Better Running

Discussion of model and prototype wheel/rail interaction.
billbedford
Posts: 737
Joined: Tue Oct 02, 2012 2:40 pm

Better Running

Postby billbedford » Sun Oct 27, 2019 8:21 am

I keep getting told that some stock will run 'better' than others, and I have to admit that I have been sceptical. Then I found this:

image.png
image.png (102.97 KiB) Viewed 4185 times


Which begs the question; 'Just what can I measure to prove that the way the stock runs has been improved?'

Let's keep it simple and say that I am working on replacement bogies that should 'just work'. I'm not, for the moment, interested in locos.
Bill Bedford
Mousa Models
http://www.mousa.biz

User avatar
Guy Rixon
Posts: 828
Joined: Sun Mar 27, 2011 6:40 pm

Re: Better Running

Postby Guy Rixon » Sun Oct 27, 2019 9:18 am

The trackholding experiments published previously in MRJ all involved a vehicle free-wheeling down a plane through various types of pointwork and imperfect track. One hundred runs were done for each vehicle configuration and the number of successful runs tabulated. I see two improvements to that.

One hundred runs is not enough to show up the rarer problems. In the tests of 4mm-scale wagons, the sprung wagons of different formats all scored 100%, so we never found out which springing system worked better. Also, a success rate of 99% translates to quite a low MTTF in actual running, if every passage of a turnout introduces one chance to fail. One would need 1000 or more runs to get decent statistics, and this would be very tedious on a hand-worked gravity-test.

Rolling a vehicle down a slope under gravity sounds like it simulates hauling that wagon through the questionable track. Propelling the vehicle would be more testing for the suspension.

Therefore I suggest that you need a level test-track with an automated loco to push and pull the stock under test. It would be easiest to work with if it could alert you when the stock derailed (some kind of AC track-circuit?). A web cam to record the derailments would be good.

User avatar
Guy Rixon
Posts: 828
Joined: Sun Mar 27, 2011 6:40 pm

Re: Better Running

Postby Guy Rixon » Sun Oct 27, 2019 9:26 am

You could also try and assess the smoothness of ride from the vibration spectrum. You'd need to know, or discover, the signature of a lurch in the frequency spectrum, and you'd need an on-board accelerometer to obtain the spectrum. Small accelerometers with wireless connection are available.

I suggest that high-frequency vibrations don't matter in models, even though they matter a lot in full-sized trains. The metric would be the elmination of lurches and of visible, sustained wobbles around 1Hz. You could also try tuning the stock to eliminate noise by reducing vibration in the 100-100Hz range, but I'd consider that a secondary goal, as I've never been bothered by bogie noise from models.

You could also assess the ride with a sideways-mounted camera on board the train, but it would be subjective.

User avatar
jim s-w
Posts: 1864
Joined: Wed Jul 30, 2008 5:56 pm

Re: Better Running

Postby jim s-w » Sun Oct 27, 2019 10:00 am

I think the easiest way is to just run stuff. A weekend exhibition* will show up problems. Of course it’s less controlled and there’s a lot of variables so you could argue it’s less scientific but it’s also more ‘real world’

Of course if something falls off and you just plonk it back on and carry on it’s not going to tell you much. As I’ve massively oversubscribed the stock for my little layout I have the luxury of sending stock to the naughty step if it falls off.(discounting operator error) So far my to do list has reduced after each show. I think there was half a dozen wagons and my 03 (first show and the COG was off) in the box after DEMU which was not only less than scaleforum but also significantly less than the amount of new stock that I took along. I do try to run all new stock across all track before it’s signed off so to speak.

It’s not really scientific in the way I think Bill is looking for but it seems to be working for me.

Jim

*of course a weekend exhibition doesn’t have to be in front of the public.

davebradwell
Posts: 480
Joined: Fri Jan 18, 2019 3:48 pm

Re: Better Running

Postby davebradwell » Sun Oct 27, 2019 12:59 pm

A very good question and logical responses which have already split between numbers of derailments and smoothness of ride - might there be a link between the two? What type of suspension do you fail most often, Jim?

Suggest that you would need to push sufficient vehicles to compress the buffers a little as this is where springing is supposed to be an advantage - a gradient would help here. Also noise is important to us so should be an added factor as there is an opinion that sprung vehicles are quieter. As the loco noise might make such measurements on the test wagon difficult, then we might be on to cable haulage with the motive power off-piste.

It might be difficult to differentiating between springing systems because, after all, a spring is a spring unless it's fully compressed when it's not a spring. Tests could give an idea of optimum spring rate but suspect this is influenced by how bad the obstacle course is.

If you're looking at the frequency response of a suspension then it's damping that's the important factor. Unfortunately we've only a bit of (very random) friction in the axlebox guides to do this although Chris Pendlenton's bogie design does include a sort of adjustable damping on the (he thinks more important) secondary springs. You might be into running the tests at different speeds here.

The MRJ tests were on 7mm scale wagons and in 4mm scale the results would be more prone to sticking guides and poor setting. All in all a big job and quite difficult. It would be very nice if someone else did it.

DaveB

User avatar
grovenor-2685
Forum Team
Posts: 3507
Joined: Sun Jun 29, 2008 8:02 pm

Re: Better Running

Postby grovenor-2685 » Sun Oct 27, 2019 1:08 pm

If these are coach bogies then I suspect that a free rolling test as described by Guy, and with each bogie variant tested on the same vehicle, is the best you can do. As soon as you couple it to another vehicle you bring in a whole load of additional variables from interactions through the couplings, buffers and gangways that also need to be controlled and measured. Equally important for reliable operation of course but you do need to isolate measurement to the parameter you are trying to improve.
Regards
Keith
Grovenor Sidings

User avatar
jim s-w
Posts: 1864
Joined: Wed Jul 30, 2008 5:56 pm

Re: Better Running

Postby jim s-w » Mon Oct 28, 2019 8:28 am

davebradwell wrote:A very good question and logical responses which have already split between numbers of derailments and smoothness of ride - might there be a link between the two? What type of suspension do you fail most often, Jim?
DaveB


There’s no real pattern. Other than to say anything that’s not set up right, axles not parallel or skewed will likely cause a problem regardless. Most the stock is rigid, some sprung but I only use rocking compensation (inside type) on brake vans (I want the extra resistance).

I’ll also throw in that sprung buffers that aren’t smooth cause far more problems than rigid ones.

Jim

Philip Hall
Posts: 1632
Joined: Mon Aug 10, 2009 7:49 pm

Re: Better Running

Postby Philip Hall » Mon Oct 28, 2019 1:13 pm

I had a little experience of this last weekend, running my converted Bachmann Dukedog pulling a very heavy (860g) four and a half car rake around the Epsom Club’s ‘Wadhurst’. All vehicles - four Hornby, one Bachmann PMV) were marshalled (sprung) buffer to buffer including the engine tender. The whole train reliably glided around (as did several others) with the train ‘of a piece’, not a shimmer or a shake, taking curves and slight gradients and very complicated pointwork in its stride. None of this train was suspended in any way, apart from maybe the give in the plastic bogies. However, all wheels were dead concentric, either supplied that way or fettled.

Which leads me to think that in many cases getting the basic engineering right is the most important route to reliability. When I finally get some of this running at home I will be able to truly find out whether my money is where my mouth is!

Philip
Last edited by Philip Hall on Mon Oct 28, 2019 8:05 pm, edited 1 time in total.

User avatar
Tim V
Posts: 2523
Joined: Tue Jul 29, 2008 4:40 pm

Re: Better Running

Postby Tim V » Mon Oct 28, 2019 2:33 pm

Agree there Philip, decent wheels are key to good running. Unfortunately, most modellers seem to want to race to the bottom, and won't pay more than the bare minimum for wheels, but then moan when they can't get good running.
Tim V
Scalefour News Editor

Proton
Posts: 53
Joined: Tue Apr 23, 2019 10:54 pm

Re: Better Running

Postby Proton » Mon Oct 28, 2019 3:55 pm

I agree with Jim's approach, run as many trains as you can. When I am in this mode, I triple head very long trains of up to 20 cars, and the odd vehicle which is below par shows up after a short time. It also highlights those with kadee couplers that are too tight!
PBM.

User avatar
Le Corbusier
Posts: 1588
Joined: Wed Feb 11, 2015 3:39 pm

Re: Better Running

Postby Le Corbusier » Mon Oct 28, 2019 4:43 pm

Philip Hall wrote: However, all wheels were dead concentric, either supplied that way or fettled.

Philip


Philip,

I would be interested to know your fettling technique? ... particularly given Gibson Wheels as a starting point. :thumb
Tim Lee

Philip Hall
Posts: 1632
Joined: Mon Aug 10, 2009 7:49 pm

Re: Better Running

Postby Philip Hall » Mon Oct 28, 2019 6:43 pm

Tim, essentially it involves using a lathe to (where necessary) skim the wheel centres and gluing the tyre back on with epoxy.

Philip

billbedford
Posts: 737
Joined: Tue Oct 02, 2012 2:40 pm

Re: Better Running

Postby billbedford » Tue Oct 29, 2019 9:15 am

Philip Hall wrote:Tim, essentially it involves using a lathe to (where necessary) skim the wheel centres and gluing the tyre back on with epoxy.

Philip


If you are skimming the centres, how do to ensure the rims are going on concentrically?
Bill Bedford
Mousa Models
http://www.mousa.biz

Philip Hall
Posts: 1632
Joined: Mon Aug 10, 2009 7:49 pm

Re: Better Running

Postby Philip Hall » Tue Oct 29, 2019 9:30 am

The tyres are usually quite a tight fit on the centres, so just a little skimming will produce a sliding fit rather than a very stiff one. The tyre is then held with epoxy.

The definitive article about this has already been written by Chris Pendlenton in MRJ 219 and 221, I just used some of his ideas.

Philip

User avatar
iak
Posts: 504
Joined: Thu Nov 05, 2009 10:28 am

Re: Better Running

Postby iak » Tue Oct 29, 2019 11:03 am

I just apply a bit of persistence, patience and reasoning.
Should an item misbehave, why? Work the issues and make changes.
Be prepared to pay for wheels, proper ones.
Test run extensively; the most recent items spent 2 hours running around Mostyn. Now nothing misbehaved but if anything had, it would have been examined and trouble shot...
I am no engineer but 35 years as a laboratory technician/manager taught me to always follow a logical, problem solving protocol. Work the issues and understand why they happen...
I don't suffer from insanity, I enjoy every minute of it....

Perfection is impossible.
But I may choose to serve perfection....
Robert Fripp


https://www.facebook.com/groups/PadgateWorks/

Philip Hall
Posts: 1632
Joined: Mon Aug 10, 2009 7:49 pm

Re: Better Running

Postby Philip Hall » Tue Oct 29, 2019 4:28 pm

Iain’s comments are exactly the way I feel we should approach things. One of the reasons I set much store by the late Walter Ward-Platt’s article, reprinted a little while ago in Scalefour News. Testing extensively and settling on a way of doing things that feels right in an engineering way (even though I am no engineer!) is working for me. And I find, it is more satisfying and less hassle to assemble components that are right to begin with.

Philip

User avatar
grovenor-2685
Forum Team
Posts: 3507
Joined: Sun Jun 29, 2008 8:02 pm

Re: Better Running

Postby grovenor-2685 » Tue Oct 29, 2019 4:51 pm

Whilst such intensive running methods are fine for those of us with layouts they don't really help Bill with his question, how does he confirm that a design change has made an improvement? As a manufacturer time is money.
Regards
Keith
Grovenor Sidings


Return to “Vehicle dynamics and trackholding”

Who is online

Users browsing this forum: No registered users and 2 guests