Oxford Rail Annouce 4-Wheel Toad

[Guy Rixon]

China is a long way away and someone would need to be out there on the spot all the time checking progress. Manufacturing life isn't as simple as you might think.

 

 

China is a long way by ship or air, sure, but it's really easy to exchange CAD files by internet to check the design. Checking it before the tooling is started should cost a week or so delay and next to nothing in money - if the design contract was drawn up sensibly, which could be the root of the problem.

 

I agree that checking the realisation of the product is much harder, and the contractor might charge for reworking. But neglecting to check a 3D CAD before tooling is just slack.

Edited August 16, 2017 by Guy Rixon

[autocoach]

China is a long way by ship or air, sure, but it's really easy to exchange CAD files by internet to check the design. Checking it before the tooling is started should cost a week or so delay and next to nothing in money - if the design contract was drawn up sensibly, which could be the root of the problem.

 

I agree that checking the realisation of the product is much harder, and the contractor might charge for reworking. But neglecting to check a 3D CAD before tooling is just slack.

Doesn't Oxford own the factory in China?

[mikeharvey22]

I think Oxford have a lot of room to improve their processes. If I understand it correctly they announce items when they start tooling, and miss out on informed input from RMWeb and elsewhere at the CAD stage. But I am acutely aware that the combined expertise of RMWeb spends months iterating on each model, with some commentators better informed than others. Oxford has already had several duplications of others' new items or near duplications and have not pulled back, because they are already in tooling. Of course it would be great if they used a wider range of well informed experts at the outset, and I really believe the trawl for expertise is the stage which needs to be improved.

 

On a cautionary note I once built a model bus based on a well authenticated photo taken in 1950. The bus carried a livery which pre-dated the general use of the livery on the fleet. My model was berated publicly by a fleet expert who condemned it for the livery error which he asserted was never used before 1957. He never apologised even when I referred him to the same photo which he was selling from his extensive Fotopic site. His expert status took a bit of a knock!

Edited August 16, 2017 by mikeharvey22

[dibber25]

China is a long way by ship or air, sure, but it's really easy to exchange CAD files by internet to check the design. Checking it before the tooling is started should cost a week or so delay and next to nothing in money - if the design contract was drawn up sensibly, which could be the root of the problem.

 

I agree that checking the realisation of the product is much harder, and the contractor might charge for reworking. But neglecting to check a 3D CAD before tooling is just slack.

Ever tried doing it? First you need the right software. Then you need it to work on the computers you've got. Then you need to actually be able to understand and interpret what you're seeing. Yes, if you're set up to do it on a range of your own products it should be easy enough to learn but with commissioned models, believe me, it can be a real headache. Once you've reached tooling it's too late for all but the smallest of changes, so it really is vital to get the CADs right. However, there's a law with most types of proof that says no matter how carefully you check it, you'll always spot a problem just after you've given the OK! (CJL)

[tomparryharry]

Ever tried doing it? First you need the right software. Then you need it to work on the computers you've got. Then you need to actually be able to understand and interpret what you're seeing. Yes, if you're set up to do it on a range of your own products it should be easy enough to learn but with commissioned models, believe me, it can be a real headache. Once you've reached tooling it's too late for all but the smallest of changes, so it really is vital to get the CADs right. However, there's a law with most types of proof that says no matter how carefully you check it, you'll always spot a problem just after you've given the OK! (CJL)

Yes. Been there, done that. One die tool we were looking at, got pulled for re-modelling when the test impression looked like 'rude lady parts' on the alloy wheel we were running.

 

Oh, we laughed & laughed (not).

 

Ian.

[Guy Rixon]

Ever tried doing it? First you need the right software. Then you need it to work on the computers you've got. Then you need to actually be able to understand and interpret what you're seeing. Yes, if you're set up to do it on a range of your own products it should be easy enough to learn but with commissioned models, believe me, it can be a real headache. Once you've reached tooling it's too late for all but the smallest of changes, so it really is vital to get the CADs right. However, there's a law with most types of proof that says no matter how carefully you check it, you'll always spot a problem just after you've given the OK! (CJL)

 

In a limited senses yes: I exchange STL files with Shapeways. When there's a printing problem, I can see on their web-site what they think the model should look like.

 

I appreciate that STL files are not "CAD files" in the strict sense, but they are a portable representation of a solid model. In this sense 3D CAD is better served than 2D CAD.

 

In the absolute worst case of software incompatibility, the remote designer could render from CAD a set of views of the virtual model and send them back to the customer.

[Neil]

In the end do any of the flaws really matter? Some will buy the van and be happy with it 'as is', some will buy it and do some work on it until it satisfies them (EM and P4 chappies will do this as a matter of course) and some will decide that neither of these options is for them and will spend their money elsewhere. I'm going to stick my neck out here and offer the thought that the third group will be heartily outnumbered by the first and second ones. The Oxford  Rail cattle wagon, with its flaws, seems to have sold well enough. The only truly serious implications are for Oxford Rail and their staff if the vans fail to sell.

[The Johnster]

I reckon both the 4 and 6 wheel vans will sell, whatever their inaccuracies,  in about the quantities Ox reckon they will as long as they look as if they are accurate to people who don't know that they are not; i.e. if a fair level of detail is visible and especially if separate handrails are included.  This is as much about fine detail, printing, and finish as it is about accuracy; the trick is to provide a model that looks good so that it's owner can show it off, and the huge majority of the people he/she shows it off to will be impressed but basically unconcerned if it is incorrect.  Isn't he clever, look, it's even got the depot printed on it, isn't he clever to know that the GW had six-wheel brake vans, that's a bit unusual isn't it, he must know loads about the GWR, they will think about the model's owner, as if being able to buy something is clever (this seems to be the principle behind conspicuous consumption), I'd love to build a model railway one day, and then off they go and forget the whole thing.

 

If they do build a model railway one day, they quickly realise that they don't know much about real railways, after all why should they, and blather on about fine detail and rivet counting not mattering to them.  This is a prevalent attitude and I'm not having a go at the approach, honest, it enables large numbers of people to gain a massive amount of enjoyment from the hobby, but it annoys me a little when Ox, who's strapline is 'in pursuit of excellence', make errors that could be eliminated if their research and production methods were altered to allow it, and tell people that the models are accurate scale renditions of reality.

 

I am aware of and live with major faults in some of my models, but am at least aware of them; I am unable to say if the awareness is preferable to blissful ignorance, but I believe that it is.  Anyone who habitually runs a Hornby 2721, even if it is worked up a bit, can hardly be described as a rivet counter or obsessed with detail, but Hornby did not profess to be in pursuit of excellence or indeed anything but an honest profit as a toy maker when they introduced the model all those years ago.  I will be buying either a 4 or 6 wheel toad in early BR livery whatever it's faults; as a general impression of a small toad it is a fairly decent model and at a very reasonable price.  But I'll only buy one, and I will be buying  several more Hornby toads, firstly because I know that to be a more accurate model and secondly because the larger toad it represents was more typical of the area and time I have set my layout in.  No rtr manufacturer is going to make much of a living out of my tiny blt's requirements, and thus I do not expect them to take a lot of notice of my ramblings on sites such as this.

 

I am a bit of a cynic, but it doesn't do any good...

[coachmann]

Some folk appear to be doing their damnedest to distance themselves from the silent majority by criticizing them. I wonder why.  Banging away with a headache ball is pointless and using Oxfords strapline is well past it's sell-by date. It should be obvious by now that Oxford is mass-producing models that lack absolute accuracy but nevertheless sell! I wonder if sales of Oxford's GWR Toad will ever catch up with the works of fiction offered by Bachmann, Hornby and Dapol and which many of us already have on our layouts...?  

[The Stationmaster]

China is a long way by ship or air, sure, but it's really easy to exchange CAD files by internet to check the design. Checking it before the tooling is started should cost a week or so delay and next to nothing in money - if the design contract was drawn up sensibly, which could be the root of the problem.

 

I agree that checking the realisation of the product is much harder, and the contractor might charge for reworking. But neglecting to check a 3D CAD before tooling is just slack.

 

Very easy to exchange CAD files by email however all you are doing is emailing a file - you then of course need the software to open and view it (that's also easy and it can be downloaded off the 'net at no cost).  But all of that is the simple bit - no doubt any computer savvy 12 year old (or 9 year old) could do it in a twinkling of the eye.

 

But then you actually have to look through the CAD file and that takes three things  - patience, application, and knowledge (of the prototype you are looking at and which reference sources you should be using and just how reliable or not those reference sources might be).  If you happen to know a bit about the subject the last part becomes a bit simpler, if you know quite a lot about the prototype it is a bit easier than that but you need to know enough to know that published photos might be wrongly dated (surprising how often that happens) and to know the exceptions from the 'run-of-the-mill' and not to necessarily trust 'preserved' examples (when even a particular NRM loco has been restored with a minor livery error - which a manufacturer then copied).  

 

In other words it's no good looking at a CAD unless you know what you're looking for and where and how to look for it.  Then you need to get it correctly corrected by the designer back in China - who might well have a reasonable understanding of English but doesn't necessarily know the language as well as you do so might go a little bit awry.  Then the CAD comes back for the corrections to be checked, which has to be done just as thoroughly as the previous check.  And that could easily happen several times all requiring the same degree of thoroughness = time spent = cost = possibly production slot lost = extra interest on the money borrowed to fund development = an ultimately delayed release with extra cost to recover in the retail price and extra costs because the £ has slumped.  Which leads to loads more 'consumer' moans because of the delay and higher retail price.

 

It would seem an awful lot simpler and quicker to send a CAD file and a pile of photos to a designer in China and let him get on with it enabling you to meet your originally planned production slots and sell the result at a lower price. (Or maybe it wouldn't if it gives threads like this one a long life?).

Edited August 18, 2017 by The Stationmaster

[BG John]

I still don't understand why there are so many different versions, if the checking/correction stage is so difficult, and most will be bought by customers who don't know/care.

[Bernard Lamb]

It would seem an awful lot simpler and quicker to send a CAD file and a pile of photos to a designer in China and let him get on with it enabling you to meet your originally planned production slots and sell the result at a lower price. (Or maybe it wouldn't if it gives threads like this one a long life?).

To go a step further why not get an undergraduate in engineering to do the drawing as a project here in the UK?

You might even find one who has an interest in railways.

Back in the day we had two Universities willing and able to supply students to make prototype drawings and to undertake various other tasks.

Having worked at the sharp end all my life, with constant deadlines, I do find it strange how things take so long these days.

Or do the makers of model trains lack any clout in the wider world?

Bernard

[rovex]

I think I'll wait until the models are in the shops and then see who is first to post a thread about modifying them to produce an accurate version before deciding if I want to go to the trouble

[tomparryharry]

I think Oxford have a lot of room to improve their processes. If I understand it correctly they announce items when they start tooling, and miss out on informed input from RMWeb and elsewhere at the CAD stage. But I am acutely aware that the combined expertise of RMWeb spends months iterating on each model, with some commentators better informed than others. Oxford has already had several duplications of others' new items or near duplications and have not pulled back, because they are already in tooling. Of course it would be great if they used a wider range of well informed experts at the outset, and I really believe the trawl for expertise is the stage which needs to be improved.

 

On a cautionary note I once built a model bus based on a well authenticated photo taken in 1950. The bus carried a livery which pre-dated the general use of the livery on the fleet. My model was berated publicly by a fleet expert who condemned it for the livery error which he asserted was never used before 1957. He never apologised even when I referred him to the same photo which he was selling from his extensive Fotopic site. His expert status took a bit of a knock!

I'd suggest that it depends what sort of feedback you get. If it's a sort of feedback where the commentator 'suggests' modifications & improvements, with backup documentation, then you're in with a chance. If it's a hectoring tone, whereby "you must do this, you must do that", then things start to go wrong. I do think things like flaws are very important, I'd (respectfully) suggest. It's all about product perception. You're expecting something to resemble, and it doesn't match up. Either your next product improves or, your product diminishes. Then, your personal expectations get pre-conditioned by your historical expectations. It's true that Hornby suffered to some extent, and thoroughly overcame the problem with some first rate stuff.

 

There's plenty of room in the market for Oxford to turn out some little crackers, and I don't see why they shouldn't do so. I'd dearly like to Oxford get it together, and provide a third (or 4th) major player.

 

Time will tell.

 

Ian.

[mikeharvey22]

One further consideration in the CAD work is that the resulting computer model has to be suitable for manufacture, the CAM part of the modelling. The bit we tend to get excited about is the outer skin we see, but beneath that is a mechanism (powered or not) which has to work as well, and all the components have to be suited to moulding in plastic, casting in metal, or etching or what have you. Having the CAD/CAM physically close to the manufacturing seems to be the way that most of the players work.

[Gerald Henriksen]

In the end do any of the flaws really matter?

 

To Oxford, likely not given that this isn't their first model with issues.

 

On the other hand, it can hurt the hobby if by releasing an inaccurate model they remove the market for an accurate model to be made by someone else - for a likely example look at the Mk3 (which in fairness hasn't been released yet, but then again about that track record...)

[Neil]

I can see where you're coming from Gerald though duplication already exists of other basically accurate models.

[BG John]

I can't see anyone else producing a GWR brake van suitable for the late 19th/early 20th century though. I'm quite surprised that Oxford are. So whatever its faults, it's the best we're likely to get.

[Garethp8873]

Did manage to get a few shots of GWR 56400 at Didcot on Wednesday...

 

Didcot Railway Centre - August 2017

 

 

[lofty1966]

I can't see anyone else producing a GWR brake van suitable for the late 19th/early 20th century though. I'm quite surprised that Oxford are. So whatever its faults, it's the best we're likely to get.

You are probably right, so I have my tube of filler , clean putty knife and sharp scalpel waiting !

[The Johnster]

I can't see anyone else producing a GWR brake van suitable for the late 19th/early 20th century though. I'm quite surprised that Oxford are. So whatever its faults, it's the best we're likely to get.

 

Probably not, but Triang did, in bauxite.  I had one and thought it was inaccurate because it was too short, which it wasn't whatever else was wrong with it; it certainly sat up and begged too much, but all Triang's stuff did.  I was just used to a reality now long gone where toads were longer and lower than that and thought that that was all there was; it was some time before I realised that I was wrong.  It had the advantage for a train set toy of easily negotiating 13 inch curves without buffer locking; the HD 'proper length' (as I thought) one did as well but the couplers swivelled on this.

 

These still appear on eBay sometimes if anyone wants a go at working one up.  No idea about dimensions or scale accuracy, but it might not be all that bad; some of their stuff from the 60s was about as good as anyone had a right to expect from a toy, and I still have a Southern Utility Van awaiting decent bogies and a new roof from that era.  Whatever you say about the upcoming Ox, though, it'll be better than an old Triang...

Edited August 18, 2017 by The Johnster

[Gerald Henriksen]

I can see where you're coming from Gerald though duplication already exists of other basically accurate models.

 

I was referring more generally (and it also doesn't always apply to Oxford).

 

Yes, many models have duplication.  But there are also many prototypes that will never be popular enough to allow for duplication, in which case a bad model kills off the chance of an accurate model.  Or cases where either both or all models have issues meaning an accurate model isn't feasible.

[BG John]

Probably not, but Triang did, in bauxite.  I had one and thought it was inaccurate because it was too short, which it wasn't whatever else was wrong with it; it certainly sat up and begged too much, but all Triang's stuff did.  I was just used to a reality now long gone where toads were longer and lower than that and thought that that was all there was; it was some time before I realised that I was wrong.  It had the advantage for a train set toy of easily negotiating 13 inch curves without buffer locking; the HD 'proper length' (as I thought) one did as well but the couplers swivelled on this.

 

These still appear on eBay sometimes if anyone wants a go at working one up.  No idea about dimensions or scale accuracy, but it might not be all that bad; some of their stuff from the 60s was about as good as anyone had a right to expect from a toy, and I still have a Southern Utility Van awaiting decent bogies and a new roof from that era.  Whatever you say about the upcoming Ox, though, it'll be better than an old Triang...

1) Buy one on eBay

2) Discover the body is too high, verandah too long, it's too wide and the ends are rubbish

3) Perform major surgery on it, and also on an Airfix van to get better ends and roof. Plus prepare for surgery on the Airfix chassis.

4) Oxford announce 6-wheeler, that looks as though it just needs the centre wheels removed. Pre-order one, and give up on bashed Tri-ang/Airfix one

5) Oxford announce 4-wheeler. Change pre-order, as this looks even easier

6) Wait to see what turns up!

[No Decorum]

… a bad model kills off the chance of an accurate model.  Or cases where either both or all models have issues meaning an accurate model isn't feasible.

A very good point to make. It is the cause of a lot of frustration.

[The Johnster]

1) Buy one on eBay

2) Discover the body is too high, verandah too long, it's too wide and the ends are rubbish

3) Perform major surgery on it, and also on an Airfix van to get better ends and roof. Plus prepare for surgery on the Airfix chassis.

4) Oxford announce 6-wheeler, that looks as though it just needs the centre wheels removed. Pre-order one, and give up on bashed Tri-ang/Airfix one

5) Oxford announce 4-wheeler. Change pre-order, as this looks even easier

6) Wait to see what turns up!

DSCF8003.JPG

 

(cod Italian accent) 'My beautiful plan, she is all a-gone smash!'.

Edited August 19, 2017 by The Johnster