Interesting article in Bachmann Times re quality control

[Elvinley]

Reading through the recent Bachmann Times there is an article about the process of producing their models. Most of us these days assume that quality control is non existent but according to this article, the products are tested as the process progresses including test tracks and electrical testing.

[Chris Morris]

Haven't seen that article. I'm sure that the vast majority of purchasers have a great product and don't post anything on forums because they are completely satisfied but the very small number who do have problems say so on forums such as this. It is therefore likely that forums show a very distorted picture. I still haven't returned either a Farish or Dapol item. The Farish items seem to be very well made and as solid as anything of that size and with such fine detail can be. It feels to me that Farish products are a little more resilient than Dapol products in terms of not having minor problems during reasonably heavy use. All my Farish items run very smoothly and without any drama even though they get quite a hard life at exhibitions; I just wish the Castle had traction tyres.

Anecdotes I have heard suggest that a fair number of the returned items are not due to manufacturing or design issues.

[Webbo]

Chris

I'm happy for you that you've had no problems with Farish or Dapol products. I really wish the same was true for me. Anecdotes supporting one position or the other regarding locomotive reliability are not good enough. I would suggest that the assertion that the number of customers who have had problems is very small can't be substantiated by you or by anyone else.

The truth is out there and is not going to be determined by force of proclamations.

Webbo

[Bealman]

If you haven't  seen the article, Chris, should you make comment? 

[Chris Morris]

If you haven't  seen the article, Chris, should you make comment? 

Just saying as an end user that the quality of Farish products I have bought feels good which means there must be good manufacturing process including quality controls.

[NinOz]

If you haven't  seen the article, Chris, should you make comment? 

Why not?
His comments don't address the article content.  Just presents his opinion and supposition regarding quality and failures. Same as everyone else on this site.

5,390+ members; how many active readers; how many have problems with new locos; how many of those can be bothered to write?  Damned if I know!
Unless one has access to hard data from the manufacturers then all else is just dust in the wind.  Some just like to assume, suppose and pontificate.

For me in the last 7 years :
Faulty Farish = 0/8
Faulty Dapol = 1/9 (one of the cheap returns, burnt out led)
Don't know if I'm lucky, it is normal experience or perhaps more careful than most.

I detect an inference that some of the failures and returns may be due to operator error.  My experience of ham-fisted users in a variety of activities would support this.

CFJ

[Snowwolflair]

Chris

I'm happy for you that you've had no problems with Farish or Dapol products. I really wish the same was true for me. Anecdotes supporting one position or the other regarding locomotive reliability are not good enough. I would suggest that the assertion that the number of customers who have had problems is very small can't be substantiated by you or by anyone else.

The truth is out there and is not going to be determined by force of proclamations.

Webbo

A pertinent observation is the significant fall off of reworked items by Farish at trade shows.  Gone are the days of a counter of reworked returns at a discount.

I asked at last years TINGS and was told the number of returns of new models have fallen significantly and as older less reliable models go out of production their returns are falling as well.

[silly moo]

I think a quite few dud models still manage to get though QC!

One thing manufacturers can't control is the way that modellers use their models, I've seen some pretty rough and clumsy handling and dodgy track and am surprised that some of them run at all. (The dodgy track usually belonged to members of our club who asked for help)

I once went to the Bachmann headquarters in Barwell to return a faulty loco in person (they don't usually let you in unannounced or at all but they made an exception because I said I was on my way back to Africa) It was just a warehouse with offices but at that stage they did have someone who tested returned locos.

I chatted to him while I waited for my replacement loco. He said that they got locos full of dog hair, carpet fluff etc. They also got locos with no return address and ones that they could find absolutely no fault with, those they put down to the customers using faulty controllers or some other electrical problem.

By the way, the loco I returned was a rebuilt Merchant Navy which although it ran well, had damaged coupling rods that had been bent into the correct profile with pliers either by the factory or the retailer.

[Carmont]

 @Elvinley may be able to clarify since he has the article, but my understanding of QC in these situations is to take a sample of items, usually at random and test them, rather than test every item individually (a mammoth task, and costly, for batches in the thousands). It's inevitable, therefore, that some faulty items will pass through the system.

What would be interesting to know is the failure rate of the random items that they tested.

[silly moo]

I have also read that manufacturers of almost everything use batch testing as a method of quality control. I was told that some manufacturers of washing machines test one in ten while others test each one.

Another factor is good design of the article in the first place, model locos have to look like the prototype and work as miniature models. I do think that some of the Dapol models which look exquisite detail wise are rather fragile and need to be handled carefully.

[austinbob]

@Elvinley may be able to clarify since he has the article, but my understanding of QC in these situations is to take a sample of items, usually at random and test them, rather than test every item individually (a mammoth task, and costly, for batches in the thousands). It's inevitable, therefore, that some faulty items will pass through the system.

What would be interesting to know is the failure rate of the random items that they tested.

Sampling inspection and testing is a mathematical science based on a given sample size reducing failures to below a defined level of failures.
The sample size is selected depending upon batch size, previous failure levels and what is considered to be an acceptable failure rate.
If previous failure rates are high enough then it may still be necessary to check 100% of products to achieve the chosen acceptable failure rate.
Its also worth noting that if, say, the acceptable rate is 10%, then there is a high probability that 10% of customers will receive faulty product.
These are the customers who, justifiably, complain about poor quality.
It is also worth noting that if the manufacturer does not stick to the sampling rules (based on statistical sampling tables) by reducing samples below the required rate, there's a risk failure rates will be much higher.

[Snowwolflair]

Samsung do one test on TVs they power them up, look at the picture and measure the current to uA accuracy.  If the current is correct and the picture correct they are passed.

They have virtually no failures due to manufacturing in the field.

[Elvinley]

The way the article reads, it seems everything is tested.

[austinbob]

The way the article reads, it seems everything is tested.

That would make sense. They probably have relatively small batches/production runs and may not be able to establish enough failure history for a new loco to be able to determine an effective sampling plan. Although they might be able to use the failure history of similar Models.
I must say I've not had any dodgy Farish locos for some time now. I've certainly had a few in the past.

[silly moo]

The last ten Farish steam locos I have bought (over four years) have all been excellent.

Two were Bachmann returns and one was bought recently second hand. The only problem I have had was self inflicted.

I can remember when there were pages and pages of complaints about any new model, that doesn't happen any more.