Happy Birthday Frank (Hornby)

It was Frank Hornby’s 150^th anniversary this week. Google celebrated with a special “Google Doodle” on the UK website. It set me off wondering what Hornby meant to me.

There is no question that Frank Hornby was an inspired entrepreneur. He popularised mechanical engineering to a whole generation of kids by inventing Meccano. Yet by the time I was in my childhood, Meccano was being trumped by Lego – you could build fantastic things in Lego much quicker, and much more easily. And I think that the wonders of engineering had been all too readily accepted by then. The wonder material through the Victorian and Edwardian eras had been iron and steel, but by the 1950’s the new wonder material was undoubtedly concrete. Meccano looked like the Victorian great structures such as the Forth rail bridge, but Lego looked like the South Bank centre and the National Theatre in London.

Of course, whilst it is well known that Frank Hornby invented Meccano, it is the model rail system named after him that is more famous today. There is a link in his development of the Meccano idea to early trains, which led his company onto the Hornby Double O system. Looking back at the various books and web resources. you can see some great model trains which were developed through to the 1960’s – models I look at with a bit of longing in my mind.

My first loco - a Playcraft clockwork 0-4-0

My first train set wasn’t Hornby set. By the time I was old enough to get involved in trains, Tri-ang had come along and set up in competition to Hornby. Tri-ang was part of the great toy empire of the Lines Brothers, and they had access into every toy shop in the country. But the launch of Tri-ang set other people’s minds in motion, and Playcraft did a deal with the French firm of Jouef to introduce a range of British look alike train sets, sold mainly by Woolworths. These sets were much cheaper than either Tri-ang or Hornby, the cheapest was one penny less than a pound – equivalent to around £20 at today's values. You can’t find anything equivalent to that sort of value even today. 

Playcraft sets were great, but the range was never that developed. As I grew older, Playcraft became harder to find, but Tri-ang was everywhere. I soon became a convert to Tri-ang. There were problems though – the rails didn’t match as the rail connectors were on different sides (when you put a playcraft rail up to a Tri-ang rail, both connectors were on the same side, leaving the other side without any connectors). Although it wasn’t obvious to me as a child, the height of the rails were also different.

The underside plate to my loco with Playcraft's name and Jouef diamond logos

What saved the day from my point of view was that Tri-ang merged with a firm I had only vaguely heard of – Hornby. As part of the merger, special adaptor rails were introduced to let Hornby track interconnect with Tri-ang – and guess what; they fitted Playcraft track as well. 

So from a very personal point of view I’m grateful to Hornby, but in a very backhanded compliment sort of way – it was only the demise of his train system that helped me out!

Being more than a little older now, I can look back and start to appreciate, both the imagination which led him to launch model train sets in the UK, and also the attention to detail they still show to this day. Of course history moves on and much later, the Lines Brothers' toy empire fell, and Tri-ang Trains were sold off. They were re-branded as “Hornby” and the modern system which bears Frank’s name is arguably much more a product of Tri-ang than of the original Hornby model system. And in their new guise Hornby have been very creative, they have ramped up the accuracy, and have tried very hard to respond to what their buyers want. For the current generation of children, Hornby means model trains, and a great inspiration they are too!.

Wheels of Progress

After the Railway Maintenance Working Party had done such sterling work in wheel cleaning Al had taken the cleaned locos and run them around our layout. One of our pride and joys is an old Tri-ang R653 loco, a 2-6-2 SNCF black steam model (a Continental “prairie” model from the 1960’s). It’s the most majestic looking loco we have (Al would probably vote for the Hitachi bullet train, but I am an old enough fogey to feel nostalgia for the steam engine!). It was the last of the three cleaned locos to be test run, and he called me in to show it running. But - it stopped. 

A Tri-ang R653, a Continental Prairie in French livery

Now there are lots of possible reasons for the stoppage. It was stopping near to some points. One of the constructional differences between original Tri-ang and modern Hornby is that the track is smaller in height. It is more “finescale” in the jargon. The flanges on the wheels (the bigger bits that drop down to the side of the rail) were made to suit older higher track. When a Tri-ang loco goes over a point there is a piece in the middle – called the “frog” where the two running lines come to meet. The groove in these frogs is not deep enough for the depth of the flanges on most Tri-ang locos. This means the main wheel is lifted up and loses contact with the rail – reducing the transfer of power from the rail to the engine. This usually results in a “stutter” in the running of the engines around the track.  Since the problem appeared to be happening just after the points, I wondered if that was the cause. We ran the loco a few more times.

If you look carefully you can see the dirt on the wheels.You can
also see the deep flange which makes it difficult to run on modern track

Yes that’s right – we ran it again – even though it had stopped. What we found was that if we turned off the Hornby power controller for about a minute, it seemed to reset things so that it could run the loco again. And on the next test run , the loco stopped again, but nowhere near the points. The way our weekends go we didn’t have time to do any more testing, so the problem remains a mystery at the moment. I am wondering if the engine is somehow building up electrical resistance so that eventually the power controller cuts out. This does seem to be a feature of the Hornby rail controller. More research is called for – but that’s going to mean finding some more time to spend at the trackside!

Geometric Designs

In a previous post I talked about track geometry. I explained about the problem I had had with a full straight track section being used instead of two quarter sections. Here is an illustration which shows the difference between a quarter section (R610), two quarter sections joined together and a full straight (R600) - all of these are Hornby current track sections.

In looking back, I remembered that it took me some time to realise that “track geometry” was the magic phrase (the one to put into google) to help me work out how to get a layout working properly. I suspect too many of us would rather leave “geometry” as a vaguely remembered school topic long since relegated for much more useful things.

As a kid I remember pulling out my trainset and fitting all the track together from time to time, and being mystified why some curves seemed to end up in one place and others went somewhere else. With all the ingenuity of a 10 year old, this was seldom a problem – pushing it a bit usually worked, after a fashion!

As a now adult helper to Al and E, this “push until it does fit” approach no longer washes. I had to find out what was going on. It fairly quickly became apparent that Hornby, and Tri-ang before them, make curves that allow you to build three concentric circles (circles within circles). In fact Hornby introduced a fourth circle not so long ago.  These go by the names of first, second third and now fourth radius. In each radius it takes eight standard curve sections to create a circle, and if you had all four circles they would sit one inside the other with 67 mm between them - enough clearance to allow trains to go around them without interfering with the trains on the next circle. Hornby have a useful diagram showing how the different pieces fit together; you can find it on the Hornby website, but its not that obvious so you might like the direct link just to the diagram, which is

http://www.hornby.com/_assets/files/Track-Geometry-PDF.pdf

A Lofty Find

I was clearing out the loft over the weekend and came across a box of train bits and pieces. Mainly it had lots of empty boxes, but in amongst them I was delighted to find a Tri-ang instruction leaflet on the Overhead Power Supply system. I wish I had had it when the R257 wasn’t working! Anyway, here is one of the illustrations which shows how the catenary system works better than I could!

The leaflet is actually based around the first catenary system Tri-ang introduced. This has catenary wire made to look like its suspended from a curving cable. The Catenary we’ve got is just a single, plain wire. However from what I’ve seen, the rest of the system is just the same.

Train delayed - Haggis on the track!

We survived the Burns Night, but our test layout didn't. (It was after all on the dinner table, and I didn't think it was quite up to hauling the haggis around the table!). I did manage to catch a few quick camera shots, before it all disappeared. 

Tri-ang R257 on straight section showing catenary

 The first pic shows the train on a straight. You can see how the catenary (overhead wire) sits nicely over the middle of the pantographs. The trouble starts when you drive your train around the bend. 

diagram showing how straight catenary would run on
curved track

Here is a diagram drawn using XtrakCad which shows how a line drawn from centre line at one end of a curve to the centre line at the other end (the black line) actually veers off to the inside of the curve.  The pantographs won’t be able to reach across to this line and so won't collect any power. One alternative would be to split the curve into two – see the red lines. But even this line gets close to the edge line of the pantograph. (This would be the equivalent of the Marklin recommendation for a post every 22.5°).

Tri-ang R257 on curved track showing catenary

The second pic shows the train on a curve. You can see how close the wire is to the edge of the pantograph (where I have added a small yellow arrow). What happens in practice is, unless you get it right, the pantograph slips off the wire. As it is sprung, it extends to its full height, and often continues to pick up power by scrapping along the side of the wire – until CRASH – it hits a post.

As you can see from the video of a couple of posts ago we did manage to get it right. First I had to straighten the individual strands of wire I had bent as a kid. One trick I picked up on the internet was to roll the wire – this shows where the bends actually are. It's quite a painstaking job to bend them back. Our test circuit was built using Tri-ang R483, first radius curves, with three straight sections (R480/R481) on either side of the oval. Without getting every “straight” section of wire really straight, the wire didn't quite meet in the middle. In the end I was able to ease some sections of the wire apart inside the connecting clips. This isn't ideal of course – but it worked!

Tri-ang catenary mast (R419)

One other problem I had was with the mast clips – the little pieces of nylon which hang down from the top of the post. Here is a picture of one of the problem mast clips - if you look carefully you can see the wire is sitting behind the clip. The idea is that you push the wire up into the clip and the nylon springs closed around it, leaving the bottom edge free to touch the pantographs. But several of my old masts - like this one - had clips which just pushed the wire back out and wouldn't grip it. If anyone has any ideas how this might be solved I'd love to hear them.

We now have to work out just how to link in a Tri-ang overhead powered track section can fit into our main layout. We'll let you know if we solve that one!

And in the meantime, the words of Rabbie Burns’ Address to a Haggis are still ringing in my ears – “great Chieftain of the pudding race” – the proof of our pudding was watching the R257 zip around the track using overhead power with head lights blazing!

Keeping on the straight and narrow

Where we live, trains are run on electric power which comes from a third rail on the ground – the good old Southern Electric Railway, as it used to be in my youth. But for most people, electric trains means overhead wires. That's what makes catenary on a model layout so obvious. Yet as I discovered, even getting electricity to run smoothly around metal track is hard enough – adding in the complexity of suspending wires above the track makes life much more difficult. My guess is, that is why Hornby dropped catenary, and now only produces models with non-functioning pantographs (the roof top power collectors).
  
It wasn’t always like that. Tri-ang had two attempts at working catenary. Nowadays, Marklin produces a working catenary set, and I've seen some ads suggesting Dapol might have something as well. I haven't tried either. But when I was a kid, I had a set of the second generation of the tri-ang catenary system. After all I had a train which needed it – yes as far as I was concerned – it was needed. And so I am the proud owner of an R416 15” catenary set dating from the mid 1960’s. But could I find it? Of course not. And I also remembered just what a pain it had been trying to get it to work properly when I was a kid. Back in those days, we didn’t have space to have a permanent layout – so every time I set up the trains, it meant relaying the track, and re-installing the catenary.


My main memory about it was the problems I had in trying to get the catenary to run around curves. As a kid I was bending the wires to follow the curve. But the more I researched it now, the more I became convinced that that had been the problem when I was a kid.
Real wires on real railways do not turn bends in mid-air. If they curve its because the weight in the middle between two support posts causes a downward curve. Between each support post they run straight. This is how the Marklin system works, where they recommend a support every 22.5 degrees of turn.


Well I had the engine with the pantographs. I had my son ask “what are they for?” whilst my daughter watched me closely for the answer. But no catenary to show them. Ebay to the rescue. I managed to buy another R416 set of Tri-ang catenary. Time then passed – work has this habit of getting in the way of a smooth running model train set! And before I could find the time to set up the new set, my son discovered my original set, hidden away in some boxes. Now we've got two sets of 15 feet each – and we'll have to find some place to test them! (see the video in the last post)

Live wire running!

Electric Overhead

So our test run worked - finally. In the shot is our Tri-ang R257, capable of running powered from the track or from catenary - the posh name for overhead wires!


Its a great model - but frankly a bit unlikely. One of my discoveries in coming back to train sets is the great divide - between modellers who strive for exact replication of real life, as against most of the kids I know (old pretend kids as well as real kids) for whom the toy's the thing. This model was based on an Australian diesel engine, but Tri-ang added the "pantographs" - the roof top collectors - to make it work with overhead power supplies. 


A couple of weeks ago we took this model to a local train fair. I had finally given up trying to get it to run. There we met Dr John the Train Doctor - very clever at fixing model trains. A friend of his told me that this model was altered by Tri-ang to try and break into the American market, where overhead power was considered essential. He said the Americans didn't take to this model and surplus stock was brought back to the UK. Meanwhile Dr John was checking out the inner workings of the model. The problem, it turned out, was simple - inside the motor there is a wire spring, one side of which is sheathed. The sheath prevents a short circuit - but I hadn't spotted that the sheath had worked itself back so the spring was shorting out the  two sides of the motor. Easily fixed - if you know how. (I plan to come back to the inner workings of the R257 in another posting).


But our problems didn't just end with the train itself. We had enormous problems with setting up the catenary. The wires have to sit above the centre line of the track. That's easy when the track is straight, but on a curve you have to bend the wire by hand to match the radius of the curve. As the train moves around a bend, the pantographs move away from the centre line. If the wire is too close to one side or the other, the pantographs slip past the wire and then - CRASH - the pantographs hit the next catenary mast. I seem to have spent days bending and re-bending the wires to get the curve just right. And now - it works - and its a marvel!


The Tri-ang R257, with "Tri-ang Railway" markings is said to be very rare these days. Catenary sets from Tri-ang are also a bit difficult to track down - ebay had a couple in the last three or four days, but both were incomplete sets. I have always loved this train - and there is something special about seeing it powered from above. My son and daughter have loved working with me to get the whole setup running - and there is something special about that as well!