Yet another Dapol moan thread

[NinOz]

Stamp collectors WANT and NEED faults in their stamps cos they're worth more!! 

  Now, if we could just capture that mindset and install that way of thinking into model railway enthusiasts.

Could be on a winner there.

CFJ

[daffy]

I firmly agree that if a new loco - or anything for that matter - is faulty out the box, or shortly after, then back it goes, tout suite. To do otherwise is to allow manufacturers to continue to churn out sub-standard goods without censure.

I have not bought Dapol nor will I, but in my short time here on the Forum I have noted two things: firstly, the number of posts that refer to faulty locos from them, with poor build and bad wiring and drive connections seemingly all too common; secondly the number of posts from folks defending them gamely.

Okay, so I get that the exterior detail and quality of their models is considered very good, but if what's on the inside needs attention from new in as many instances as get noted here then I would not consider them worthy of defence.

I have bought Swiss outline from Fleischmann, Kato, Hobbytrain, Marbartren, Liliput, Sowa-N, Trix and others, but so far all the items bought new have functioned perfectly, have had no discernible faults, and therefore have been worth the extra cost these manufacturers charge. I'm not saying they are perfect, and others no doubt will have had different experiences with these brands, but I don't find a wealth of criticism matching that relating to Dapol here or on other Forums and webpages.

If loyalty to a maker like Dapol means buying sub-standard, faulty goods and accepting that as normal or perhaps part of the fun, then I don't get it. If my new car arrived with wiring that broke and gears that split, or a dashboard light that doesn't work, would I just patch it up? I think not!

Shoot me down if you like, but that's just how I see it - with all manufacturers, be they model train makers, car manufacturers, or the folks who build this magical little electronic box of tricks I'm typing on - which has always worked just fine, right out the box. (if it didn't I'd have sent it back)

[NinOz]

But when Farish produced less sophisticated/lower detail models they still managed to get split gears and 30 years later they have not managed to fixed it. Maybe its just another profit line for them selling replacement gears

Just finished replacing a split gear, other bogie feels a little "gritty" so probably another replacement on the way.
Such a simple thing, a bit of plastic on a bit of metal - how can one get it wrong for so long?

CFJ

[Nik96]

I've only had a handful of Dapol products but never had a single problem with any of them I couldn't fix in 30 seconds.

Long may Dapol prosper and keep delivering quality I've come to know.


Its the nature of anything manufactured... They wont be perfect and when you deal with small volumes, these issues are just amplified. There isn't the ability to tack 10% extra volume to an order because you know you'll have faults. To expect everything to work faultless first time out of the box is quite frankly ridiculous. This is the purpose of warranties because this isn't an ideal world...

This is the issue of a niche market. That said if you do want to take up a fault free hobby there's always stamp colllecting 

All products should work straight out of the box. That should be what Quality Assurance is for. The warranty is for if anything goes wrong after you have started using it which worked correctly when you first started using it. This applies to all products not just model rail.

Personally I have 13 Dapol locos and the have all worked out of the box. The only failures I have had are with two Farish locos that arrived with split gears which were returned for replacement

But what if these models worked at QA then were shipped out to customers and modellers to then find out they don't work? Once a locomotive has been run in it should spend a long time providing faultless running before wear starts to cause issue. This follows the comment you make that "The Warranty is for if anything goes wrong after you have started using it which worked correctly when you first started"

[PaulCheffus]

But what if these models worked at QA then were shipped out to customers and modellers to then find out they don't work? Once a locomotive has been run in it should spend a long time providing faultless running before wear starts to cause issue. This follows the comment you make that "The Warranty is for if anything goes wrong after you have started using it which worked correctly when you first started"

Hi

I doubt very much every loco is put through the QA process as it would cost too much, it would be more likely random sampling.

Cheers

Paul

[Nik96]

But what if these models worked at QA then were shipped out to customers and modellers to then find out they don't work? Once a locomotive has been run in it should spend a long time providing faultless running before wear starts to cause issue. This follows the comment you make that "The Warranty is for if anything goes wrong after you have started using it which worked correctly when you first started"

Hi

I doubt very much every loco is put through the QA process as it would cost too much, it would be more likely random sampling.

Cheers

Paul

It's the most cost effective approach. It's probably one of these scenarios where all the parts are checked individually at manufacture found to be working and the assembly is actually what throws up issues.

[austinbob]

But what if these models worked at QA then were shipped out to customers and modellers to then find out they don't work? Once a locomotive has been run in it should spend a long time providing faultless running before wear starts to cause issue. This follows the comment you make that "The Warranty is for if anything goes wrong after you have started using it which worked correctly when you first started"

Hi

I doubt very much every loco is put through the QA process as it would cost too much, it would be more likely random sampling.

Cheers

Paul

It's the most cost effective approach. It's probably one of these scenarios where all the parts are checked individually at manufacture found to be working and the assembly is actually what throws up issues.

However the number of products sampled should be dependent upon the number of faulty products found. The more faults (including warranty returns) then the bigger the sample has to be for confidence that the shipped product is good. In the limit, too many faults requires 100% inspection.
This is standard test/inspection sampling planning. I doubt Dapol (or Farish) do this properly.

[Nik96]

But what if these models worked at QA then were shipped out to customers and modellers to then find out they don't work? Once a locomotive has been run in it should spend a long time providing faultless running before wear starts to cause issue. This follows the comment you make that "The Warranty is for if anything goes wrong after you have started using it which worked correctly when you first started"

Hi

I doubt very much every loco is put through the QA process as it would cost too much, it would be more likely random sampling.

Cheers

Paul

It's the most cost effective approach. It's probably one of these scenarios where all the parts are checked individually at manufacture found to be working and the assembly is actually what throws up issues.

However the number of products sampled should be dependent upon the number of faulty products found. The more faults (including warranty returns) then the bigger the sample has to be for confidence that the shipped product is good. In the limit, too many faults requires 100% inspection.
This is standard test/inspection sampling planning. I doubt Dapol (or Farish) do this properly.

Your quite right failure rates should dictate inspection rates.

Yet the figures manufacturers have for Mean Time Between Failures or Percentage Faults to batch size is distorted unless every fault is reported. In a hobby where make do and mend is very common...

[ntpntpntp]

Split gears aren't unique to Farish by any means. 

Over the years I've had Arnold locos with the same problem.  Typically the final axle drive gear on bogie diesels/electrics, where it's a small size and too tight on the shaft and the plastic has failed over time.   Arnold spares can cost a fortune (if you can find them), but fortunately I've been able to source suitable replacements from MicroAntriebe.  You just need to count the teeth and work out the MOD value and thickness, and see what they've got in standard stock or can make to order.

I know some folk go on about not liking brass gears because they're noisy and they wear out leaving metal in the mechanism, but to be honest I'd still rather have brass gears. Properly maintained they last just fine.

[Sprintex]

I have bought Swiss outline from Fleischmann, Kato, Hobbytrain, Marbartren, Liliput, Sowa-N, Trix and others, but so far all the items bought new have functioned perfectly, have had no discernible faults, and therefore have been worth the extra cost these manufacturers charge.

And therein lies the truth of it all, the models you have bought have been "worth the extra cost".

. . . in my short time here on the Forum I have noted two things: firstly, the number of posts that refer to faulty locos from them, with poor build and bad wiring and drive connections seemingly all too common; secondly the number of posts from folks defending them gamely.

So presumably the THIRD thing you might have noticed is that there's also a lot of people that continually moan and groan about the price of UK outline models even though they are cheaper than those you have mentioned, AND that a lot of those moaning about said prices are the same people who complain about lack of quality!

Some want cheap AND 100% reliable and that just isn't going to happen, outside of Union Mills who are still in the 1980s with regard to detail finish.


Paul

[austinbob]

I for one, as I've indicated before on this forum, would be more than happy to pay more if that's what it takes for a properly working and reliable loco.

[ntpntpntp]

I have bought Swiss outline from Fleischmann, Kato, Hobbytrain, Marbartren, Liliput, Sowa-N, Trix and others, but so far all the items bought new have functioned perfectly, have had no discernible faults, and therefore have been worth the extra cost these manufacturers charge.

And therein lies the truth of it all, the models you have bought have been "worth the extra cost".

Yet are they really that much more expensive?  Granted there's a bit of a hike recently with the exchange rate fluctuations etc. and some of the prices for DCC and sound fitted stuff is eye-watering (not something I'm interested in), but I can still find new locos of those makes mentioned  for around the £90 - £120 mark, and multiple units in the £150 range.

The contintental N brands don't always get it right:  I've had motors fail on Liliput FLIRT emus, a known fault with these if you read german forums, and I had to order replacements at cost.  Of course Liliput is part of the Bachmann/kader empire these days.     

[N-Gauge-US]

However the number of products sampled should be dependent upon the number of faulty products found. The more faults (including warranty returns) then the bigger the sample has to be for confidence that the shipped product is good. In the limit, too many faults requires 100% inspection.
This is standard test/inspection sampling planning. I doubt Dapol (or Farish) do this properly.

That isn't how QC works. All products receive visual QC at the factory and mechanical QC testing is based on failure rates. Acceptable failure rates are given by the factory and get exponentially more expensive as you reduce the likelihood of failure (i.e. If you agree to a failure rate of say 0.1% it might be 100 times as expensive as a failure rate of say 1%). Moreover, people often talk about how rising costs are tied to Chinese factory worker pay increases; the cost of QC is ALL labor. The type of 100% reliability people are talking about wanting to pay for, they can't even begin to fathom affording. We are talking about a shift from 125£ for a newly tooled loco to a cost of 400+£. Just not a feasible market shift.

[austinbob]

Common procedure for final inspection...
Choose an AQL (acceptable quality level)
Choose batch size
note failure rate from previous batch.
Look up sample size from aql tables
Look up number of failures allowed in sample
Carry out sample inspection/test
If failures less than allowed then ship batch and maybe reduce sample for next batch
If failures more than allowed then increase sample size as necessary. Maybe increase sample size for next batch.
Examine reasons for failures
Make improvements (design, components, operator training, test methods - whatever)
The limit for sampling depending on sample failures, could be zero for good quality and be 100 percent for poor quality
Yes it will cost more to have a bigger sample and a lot more to inspect 100 percent the whole point is to reduce customer exposure to faulty product and have a process of continuous improvement in design and production to achieve this and reduce failures and costs.

[Chris Morris]

But what if these models worked at QA then were shipped out to customers and modellers to then find out they don't work? Once a locomotive has been run in it should spend a long time providing faultless running before wear starts to cause issue. This follows the comment you make that "The Warranty is for if anything goes wrong after you have started using it which worked correctly when you first started"

Hi

I doubt very much every loco is put through the QA process as it would cost too much, it would be more likely random sampling.

Cheers

Paul

It's the most cost effective approach. It's probably one of these scenarios where all the parts are checked individually at manufacture found to be working and the assembly is actually what throws up issues.

However the number of products sampled should be dependent upon the number of faulty products found. The more faults (including warranty returns) then the bigger the sample has to be for confidence that the shipped product is good. In the limit, too many faults requires 100% inspection.
This is standard test/inspection sampling planning. I doubt Dapol (or Farish) do this properly.

Your quite right failure rates should dictate inspection rates.

Yet the figures manufacturers have for Mean Time Between Failures or Percentage Faults to batch size is distorted unless every fault is reported. In a hobby where make do and mend is very common...

You cant inspect quality into a product. It needs to be built right first time every time. As I have said before, Dapol are minnows compared to the Chinese manufacturers and have no business clout to force the factory to do better. Yes they could move their business somewhere else -but where. And would the current factory be the least bit worried about losing the Dapol contract - of course not because it is such a small part of their total business. Forcing other companies to do what you want is all about size and spend.