Model Design Errors

[Ian J.]

So, yet again it appears a model's CAD work errors have reached pre-production with the Heljan 45 (see that model's thread).

 

I raised the question of why such errors occur in the Accurascale Class 55 thread, but as it was a little hidden there and didn't in any way apply to their model of the Deltic, I thought I would start a thread for discussion of why these keep happening and what practical steps could be taken to avoid them.

 

All the major 'manufacturers', and a few of their commissioners, seem to have this problem. Sometimes a model is made that is pretty much perfect, then the next one is full of issues. Some easy issues get fixed, but basic body shape problems seem to just get into the CAD work, and then, even if pointed out, never get corrected. Why is it such a problem for these things to be sorted out before any tooling is cut? Surely the 'manufacturers' know that a model with errors won't sell as well, while one that is good sells better? An example of a model continually selling - the SJW Class 24. An example of one not selling, the Hornby Mark 2E. There are countless examples of both, from the same 'manufacturers' so I can't list them all.

 

Does it really just come down to setting a budget that's too low, running out of money, and the 'manufacturer' committing to tooling on a design with known flaws, just to save 'pennies' compared to getting it right and making more sales?

 

What are the systematic and systemic issues that need to be addressed to make sure that the errors don't get past CAD?

 

TIA

[tomparryharry]

Perhaps they (the manufacturers) don't know that there is a problem. Or, the raised questions is/are put over in such a way as for said manufacturer to say "s*d you", and carry on anyway.  Price?  Certainly price. Making 'your' model different enough so it's not copied. 

 

Lack of understanding about how the model functions in real life. 

 

The compromise that a manufacturer has to take between a scale model, and a toy.  Fragility? Playability?

 

Just some of the considerations that are accounted when (probably anything) is made on a commercial basis. 

[Ian J.]

1 minute ago, AlexHolt said:

If people are going to get so wound up about every single little error perhaps they should move to EM or P4 so that the wheels are the correct gauge. 

 

The issue of gauge accuracy has (for the most part) no bearing on the accuracy of body shape. Certainly things like body side to roof transition (as is the issue with the Heljan 45); front cab window position (Hornby VEP and Kernow/Bachmann 4TC); cab shape (Heljan Western; Dapol Western); and cab cut out shape (Oxford Dean Goods; Dapol B4) are not affected by the gauge.

 

So if possible I'd like to keep this thread free of the gauge inaccuracy issue.

[Phil Bullock]

I was pondering these issues myself given the dialogue re the Heljan 45.

 

The main issues - basic body shape, absence of triangular framing behind body side grilles, over prominent rivet/bolt heads - dont have any fragility/playability issues so in this specific instance we can discount that issue. That largely leaves cost and if a manufacturer has a pre-existing model of the same prototype in another scale then there might be savings by using research and CADs from that model but that risks perpetuating pre-existing errors. However looking at pictures of the O gauge model its only the bolt/rivet heads that appear to be an issue on that - the triangular framing and body shape issues being described with the OO model dont seem to exist there.

 

Are the manufacturers victims of our own desires for visibility of progress? I for one find CADs difficult to assess but plainly that is the stage at which errors need rectification as once metal has been cut for the dies then its pretty well too late. Surely the question has to be how to ensure all is correct before metal is cut....are there key characteristic points for a model that could be listed and checked?

 

Input from experts in fora such as this one ought to be key but has already been pointed out the "s*d you" factor comes in to play if this is not done appropriately. Which then takes us to the last point in the process - to buy or not? Entirely a personal decision - if the faults jar sufficiently then dont buy the model, or buy and rectify the issues with some modelling.

 

Will the Heljan 45 be a significant advance over exiting models? If these faults are not rectified possibly not but that does not deal with the disappointment of those hoping for the definitive sealed beam/AB/ETH loco (delete as appropriate) and the probability that it will be a long time before anyone contemplates releasing another new version of this model. Not me! Steam heat and a few DB locos but mostly VB is my bag.

 

For me the jury is still out.... and whilst Heljan have been a mixed bag over the years from my experiences the Hymek in particular is a cracking model....heres hoping theres still time to get this beast to the same standard

[Steamport Southport]

There is also the problem in that when you do point out an error there is often a chorus of "rivet counter" or "It's only a train" even at an early stage.

 

Some people do know a vast amount about locomotives and stock on this website, but they are often shouted down by those that don't. I've even recently noticed an etched kit manufacturer being virtually told he doesn't know what he is talking about in a thread about a prototype he has spent years researching and has already manufactured a kit of.

 

Remember "splashergate" when someone pointed out the erroneous splashers and was ignored and told "It's only a CAD they'll get it right in the end". When the model came out everyone complained about them and it almost certainly hit sales....

 

Also don't belittle those that do know what they are talking about. I recently withdrew from a thread due to some obnoxious individual telling me to "do the research" in a very patronising and abusive manner when I was trying to point out that somebodies post was wrong. I was actually doing that research at the time, which wasn't even relevant to me. So I binned it all.

 

 

 

Jason

[Wheatley]

2 hours ago, Ian J. said:

 

The issue of gauge accuracy has (for the most part) no bearing on the accuracy of body shape. Certainly things like body side to roof transition (as is the issue with the Heljan 45); front cab window position (Hornby VEP and Kernow/Bachmann 4TC); cab shape (Heljan Western; Dapol Western); and cab cut out shape (Oxford Dean Goods; Dapol B4) are not affected by the gauge.

I think Alex's point was that the majority of OO buyers, being unconcerned about the obvious gauge error, are less likely to be concerned about minor shape errors either.

 

According to its designer there is nothing wrong with the Dapol Western, it's a really great model, beautiful model, perfectly accuate, no errors, and all the naysayers (and there were some) were merely spreading fake news. He even had a go at MR for not giving it 100% in their review. So not so much a case of "sod you" as "ner ner ner I'm not listening". As for Oxford, they are definitely designed down to a price, see the LNER cattle wagon with one of the sides back to front. They keep selling so clearly most people don't care. 

 

It is annoying though.

[Phil Parker]

There is an assumption on many forums (Facebook being the worst) that manufacturers don't bother consulting anyone before putting out a model. What the people typing usually mean is, "Why didn't they consult ME!!!" - when the me in question will probably be unknown to anyone up to that point. It doesn't help that you then get arguments between the "experts" that descend into a slanging match. Every manufacturer does their research and consults people they consider to be experts before launching anything. They do look at forums, especially this one, but asking for comment means someone watching the resulting thread and trying to disentangle the facts from invective.

 

Ultimately, if a model was only launched when everyone on ever forum was happy with it, we'd never see another model again.

 

From the customers point of view, the solution is simple, if you don't like a model, don't buy it. I've said this many times - these are toy trains NOT an essential that you can't live without like food and shelter. Get as angry as you like, there's a lot of factors in play when designing a model, many not obvious to people at a keyboard. It may simply be a calculation that there are enough people who are going to be happy with the model to sell it and make a profit.

 

Personally, when someone starts arguing over minutiae, I want to see their layout because if the thing that are ranting about is going to make that model the worst thing on the layout, it must be breathtakingly impressive. No manufacturer sets out to deliver a less then perfect product, but bringing a loco to market is a very difficult process - look at the last year to see new entrants who ultimately failed to do this despite much enthusiasm on here for what would be models setting new standards.

[Ian J.]

8 minutes ago, Phil Parker said:

...

It may simply be a calculation that there are enough people who are going to be happy with the model to sell it and make a profit.

...

 

This is what worries me. It seems to be a settling for less profit than could otherwise be made from a model.

 

Surely getting basic body shape right means that a model will have a better chance of selling more and therefore making more money for the producers of said model? Especially so given these days of online communication where when a significant error is noticed such that the buyers cancel their orders?

 

There must be ways of checking a model's CAD against prototype to reduce the risk of error? I suggested in the Accurascale Deltic thread that:

 

Quote

...Surely someone with knowledge and vested interest is checking to make sure that the CAD looks right; that all GA and critical dimensions and potentially problematic areas are noted and checked to make sure they are done correctly; a 3D print done to ensure output from CAD is working right and then all GA and critical dimensions and noted areas checked again; the first tooling done and an EP received and again, all GA and criticial dimensions and noted areas checked again...

 

[Phil Parker]

19 minutes ago, Ian J. said:

This is what worries me. It seems to be a settling for less profit than could otherwise be made from a model.

 

Surely getting basic body shape right means that a model will have a better chance of selling more and therefore making more money for the producers of said model? Especially so given these days of online communication where when a significant error is noticed such that the buyers cancel their orders?

 

Your whole argument assumes no-one checks the CADs, this isn't true. Mistakes will still happen though - if you've never worked somewhere where things go wrong despite everyone's best efforts then you work in a pretty unique industry. When you find the mistake, a decision then has to be made about how you deal with it. For the 45, fixing the body shape means (as it says on the thread) a new mould. Cost, tens of thousands of pounds. Would this mean that many extra sales? That's a decision for the person who is paying for it.

[Enterprisingwestern]

1 hour ago, Phil Parker said:

Personally, when someone starts arguing over minutiae, I want to see their layout because if the thing that are ranting about is going to make that model the worst thing on the layout, it must be breathtakingly impressive.

 

Most of us aren't professional full time layout builders though, just customers buying sub standard (to varying degrees) items, generally not at giveaway prices.

 

Mike.

[Phil Parker]

25 minutes ago, Enterprisingwestern said:

 

Most of us aren't professional full time layout builders though, just customers buying sub standard (to varying degrees) items, generally not at giveaway prices.

 

Mike.

 

What is the standard they are sub? I'd suggest that if the model is as good as the layout it will operate on then it will look fine. A mega-model sat on a layout ballasted with pea-gravel doesn't make the layout look better, it makes it look worse in comparison IMHO.

 

However, is a model is really "sub standard", then simply don't buy it. Market forces will then take effect and either the firm will do better or go bust.

[Ian J.]

OK, so can we keep to the costs up to signing off CAD? I fully understand that a mistake getting through to tooling would be expensive to fix - but this thread is mostly about catching those errors before signing off CAD, when they are far cheaper to fix.

 

As for 'errors happen', I'd agree, except that what I'm worried about is repeated errors, not just the infrequent one off minor error. The errors I'm seeing (and it's not just Heljan) are coming about repeatedly across different models. This suggests strongly that at the CAD stage there is something systemic and/or systematic that is 'wrong'.

 

If a CAD is correct when signed off, then tooling errors should be corrected by the tooling maker for no more cost than initially agreed, as it's they who have not followed the CAD. That shouldn't be something the originator should have to pay for.

 

I also feel that it's unfair to compare what an amateur modeller is capable of producing against a professional model maker's output. I don't ever expect to sell my efforts to anyone else and even if someone wanted to buy something I'd designed, I'd be adding caveats all over the place to say I'm not a professional designer, I have no design training, and no significant design experience, so any errors will have to be accepted.

[D9020 Nimbus]

In the case of the Heljan 45, it isn't going to sell much if the standard it is "sub" is that of the existing Bachmann 45.

 

Personally, not having much room I've no intention of purchasing a class 45 from anybody—my main interests are N gauge and narrow gauge (009 and HOe). In that area, the money I "invested" in a couple of Heljan Manning Wardles has been effectively wasted. I think my pair are second-generation chassis. While they don't have the issues some of the first batch had they won't negotiate points of approx. 12" radius and the coupling pockets are at the wrong height to couple to PECO stock. Odd, when you consider the two companies were said to have collaborated.

 

One of the two is inclined to sit on the track with its driving wheels spinning. I note that PECO are collaborating with Kato on the Ffestiniog models, not Heljan…

[andyman7]

I'll wager Hornby sold a lot more Mk2Es than SLW have sold Class 24s. The SLW model has a narrow but very well identified market which it has served very well, but it's not the same market as Hornby. 

 

To a greater or lesser extent we all have our personal prejudices and red lines where models are concerned but I think there a very few models that have underperformed in sales terms due to design shortcomings, I think it is a false attribution by those who are personally frustrated when an anticipated model fails to pass (their) muster.

 

To give you a few personal examples where my opinion is probably at odds with others (and that deosn't mean either of us is 'right' or 'wrong'):

The Heljan Class 86 was.....OK with me! And I knew the real thing very very well indeed.

The original Heljan Class 33....honestly, that roof shape thing is barely noticeable

Bachmann Deltic - works for me

The Bachmann Mk2a BFK with incorrect roof vents - now that really annoys me, especially as the Triang model is correct

The Hornby Class 60 Railfrreight with incorrectly positioned logos - I've held off buying until the corrected version appears

The Bachmann blue 2EPB with transposed vehicle numbers - again unacceptable for me and I got hold of the replacement bodyshells

 

I'm sure other lists would differ, and as Phil Parker says above, these days I'll buy the things that meet my needs and leave behind the ones that don't

[Enterprisingwestern]

35 minutes ago, Phil Parker said:

 

What is the standard they are sub? I'd suggest that if the model is as good as the layout it will operate on then it will look fine. A mega-model sat on a layout ballasted with pea-gravel doesn't make the layout look better, it makes it look worse in comparison IMHO.

 

However, is a model is really "sub standard", then simply don't buy it. Market forces will then take effect and either the firm will do better or go bust.

 

Rubbish.

So we can only buy stock that is commensurate to our modelling standards.

Might as well re-introduce Budgie, Trix, Hornby Dublo etc.

 

Mike.

[AY Mod]

8 minutes ago, Enterprisingwestern said:

So we can only buy stock that is commensurate to our modelling standards.

 

Publishing the images thereof in black and white seems to improve their accuracy in the eyes of some. Do away with colour and there'll be an instant improvement in quality for a percentage of modellers it seems.

[Matt C]

Sticking head above parapet.  If you want perfection you buy a Rolex and pay a premium price accordingly

 

If you want a very good watch  you pay a lower price, its a very good watch, but its NOT a Rolex

 

If you want something that just tells the time, you can buy a cheap watch

 

What you are saying is you want  someone to keep on re designing until they produce a Rolex, but at the lower price.  ( The cost of constantly re designing to eliminate every imperfection would mean a loss on every sale leading to quickly going bankrupt)

 

Most are happy with a very good watch, at a lower price, KNOWING. Its not a Rolex and accepting that, but still having a very good watch

 

If you want that perfection, go commission a Rolex

[tomparryharry]

*

Edited June 22, 2020 by tomparryharry
Deleted, not relevant.

[Phil Parker]

14 minutes ago, Enterprisingwestern said:

 

Rubbish.

So we can only buy stock that is commensurate to our modelling standards.

Might as well re-introduce Budgie, Trix, Hornby Dublo etc.

 

 

If your layout is full of badly made card kit buildings, then hours spent arguing over tiny errors on a loco is IMHO, a waste of time. A layout with a consistent level of detail will look better than one where you have a couple of star items and the rest is mediocre. I suspect that's why a lot of people are happy with models that some people will spend hours criticising. Every one of us is different.

[Enterprisingwestern]

4 minutes ago, Mattc6911 said:

Sticking head above parapet.  If you want perfection you buy a Rolex and pay a premium price accordingly

 

If you want a very good watch  you pay a lower price, its a very good watch, but its NOT a Rolex

 

If you want something that just tells the time, you can buy a cheap watch

 

What you are saying is you want  someone to keep on re designing until they produce a Rolex, but at the lower price.  ( The cost of constantly re designing to eliminate every imperfection would mean a loss on every sale leading to quickly going bankrupt)

 

Most are happy with a very good watch, at a lower price, KNOWING. Its not a Rolex and accepting that, but still having a very good watch

 

If you want that perfection, go commission a Rolex

 

A somewhat irrelevant argument.

We're not talking of aspirational goods, just toy trains.

At the end of the day, as with most things in life, you get what you pay for.

As others have said, you pays your money and takes your choice.

 

Mike.

[Enterprisingwestern]

1 minute ago, Phil Parker said:

 

If your layout is full of badly made card kit buildings, then hours spent arguing over tiny errors on a loco is IMHO, a waste of time. A layout with a consistent level of detail will look better than one where you have a couple of star items and the rest is mediocre. I suspect that's why a lot of people are happy with models that some people will spend hours criticising. Every one of us is different.

 

Stop belittling the achievements and skill levels of other modellers, being unable to produce Downes like buildings, Hewitt style signals and such doesn't mean you have to !ower your expectations when it comes to rolling stock

That modeller with his badly built layout is probably working to the limit of his capabilities and would expect manufacturers to do likewise.

 

Mike

[Phil Bullock]

6 minutes ago, Phil Parker said:

 

If your layout is full of badly made card kit buildings, then hours spent arguing over tiny errors on a loco is IMHO, a waste of time. A layout with a consistent level of detail will look better than one where you have a couple of star items and the rest is mediocre. I suspect that's why a lot of people are happy with models that some people will spend hours criticising. Every one of us is different.

 

Whilst I dont disagree with those points Phil how many of these models ever end up on a layout to put their quality in context?  Viewed in isolation in a cabinet or box the errors probably assume greater proportions.

[Legend]

19 minutes ago, Phil Parker said:

 

If your layout is full of badly made card kit buildings, then hours spent arguing over tiny errors on a loco is IMHO, a waste of time. A layout with a consistent level of detail will look better than one where you have a couple of star items and the rest is mediocre. I suspect that's why a lot of people are happy with models that some people will spend hours criticising. Every one of us is different.

Not sure I agree with that . I know people that run the most detailed models  , yet run them on a layout with no scenery .  You can get pride in the appearance of individual models  Even if you’ve got nothing but a test track to run them on . But in a way you are right . We are all different. What pleases one doesn’t another

Edited June 21, 2020 by Legend

[The Johnster]

I have several slightly questionable models on my layout, and would replace them if better RTR versions were available but will happily live with their faults for now.  A 3 coach set of semi-fictional Triang GW shorty clerestories, a Limbach 94xx representing a loco that is not one of the GW build first 10, and a Triang Hornby 2721, all of which have been worked up to some extent.  The worst offender is probably the 2721; I have retained the original chassis because a) the splashers sit over the wheels properly and b) Photographic dated evidence of my prototype shows that it should have fluted straight coupling rods.  A 'correct' Bachmann 57xx chassis would have the wheels out of alignment with the splashers and fishbelly coupling rods, and a Bachmann Jinty/1F chassis to replace the generic incorrect Jinty Triang Hornby chassis would also have incorrect wheel positions that would not match the splashers, as the generic Jinty R52 chassis is incorrect for a Jinty.

 

This sort of thing, a manufacturer getting it wrong in a situation that wasn't any harder to get right, is annoying.  Apparently, the incorrect R52 chassis came about because Rovex wanted to use the Princess coupling rod, but were happy enough to make new ones for the L1.  This error was kept through several retoolings of the generic chassis for Jinties, 2721s, cod 08s, J83s, and still turns up brand new under train set 08s.  Once an error has been established, it has a tendency to propagate into new toolings and even other company's models in different scales.  When Lima went into 0 gauge, they simply put their 00 coaches through an embiggening ray, so that the plastic sides were several mm thick!  It looked bad in 4mm, appalling in 7mm.  Hornby are repeating the B1 bogies on shorty clerestories in current production, despite having suitable bogies in the range.  It's daft.  It was daft of Hornby to perpetuate the 2mm too high sit of models for so long, but at least there was a reason for it, to allow stock to cope with the transition angle at the bottom of train set gradients.

 

 

[Matt C]

 

 

20 minutes ago, Enterprisingwestern said:

 

A somewhat irrelevant argument.

We're not talking of aspirational goods, just toy trains.

At the end of the day, as with most things in life, you get what you pay for.

As others have said, you pays your money and takes your choice.

 

Mike.

 

 

 

 

How is it irelevant ? One Mans Rolex is another mans Toy train.  This thread is about Model design errors. 

Perfection = no errors  Perfection comes at a much higher price than most are prepared to pay  Hornby, Bachmann, Heljan etc to produce. I feel that as models get ever more accurate, detailed and complex we lose sight of the basics, does it look reasonable, does it actually RUN reliably, at the price I'm paying THATS what I'm looking for, not " are those brake shoes one millimetre out of alignment"  after I've picked it up and plonked it down half a dozen times they have fallen off anyway.  

 

Get the basics right, and make it run well,. Let others add the micro detail as an addon pack if they want 'perfection'