What back to back setting do you use?

[Le Corbusier]

and the recommendation of 17.75 min B to B to ensure good running?

Nothing new there, the same recommendation is in the Standard and is explained by Philip just a couple of posts back.
Regards

Maybe I misunderstood .. I thought that the article's contention was that the digest range was too wide and that 17.67min was too close ... and rather than being the maximum within the P4 range 17.75 should rather be the minimum?

[Martin Wynne]

The article makes the same mistake as modellers down the ages. The belief that the back-to-back dimension is somehow sacrosanct and should be the same for all wheels.

It doesn't matter a damn what the back-to-back dimension is, provided that it is greater than the minimum.

The far more important dimension is from the back of one wheel to the front of the flange on the other. Which must never exceed the maximum, but for the best running should be close to it. What that makes the back-to-back depends on the effective flange thickness of the actual wheel.

Throw away your back-to-back gauges.

Using the check gauge tool, make yourself a simple back-to-flange testing fixture. For details, see:

http://4-sf.uk/

Sroll about half-way down.

Martin.

[Philip Hall]

I found Roger’s article interesting and gave me food for thought. I have always used b-b gauges set at 17.7mm, right in the middle of the minimum and maximum specified dimensions of the P4 standard. As I have said in previous posts, I prefer a little more play between the railheads to allow for a little wobble in wheelsets that might develop.

It does occur to me that this unintended wobble might possibly be the reason that running with a b-b at the top end of the standard has improved running on some layouts; wheels could wobble above but also below the standard, and below would really give problems. I have always found that getting a b-b gauge to sit nicely just between flanges is not easy, and the gravity gauge is most useful here, if the wheelset is as far as possible completely wobble free. This Nirvana is hard to reach, though. So I still feel that my mid position is a good starting point. Certainly I have (so far in 30 plus years) had no problems; also no one that I have built engines for in P4, running on many different layouts, have said that they don’t stay on the rails. I guess that proves something...

Philip

[steve howe]

If it is found that the back to back (sic) is tight on a loco which otherwise runs well, is there any device, implement or instrument other than brute force, that will gently ease the wheels apart in a measured controlled way, without upsetting the quartering or inducing wobble into the wheelset? I believe GW makes a wheel puller, but I don't know how effective it would be working with wheels in situ

Steve

[Philip Hall]

The GW wheel puller works on the tyres rather than the centres so there is a problem if you are working with Alan Gibson wheels; as the tyres are not locked on, just a push fit, it is possible that the tyre will pull off. Ultrascale centres are moulded into the tyre so will not be a problem.

When I build an engine I often find, even when I use the GW wheel press and get the axle length as near correct as I can, that the B-B is not quite right. I don’t want to disturb the quartering, so my method is to support one wheel of the assembled set between two square metal strips (small vice jaws are ideal) and tap the end of the axle with a small hammer and pin punch. Depending on the way you tap the axle you can reduce or increase the B-B setting. This sounds quite brutal but because a straight tap is involved the quartering should not change - it never has on all the occasions I have done this. If you have any doubt then a line scribed across the wheel and axle before you adjust will show if the quartering has altered.

This is trickier when the wheels are mounted in the chassis but with thinner strips of metal it can be done.

Philip

[steve howe]

Thats really useful, thanks Phil.

[Julian Roberts]

But why make it "right" if it runs well? If it ain't broke don't fix it is a phrase that springs to mind. I certainly don't want to contradict Philip's advice in principle if you are going to re do them, but I can't see any reason why it shouldn't work narrow to gauge down to the mid 17.5 s as I have found in practice with some 2 or 3 of my earlier locos through sheer incompetence fixing on the wheels . They still run perfectly!!

[steve howe]

[quote="Julian Roberts"]But why make it "right" if it runs well? If it ain't broke don't fix it is a phrase that springs to mind.=quote]


I meant the locomotive was running well mechanically but one of the wheelsets was tighter to gauge than the others.

[Paul Townsend]

Snooze article from Roger re-focussed my mind on the selection of B2Bs I have accumulated.
I have also found significant variations and agree a preference for the fattest ones.

I like the idea of Scaleforum having a test and measure service and will be happy to help with that at S4North and part-time at next Aylesbury ( Bristol Group will be pretty busy there ).

Tomorrow is the next meet of BS4 ( Bristol AG) and I have circulated a suggestion that we all bring our gauges and measurers.

Of course I am only a humble Group Member. Really Tim Venton should be issuing this as an edict for BS4......watch this space.

[Martin Wynne]

But why make it "right" if it runs well? If it ain't broke don't fix it is a phrase that springs to mind..... I can't see any reason why it shouldn't work narrow to gauge down to the mid 17.5 s as I have found in practice with some 2 or 3 of my earlier locos through sheer incompetence fixing on the wheels . They still run perfectly!!

Correct. The maximum check span in the P4 specs is 17.47mm. You need a running clearance above that, say 0.08mm, which makes the minimum back-to-back around 17.55mm. Anything above that will run fine, it doesn't matter if it is tight on your back-to-back gauge.

Admittedly it might run fractionally smoother through crossings if widened to give a back-to-flange dimension nearer the maximum of 18.15mm, but it's not worth doing if it is going to damage the quartering, paintwork, or introduce wobble.

regards,

Martin.

[Martin Wynne]

A proper wheel-setting gauge. As far as I know only the Gauge 0 Guild supply this type of gauge (etched). It could be laser-cut or produced as a roller. This a rough sketch, the max profile needs proper dimensioning to match the rail-head. The maximum back-to-back will vary according to the actual flange thickness of the wheel.

p4_b2b.png

Martin.

[David Knight]

A proper wheel-setting gauge. As far as I know only the Gauge 0 Guild supply this type of gauge (etched). It could be laser-cut or produced as a roller. This a rough sketch, the max profile needs proper dimensioning to match the rail-head. The maximum back-to-back will vary according to the actual flange thickness of the wheel.

p4_b2b.png
Martin.

An interesting post Martin, in that what you have shown is not that far off the standard NMRA gauge which performs much the same function and also checks flange depth but with possibly less slop depending on the width of the slots you have shown. In moving from HO to P4 I do confess to being puzzled by the use of back to back rather than flanges for setting wheels, but horses for courses I suppose.

Cheers,

David

[Martin Wynne]

An interesting post Martin, in that what you have shown is not that far off the standard NMRA gauge which performs much the same function and also checks flange depth but with possibly less slop depending on the width of the slots you have shown. In moving from HO to P4 I do confess to being puzzled by the use of back to back rather than flanges for setting wheels, but horses for courses I suppose.

Hi David,

There is no slop -- the wheel on the left is always held firmly against the gauge. The slot on that side is simply to aid resting the wheel on it. The gauge could be cut off flush with the back of the wheel on that side, but it would be more difficult to hold and use.

regards,

Martin.

[grovenor-2685]

In moving from HO to P4 I do confess to being puzzled by the use of back to back rather than flanges for setting wheels, but horses for courses I suppose.

If you read the P4 standards they require the use of a standard wheel profile, and with that as a given then back to back gauging is satisfactory.
It is, afterall, how the prototype usually does it.
Where you have to deal with a variety of wheel profiles as in 00, then Martin's gauge is the better way to do it.
Regards

[Le Corbusier]

If you read the P4 standards they require the use of a standard wheel profile, and with that as a given then back to back gauging is satisfactory.
It is, afterall, how the prototype usually does it.
Where you have to deal with a variety of wheel profiles as in 00, then Martin's gauge is the better way to do it.
Regards

Phew !... thats how I understood things.

I just received my brand new DCC concepts B to B gauge in the post today ... it measures 17.73 (edit - mistake corrected) and is very shiny

gauge.jpg

I like the shape which means you can take out a degree of wobble whilst setting the b to b

[grovenor-2685]

I like the shape which means you can take out a degree of wobble whilst setting the b to b

same goes for the L shape ones.
But 17.3 does not cut it, well below P4 minimum, get your money back.
All 3 of my L shape ones measure 17.7 as does my Bernard Weller version.
regards

[Le Corbusier]

I like the shape which means you can take out a degree of wobble whilst setting the b to b

same goes for the L shape ones.
But 17.3 does not cut it, well below P4 minimum, get your money back.
All 3 of my L shape ones measure 17.7 as does my Bernard Weller version.
regards

Sorry keith,

My usual incompetence left out a critical digit .. I should have said 17.73 -now corrected (it's advertised at 17.75 but that's not what my calipers say!)

[Julian Roberts]

I think we should talk about weight.

One loco below has assymetric compensation biased to the leading wheels as per my article in Snooze 199, approx double the weight on outer wheels compared with centre.

[youtube]/6-LeQjdCmx8[/youtube]

The other (an 0-6-2) has even springing (not CSBs) on each wheel.

[youtube]/jQWMhyn3-10[/youtube]

Both worked perfectly in the reverse direction. The pony wheel is sprung to take weight and give directional control.

BB on compensated loco is 17.8ish on front, 17.6ish on rear.

BB on sprung loco is less than 17.75 on front drivers and pony wheelset.

Compensated loco weighs about double the sprung one. Thus the weight loaded on the front wheels of the loco that stays on the track is about 4x the other one.

The same roadholding was the case on many repeated tests on this piece of track. The sprung loco otherwise behaves perfectly (as does the compensated one).

Anyone think the difference of behaviour can be caused by microscopic changes of wheel back to back? Or anyone think it could be caused by differences in magnitudes of the axle weight?

Dunno if the videos will load from u tube

[Martin Wynne]

Dunno if the videos will load from u tube

Corrected youtube links:

[Martin Wynne]

If you read the P4 standards they require the use of a standard wheel profile, and with that as a given then back to back gauging is satisfactory.

It strikes me as optimistic at least to assume that every P4 wheel in existence has the same flange thickness.

Even if the standards do so require.

The frequent discussions on here and in print, including the latest article, and now again in this topic title, about what back-to-back different folks use are largely meaningless without knowing the flange thickness of the wheels they are using.

We had all that with the long discussions on here about using EM wheels for P4. Some with thinned flanges, some not. Back-to-back this and back-to-back that, without anyone ever measuring back-to-flange. Admittedly it is tricky to measure, but it is in fact the controlling dimension.

regards,

Martin.

[Julian Roberts]

Thanks for correcting the video link Martin.

The loco that has 17.8ish BB on the front has once derailed on a crossing V. It is outside (above) the P4 spec despite being regauged because my BB gauge measures 17.75 but I didn't know that when I bought it from the Society. I thought it was 17.67. The wheels wobble of course and at the maximum are around 17.82. Just once to my knowledge that maximum wobble has coincided with a crossing V. So now I know I have to press hard against the gauge to get a 17.75 MAX.

In contrast none of my below spec locos have ever derailed in normal use. It is OK going narrow but not wide. I have one where the maximum wobble takes it to a minimum 17.47 but that is one of the favourite runners for its club exhibition outings.

All this stuff about a magic 17.75 is hocus pocus compared with thinking about maximising weight on guiding wheels and making sure the suspension or whatever actually works, IMHO. Similarly (again IMHO) trying to eliminate minute changes of level in the track is a waste of time compared with putting in effective suspension, as Mike Sharman was proving with his Flexichas concept.

BTW Martin the P4 flange spec says 0.35-40. Do you think that is not in practice what we get when we buy P4 wheels?

[Le Corbusier]

Even if the standards do so require.

The frequent discussions on here and in print, including the latest article, and now again in this topic title, about what back-to-back different folks use are largely meaningless without knowing the flange thickness of the wheels they are using.

We had all that with the long discussions on here about using EM wheels for P4. Some with thinned flanges, some not. Back-to-back this and back-to-back that, without anyone ever measuring back-to-flange. Admittedly it is tricky to measure, but it is in fact the controlling dimension.

regards,

Martin.

Presumably though, if you stick to one wheel manufacturer, then (within the wheel manufacturing tolerances which must be well within B to B tolerances?) it should be possible to set a B to B which works for you?

I have found that by re-setting my wheels with the B to B gauge I posted the image of above I appear to have as a by product taken all the visible wobble out of the wheels .... due I presume the large mating surface. Something I failed to achieve with my L shaped gauge (despite Keith's comments) Now it may be my limited skill set or the limited range of tools I have, but I am not sure the process would be so easy or foolproof using the more accurate/versatile jig you propose.

For people at my level the B to B might therefore be the better compromise?

My experiments on Monsal Dale suggest that for me it is working so far ....running through the crossover I have just built (my first ever). It also suggests that being at the wider end of the B to B range is smoother and more reliable where I am concerned as Roger Sanders suggested it might- the derailments I was getting with my previous gauge would appear to have totally disappeared (previous gauge 17.65 measured with my calipers).

I do wonder if each one of us has different foibles which affect the way we build our locos and stock, and as such differing solutions are required to iron out these foibles ? That is to say that one size doesn't fit all.

[Martin Wynne]

Presumably though, if you stick to one wheel manufacturer, then (within the wheel manufacturing tolerances which must be well within B to B tolerances?) it should be possible to set a B to B which works for you?

Hi Tim,

Yes, hopefully, if all your wheels were purchased at a similar time. But discussions on here could be referring to any of the entire installed stock of P4 wheels going back 40 years.

Also, the critical dimension is effective flange. You measure that by holding the wheel against the rail, and measuring from the back of the wheel to the rail. That is affected by not only the raw flange thickness, but also the root profile, and also of course the rail head profile.

So when someone says "I use X back-to-back and it works fine" it is no help to anyone unless it is accompanied by some information about the wheels (latest delivery from Ultrascale, or some you just found in a stockbox from about 1987, or on a loco you built in 1975?), and the rail head profile (latest from the Stores, or running on your layout built from Ratio rail about 1980?).

I recall some code 75 bullhead rail from I think Millholme Models(?) in the 1970s which had such a sharp top corner profile that you could cut your fingers on it while handling it. The effective flange thickness of wheels running on it was significantly increased.

Martin.

[David Knight]

In moving from HO to P4 I do confess to being puzzled by the use of back to back rather than flanges for setting wheels, but horses for courses I suppose.

If you read the P4 standards they require the use of a standard wheel profile, and with that as a given then back to back gauging is satisfactory.
It is, afterall, how the prototype usually does it.
Where you have to deal with a variety of wheel profiles as in 00, then Martin's gauge is the better way to do it.
Regards

Fair points all. As to the prototype it was the picture at the end of Roger Sander’s article in Snooze 205 that made me think of the NMRA gauge because that’s exactly how it works. Maybe it’s just an S&D thing?

Cheers,

David

[grovenor-2685]

Fair points all. As to the prototype it was the picture at the end of Roger Sander’s article in Snooze 205 that made me think of the NMRA gauge because that’s exactly how it works. Maybe it’s just an S&D thing?

The writing on that gauge only identifies it as a carriage and wagon standard, to me it looks to be a back to back gauge and includes a flange check as well. These days it is more usual to have separate gauges for the flanges and the BB. The nmra gauge works as per Martin's diagram except that it does not include a rail gauge corner. Difficult that as that requires knowledge of the detailed rail profile that is in use, which varies as mentioned by Martin above.
Regards