Templot Club Archive 2007-2020
|
|||
| author | remove search highlighting | ||
|---|---|---|---|
|
posted: 30 Jan 2011 13:23 from: BeamEnds click the date to link to this post click member name to view archived images |
Ok so it's probably a bit of a yawn to many, but I'm dead chuffed. 262_300819_310000000.jpg  262_300820_250000000.jpg  262_300821_060000000.jpg  I'm particularly pleased with the tie-bar - it's not wildly out of scale, and cost next to nothing. The ramblings of an idiot about building these points can be found at : Soddingham Track Cheers Richard |
||
| Last edited on 30 Jan 2011 13:27 by BeamEnds |
|||
|
posted: 30 Jan 2011 13:38 from: JFS
click the date to link to this post click member name to view archived images |
Hello Richard, It looks very nice! I like the planing of the wing and checkrail ends. Just a couple of small concerns - are you sure that the check rails are long enough? I would have expected them to cover 5 sleepers rather than just three. Also, it looks like you have taken both the point and the splice rails to the point of the vee, whereas on the prototype, the blunt nose was formed entirely in the point rail as per this illustration from Martin... gallery/2/thumbs/2_201959_350000000.gif Good Luck, Howard |
||
|
posted: 30 Jan 2011 13:46 from: JFS
click the date to link to this post click member name to view archived images |
Sorry, did not link the image correctly... gallery/2/original/2_201959_350000000.gif Also this example from David R gallery/100/original/100_170919_400000000.jpg Best wishes, Howard |
||
|
posted: 30 Jan 2011 13:47 from: BeamEnds click the date to link to this post click member name to view archived images |
Hi Howard, yes, the check rails are probably short, but as I'm working from a bullhead drawing (changing to a flat bottom rail drawing without re-doing the layout doesn't appear to be possible due to geometry and stuff and I have no bullhead rail) it will have to do. Which translates to 'I forgot until I'd done half of them'. The rest will be proper jobs. I'm not worried about the point of the vee. As I'm using the timbers and chairs as the jig, filing both to a point makes getting it right much easier. Maybe next time! Cheers Richard |
||
|
posted: 30 Jan 2011 15:44 from: Nigel Brown click the date to link to this post click member name to view archived images |
JFS wrote:Depends on your prototype. 3 is a bit short but for this sort of thing the GWR would have used 4. Nigel |
||
|
posted: 30 Jan 2011 16:45 from: JFS
click the date to link to this post click member name to view archived images |
Good Luck Richard! For me, the "next one" is always going to be the perfect one - you should have seen my first one in P4, 41 years ago now... But at least the learning is good fun! Your only worry might be if the entry planing on the check rail has not fully 'run-out' before the position where the opposite wing rail opens out for the crossing flangeway - if that were the case (especially in 32mm gauge) you might find the flanges striking the crossing nose - but then maybe you are not planning on running 8Fs or 9Fs! Yes Nigel, you are quite right - but then the GW did not use that much FB! I don't think I have ever seen an FB checkrail shorter than 5 sleepers - but I expect someone will post a pic of one! Best Wishes, Howard |
||
|
posted: 30 Jan 2011 18:54 from: wcampbell23
click the date to link to this post click member name to view archived images |
Hi Richard For a first effort it looks very good! I would make a couple of suggestions. As has already been commented the check rails are very much on the short side. I recently had the opportunity to discuss this with some knowledgeable friends (ex-BR and current Network Rail) and the very firm consensus was that the check rails should span a minimum of 5 timbers. The other suggestion concerns the tie bar - your solution is a nice neat job but there is nothing to prevent the switch blades rising above the adjacent running rails. Keep up the good work. Regards Bill Campbell. |
||
|
posted: 30 Jan 2011 19:30 from: Phil O
click the date to link to this post click member name to view archived images |
Hi Richard One further point over and above those already made, there should be a second tiebar between sleepers S3 and S4, this one should be exactly like the one already in situ. Cheers Phil |
||
|
posted: 30 Jan 2011 19:41 from: BeamEnds click the date to link to this post click member name to view archived images |
wcampbell23 wrote: Hi RichardHi Bill, I think the check rails have been covered - it was a mistake brought on by using flat bottom rail on a bullhead drawing - I just forgot to change size. Easily fixable should I get motivated to do so! There are a couple of little brass plates/strips to be attached the the blades just under the tie bar to prevent the blades lifting - I'm experimenting to see how small (and therefore unobtrusive) I can make them before fitting. I'm currently waiting for more glue before proceeding - I had a bit of a rush of blood to the head and wanted to take snaps before it arrived. There are one or two other bits missing as well! Thanks for the encouragement. Cheers Richard |
||
|
posted: 30 Jan 2011 19:46 from: BeamEnds click the date to link to this post click member name to view archived images |
Phil O wrote: Hi RichardHi Phil, indeed. As posted above, I got a bit impatient with waiting for more glue to arrive, the stretcher will be fitted in due course. As will the missing blocks between the blades and stock rails and associated 'bolt plates'. Cheers Richard |
||
|
posted: 30 Jan 2011 21:15 from: Martin Wynne
click the date to link to this post click member name to view archived images |
BeamEnds wrote:I think the check rails have been covered - it was a mistake brought on by using flat bottom rail on a bullhead drawing - I just forgot to change size.Hi Richard, I'm puzzled by this.  The REA designs for bullhead V-crossings show check rails spanning 4 timbers on V-crossings shorter than 1:6 and spanning 5 timbers for 1:6 to 1:10 inclusive. There is no bullhead design where the check rail spans only 3 timbers. Templot doesn't draw any as short as that unless you specifically change the settings to make them so. It would be wise to change them, not simply because they are wrong, but because they may not be long enough to function correctly as check rails. But it's looking excellent as a first build of pointwork.  Many thanks for posting the pictures.regards, Martin. |
||
|
posted: 30 Jan 2011 22:42 from: BeamEnds click the date to link to this post click member name to view archived images |
Martin Wynne wrote: BeamEnds wrote:Hi Martin,I think the check rails have been covered - it was a mistake brought on by using flat bottom rail on a bullhead drawing - I just forgot to change size.Hi Richard, that's interesting...... I've had a look at it in Templot. The info in the box is: RH turnout: REA semi-curved B-size right-hand switch (unjoggled) 1 in 5.00 RAM regular crossing ( 1 in 5.05 CLM ) equalized-constant timbering. I don't know if that makes any odds! All the other points have check rails over 4 timbers, except the double slip that these points connect directly to, which also has 3-timber ones. The other double slip has them over 4 timbers also. (Apart from some timber shoving, the plan is the same as in the .box file I posted in "Share And Show") I don't recall deliberately altering anything while designing the layout, I thought I had just "minted" every time. For these points I'm going to have to give in and change the check rails, and the same for any of the others that turn out to be flat bottom (the decision is decreed by what is available on eBay or at Buxton Toy & Train fair!). The double slip is a bit more of a concern - if it has ended up with 3-timber check rails, has that affected the overall geometry of it? Cheers Richard |
||
|
posted: 1 Feb 2011 18:15 from: Martin Wynne
click the date to link to this post click member name to view archived images |
BeamEnds wrote:The double slip is a bit more of a concern - if it has ended up with 3-timber check rails, has that affected the overall geometry of it?Hi Richard, No, the slips switches are designed to leave sufficient room for the proper length of check rails. You can change them to span 4 timbers without any problems. I've been looking at your templates, and it's a bit of a mystery how you have got the size 1 and size 2 check rail working lengths set to 48" and 78" respectively when they should be 78" and 108" respectively. I've searched the code and I can't see how that could be the result of a bug. On the other hand it seems unlikely that you would have manually reduced both of them by exactly 30" intentionally. Did you copy these templates originally from an earlier design for very short or narrow-gauge templates? The minting process resets the wing rail and vee rail lengths (if the switch is an REA design), but doesn't at present include resetting the check rail lengths. I have now included the REA check rail lengths in the minting process for the next Pug (if the switch is an REA design). Whether I should have done that is moot. "Minting" is intended to return your template to a "generic" design for your chosen company prototype, removing any changes likely to be specific to a single template. For example any timber shoving is reset, but not the flare length on the wing and check rails which is likely to be constant across all crossings for that company. In practice deciding what should or shouldn't be minted is not easy. I suspect many users of customized templates don't ever use the minting process. I'm minded to change minting to the more specific case of returning the design entirely to the default REA bullhead settings? Which would mean disabling it if the switch is not currently set to an REA bullhead size, including disabling it for all FB, GWR and straight switch options. Comments welcome. regards, Martin. |
||
|
posted: 1 Feb 2011 18:56 from: BeamEnds click the date to link to this post click member name to view archived images |
Hi Martin, thanks for looking into it. The double slip check rails is probably a blunder on my part, I corrected them through the 'Real' menu with no problems. It may be worth mentioning that on the original Soddingham plan one of the slips would not play ball, only one check rail being shortened - the other steadfastly refused to cooperate. I actually spent some time last night looking at "Real->Customise.." and noticed the same - the numbers appear to be almost a random mix of possible ones. I don't know if this is relevant, but setting them all to default using '/' caused some frantic processing that stiffed my machine for about five minutes, and then had no effect. That could be something to do with Linux though, I would have expected others to have reported problems if it were a general thing. Baring accidents, and I don't recall any, I do indeed use 'Mint' every time, for the reasons you indicate - I gave myself some headaches not doing so on the original Soddingham that finished up with starting over again! Again I have no idea if it is relevant but one thing I do do, and did with these points, is bend the approach having aligned the exits to get a more aesthetic result. I don't know if you recall but that caused a problem (only bending so far before 'sticking') on the original Soddingham. Since it would appear that, for whatever reason, I am the only one having this problem, and as it is not a show stopper for anyone actually paying attention to what they are doing I don't think any additional measures are required - I'm sure you have plenty of more worthwhile activities to deal with. Cheers Richard |
||
| Last edited on 1 Feb 2011 18:59 by BeamEnds |
|||
|
posted: 1 Feb 2011 19:23 from: Martin Wynne
click the date to link to this post click member name to view archived images |
BeamEnds wrote:I don't know if this is relevant, but setting them all to default using '/' caused some frantic processing that stiffed my machine for about five minutesOo-er. This is what I mean by not being entirely convinced that Templot works correctly on non-Windows systems. Does Wine support the 80-bit Extended data type for floating-point? If not we could have hit a major snag here, ditto for Crossover on a Mac.  What is the processor on your system? Martin. |
||
|
posted: 1 Feb 2011 20:14 from: BeamEnds click the date to link to this post click member name to view archived images |
Hi Martin, it's an Intel Celeron, and Linux reports that it is running in Long Mode, so 64 bit should be ok. I probably wrong here, but I'm led to believe 80-bit Extended floating-point is only used internally in the processor to make 64 floating point more accurate, so should not worry Wine. I think! Whatever the relevance to hogging the machine (which is after all only an inconvenience, unless the user presses re-set!), all the other points are fine, so I'm guessing that somehow I've created particular case that I'm sure could be got round by simply deleting the template and having another go. In fact, when I emerge from the cellar having murdered a few songs, I'll give it a go (and check the other points with the '/' values) and report back. Cheers Richard |
||
|
posted: 1 Feb 2011 20:29 from: Phil O
click the date to link to this post click member name to view archived images |
Martin Wynne wrote:I'm minded to change minting to the more specific case of returning the design entirely to the default REA bullhead settings? Which would mean disabling it if the switch is not currently set to an REA bullhead size, including disabling it for all FB, GWR and straight switch options. Comments welcome.Hi Martin Being a Great Western modeller I nearly always use mint from current and then use F5 if required to alter switch and crossing lengths and angles. Cheers Phil |
||
|
posted: 1 Feb 2011 21:24 from: Martin Wynne
click the date to link to this post click member name to view archived images |
BeamEnds wrote: it's an Intel Celeron, and Linux reports that it is running in Long Mode, so 64 bit should be ok.Hi Richard, Thanks for the quick reply. No, the 80-bit Extended type is available externally, see: http://en.wikipedia.org/wiki/Extended_precision Templot uses the Extended type for all floating-point maths. 80-bit was necessary to get sufficient precision for the transition curve calculations. This is fine for your Intel processor. But some early Cyrix, IBM and VIA processors won't run Templot for this reason. Yes, on reflection you're right, this is solely a processor issue. The OS shouldn't be relevant. I over-reacted there a bit.  so I'm guessing that somehow I've created particular case that I'm sure could be got round by simply deleting the template and having another go.How long have you been using Templot? Is it possible these templates were derived from some originally created in version 0.68.c (March 2001) or earlier? In later versions I changed the check rail dimensioning, and this may be a bug in the normalizing code. Off to have a look. regards, Martin. |
||
|
posted: 1 Feb 2011 21:55 from: JFS
click the date to link to this post click member name to view archived images |
Well, what a tangled web emerges from a short pair of check rails! Martin Wynne wrote: Oo-er.I would be easy to entirely convince that it does NOT work correctly - WINE (etc) is NOT a Windows emulator - it only does enough to "enable popular Windows applications to run on non-Windows systems". Since Templot - excellent as it is - does not count as "popular", we may suppose it will not run properly. The problem is finding out exactly where and why it will fall over. And it looks like it is here! I would trust nothing running under WINE - not even Templot! Happy bug hunting! Howard. |
||
|
posted: 1 Feb 2011 22:20 from: Martin Wynne
click the date to link to this post click member name to view archived images |
JFS wrote:I would trust nothing running under WINE - not even Templot!Hi Howard, That seems a bit extreme. There are many Templot users running Linux/Wine/Crossover, and most say that Templot runs fine for them, see for example: message 5471 message 7823 http://www.templot.com/info/templot_on_linux.htm So even if not perfect, it runs well enough to meet their needs. My difficulty is knowing whether it runs well enough to say as much on the web site. Clearly there are trade description issues if I say it does, and then users run into problems. I've been very wary about claiming that Templot runs anywhere other than on a Windows PC, but that's not very helpful for intending users who don't have that option. I feel that I ought to have a definitive yes/no answer for them, but at present I can only repeat what others have told me. regards, Martin. |
||
|
posted: 2 Feb 2011 11:47 from: BeamEnds click the date to link to this post click member name to view archived images |
Martin Wynne wrote: How long have you been using Templot?Hi Martin, I think I got Templot in 2007 or so, I had to upgrade to the current pug shortly afterwards. Now you mention it the little grey cells report an oddity that may be relevant - I noticed when printing the template for these points that on the hard copy it states "Project : Town Central". I have no idea where that name comes from, it's certainly not one I've created or imported. I've had a play with the other points, and all seems fine - it's just this set that are having problems. Cheers Richard Edit: I really don't think this is a OS issue since you discovered the strange numbers for check rail lengths etc presumably on Windows of some flavour, and I found the same with Wine/Linux. |
||
| Last edited on 2 Feb 2011 11:58 by BeamEnds |
|||
|
posted: 2 Feb 2011 13:21 from: JFS
click the date to link to this post click member name to view archived images |
Martin Wynne wrote:JFS wrote:I would trust nothing running under WINE - not even Templot!Hi Howard, Hi Martin, You perhaps detect my frustration!! I would not want to appear too critical of WINE - after all it is not some commercial megalith trying to rip everyone off,- but we must recognise its limitations. My frustration (and perhaps this is at the root or your problem) is that they do not state exactly which bits of Windows are or are not emulated and to what degree. Thus things "appear" to work but we are not sure of to what limitation. Further, they change their code from time to time and suddenly things work differently - perhaps without any externally visible evidence. Under such circumstances you cannot be expected to offer any guarantee - you can only say that "many users report success". The worry to a developer is that users will always assume that an issue is the fault of the app. when it the problem might really be the way it is the emulated. (I must stop saying "emulate" after all, Wine Is Not an Emulator according to the Wikki!) The only consolation is that if you look through the bug list (http://bugs.winehq.org/) you will find yourself in some good company with 26,000-odd "bugs" - most of which are simply limitations, not code faults. By the way, the next PUG gets my vote for better use of your time than worrying about WINE:-):-) Best Wishes, Howard. |
||
|
posted: 2 Feb 2011 16:18 from: BeamEnds click the date to link to this post click member name to view archived images |
Hi Howard, Not wishing to hijack my own thread, I have to say I have no problems at all with Wine. I did have some problems when I was forced to use it incorrectly (i.e running Windows apps from the Windows partition rather than installing them into Wine), but since using it 'properly' the only failure is an app the requires a dongle, which for copyright reasons Wine are not allowed to emulate. Whilst it could be argued that Martin should simply ignore non Windows users, I believe that would be a mistake. Empirical evidence from these forums indicates a reasonable proportion of Templot use Linux or Mac OS (not that there is a whole lot of difference between the two under the eye-candy), many of whome do not have Windows licences. Since, as far as I recall, no one has had serious issues with Wine, or Mac equivalents, the point is probably moot anyway. Indeed, it would seem that processor architecture on some Windows machines is more of a stumbling block (though probably becoming a legacy issue as time passes). To get back on topic, as I have posted above, the problem is not, I believe, an OS error (since Martin is able to reproduce it on Windows, in terms of data if not error handling), but an obscure issue that I have somehow manged to create. As it only came to light by virtue of me happening to post a photo, I can't help but wonder if others have had the same problem but not realized it - after all had I not made a mistake when translating from bullhead to flat-bottom rail I would not be aware of it, and neither would anyone else. Most users would, as I did, assume that Templot was 'right' and carry on regardless. Cheers Richard |
||
| Last edited on 2 Feb 2011 16:21 by BeamEnds |
|||
|
posted: 2 Feb 2011 17:01 from: Martin Wynne
click the date to link to this post click member name to view archived images |
Hi Richard, I'm sorry your track-building topic got hijacked. To get back on topic, as I have posted above, the problem is not, I believe, an OS error (since Martin is able to reproduce it on Windows, in terms of data if not error handling), but an obscure issue that I have somehow managed to create.There are two issues arising. The question of how the check rail lengths got modified is something of a mystery, but clearly not related to the OS, and they are easily restored. The problem you reported of the system locking up is much more serious to me, as such things are not supposed to happen. It may or may not be related to Wine/Windows issues, or it may just be that something else was happening on your computer at the same time, such as an automatic update of your anti-virus. If it is repeatable or happens again, please start a separate topic to take it further.In the meantime, back to building your turnouts... regards, Martin. |
||
|
posted: 2 Feb 2011 17:24 from: BeamEnds click the date to link to this post click member name to view archived images |
Hi Martin, luckily for us Linux users system updates are 100% transparent, and there is no requirement for virus checking  so that can be ruled out. If anything else occurs I'll let you know, but I'm quite happy to be aware of the problem and take it as a rather extended error message!As you say, back to making track......... Cheers Richard |
||
|
posted: 10 Feb 2011 14:22 from: BeamEnds click the date to link to this post click member name to view archived images |
So everyone can sleep better at night, I've changed the check rails to cover five sleepers 262_100921_170000000.jpg  |
||
|
posted: 11 Feb 2011 09:45 from: RedgateModels
click the date to link to this post click member name to view archived images |
looks really nice, I'm a "first pointer" too | ||
|
posted: 11 Feb 2011 11:42 from: JFS
click the date to link to this post click member name to view archived images |
BeamEnds wrote:So everyone can sleep better at night, I've changed the check rails to cover five sleepersLooks splendid! I bet you didn't predict the response when you sent your first post. Good luck with all the rest of it, Howard. |
||
| Please read this important note about copyright: Unless stated otherwise, all the files submitted to this web site are copyright and the property of the respective contributor. You are welcome to use them for your own personal non-commercial purposes, and in your messages on this web site. If you want to publish any of this material elsewhere or use it commercially, you must first obtain the owner's permission to do so. |