of up to an order of magnitude greater than 0.1mm".

I thought that Bill Bedford's print-out CSB plots were a great idea and potentially very helpful, but sadly there wasn't one for the prototype I want and, indeed, none at all for tenders.

I have to confess that I'd started this thread in the belief that a tender would be a fairly simple way of starting with CSBs. I'm now realising that that was probably a mistaken apprehension and I'm beginning to think that I'm perhaps being a bit ambitious and that sticking with the conventional method of a fixed rear axle with the front two compensated on a single centre beam is the better option. Or of course there's also Tim Lewis's solution in the earlier thread "Tender Chassis Construction" which may not be pretty but appears to work well without any critical sensitivities!

DT

[grovenor-2685]

I have to confess that I'd started this thread in the belief that a tender would be a fairly simple way of starting with CSBs. I'm now realising that that was probably a mistaken apprehension

I think you are worrying over much, its very easy to adjust the C of G of a tender when adding the lead weight, so in your place I would just put my fulcrums such that the span on the centre axle is a bit longer than on the end axles, and if it doesn't sit quite level when finished put a little more weight in the high end.
The end fulcrums will have to be 2mm from the buffer beams so positions from the front would be 2, 22, 48 and 68. giving spring spans over the axles of 20, 26 and 20.
I would suggest an 11 thou wire then weight to get somewhere between 0.25 and 0.5 mm spring movement.

Note that tenders are much less critical than locos as there is no traction requirement.
Regards
Keith

[Will L]

I’m sure Keith’s right, you can get in a full set of fulcrums in, no trouble.

I’m interested in this plot as I have a few Holden tenders of my own to do, so I’ve checked it out with the spread sheet.

I’m sorry Keith but I think your first fulcrum point could, with advantage, be moved in 1mm to 3 from the end, giving spans of 19, 26 and 20. This will allow for the asymmetric nature of the wheel base without needing to fiddle with the CofG location.

The easiest way to define CSB fulcrums is relative to the axles, otherwise I find the numbers get confusing. With F= fulcrum, A = Axles and dimensions in millimetres, the amended version of Keith’s plot is
F, 8, A, 11, F, 15, A, 11, F, 11, A, 9, F

As modified it is as well balanced as you are likely to get. You want the CofG of the body between the wheel base centre point and the centre axle. To give a 0.5mm deflection on 11thou wires you need a body weight of 240 grams, for 10 thou wires the equivalent body weight is 170 grams.

A paper and pencil method

While I used the spread sheet to do these checks, because I can, there is a method which needs no computational assistance and which will produce a good solution. It also happens to produce a plot with shortest practical CSB wire length which might be preferable on tenders with little overhang at the ends.

It goes like is

Work out the full wheelbase (59-11 = 48)

Set the inner fulcrums symmetrically between the outer two axles. To do this on a 3 axle chassis divide the wheelbase by 4 (48/4= 12) and set the two middle fulcrum point 12mm in from the outer axles. The middle axle won’t be central between the middle two fulcrums but this doesn’t matter.

To work out the end fulcrums use the inner fulcrum distance (12) multiplied by 0.56 (The 5.6 number has been arrived at empirically, but it works) and round to the nearest 0.5mm i.e. 6.5

Finally, as the axle spacing’s are significantly different, 26 at one end 22 at the other, you should soften up the springing at the end with the axles closest together by moving the outer fulcrum at that end out by an additional 0.5mm. This gives the plot

F, 6.5, A, 12, F, 14, A, 10, F, 12, A, 7, F

As the CSB wire is a bit shorter, a 0.5mm deflection needs 290 grams of body weight with 11 thou wires and 200grams with 10 thou.

Neither of these two plots is necessarily any better than the other, unless something else about the chassis makes one solution preferable, like chassis spacers or brake gear hangers getting in the way.

Will

[David Thorpe]

Thank you very much for these extremely helpful replies. It never ceases to amaze me how the more experienced modellers on this forum will take so much time and trouble to help the rest of us out. In this case, you've certainly given me the confidence to go ahead and try CSBs on this tender, although I was a liitle bit horrified at the weight required for the tender - hope the loco is up to it! And Will, I was exceedingly pleased to see a paper and pencil method - do I take it that that would apply to 0-6-0 locos as well as this tender?

Needless to say, before I proceed there are one or two odds and ends I'll have to acquire - am very envious of those of you who are going to Scaleforum this month (can't find a "smiley" for serious envy).

DT

[grovenor-2685]

although I was a liitle bit horrified at the weight required for the tender

Although 0.5mm deflection is seen as desirable, I think you should be OK with a lesser deflection on a tender, my 4F tender has been test run without any added weight and seems to be happy. So use the smaller wire and add weights incrementally to see how it goes.

A lot is going to depend on the running conditions, if you have to cope with heavy loads on grades and especially curved grades then you will probably need more weight. If its only light trains on the level then a lightweight tender will usually be fine.
Regards
Keith

[Will L]

...you've certainly given me the confidence to go ahead and try CSBs on this tender, although I was a liitle bit horrified at the weight required for the tender - hope the loco is up to it! And Will, I was exceedingly pleased to see a paper and pencil method - do I take it that that would apply to 0-6-0 locos as well as this tender?

That's good as I think passing on the knowledge is exactly what P4 society is, or should be, all about. And yes the paper and pencil method applies just as well to a full loco chassis.

Weight

Sorry yes these are a bit big for a tender I'm used to thinking about loco's.

Keith is right to say that getting you full 0.5mm is less important on a tender with a reactively short wheelbase. None the less, on Keith's plot (F, 8, A, 11, F, 15, A, 11, F, 11, A, 9, F) with an 8 thou wire (smallest guitar strings general available), you get the 0.5mm deflection on just 65 grams. Because the P&P method gives a shorter wire it comes in at 80 grams.

Caviat. Guitar wires are not all that accurate to the size printed on the packet and seem as a rule to be a bit less than the given size. I have wondered in the past if this is the size of the die they were pulled through not the actual size of the wire. The truth is I genially don't bother with the maths, just fit bigger or smaller wires till the vehicle sits at the right hight and runs OK. The method is robust and nothing nasty is going to happen if your a size out!

Will

[David Thorpe]

I've ordered a selection of guitar wires and some Markits handrail knobs, so once again there's going to have to be a delay, but I can hopefully mark out the frames in the meantime. I shall also devote some time to re-reading (thoroughly!) Will's thread on "Loco Suspension, Fitting CSBs" which I should have remembered includes, among other things, notes on setting up CSBs on a 6-wheel tender.

Am I right in understanding that the term "deflection" used in this thread refers to the maximum vertical movement allowed to the axle by the springs? 0.5mm doesn't seem an awful lot, although I suspect that I have been allowing far too much room for vertical movement in my previous Sharman-type compensation efforts.

DT

[Will L]

...Am I right in understanding that the term "deflection" used in this thread refers to the maximum vertical movement allowed to the axle by the springs? 0.5mm doesn't seem an awful lot, although I suspect that I have been allowing far too much room for vertical movement in my previous Sharman-type compensation efforts.

The deflection we are talking about, correctly called the static deflection, is the amount the CSBs are deflected upwards when the loco is standing still on flat track. In this state, the wheels are carrying the full body weight, the axle centre line is where it is designed to be, and the loco is standing at the designed hight.

The maximum vertical movement allowed when running, called the dynamic deflection, is a further 0.5mm upward. Therefore the overall movement allowed between the wheels unloaded and the CSBs strait, and the horn block hitting the top of the horn guides with the CSBs fully defected, is 1mm. Even this doesn't seem much when you have got used to floppy compensated chassis, but experience has shown it is quite enough. It's those floppy chassis that are wrong. Incidentally, I am given to understand that many real locos where designed to allow 2 inches dynamic deflection, so our 0.5mm is not just practical but also in line with prototype practice. P4 lives!

Will

[Paul Townsend]

Incidentally, I am given to understand that many real locos where designed to allow 2 inches dynamic deflection, so our 0.5mm is not just practical but also in line with prototype practice. P4 lives!

Thats an interesting figure that I have not seen before.
2" movement of a single protype wheel seems huge!
You say many locos were designed that way, what deflection was allowed on the rest?
I hope prototype track joints never reach a 2" step....ouch, that was my coffee spilling!

Presumably designed cant, unwanted subsidence, length of wheelbase etc all play a part.

Were there ever Board of Trade or in Company guidelines to CME's re this ?

If considered too OT mods feel free to start a new topic with this.

[Will L]

Thats an interesting figure that I have not seen before.

Not surprised I've miss quoted my self. I should have said 1.5 inches, which is 0.5mm in 4mm scale and makes much more sense. Sorry.

2" movement of a single protype wheel seems huge!
You say many locos were designed that way, what deflection was allowed on the rest?
I hope prototype track joints never reach a 2" step....ouch, that was my coffee spilling!

Presumably designed cant, unwanted subsidence, length of wheelbase etc all play a part.

Were there ever Board of Trade or in Company guidelines to CME's re this ?

The full size like the model, needs the full range of movement to cope with more subtle changes that occur over the length of the wheel base, rather than the need to deal with extreme single steps. I'm sure there are guidelines for the maximum acceptable changes in cant. I've posted on this topic before but it didn't raise a lot of interest.

That said, don't I remember an American(?) video clip showing repeated failed attempt to cause a test derailment by removing increasingly long chunks of rail?

Will

[barhamd]

The problem I can't seem to get over is that most tenders have the outside axles very near the ends of the chassis, often considerably less than the half the distance between the axles. This seems to me to imply that if you put a pivot point on the centre point between the axles you can't get anything like an equal length of wire on the end pivots.

for example my GER tender has a wheel base of 12 foot but the over hang on the ends is 3 foot 3 inches to the outside of the buffer beams and 3 foot to the front of the tender. Given that the chassis is an inside frame unit which in the PDK kit doesn't extend the full length of the tender I ended up with the option of either very lobsided springs or as I ended up doing threading the wire through an extra handrail knob so that there were two pivot positions between the axles.

I'm sure this will appall people or am I missing something?

David

[Will L]

The problem I can't seem to get over is that most tenders have the outside axles very near the ends of the chassis, often considerably less than the half the distance between the axles. This seems to me to imply that if you put a pivot point on the centre point between the axles you can't get anything like an equal length of wire on the end pivots.

That's not a problem as you don't need that much length for the outer fulcrums. Look at the plots above and you will see the outer two fulcrums are much closer to the axles than the inner two, so the outer fulcrums on a GER tender fit with 2 or 3 mm to spar depending on which fulcrum plot you chose.

It works like this because you have a single spring wire and the inner and outer segments interact. The setting of flucrum points isn't intuitive, which is why you need a method to calculate them or ask somebody who can to do it for you.

Will

[David Thorpe]

Well, I've now largely finished the tender using the F, 8, A, 11, F, 15, A, 11, F, 11, A, 9, F scheme suggested above, and am very pleased with the result. As originally built the tender had a fixed rear axle, with the front two axles on a compensated beam, and it never ran well. I removed all that but decided to try to use the existing MJT hornguides and blocks in situ, adding a third set to where the rigid axle had been. Trouble with the MJT hornguides in a CSB context is that they have a flap on the top, and that had to be removed, as did the top mm or two of the side bearing guides. The hornblocks themselves were drilled to take Markits Original handrail knobs which I then glued in place with epoxy.

I also used Markits handrail knobs on the chassis frames, using the dimensions recommended on the CLAG site (http://www.clag.org.uk/beam-annex4.html) This was slightly problematical as, the Gibson chassis frames being shallow, the holes for the handrail knobs had to be drilled only 1mm up from the bottom of the frames, and, sod's law being what it is, I need hardly add that some of the required positions were within a gnat's crochet of the predrilled holes for the brake gear. Anyway, they too were glued in place.

I used Gauge 9 guitar wire for the springs. Once they were fitted and everything put back together, the tender weiged 57g and sat very slightly too high. I added 20g of lead, and that brought it down to the correct height. The tender now runs smoothly and sure-footedly, a great improvement on its previous performance. I've already earmarked another tender for similar conversion, and will try the High Level CSB hornblock system on that.

I'm now wondering if I should have used CSBs on the loco (an E4). It's got a fixed rear axle, with the front drivers and the pony beam compensated, and althougb all this runs very smoothly the pony is apt to fall off the track from time to time although, to be fair, it does only do that at not-very-good bits of track .

My thanks to Keith and especially Will for their invaluable help in this.

DT