Le Corbusier wrote:Julian Roberts wrote:Edited previous post
Tim yes it must have been even further back!
Individual guitar type wire springs for each wheel
Julian,
I am interested to discuss this further.
Did you measure the weight at each wheel in your set up? I would be very interested to know what is going on at each axel. I assume from what you say the C of G may be somewhat incidental as according to your snooze article hypothesis it is the increased weighting over the front ( and presumably rear) axels along with the increase in weight over all that is the critical factor. Why were the 'individual springy beams' to the front wheels beefed up? Was this to do with the increased weight added at the front? If so was there any issue with the loco sitting level? Were there any changes made to any of the other beams? I assume that the weight at the rear over one or other of the axels would be even greater than at the front given the C of G? Presumably prior to adding the weight the front of the loco was very lightly loaded with the C of G even further back and perhaps explains the derailment almost on its own?
Do you think the change to the B to B had any effect?
CSBs ..... I might be being a little thick here and perhaps Will or Ross or anyone really could advise?
If I understand CSBs correctly then the C of G needs to be placed at the point to coincide with the spread sheet ... which is often near as damn it central.
However, if to achieve the correct overall balance the weight is (of necessity because of the limited positions to apply weight) placed eccentrically within the loco such that the front end is loaded and the rear is loaded even more (but nearer to the fulcrum) such that the whole is in balance as far a C of G is concerned but eccentric as far as loading at each wheel is concerned (prior to springing) with the centre wheels having less weight .... would the CSB springing (being continuous) equalise this out such that the weight is evenly distributed across the wheels? Or do loadings still increase at the different wheels according to the distribution after the springing? If this is the case would the weight have to be re-adjusted to ensure an even distribution so depression of the beam is not eccentric and the loco sits level?
Sorry this is straying a little from Back to Back dims.
Hi Tim
The builder of the loco who I will call Mr D is a more typical member of the Society reticent about putting pen to paper in a digital way. Also I'm trying not to say the chassis kit maker's name as I don't want to in any way blacken his reputation. In fact to me the fact that every wheel has its own spring that can be height adjusted as well as have different thickness of guitar string to give different spring character is the best possible advert for his style of suspension.
I can't fully answer for Mr D's every thought but we had an email conversation following the meeting where I took the video. Basically I suggested to him the problem might be that the weight wasn't concentrated enough on the front wheels. At first I suggested heavier springs on the front but then realized lighter centre ones on the centre axle would be a better option. However by then he had changed the front springs. I suggested to him this might not work except in combo with more weight at the front - which it was possible for him to increase. He also, being aware of this thread, increased the BB. (I had not appreciated the CofG was so far back till the second test which was more than a week later.)
So whether any one of those actions would have achieved the desired result is open to question.
So as I said previously the C of G is obviously important/vital, but so blindingly obvious it scarcely needs to be mentioned (except
that's disproved by the fact I'd not thought about it with Mr D's loco at first!). But a full understanding does help, I have realized from the Articulated Beams thread. My lack of full understanding of it doesn't invalidate my article, because I was writing there what I had been doing for ten years achieving what I wanted to achieve. Of course scepticism is perfectly legitimate as anyone can claim any darn thing they want and people may believe them. I wanted people to know what I had found because I thought it might help!!!!
I don't think the sitting level of the loco has been changed. No other beams were changed. Yes I think the load would have been greater on the rear wheels than the front with the CofG so far back (it must have been more or less at the rear wheel prior to extra weight being added at the front) and the imbalance of the loco could have explained or gone a long way to explaining the derailment prone nature of the loco as was.
Trying to get my head around what you think might be happening. Would I be right in thinking that if the weight is different at each axel, then the spring at each axel needs theoretically to be a different thickness to counter the loading and ensure the loco sits level (or at least front and back as there might be a degree of bridging at the middle)
Yikes this is getting a bit headbanging for me! Getting the loco sitting the right height is difficult with compensation and another ball game with springing. Maybe have a look at my ramblings on my Crab build thread but that's work in progress still (being somewhat delayed by this indulgence in the immediate gratification of seeing visual results so quickly on a digital screen.) I'd basically suggest the CofG needs to be at the centre of the fixed wheelbase on a model steam loco, so the springs would be equally strong on my "guiding" wheels (and in my terms lighter on the centre ones. Whether that is a practical proposition with CSBs is of interest to me and where Will disagrees with me...which is where the other thread talking about weight on wheels is interesting to me (Engineering Injuneering and Knitting)). Normal CSB theory is I think equal weighting on all driving wheels to maximize haulage power - which as this thread has revealed is another commonly misunderstood business in our little world. As Will has said, it maximizes haulage power for a given weight, but it's the amount of weight that matters most for actual amount of pull haha. As my video there shows
https://m.youtube.com/watch?v=Q3JINX9Gnv0 each wheel on the real thing weighs enough to depress the length of rail between three sleepers by an inch or two if there is a void under one of them. That could not possibly happen in our scale. So we have far less weight on each wheel that IMHO we really need for properly reliable track holding that will not rely on perfect track (in my opinion a uselessly idealistic objective), so IMHO we need to maximize it on the guiding wheels, though intermediate ones still need
enough to keep them on the rails too as they are needing to move sideways on curves.
I think that has answered your last paragraph. Hope so. Oh sorry - well I don't know if the BB alteration on its own would have done the trick. Mr D did all the alterations at once so I couldn't tease all this out. Others may have a go. But the thing about BB is that every time you fiddle about with it the wheels get looser so it's going to become less and less reliable, as they may alter themselves - for all the clever dick work getting them to a precise 17.75 (at one point of their circumference - bet it won't be all the way round!) With a steam loco the quartering will be likely to be spoiled with this fiddling about, and if wheels are at all loose likely to be out soon, if not already made so by the fiddling about. Only on the new (as of a few years ago) design Exactoscale wheels can the BB be adjusted without fear of this problem. I lock the wheels on a steam loco with Araldite and steel pin as soon as they're on the axle and I just don't believe all this stuff about BB for the reasons I have already said, and my proven experience which arose from originally having a below spec BB gauge. - So long as the BB is not more than 17.75. (The 6ft 6ins rear wheel on my 4-4-0 is larger than that at one point of its wobbly state of being, so one day it will come off when going tender first when it happens to coincide with a crossing V, but that hasn't happened so far...)