Axle weighing - first experiments

[oldnorton]

I have reread what was written and my apologies to robmort for inferring there was anything incorrect in what he said. To be fair he didn't say steel was not elastic, he just correctly said that for other very elastic (and plastic?) materials they soon develop lots of surface shape.

Roger, if one cannot use simple coefficients of friction to determine adhesion of wheels on rails (which we are all guessing is the case) then what is modifying that coefficient, and can anything help explain why eight wheels might be better than six, with the same total mass?

Or perhaps eight wheels are not better than six for grip. One reason for eight is to have smaller diameters and thus greater tractive effort.

[Deleted]

May 16, 2021 9:15:07 GMT robmort said:

Do we accept that?

Do we know that 2 wheels can't provide the same grip as 6 with the same total load?

Friction is linearly proprtional to the weight applied, and is independent of the area of contact. So 2 wheels should be the same as 6 with the same total weight.

Full-size locos had more wheels to limit each axles load not to provide better adhesion.

[/quote]

In my own head, 6 wheels should always have more grip than 2, surely as well as the weight applied, you also need to consider the amount of contact with the rails. 6 contact points being better than 2? It doesn't make sense to me that contact area can be independent?

Pete

[Deleted]

Looking at this again.. surely contact being independent only applies to a part that's sliding, not rolling? I don't think that friction is the correct rule to be applied here? So for me.. a powered set of 6 wheels has to have more grip than 2???

[uuu]

We're still talking "static" friction in the rolling situation. But you're right in your earlier post - if you have more wheels, you can increase the weight of the loco, and remain within axle limits. And the extra weight is going to give you more traction.

Modern traction systems have controlled slip. It's fascinating to see a diesel pulling away with a big train behind - each wheel is going slightly faster than looks quite right.

Wilf

[Roger]

May 17, 2021 8:43:10 GMT oldnorton said:

I have reread what was written and my apologies to robmort for inferring there was anything incorrect in what he said. To be fair he didn't say steel was not elastic, he just correctly said that for other very elastic (and plastic?) materials they soon develop lots of surface shape.

Roger, if one cannot use simple coefficients of friction to determine adhesion of wheels on rails (which we are all guessing is the case) then what is modifying that coefficient, and can anything help explain why eight wheels might be better than six, with the same total mass?

Or perhaps eight wheels are not better than six for grip. One reason for eight is to have smaller diameters and thus greater tractive effort.

To be honest, I don't know the answer to those questions. If the greatest tractive effort is achieved at the point it's actually slipping, then it sounds like these things have been determined empirically. As you rightly point out, you have to consider the diameter of the wheel as well as the loading, because the contact patch will be a different size for a different wheel. This is presumably one reason why you see lots of small wheels on modern locomotives and not less large ones.
The contact patch area versus the load is obviously not a straight line relationship, so adding ever more load won't increase the friction linearly. Worse than that, the pressure across the contact patch is going to vary enormously. The edge of the contact patch is going to be experiencing much less pressure than the centre.
I did find this document online which gives some insight as to the complexity of this subject.

I don't think you can necessarily equate full size results to ours either. The grades of Steel used on the tracks is different, if it's Steel at all, and many wheels are Cast Iron, which is a poor choice to begin with.

It would certainly be interesting to measure the force it takes to drag a 6 wheeled locomotive and then try it with the middle wheels carrying no load.

[andyhigham]

Something that I believe is of utmost importance. Under normal running conditions the suspension should be around 1/4-1/3 compressed. This will allow the wheels to follow dips and hollows in the track. Also when a wheel hits a high spot it avoids lifting. the next wheel along off the track.
As I pointed out in the other thread, centre of gravity is also extremely important

[coniston]

This is all very interesting theory however completely impractical, for example an 0-2-0 loco would not go anywhere so as a minimum a set of trailing or leading non powered wheel would need to be added so why not couple them as an 0-4-0? (Rocket vs Invicta?)

I think we can all agree that more wheels are added to a locomotive chassis for other reasons than just friction, hence traction. Axel loading being probably the most critical along with eliminating overly long wheel bases (think of an 0-4-0 version of a 9F) Structural integrity of the frames and other structure would be a significant problem when considering the UK loading gauge, so more wheel sets spreads the heavier load, not necessarily increasing the total adhesion, but as Greenglade says it possibly does.

Yes Andy you are correct that getting the CoG as close to the centre of the driving wheels enable the most even spring and hence axel loading but one must not forget that any leading or trailing wheel sets also need to carry enough weight to do their job, so maybe the ideal is to get the CoG as close to the longitudinal centre of the loco is ideal? That is what I am trying to achieve with my B1 by adding lead to the rear of the loco.

I have now got as far as creating the space and am now 3D printing some blocks of peculiar shapes to fit around all the various pipes, valves, levers etc to make the most of the available space. These will then become the moulds for casting the lead hopefully using plaster for the moulds, not forgetting to oven bake them before poring the molten lead.

[lesstoneuk]

The friction between the wheel and rail is finite. So an 0-2-2 will not be able to move huge loads if fitted with a 250lb boiler and 21 inch cylinders. But put more axles on and once you get within the acceptable parameters (axle load, no of axles, wheelbase) you can take it to the limit. Also, the GWR proved this many times on loco exchanges, on acceleration, there is a weight shift rearward. If the rear wheels are drivers then extra weight pushes down to increase traction, but if the rear wheels are carrying wheels then that useful weight shift is lost and the available tractive weight goes down

[Roger]

May 18, 2021 7:11:20 GMT lesstoneuk said:

The friction between the wheel and rail is finite. So an 0-2-2 will not be able to move huge loads if fitted with a 250lb boiler and 21 inch cylinders. But put more axles on and once you get within the acceptable parameters (axle load, no of axles, wheelbase) you can take it to the limit. Also, the GWR proved this many times on loco exchanges, on acceleration, there is a weight shift rearward. If the rear wheels are drivers then extra weight pushes down to increase traction, but if the rear wheels are carrying wheels then that useful weight shift is lost and the available tractive weight goes down

I guess the maximum tractive effort is going to be when all of the wheels are carrying the same load at the point of slipping, which includes the shifting of weight to the rear. Just how much a locomotive should be 'nose heavy' is one for experiment. I guess you could anchor the locomotive to a fixed point while turning the wheels to the point where they slip. I could do that on my setup and watch the weights on the axles, adjusting the amount of weight on the nose until they all read the same weight as they slip.

[andyhigham]

The problem factoring weight transfer into the design (in full size) is that locomotives are used in both directions. Designing in beneficial weight transfer for traveling chimney first will become a handicap when travelling bunker/tender first

[Roger]

May 18, 2021 8:47:25 GMT andyhigham said:

The problem factoring weight transfer into the design (in full size) is that locomotives are used in both directions. Designing in beneficial weight transfer for traveling chimney first will become a handicap when travelling bunker/tender first

True, but for us it could make quite a difference.

[Deleted]

From what I've read the general design was to have all main drivers supporting the same weight, the exception to this was 1470 Great Northern which was originally drawn to have 24 ton over her leading driver with 20 ton over the crank and rear drivers. This was later changed to IIRC 22 tons over each, most likely due to gauge as 24 tons was above the allowed limit for the track and civil engineering allowance.

To add some real context to the discussion I can give figures for the GWR 111 Great Bear, NER Raven, and GNR 1470 Great Northern:


111: GWR

Front Bogie: 19 tons 18 cwt

Leading driver: 18 tons

Middle driver: 18 tons

Rear driver: 18 tons

Trailing wheels: 17 tons 8 cwt

Total weight: 97 tons


Raven: (under construction at the time) NER

Front bogie: 19 tons

Leading driver: 20 tons

Middle driver: 20 tons

Rear driver: 20 tons

Trailing wheels: 18 tons

Total Weight: 97 tons



1470: (as drawn) GNR

Front bogie: 17 tons 1 cwt

Leading driver: 24 tons

middle driver: 20 tons

Rear driver: 20 tons

Trailing wheels: 15 tons 8 cwt

Total weight 92 tons 9 cwt

For my own model, I plan to use full-size weight percentages over each wheelset as a starting point and see how things go from there. Like Roger and Coniston I can weigh each wheelset or individual wheel to achieve this.

One interesting thing to note on the above weights is that other than the 'Raven' which was still under construction so perhaps not tested yet, the weights don't add up? So were the axles loaded up to give more grip? the overall weight wouldn't change but is it possible that loading up an axle was a deliberate act in distributing weight? How does that even work? I'm afraid that I haven't looked into this yet but the figures given do beg the question?

Pete

[andyhigham]

That gives a good idea of where we lose out on scale. A 5" version (approx 1/12 full size) of the 97 Ton loco weighs a lot less than 8 Ton

[Roger]

Hi Pete,
Those are interesting figures. It will be interesting to see what you end up with. My gut feeling is that you only want enough weight on the undriven wheels to make the locomotive safe. Any more than that is going to cost you dearly in grip. The relative roughness of our tracks mean that putting too much weight on the bogies is sometimes going to leave the driving wheels carrying very little weight. That situation is made worse if you have stiff springing.

[uuu]

I think "Bridget" was designed to have about 25% on the rear pony truck.

Wilf

[Roger]

May 18, 2021 10:42:30 GMT andyhigham said:

That gives a good idea of where we lose out on scale. A 5" version (approx 1/2 full size) of the 97 Ton loco weighs a lot less than 8 Ton

Did you mean that? A 5" gauge version isn't approx 1/2 full size, is it...

[Deleted]

May 18, 2021 10:42:30 GMT andyhigham said:

That gives a good idea of where we lose out on scale. A 5" version (approx 1/2 full size) of the 97 Ton loco weighs a lot less than 8 Ton

ah.. scaled weight.. above my pay grade.. I think you cube the scale and divide by 100... I think? if so 5" or 11.3 cubed is 1442 (I've rounded this) 97/1442 gives 0.0672 or have I that the wrong way around? 1442/97 gives 14.865? I'm sure that I'll wake up later..

Pete

[Deleted]

May 18, 2021 10:52:47 GMT Roger said:

Hi Pete,
Those are interesting figures. It will be interesting to see what you end up with. My gut feeling is that you only want enough weight on the undriven wheels to make the locomotive safe. Any more than that is going to cost you dearly in grip. The relative roughness of our tracks mean that putting too much weight on the bogies is sometimes going to leave the driving wheels carrying very little weight. That situation is made worse if you have stiff springing.

I think judging the weight over the front bogie is crucial for safe driving, clearly, full size put a lot of weight over the front bogie, not far off that for the drivers? I wonder why 1470 was originally drawn to have more weight on the front driver than the middle and rear? I mean, what was the thought process there? I've not seen that for other designs.

[ettingtonliam]

May 18, 2021 9:48:57 GMT Roger said:

May 18, 2021 8:47:25 GMT andyhigham said:

The problem factoring weight transfer into the design (in full size) is that locomotives are used in both directions. Designing in beneficial weight transfer for traveling chimney first will become a handicap when travelling bunker/tender first

True, but for us it could make quite a difference.

How many of our size locos are driven bunker/tender first?

[uuu]

Using 5" to be one eleventh scale, then 97 tons scales down to 163 pounds.

Wilf