Hacker News Reader

Comment by modeopfer | Hacker News Reader

I also studied engineering and I think your professor missed the point. You don't overengineer stuff anymore nowadays, not because you want to scam your consumer out of money (aka planned obsolence) but because the design tools got better and every gram of metal or plastic and every screw and every operation like welding, glueing and so on costs money. Nowadays simulation tools, new manufacturing processes and experience allow very cheap designs that endure the predicted amount of stress but not much more. That's the trade-off.

Not every customer needs a power drill that lasts 5000 hours. Most people are okay with a cheap drill that lasts a cumulative 10 hours because that's the amount of lifetime use they get out of it.

View on HN

Fundamentally, you can't serve a turd and call it steak.

Power tools are an example of how to get it right. If I go to Home Depot and look for a drill, it's obvious from the branding, battery, warranty and physical characteristics which DeWalt drill is the "ok" consumer one, and which is a professional tool. (I use this as an example because I bought one yesterday)

The problem with consumer appliances is that they use dark patterns to sell stuff. There is no meaningful signaling about what is garbage and what is not. No facts are obvious that tells me what the expected lifecycle of a washing machine is... can it handle 350 loads a year? 100? 50? 1000? No fucking clue. Counter-intuitively, many of the premium priced units are worse than their cheaper counterparts!

Considering that Black & Decker own DeWalt, I wouldn't put much hope into your drill being a long lasting professional tool. I put my vote with Makita.

Author here, you are absolutely right. Want to see me angry? You should have seen me when I found out my 90 year old neighbor lady is on her second Whirlpool made vertical modular washer in the past 5 years. (top loading washer with the led lights on the control panel) Not everyone can afford to be ripped off and sold garbage washing machines that WILL break within 3 years.

I think your power tool analogy is a good one. Same with most consumer goods, people know that when they are buying a plastic version of something that it likely won't last, so they wrestle through the tradeoffs.

When it comes to tools, I usually buy the cheap version at Harbor Freight first, and then if I end up using it enough that it breaks, I'll research and get something more reliable.

Also, thanks for the Krylon tip. I bought a house that came with a Hamilton clothes dryer from 1970. Works fine, no plastic parts, though it is starting to rust. (My main worry is that the heating element will eventually rust away and they don't make replacements.)

To your point, I saw or heard an interview with a Ford engineer a while back where he basically said look we know how to make parts that last forever, but they are heavy and expensive. He went on to say people want more affordable cars (cheaper parts) and we need to make cars lighter for fuel efficiency standards (lighter parts), so they set an internal goal of 10 years of longevity for a part, and then try and make the cheapest/lightest part that meets that goal.

It's a fucking washing machine, 10kg extra isn't going to make any difference.

I seen an article on the ring pulls on top of coke cans. They are hollow because it has a huge impact on raw materials usage (over the course of 10 million cans). I think every gram counts to the bottom line of the share holders no matter how mundane. Similar to the UK Construction Industry. Hit the minimum possible legal requirement and charge the maximum amount of money.

> They are hollow because it has a huge impact on raw materials usage

It's been a while, but I remember when the transition from pull tabs to "pop tabs" occurred; at first (IIRC), the "pop tabs" were solid, but it wasn't long until they became hollow as well.

Of course, that led to some problems which still exist today (though not nearly as often). The biggest one being the balance between the strength of the tab, vs the opening part (whatever it is called - closure?). In the past (and occasionally today for the odd soda), you could pull up on the tab - and it would bend or break off, without opening the soda! Simple enough to fix (do not press down on the opening with your thumb!) with a butter knife or some other similar tool, but annoying at the same time.

10kg of steel would cost about USD$10.00, and then shipping it around the world costs again. At my factory, we use the number "50 cents a pound" as an average cost for shipping pallets around the lower 48 states.

So your extra 10 kg would cost me about $20 more. At retail that would be another $100 or so. In other words, making it 10kg heavier would sharply increase the price; it would also be for things that aren't easily visible to the consumer. Your competition would destroy you.

Source: I own a factory.

Are washing machines not still being made with concrete in the base? I'm sure balancing has gotten more intelligent, but there still needs to be some mass to keep the thing from walking.

Of course if that 10kg of steel was on the drum, the stationary mass would need to be increased as well. But I doubt washing machine drums are really a high failure part, so a hypothetical "10kg of steel" is a useless in the context of a washing machine.

It's hard to make a comprehensive argument about entire machines when the problem is designers having a principle agent problem for every single part.

Having worked with their racing teams in the past, you are correct. A good example is the clutch. Most clutch pads made these days, and likely for the last 4 or so decades, are made out of composite materials like ceramics, carbon-fiber, and even paper. These clutch pads have higher heat tolerances before warping occurs, yes, but they wear out much faster under daily driving conditions. Brass clutch pads will last beyond the next 2 ice ages, but do warp at very high heat/pressures, as one may experience when first learning a stick (but then decided to continue grinding away for the next 15 years anyways). They also do not design such that it will only last 10 years, they design such that it will last until some statistical deviation longer than the warranty for that part/car (depending on the contract corporate pushes out). Example: The manual transmission has a warranty for 3 years/50,000 miles. The clutch pads are then designed to wear out at 55,000 miles/3.25 years, or whatever the algorithms says will produce the most money.

A clutch is a wear item that will not be covered under warranty, regardless.

That said, my BMW is at 147k miles without a clutch replacement, and it's lived a pretty hard life (track events, etc). In fact, in 20 years of driving, I've never replaced a clutch in a car I owned, and I've never owned a car with less than 50k miles (well, okay, I just bought a brand new car last year, so of course its clutch is still going strong).

I don't think cars are a good example. Cars last much longer and with far fewer significant problems than they used to. A car made in the 1970s would maybe last 100,000 miles, if it didn't rust out first. And it would likely have one or more significant problems in that time (transmission failure, major engine problems, etc).

A new, mass-market quality (e.g. Honda) car made in the last decade will easily go several 100K miles if given basic care, and will probably not have any major problems or be showing much if any rust in that time.

Eh, when you consider the drive-train of your average 70s American car there's nothing preventing it from going 300k. The same small-block v8s and 3spd autos were used through the 90s with very good reliability. It's the little crap going wrong everywhere else that makes people upgrade.

The other point is that you don't want to over-engineer any single part. Getting a 10,000 hour lifespan out of the motor is useless if the chuck cracks in 2,000 hours and the thing gets thrown away. Ideally you'd have the entire thing fall apart at once.

But, of course, failures are statistical. Worsening the quality of the most over-engineered part is usually beneficial, since in most cases you're paying for nothing, but occasionally that'll still be the part that gives out. So you come out ahead by making a more cost-effective product even while failure rates rise.

Of course, that's where the article's point about insufficient competition comes in... Standardizing quality throughout the device is sensible, but if those savings aren't passed to the consumer then they're losing money.

That lifetime perspective is really on point. I recently bought some used barber supplies from a guy who's mom is in the styling business. There's been a shift for them to go cordless/battery powered for convenience and I wound up buying some trimmers and clippers from him for a fraction of the cost. The lifetime these tools were designed at far exceeds the lifetime of my usage with my family and stuff; they'll go the distance for sure. I've since come to learn that this little market does appear to have pretty serviceable parts as you can see with a lot of exploded diagrams they put out https://www.proproductsandmore.com/images/andis_sl2_trimmer.... I'm trying to remember what the lifetime quote to me was, something like these clippers were meant to cut 20 heads a day, day after day for some number of years. I realized at the rate I was going they'd be heirlooms to pass on to my son and stuff.

If products are failing before their owners were "done" with them, then the analysis you describe is inferior to whatever design techniques preceded it, regardless of how sophisticated it is, because, well, products are failing before their owners are done with them.

You can explain this as "planned obsolescence" or poor engineering (or a poor understanding of the market's needs), but not both, really. I think what you say about consumers often wanting a "disposable" version of a product is true for a lot of products but by no means every product. This is all also tied in with the cycle of trendy new electronic gadgets, which seems to drive consumers' desire to replace perfectly good products they already own.

Reply on news.ycombinator.com