Wright writes.....

[gr.king]

An Isinglass drawing that endeavoured to show three different variants of a small family of ECJS & GNR carriage designs on the same single set of main views, with some small scrap views thrown in too to show additional variations due to rebuilds, was "fun" to work from. Two different possible bogie types, three different possible spacings of bogie centres, beadings on the carriage sides not all in the same places as those in the only photograph I could find, plus one or two other little quirks......

BUT, it was the best drawing I could obtain, from sources known to me, without going to a lot more trouble and possibly spending significantly more money, and it did include information in the notes that I had not seen elsewhere. Where that information came from originally seems to be something that was only known to those who are no longer with us.

An authentic, fully dimensioned, specific works drawing would have been preferable, but what if none seems to exist? It would be wonderful to know in every case where official drawings now reside, although I gather that even some Doncaster Works drawings for certain items, such as rebuilt older classes of locos that were not originally all of the same type, and various rebuilt tenders, were somewhat sketchy, with new (and not always tidy) lines or figures superimposed on older drawings and the same general diagram sometimes meant to cover locos or tenders with slightly varied dimensions. Perhaps other works were just the same, or worse in some cases?

[St Enodoc]

8 minutes ago, Dunsignalling said:

Both could therefore be OK, but not on the same model

Can we be certain of that?

[Barry O]

GA drawings are not be build to but to get it to look OK. 

Pipework drawings of cab pipework are a figment of the draughtsmanship minds. Give a sketch map to a pipe fitter.  A set of parts, pipes and tools and..each one would be completed safely..but never quite the same as any other one.

 

Baz

[richard i]

Drawings, you are lucky. 
model far enough back and there may or may not be a drawing from the time, there may be one someone has tried to work out from photos and written information. Sometimes only photos and some known dimensions to work from, sometimes with a photo only of one side clearly shown. The other being guesswork based on what is known/ thought to be the layout inside. 
As has been said before, it is the challenge which makes the hobby fun. 
Richard 

[t-b-g]

43 minutes ago, Barry O said:

GA drawings are not be build to but to get it to look OK. 

Pipework drawings of cab pipework are a figment of the draughtsmanship minds. Give a sketch map to a pipe fitter.  A set of parts, pipes and tools and..each one would be completed safely..but never quite the same as any other one.

 

Baz

 

Malcolm once told me that some loco classes had their full GA drawings done after they were built. The workshops would have the individual drawings for all the bits and lots of experience of building locos. It was enough! 

[PJT]

8 minutes ago, t-b-g said:

Malcolm once told me that some loco classes had their full GA drawings done after they were built. The workshops would have the individual drawings for all the bits and lots of experience of building locos. It was enough! 

 

Quite so.  Same thing happens in the automotive industry, too.  There, a GA drawing is (should be!) regarded as no more than a list of all the sub-assembly and detail drawings that you'll need to search out to obtain your required information.  At the top dead centre of all GA and sub assembly drawings (and almost all dimensioned component drawings too) is stated 'Do Not Scale'. 

 

Pete T.

 

Edited May 19, 2020 by PJT
missed out 'to'.

[t-b-g]

1 hour ago, richard i said:

Drawings, you are lucky. 
model far enough back and there may or may not be a drawing from the time, there may be one someone has tried to work out from photos and written information. Sometimes only photos and some known dimensions to work from, sometimes with a photo only of one side clearly shown. The other being guesswork based on what is known/ thought to be the layout inside. 
As has been said before, it is the challenge which makes the hobby fun. 
Richard 

 

You are in good company!

 

Most of Peter Denny's models were built from a known dimension like a wheelbase or wheel diameter and a photo. He would do his own (rather lovely and very nicely drawn) drawing.

 

Nobody can claim they are 100% accurate dead scale models but that just doesn't seem to matter. They look like what they are supposed to be.

 

The phrase "better than the one we haven't got" comes to mind.

[Tony Wright]

2 hours ago, t-b-g said:

 

Malcolm Crawley taught me how to pick the bones out of a GA drawing. I was like you, I used to see a mass of lines and my head would spin. He taught me to ignore most of the lines and look for the written dimensions. I can remember him saying now, "Never measure off a drawing if there is a written dimension there and if you have to measure off a drawing, check a written dimension near where you are measuring and in the same plane first to make sure that the printed drawing hasn't stretched or been distorted"

 

Drawings are a minefield but I would think that using something like an Isinglass drawing and a photo would soon tell you which cab side handrail was in the right place, or allow you to get it so close that any error doesn't show.

Good afternoon Tony,

 

When Crownline produced its A2/3 it, the designer must have used the side which plotted the rail in the lower position (if he/she used the Isinglass drawing). Meaning it was incorrect - too low, 'squashing' the proportions of the lower cabside which impacted (when painted) on the positions of the lining, numbers and worksplate. 

 

Fortunately (by scrutinising photographs as you suggest)), I was able to alter the position at source - by plugging the etched holes for the pillars, then drilling more in the higher (correct) position. I had to alter the positions on the same firm's A2/2. 

 

Which makes me wonder....................... Just how many locos/items of rolling stock have been built from kits which in certain dimensions are incorrect? The bits might well fit, but the end-product is inaccurate.

 

Examples include the Millholme A2/2 and A2/3 classes - obviously designed from the Skinley drawings......................

 

Regards,

 

Tony.  

[t-b-g]

32 minutes ago, Tony Wright said:

Good afternoon Tony,

 

When Crownline produced its A2/3 it, the designer must have used the side which plotted the rail in the lower position (if he/she used the Isinglass drawing). Meaning it was incorrect - too low, 'squashing' the proportions of the lower cabside which impacted (when painted) on the positions of the lining, numbers and worksplate. 

 

Fortunately (by scrutinising photographs as you suggest)), I was able to alter the position at source - by plugging the etched holes for the pillars, then drilling more in the higher (correct) position. I had to alter the positions on the same firm's A2/2. 

 

Which makes me wonder....................... Just how many locos/items of rolling stock have been built from kits which in certain dimensions are incorrect? The bits might well fit, but the end-product is inaccurate.

 

Examples include the Millholme A2/2 and A2/3 classes - obviously designed from the Skinley drawings......................

 

Regards,

 

Tony.  

 

I am sure your are right Tony. Many kits contain either downright errors or deliberate fudges. The J17 with the wheelbase altered to fit a Triang chassis being an example.

 

I am sure your etched J17 still has a couple of minor errors even after all your heroic efforts to get it right.

 

Many kit designers are not professional engineers and are self taught, having no formal training in how to draw or design kits. They learn as they go along and mistakes are made and sometimes missed before production. Sometimes I am perhaps too harsh on such things as unless these good people produced something, we would be so much worse off in our choices. I do find it frustrating when a kit could, with very little change, be so much better than it is, either in the way it is assembled or the way it depicts the prototype.

 

So few of any particular kit seem to be sold that expecting a mistake to be rectified after production is unrealistic.

 

Hopefully most of the kits that contain the worst problems are pretty ancient and with the introduction of CAD and 3D printing for masters and resin casting (although I am not a fan - I can't solder it!), things have improved and more recent kits are certainly better than Jidenco, Ks, GEM and BEC from half a century ago.

 

There are some nice kits around that don't need a huge amount of correction. I haven't built any Finney or Mitchell kits as they don't do any that I want but I have seen some super built up ones. Fiddly and slow maybe but super results. Mike Edge of this parish is another that I would happily recommend. His go together nicely, with a well thought out and not over complex assembly and they look the part when done. I have a Galdiator GCR Atlantic and a David Andrews B3 "Valour" part done in 7mm and have hardly had to do anything other than cut the bits off the etch and clean off the cusps.

 

So thankfully there are some good ones to be found amongst the poor.

 

Of course, if you insist on building Thompson pacifics, you get what you deserve and I will leave that one there........

 

 

 

 

 

 

Edited May 19, 2020 by t-b-g

[Michael Edge]

2 hours ago, t-b-g said:

 

Malcolm Crawley taught me how to pick the bones out of a GA drawing. I was like you, I used to see a mass of lines and my head would spin. He taught me to ignore most of the lines and look for the written dimensions. I can remember him saying now, "Never measure off a drawing if there is a written dimension there and if you have to measure off a drawing, check a written dimension near where you are measuring and in the same plane first to make sure that the printed drawing hasn't stretched or been distorted"

 

Drawings are a minefield but I would think that using something like an Isinglass drawing and a photo would soon tell you which cab side handrail was in the right place, or allow you to get it so close that any error doesn't show.

I've seen plenty of GA drawings with dimensions incorrectly quoted - including the Fowler one I used last year to produce our 150hp kit. Many have pencil notes on with different dimensions to be used - sometimes you can read these and understand why they were there. Ultimately, if it doesn't look right - it's wrong, no matter what the drawings say.

[Headstock]

What a lot of hot air and poor excuses concerning drawings. The point is (as the sensible have noted) the dimensions. The dimensions of just about any steam locomotive, are available to those who chose to model them. If you don't use them, then more fool you if your creation turns out to be a Frankenstein monster. Surely, somebody could look up the relevant green book and quickly discover the pitch of the boiler on a J whatever class, it would save pages off blah and lots of soldering, unsoldering and frayed nerves.

 

This dose seem to be a developing trend with railway modellers. Recently, on a thread, there was a discussion of the ride height of different RTR carriages. It was inconceivable to those involved, that they may wish to look at the dimensions of the real thing and discover what was correct. Astonishing, as I have said before, ignorance is a virtue in modelrailway land.

[t-b-g]

8 minutes ago, Michael Edge said:

I've seen plenty of GA drawings with dimensions incorrectly quoted - including the Fowler one I used last year to produce our 150hp kit. Many have pencil notes on with different dimensions to be used - sometimes you can read these and understand why they were there. Ultimately, if it doesn't look right - it's wrong, no matter what the drawings say.

 

Absolutely. If you use a dimension from a GA and the model looks wrong, it probably is! I can often tell when a model isn't right by comparing it to a photo even when I can't get to measure it, so how the model looks is just as important as precise dimensions.

 

What you don't get on the GA are lots of wrong dimensions. So if you see a footplate marked as 8ft wide, you can check it against another similar dimension, or check it by measuring to a known scale. If the buffer beam is drawn as 7ft 11" wide and is slightly narrower than the footplate, the chances are you are on to a winner. If the buffer beam is drawn as 7ft 11" but is wider than the footplate, it needs investigating.

 

It is always possible to double check (another lesson from Malcolm Crawley).

 

You can never be 100% sure and you can never be sure that the real things were built the same as the drawing. We can only do our best and hope we end up with something that looks right.

 

 

[MarkC]

12 hours ago, manna said:

G'Day Folks

 

Hey, if you want a laugh, this is my rendition of a J17, made out of a Tri-Ang 3f, a B12 cab and tender, with bits from a King and Thomas.

 

manna

Not a bad effort at all, manna - and 'back in the day' that would have been looked on with envy by some

 

Mark

[Mike 84C]

All this talk of  drawings and the lack of accuracy  thereof  makes me wonder if I should just resell my jamieson  L&Y  Baltic kit  which just has a Skinley drawing to guide its construction.  When I checked the L&Y Society site there are lots of detail drawings for areas I do not need but no GA and a friend who volunteers at the NRM has not turned anything up yet.

  This self confessed  fan of the Great Way Round does like other things, even a huge regard for Pacifics of  the LNER but I have been reading this thread for a long time, maybe I have been indoctrinated?

[Bernard Lamb]

The last few posts do provide some sort of explanation as to why at work we always called drawings "cartoons".

For almost infinite variety take a look at the available drawings of the LMS  diesels 10000 and its- not identical- twin.

The drawings of the buffers are worthy of a study in their own right. The Beatie version is a good place to start.

 

I remember back in the bad old days when in South Africa they would "reverse engineer" a product that they were not allowed to purchase. I had a phone call from a Professor at Natal University who was in charge of a project and was having problems in roll forming a particular steel section.  When I told him that the original had never been made to the drawing but if he altered it to the figures that I gave to him the section could be produced and the Structural Engineer would be happy he was both confused and delighted. The tooling drawing and the product drawing had very little in common. The usually practise was for all modifications to be logged. However the initial cock up had been so great that all evidence had been destroyed. No problem, until an outsider tried to make their own version. Now I am not accusing anybody of deliberate foul play, but such events are by no means uncommon in the real world.

Bernard

[PJT]

42 minutes ago, Bernard Lamb said:

I remember back in the bad old days when in South Africa they would "reverse engineer" a product that they were not allowed to purchase.

 

And I remember in the car industry a large company that should have known better that reproduced a bumper bolt by reverse engineering from a secondhand sample.  The reproduction was faithful, right down to the stretched thread.  Twenty five thousand were made before the error was spotted.  They, as we say, all ended up as fishing weights.  The original dimensioned component drawings for the bolts were all still easily obtained, had they bothered to do so. 

 

Pete T.

 

Edited May 19, 2020 by PJT
clarify a point

[CXW1]

2 hours ago, t-b-g said:

 

You are in good company!

 

Most of Peter Denny's models were built from a known dimension like a wheelbase or wheel diameter and a photo. He would do his own (rather lovely and very nicely drawn) drawing.

 

Nobody can claim they are 100% accurate dead scale models but that just doesn't seem to matter. They look like what they are supposed to be.

 

The phrase "better than the one we haven't got" comes to mind.

 

I pretty much use the same technique today. The RCTS books provide known dimensions like wheel size, wheelbase, length over buffers, boiler diameter, boiler pitch etc. A decent side-on photograph and the key dimensions can then be used to produce a reasonably accurate scale drawing like the one below that I produced for a LNER N12. 

 

Like Tony says, I am well aware that this is not going to be a 100% accurate scale drawing but, with a bit of imagination and good luck, the drawing can be used to produce something that looks like an N12. The drawing looks distorted in the picture but it was just taken quickly on my phone and it has been folded up a few times. 

 

Is the model 100% accurate? Probably not.

 

Does it look like an N12? I think so, and it's good enough for me and you can't buy one from Hornby or Bachmann. 

 

My dad was a metalwork and technical drawing teacher, so I have picked up a fair amount from watching him do things when I was younger.

 

I wouldn't know where to start looking at a GA but some of the recent comments on here would make me more confident in looking at one in the future. 

 

Thanks

 

Chris 

 

[t-b-g]

17 minutes ago, CXW1 said:

 

I pretty much use the same technique today. The RCTS books provide known dimensions like wheel size, wheelbase, length over buffers, boiler diameter, boiler pitch etc. A decent side-on photograph and the key dimensions can then be used to produce a reasonably accurate scale drawing like the one below that I produced for a LNER N12. 

 

Like Tony says, I am well aware that this is not going to be a 100% accurate scale drawing but, with a bit of imagination and good luck, the drawing can be used to produce something that looks like an N12. The drawing looks distorted in the picture but it was just taken quickly on my phone and it has been folded up a few times. 

 

Is the model 100% accurate? Probably not.

 

Does it look like an N12? I think so, and it's good enough for me and you can't buy one from Hornby or Bachmann. 

 

My dad was a metalwork and technical drawing teacher, so I have picked up a fair amount from watching him do things when I was younger.

 

I wouldn't know where to start looking at a GA but some of the recent comments on here would make me more confident in looking at one in the future. 

 

Thanks

 

Chris 

 

 

You do have a big advantage Chris. Modelling things that very few people will know whether they are right or not!

 

The front splasher/sandbox is clearly not the same material as most of the loco. Is it carved from plastic? 

[Orion]

7 minutes ago, t-b-g said:

 

You do have a big advantage Chris. Modelling things that very few people will know whether they are right or not!

 

The obsession with getting things absolutely right is very unhealthy IMHO. Much safer to model a line for which you can become the only real expert. The Hull & Barnsley is probably getting close - Lancashire Derbyshire & East Coast or some other obscure line that almost no one has even heard of might be even safer. Or some obscure Siberian line in the 1950s when anyone taking photos or information might not live to tell the tale, even if they could ever get that close?  

[CXW1]

The front splasher/sandbox is clearly not the same material as most of the loco. Is it carved from plastic? 

 

Yes, it is made form plasticard. It is hollow inside rather than being carved. I have struggled to make this type/shape of splasher in the past and find that I can make something that looks more realistic from plastic. It also avoids shorts.

 

I don't think any of the N12s lasted beyond the 1930s so you are right in that not many people alive today will have a living memory of them. Plus, they weren't the most photographed of things which helps when we get into discussions about accuracy. 

 

 

[Dunsignalling]

4 hours ago, St Enodoc said:

Can we be certain of that?

No but, equally, we can't know for certain it wasn't the case. Published "modellers" drawings were often measured off locos that had long since ceased to be remotely "as built", with earlier details possibly added from memory or photographs. Many depicting quite old types seem to date from early BR days and I've long suspected that the measuring tape reached some barely ahead of the scrap-man's gas axe.

 

Steam locomotives were built by hand, and the process bore almost no resemblance to the way most things are made today by robots working to tolerances almost unmeasurable a century ago. Add in three to five decades of repairs, some carried out under running shed conditions that made a village smithy look sophisticated, and its a wonder things didn't wander about much more than they actually did.  

 

Anything that wasn't assembled on a jig was susceptible to variation for all manner of possible reasons. What if, for example, the drawing showed a handrail that was supposed to be 3' 6" long whereas the only piece to hand was 3' 2" and the loco had to be moved to the paintshop by end of shift.....

 

Drawings? If you strive for museum level "glass case" authenticity, some provide good starting points but one should  question everything and verify as much as possible using dated photographs.

 

If, on the other hand you want a layout loco and would like to live long enough to build more than one, anything that  looks the part from two or three feet away, like the ingenious and admirable Tri-ang J17 conversion pictured, or old kits where prototype dimensions were adjusted to make it fit a r-t-r chassis, is arguably "close enough".

 

Difficulty only arises if somebody presents the latter as the former.    

 

John

 

 

Edited May 19, 2020 by Dunsignalling

[t-b-g]

 

5 minutes ago, Orion said:

 

The obsession with getting things absolutely right is very unhealthy IMHO. Much safer to model a line for which you can become the only real expert. The Hull & Barnsley is probably getting close - Lancashire Derbyshire & East Coast or some other obscure line that almost no one has even heard of might be even safer. Or some obscure Siberian line in the 1950s when anyone taking photos or information might not live to tell the tale, even if they could ever get that close?  

 

Now you are really talking my language!

 

You mean.....

 

 

My LD&ECR epic, which was set in LNER days when photographed but is now backdated to GCR times.

 

My latest project is based on the SDR. Not Stockton and Darlington, not Somerset & Dorset but the Sheffield District Railway, a real and very obscure bit of railway that was part of the LD&ECR empire.

 

Edited May 19, 2020 by t-b-g

[Andy Reichert]

9 hours ago, Michael Edge said:

It doesn't help at all with track twist but it does make the frame flexible in one plane. Many of our track imperfections are simply vertical, particularly at baseboard joints, this does at least stop the loco rocking lengthways on its centre axle. The diesel example I quoted above does seem to defy all logic to me though - I walked round this loco twice to make sure of what I had seen when we measured it - but they do work.

 

Thanks. I can see your reasoning.   I'm not going to try an analyse the example either. I have too little time to model the more mundane but widely used locos I like as it is. .

 

Andy

[Andy Reichert]

6 hours ago, gr.king said:

Compensated, sprung, rigid, raised middle axle, or some mixture of arrangements? It seems to me there is no "right" or "wrong" way to do things, as the method that gives the best result is highly dependent on the kind of track over which the model is to run. There's no sense at all in anybody insisting that their method is "correct" because the their theory (with its inherent assumptions) says so. Full, unrestricted compensation relies on the assumption that the track will always support the wheels. Fine if all of your crossing gaps are small, not so good if your locos have to contend with large crossing gaps. Rigid chassis are fine so long as your track is flat enough, but if it undulates and twists excessively then derailments or interrupted electrical pick-up become a problem. Build to suit the need.

Mike Edge's method looks like a reasonable compromise for many applications, but like any compromise it won't suit every situation perfectly.

 

I can only partly agree with this.  It's the target track holding functionality for your chosen Wheels and Track Standard that matters, not the method you use to try to achieve it.  Get the functionality wrong and the model is not going to perform to your expectations, no matter how well you execute the design or construction.

 

The full target target track holding functionality in my understanding is: To have all wheels sit down on the track, carrying the same intended ideal portion of the total loco weight, statically and dynamically, at all times, and all intended speeds, regardless of the shape of the track at any instant.  

 

You can cut some (or a lot) of that functionality back for a compromise and a more flexible Standard, but not alter it to do something different or irrelevant. For example, using deep flanges allows for a rigid chassis to work satisfactorily for most 00 RTR.  Even with just three wheels actually bearing on the track at any time, with good pick-up, a sensible min. radius and top speed and the loco will run and stay on the track in just about all circumstances. 

 

Andy

 

 

[Andy Reichert]

59 minutes ago, Orion said:

 

The obsession with getting things absolutely right is very unhealthy IMHO. Much safer to model a line for which you can become the only real expert. The Hull & Barnsley is probably getting close - Lancashire Derbyshire & East Coast or some other obscure line that almost no one has even heard of might be even safer. Or some obscure Siberian line in the 1950s when anyone taking photos or information might not live to tell the tale, even if they could ever get that close?  

 

The bit about changing the gauge on all the passenger cars at the border is so well known, you can't avoid modelling it.

 

Andy