P4 vs S4 - Pro's and con's

[Knuckles]

It's all very interesting and technical, never thought creating this thread would take off but I'm glad it seems to have been utilised well.

As a bone question then, and this is without me doing any maths as is not isn't my quickest ability.

For a P4 crossing what about keeping the flangeway the same except on the check rail. Maybe making the checkrail to S4 would pull the wheels away from the crossing more?

The suggestion is probably silly though, for as a guess it in my mind seems it would do one of two things.
A) work
B) effectively become a retarder and pinch the wheels.

Worth asking though.

[Julian Roberts]

Hi Knuckles - I think your original question has turned out to be pretty interesting! I like the term "bone question"! Maths is not my quickest ability either. As for your check rail question I don't quite understand what you mean * but that seems a good one too, as (AFAIK) the S4 Check Gauge is 18.25, in P4 it used to be 18.15, and since 1998's Digest 1.2 is 18.15 - 18.20. I'm still trying to get my head round all the to-s and fro-s of the last few pages and wasn't going to write anything further till I'd understood more, and/or done some experimentation, but no one else has answered you...

As an aside re the previous conversations with Tony and Russ I asked John in our group, who built "Kettlewell" (exhibited at Ally Pally recently) about his 1:8 single slip which is central to the layout and which every train must go over, whether he has had any derailment problems on it. "Never 'appens" he said (he's from Yorkshire). I asked what BB he uses - 17.67. Preassembled acute crossings were used in its construction. I write this simply to report his experience.

*Andrew Jukes on the Track and Wheel Standards thread describes what seems to be for him a workable way of doing S4 but with P4 crossing flangeways if he wants to use the Exactoscale 0.68 chairs. His BB minimum is 17.82. Maybe that answers your question...?

[Knuckles]

"Never 'appens" - I often say that too and I'm from Nuneaton.
I'm unsure if our wires are crossed or not. My suggestion and queery was just in one area really, if we keep the crossing and wingrail with the same P4 gap but reduce the checkrail by the stock rail to S4 I was thinking that may sort the diamond issues out by pulling the wheels a wee further away from the crossijg nose.

I said it is probably a bone question as I'm not a track expert and someone will possibly do the maths and say it is a bad idea.

[grovenor-2685]

It doesn't help with diamonds as the reputed problem is at the obtuse crossings which are opposite each other and will have identical flangeways.
The so called problem only shows up in extreme conditions such as 1:8 crossings on curves, worry about it when it happens to you!
Regards

[Martin Wynne]

It doesn't help with diamonds as the reputed problem is at the obtuse crossings which are opposite each other and will have identical flangeways. The so called problem only shows up in extreme conditions such as 1:8 crossings on curves

1:8 fixed K-crossings are not permitted on curves on the prototype, so if you build one you are asking for trouble.

The solution to these problems is to build a switch-diamond instead. Easier to build, and 100% reliable.

Martin.

[Julian Roberts]

All this has been worked out donkeys years ago

So said Philip on Page 1. All these discussions took place in the past surely - is there anywhere one can read them?

Bone question from me. The accuracy of the Check Gauge tool, given its accuracy of manufacture, also depends on (1) using it correctly, and (2) the exact dimension of the rail, which Russ quoted as +/- 0.03 - right? So that's 18.12 - 18.18 variability, right?

Would it not be a way round this variability to have a spring load on the flange that bears against the check rail that is being fixed on, so that this rail is pushed against the outer flange, which is made to the correct 18.15 distance from the outside of the flange that bears against the crossing V...?

Scruffy drawing attached.

20170602_164445.jpg

Or am I missing the point re the check gauge thingy problem?

[Martin Wynne]

Or am I missing the point re the check gauge thingy problem?

Hi Julian,

No you're not -- a sprung check gauge tool is much preferable to fixed slots.

But your sprung collar design is more complex to make, because it can't be made in one piece.

There have been numerous designs over the years involving bits of bent paper-clip, hair-clips, etc.

For example:

springy_check_gauge.png

File off a portion of the inner flange on the existing turned check gauge tool. Drill a diagonal hole through it. Open out the hole to a clearance on one side. Solder in a bit of springy wire so that it is still free to move.

Martin.

[Julian Roberts]

But your sprung collar design is more complex to make, because it can't be made in one piece.

Hi Martin
Here are some track gauges not made in one piece. Seems to me the check gauge tool is a critical (once in a lifetime should be enough!) purchase and one that should not be made to be as cheap as possible but as accurate as possible for a reasonable price. I don't think any of these gauges are or were terribly expensive.

20170603_092949.jpg

Capture Protfour Check Gauge.JPG

I haven't worked out if the spring on the Protofour gauge has the same function as the design as we were discussing above - I think not? But surely a design could be devised to do so? (By the way the rail I happen to have is 0.9017, while the gap for that rail is 0.92. - measured with micrometer and feeler gauge respectively.)

I wonder what the accuracy of a tool with functional spring costing say £20 - £25 could be. Presumably 18.15 is the minimum so any inaccuracy (is tolerance the correct word?) should be in the + direction only; but could it be less than the +0.05 that the Digest realistically mentions?

[Knuckles]

Just for marketing reasons I'd be willing to pay a high price for said gauges if they truly will help produce more accurate track. As said above, only buy once.

Unless you loose or break it.

[grovenor-2685]

How accurate do you want the track?
The currently available gauges allow you to produce accurate enough track and have done for 50 years now
Regards

[Julian Roberts]

What BB do you use Keith? With 17.67 BB I can see exactly your point.

Looking back I see Martin has already said a minimum sensible tolerance manufacturing a
check gauge is +0.03.

[Jol Wilkinson]

I really wonder where this is all leading.

While it is important to question whether we are doing things in the best way, as Keith points out, the standards and tools that have been used for decades gives satisfactory results if used carefully.

With regards to manufacturing tolerances for tools, they are probably smaller than you can achieve when using them. So seeking finer tolerances may be a waste of time.

[David Thorpe]

I couldn't agree more, Jol (and Keith). Anyone not into P4 who reads this thread must wonder if it's all worthwhile, if they're not put off completely. If you use the gauges that have been tried and tested over the years, it's care, not micrometers, that lead to good running.

DT

[grovenor-2685]

What BB do you use Keith? With 17.67 BB I can see exactly your point.

Looking back I see Martin has already said a minimum sensible tolerance manufacturing a
check gauge is +0.03.

I set the BB using the BB gauge, I don't measure it.
Regards

[Martin Wynne]

How accurate do you want the track? The currently available gauges allow you to produce accurate enough track and have done for 50 years now

True, but bear in mind that conditions change over half a century. For example the rail sections available now are not the same as those which I was using to manufacture track parts in the 1970s, and we now have new entrants for bullhead rail from Peco and DCC Concepts.

For example bullhead rail from SMP Scaleway at around 0.8mm wide is significantly under scale width and very sloppy in gauges designed for 0.92mm rail. There have been reports that the stainless steel rail from DCC Concepts is around 0.83mm wide.

So a check gauge tool which can allow for varying rail width clearly has merit. I don't see much point in having a forum if we don't discuss anything which varies from 50 years of established practice.

The thing which makes me see red is when someone changes from the specified P4 dimensions, but still calls it P4. It's great that folks derive their own variant dimensions, but don't call it P4. Give it some other name of your own. If the check gauge is not 18.15mm min or the back to back is some preferred variant from the published standard, it is not P4. Call it P4-GTX or PBeansOnToast or whatever you like, but not P4.

Otherwise it is not fair to beginners trying to understand the basics, and spreads misinformation in the published record for decades to come.

regards,

Martin.

[Julian Roberts]

I set the BB using the BB gauge, I don't measure it.
Regards

Amen to that Keith but what is the dimension of your BB tool?

[David Thorpe]

If it works does that matter?

DT

[Armchair Modeller]

I am a relative newcomer myself. I just accepted the gauges as they came and built some track. It is all soldered, so subject to expansion of the gauge and rail components as they heat up. I have also used Peco Code 60 rail, which I suspect the gauges were not designed for. I fettle my crossing noses with a thin file after installation - using eyesight and finger to test how the filing is going. So far, I am not finding problems with stock derailing on the crossing noses at all, despite committing so many cardinal sins. I also have some crazy superelevation, just to add to the list. My only absolute rule is to make sure the track is never under gauge.

Admittedly I have a lot of experience at building track in other scales - most notably in 2FS. I am sure that experience does make a big difference. Even with very precise gauges, it is still possible to built track badly.

My latest trick, as I continue to test the trackwork is to push my track testing wagon as hard as I can with my hand from one end of the layout and let it roll very quickly to the far end. Despite reaching breakneck speeds I have so far been unable to derail it - except when it hit the stop at the far end too hard, did a somersault and landed on the floor. The wagon has Masokits sprung axleboxes and 40g weight. I really ought to do a video of it and show it on here.

My experience so far is that the need for absolute accuracy is greatly overrated. Admittedly it is early days though - and I haven't tried any steam locos with very long, rigid wheelbases, for example!

[Julian Roberts]

AM I am sure you are right and what really matters is the care with which you do everything. A 17.67 BB (it seems to me) gives perfectly acceptable results except to ultra discerning modellers. You missed out from what is otherwise really helpful: what dimension BB gauge tool you use?

How accurate do you want the track?
The currently available gauges allow you to produce accurate enough track and have done for 50 years now

From 50 years ago the BB gauge tool was 17.67.
Exactoscale's BB gauge tool is now 17.75 but since how long?

[David Thorpe]

So are you saying that your "perfectly acceptable results" are not acceptable to "ultra discerning modellers"? I've built quite a lot of pointwork, initially in EM, subsequently in P4. I use society jigs and an L shaped B to B gauge. I don't bother measuring these. My stock runs pefectly well over my pointwork. In what way would it be improved if I measured everything?

I agree entirely with AM - "the need for absolute accuracy is greatly overrated" and continued emphasis on it probably only serves to put people off trying P4.

DT

[Jol Wilkinson]

I really do wonder, as I hinted at before, whether all this navel gazing over track and gauge dimensions is really productive.

More importantly, some debate,discussion and even action over wheel manufacturing "standards" and fitting would be of more benefit. The question of "suspension" systems gets a regular airing (again it seem, sometimes just so some people can puff out their chest feathers) but the subject of wheels that can be fitted accurately and consistently is rarely touched upon.

Agonising over which B2B standard to use, with a variation of .08mm, may be a waste of time when you can't fit 7' (28mm) diameter wheels with out a hope of getting them to rotate with zero run out. Even the excellent Exactoscale carriage and wagon wheels, when assembled as per instructions and using the Exacoscale B2B gauge (and don't ask, I don't know what the measurement is) can exhibit a small run out.

Tom Mallard wrote an interesting piece on this in his article in MRJ on building LSWR Black Motors, Bill Bedford is continuing to develop his steam loco wheels but they are not yet available, Ultrascale wheels are not available for many prototypes, likewise the Exactoscale range - which is currently not available. So many of us are left with Sharman Wheels (on the rare occasion we can get hold of them) or AG wheels. Despite knowing all the tricks, having the right tools, etc. it isn't always possible to get these to run totally accurately.

However, despite that and using track that is made to questionable standards, using potentially inaccurate jigs, many P4 model makers are able to create models that do work. Perhaps less keyboard bashing and more model making is the answer.

[Terry Bendall]

I really do wonder, as I hinted at before, whether all this navel gazing over track and gauge dimensions is really productive.

I very much agree with Jol on this one. For some, such things are important and that is fine but as Jol wrote

using track that is made to questionable standards, using potentially inaccurate jigs, many P4 model makers are able to create models that do work

You can build track that works and things do stay on without worrying too much about 0.01mm. I have done it and so have many others and for the majority that is all that is needed.

Perhaps less keyboard bashing and more model making is the answer.

Indeed. Just get on and build things1 With a bit of care and attention to detail things will work and who knows, you may even enjoy yourself

A comment was made to me yesterday at the DEMU Showcase event from someone who works to P4 Standards and who has build a large and impressive P4 layout which works very well - "Why do those who work in P4 continually want to beat themselves up about the so-called difficulties of working in P4?" We had an interesting conversation for 10 minutes or so around that theme including the pretentious rubbish on page 4 of Scalefour News 202

Why not just get on and do things and ignore the critics?

Terry Bendall

[Armchair Modeller]

You missed out from what is otherwise really helpful: what dimension BB gauge tool you use?

At what temperature, altitude and barometric pressure would you like me to measure the track gauges? How can I be sure that my measuring instrument is correct to the tolerances you require? If I measure the track gauge with more than one instrument and they come out with marginally different results, how will I know which is the accurate one? Should I polish the track gauges to remove any oxidisation, dirt etc. - or would that invalidate the measurement?

I actually use a mixture of gauges, depending on what part of the track I am building and what first comes to hand. The dimensions probably vary, but unless I suffer lots of derailments that are likely to be due to the track, I won't even think of measuring them. Sorry!

[Armchair Modeller]

A 17.67 BB (it seems to me) gives perfectly acceptable results except to ultra discerning modellers.

The more I read this, the more it conjures up visions of P4 modellers building track in temperature-controlled vacuum chambers and refusing to operate layouts at exhibitions unless the room temperature is exactly right. I feel sure even you would feel this is a little excessive, Julian

As others are suggesting, I think you ought to chill out and get modelling. A few hundredths of a millimetre just don't matter.

[Julian Roberts]

What is runout Jol?
AM a BB gauge isn't a track gauge. It goes between the wheels
Thanks for the advice chaps.