Turnout troubles...

[John Donnelly]

Looking for some advice with a problem with my first P4 turnout built from a C&L kit.

The pic below shows the problem with the straight switch blade which is not sitting flush against the rail...



A loco will negotiate the straight road but there is a 'thump' and lifting of wheels when they come into contact with the switch blade - can this be fettled of would I be better off starting with a new switch blade?

John

[Tim V]

Looks like you need to bend it so that it sits against the stock rail.

[Knuckles]

I don't have much experience but two things I have done when encountering this is to bodge it with a mini drill. Cut a rebate or thin the blade. Currently if I can get things to work then that's fine even if it looks a bit ropey, if however you want it perfect then I don't know.

[allanferguson]

I would check that there are no obstructions (e.g. bits odf solder) preventing the blade closing up at A
Then I would file the switch tips down to a razor edge at B (at the top only -- the flanges are only 0.5mm deep), and make sure (by bending them slightly if necessary) that they fit really tightly against the stock rails
Then I would check that the gauge at C is not under gauge (and all along the switches). Slightly wide, no problem, slightly tight, bad. I'm not sure whether you have any set in the curved stock rail; lack of it may make gauge narrowing unavoidable
Finally I would check that the tops of the switch blades are no higher than the tops of the stock rails.

So, bits of fettling (a fine file and a good magnifier) and all should be well.

Regards

Allan F

Switch.jpg (76.38 KiB) Viewed 10070 times

[Martin Wynne]

Hi John,

I think you have the blade tip too far forward (at B in Allan's diagram). Prototypically the tip should always be supported on a slide chair, not hanging in fresh air between the timbers. And practically, if you do that so that the tip is forward of the set in the stock rail, it will be impossible for the tip to close fully against the stock rail. The blade tip should always be fractionally beyond the set, so that it is protected by the deflection angle from impact damage.

This diagram is the opposite hand, but illustrates the idea:



Imagine sliding the yellow part to the left -- it would cause the tip to stand proud of the blue stock rail.

I suggest that you remove the switch rail (blade) and shorten it at the other end so that you can re-fit it in the correct position.

regards,

Martin.

[Russ Elliott]

Commercial blades are unlikely to be as planed as thinly as you need.

[John Donnelly]

Thanks for all the suggestions, they are much appreciated.

John

[grovenor-2685]

You might also find a bit of guidance here, rails are the same whether glueing or soldering. Near the bottom of the first page for the stock rail set, second page for fitting blades.
Regards
Keith

[andrew jukes]

Commercial blades are unlikely to be as planed as thinly as you need.

I'm confused by this, Russ. My problem with my home-made switch rails is that they tend to be too thin, over too long a length. They then are too flexible and it becomes hard to get a big enough gap behind the open switch without an excessive opening at the toe.

Our (Exactoscale) switches don't suffer from this and in my experience are a good fit with a correctly set stock rail and do not need any thinning - even of the switch providing the curved road.

Andrew Jukes
Exactoscale Ltd.

[grovenor-2685]

Andrew,
This topic started with discussion of a turnout built from a C&L kit, so you need to read commercial in that sense
Look at the photo in the first post for the evidence.
You have raised the game with P4 track co products.
Regards
Keith

[Russ Elliott]

I seem to be in the doghouse. Again. But thankyou, Keith, for putting my post into context of the picture in the OP. My point was that an as-supplied blade having an over-thick end in a joggle-less stockrail situation is likely to cause running and gauging problems. Perhaps my post of 13 August should have read "Some commercial blades are unlikely to be as finished as you need." (I seem to remember Brian Lewis once describing his blades as 'semi-finished'.)

Andrew's point concerning the difficulty in getting enough 'mid-blade' gap behind the open switch without an excessive opening at the toe is an established problem, even with a B-switch planing length, but I'm inclined to think the number of stretcher bars (2 being desirable on a B-switch, and probably essential on a C-switch) and their method of actuation will have more of an effect in moving the planing length than the finite thickness over the last few millimetres of the blade tip.

Fig 11(c) of Digest 23.6.1 shows the final shape of the filed blade, the end tip recommended to be "about 5 thou wide". To reflect 'Allan's razor-edge fettling' described above, maybe it could be usefully changed to:

blade-filing.png (10.7 KiB) Viewed 9684 times

Here's one of Martin's pictures to compare with the OP:

[Paul Townsend]

I seem to be in the doghouse. Again.

Ah but only the sort where lots of puppies come and lick your A£$"
No nasty biters there for you.

Here's one of Martin's pictures to compare with the OP:

Having built over 100 P4 turnouts with hand-filed blades, most of which have operated trouble free without post-install fettling, except a few improved by your gauge widening advice on CLAG I feel able to comment:

I am still unsure re the advice to round the top of blade tip. As the prototype and its pics show, very little tip de-sharpening was used but your diagram and advice elsewhere seem to advise a lot of tip rounding.

I built all mine with razor sharp tips ( mostly 25-30 years ago! ) and now have a mixture of very rounded, slightly rounded and still sharp. None of the roundings really cured any troubles; where these troubles occured, Mr Mint's gauge showed up best where underguage existed in blades' area and curing that was more valuable than tip rounding.

Have any real trials been done to show that a large amount of tip rounding actually helps on a turnout where correct planing and joggling have been done? Does it reduce sensitivity to wobbly wheels and or non-parallel axles ( my current bugbears and focii of attention) ?

[LesGros]

It is interesting to observe the witness marks on the upper edge of the blades; The wheels do not appear to contact the switch rail until about three feet in from the tips. A similar observation can be made on images contained inthe Waverly accident report posted by Martin Wynne a week or two back (sorry, I've lost track of the link)
It appears that a narrow wedge is ground from the top edge of the switch so significant rounding of the top corner would be superfluous; IIRC, the report shows dimensions of the ground area in relation to the top of the rail. In the accident, a bogey wheel climbed over the switch, to follow the wrong track.
Perhaps one of the knowledgable railwaymen could elaborate.

[allanferguson]

The report concerned is at http://www.raib.gov.uk/latest_news/1208 ... ardens.cfm
There is interesting discussion about the actual shape of the switch blades and the way they work. There is much of concern about the way faults are reported and dealt with, which might concern the traveller, but not, thank goodness, the modeller!

Allan F

[grovenor-2685]

Normally there is running clearance between the flange and the rail, hence the flange will not contact the side of the switch until that clearance is used up, hence the observed 3 ft. On the straight switch not even then as the wheels do not need to be deflected from a straight path.
The exception to this comes when the switches are on the outside rail of a curve, or just after, in this situation the wheels are likely to be running up against the rail with no running clearance and hence causing sidewear of the railhead. This is where you need a joggle or set to make sure the flanges don't catch on the blade tip and climb up.
regards
Keith

[Russ Elliott]

I am still unsure re the advice to round the top of blade tip. As the prototype and its pics show, very little tip de-sharpening was used but your diagram and advice elsewhere seem to advise a lot of tip rounding.

That pic was for an blade with no tip rounding, but shown merely as a comparison to the chunky-ended C&L blade in the OP. I think the side view of that type of blade is something like this, where the wheel flange root gradually picks up the height of the carefully shaped blade:

blade-filing2.png (3.88 KiB) Viewed 9520 times

Using another of Martin's pics from a Templot forum thread, where he gives lots of good advice on blade dressing, a chunkier blade end (but probably still a scale 5 thou) with a rounded tip is shown in a joggle situation in the following pic. The tip rounding is approx a scale 1mm radius, so my quickie sketch in the previous post was probably not too far out:



The Slattock's pic shows a tiny bit of rounding at the end. Brian Lambert's 7mm thread has some good pictures of rounded blade ends.

I built all mine with razor sharp tips ( mostly 25-30 years ago! ) and now have a mixture of very rounded, slightly rounded and still sharp. None of the roundings really cured any troubles; where these troubles occured, Mr Mint's gauge showed up best where underguage existed in blades' area and curing that was more valuable than tip rounding.

I've tried to 'blend in' some troublesome blades to their stockrails using files and wet'n'dry, but the improvement was also questionable, and I agree that any underlying problem (undergauge, or the lack of applying a modicum of gauge widening in the blade vicinity) is unlikely to be significantly alleviated by tip rounding. Perhaps, in the era of commercial blades, we should be stressing the importance of considering joggles and the importance of the distance between the set and the blade tip. The Digest is noticeably lacking in many of these aspects, and I really think we should have been told about all this tricky switch business 30 years ago.

[grovenor-2685]

The tip rounding is approx a scale 1mm radius

About half that more like, the radius is essentially the head depth which is definitely not 3 inches.

I don't know what we were not told 30 years ago, but 40 odd years ago the instructions in the P4 manual worked just fine for me. Still worth following them.

And your tapering of the top surface in the diagram above is greatly exaggerated, it gives the impression you would taper off to the full head depth, this would just provide a nice ramp for the flange to run up and is definitely a no no. Taper on the real thing in this area would not be applied to the straight cut switches most of us model other than just enough to match in to the corner radius of the stockrail.
Head taper is inherent in chamfered and undercut switches but anyone modelling those does not need to read this topic.

Regards
Keith

[Martin Wynne]

the difficulty in getting enough 'mid-blade' gap behind the open switch without an excessive opening at the toe is an established problem

Hi Russ,

Unless the inner foot has been removed in the planing making the tip section too flexible, this problem is usually caused by failure to provide sufficient opening at the tip. The prototype switch opening is normally 4.1/4" which is significantly more that a rail width, and very significantly more than the flangeway gap. I've seen countless P4 layouts where the tip opening is little more than the flangeway -- the OP's pic at the start of this topic being typical. John, your switch needs to open much further than that just to be prototypical, and maybe a little modellers licence beyond that -- bearing in mind that the P4 flangeway is also overscale.

This rather dramatic pic of an open switch blade was posted on RMweb yesterday:

http://www.rmweb.co.uk/community/index. ... ntry785663

regards,

Martin.

[John Donnelly]

Gentlemen,

Thanks once again for all the advice, I've shortened the switch blade and filed it down a little at the end and locos now pass over it with nary a twitch. Just need to shorten the curved blade now...

John

[Russ Elliott]

About half that more like, the radius is essentially the head depth which is definitely not 3 inches.

Agreed. Is this a bit better?

blade-filing.png (11.23 KiB) Viewed 9504 times

I also agree that the tapering of the top surface in my other diagram is far too exaggerated, and I've edited the post accordingly. It needs a new drawing resolution to portray it properly, and it's also probably not a 4mm realistic option for the reasons you give. A red herring.

I don't know what we were not told 30 years ago, but 40 odd years ago the instructions in the P4 manual worked just fine for me. Still worth following them.

Fair enough, and I'm being devil's advocate here to an extent. I think my point is that whilst owners of the Protofour Manual were probably well provided for, the same cannot be said of the average post-1980 newbie Scalefour inmate, who had access only to the Digest. It is only in recent 'post internet' times (c 2000, say) the older material has started to emerge with a reasonable degree of transparency and Martin started publicising his educational diagrams as part of Templot. Secondly, the switch end has until extremely recently been the poor relation of any turnout construction instruction, whether Digest or forum thread, and yet I would venture that the switch end is where 80% of P4 turnout problems lie. Our focus on the crossing vee end has been disproportionate. I can't see any reason for spending money on C&L blades given they require nearly as much filing as a bit of rail costing virtually nothing. I'm not at all criticising the OP, but its picture is clear testament that some important things have been completely ignored or assumed to be correct.

Rant over.

Unless the inner foot has been removed in the planing making the tip section too flexible, this problem is usually caused by failure to provide sufficient opening at the tip.

Hmmm. This appears to be at variance with Andrew Jukes' experience (see above), with which I had considerable resonance. We seem not to have a consistent collective experience.

[Martin Wynne]

Hmmm. This appears to be at variance with Andrew Jukes' experience (see above), with which I had considerable resonance. We seem not to have a consistent collective experience.

Hi Russ,

Are we talking steel rail or nickel-silver? All my experience is with nickel-silver rail, which is noticeably harder than model steel rail, being work-hardened in the drawing process.

and Martin started publicising his educational diagrams as part of Templot.

Thanks for the kind words, but actually they were available much earlier as part of my pointwork kits and components in the 1970s. Some of my stuff was reproduced in the EMGS manual, although I'm not sure about S4.

regards,

Martin.

[andrew jukes]

Martin Wynne wrote:
Unless the inner foot has been removed in the planing making the tip section too flexible, this problem is usually caused by failure to provide sufficient opening at the tip.

I think my home-made switches have suffered from (a) some removal of the inner foot because of not really understanding how best to file the switch and (b) subconsciously thinking a more tapered switch alignment would be better, so thinning the rail too much for too far.

An insufficient opening may often be a problem, but not usually with my track. I aim for 1.4 - 1.5mm. A fair indication of whether you've got this much is whether or not the open switch uses up the full slide area on the Exactoscale P4 slide chairs.

The 'mid-blade' gap problem for me is partly the over filing of home-made switches but also (I think) the result of the rail stiffness not scaling. Stretchers on a full size turnout can be used to control the 'mid-blade' gap whilst attempting to do this on a P4 turnout often results in under gauge track because the closed switch rail is pulled away from the stock rail. This effect is compounded by stretchers that prevent the rails moving with the necessary rotation - a particular problem if a sort of stretcher 'box' is created that holds the switch rails in a particular (fixed) relationship.

If you really want to get a scale 'mid-blade' gap, you probably needed a second drive, something I plan to provide on the one D switch I need on my layout. Otherwise, I'm going to be content with a minimum 'mid-blade' gap of 0.4mm which is easily achievable with P4Track Co. switches with a 1.4mm opening at the toe. It's a far cry though from the 1.0mm minimum shown on page 4 of Digest 1.2.

Andrew

[Martin Wynne]

Otherwise, I'm going to be content with a minimum 'mid-blade' gap of 0.4mm which is easily achievable with P4Track Co. switches with a 1.4mm opening at the toe. It's a far cry though from the 1.0mm minimum shown on page 4 of Digest 1.2.

Hi Andrew,

The 1.0mm was always a nonsense. The prototype minimum is 2", which scales to 0.67mm. On the prototype it's important that wheel backs do not contact the open blade, as this can damage stretcher bars and detection gear. The report into the Grayrigg accident makes interesting reading in this regard: http://www.raib.gov.uk/cms_resources/08 ... igg_v5.pdf

Have you experimented with nickel-silver rail, which is much stiffer than model steel rail?

regards,

Martin.

[Russ Elliott]

The prototype minimum is 2",

Martin - what, typically, does the prototype actually achieve?

[Martin Wynne]

The prototype minimum is 2",

Martin - what, typically, does the prototype actually achieve?

Hi Russ,

No idea, but if much less than 2" on running lines accidents such as Grayrigg would be much more frequent. In yards and sidings of course track can get in a dreadful state and still be usable at low speeds.

Here's an extract from the RMweb pic ( http://www.rmweb.co.uk/community/index. ... ntry785663 ) with a pixel ruler superimposed. The rail widths are showing as 40 pixels with the gap as 30 pixels. So that makes the gap 3/4 of rail width = 0.75 x 2.75 = 2.06". Very approximately of course, and this is clearly a poorly maintained turnout:

switch_back_clearance.png (12.92 KiB) Viewed 9463 times

For model use a simple fix is to allow the switch rail to flex over a greater length than the prototype.

regards,

Martin.