A database of broken things to identify common failure modes and how to fix them | Hacker News Reader

A database of broken things to identify common failure modes and how to fix them

failscout.co

465 pointshubraumhugo86 comments

We never hear about broken and worn-out products. Pretty much all gear nowadays is baseline ok, it’s the negatives that really set things apart.

For once, let's turn it all upside down:

We should build a collection about how things break - review broken and worn-out products to teach how to identify cheap products. That's why I built failscout.co

It's simple: You upload your broken products and quickly describe how long you owned them, how often you used them, and where they failed.

Everything breaks eventually, but when it does, can you easily repair/fix it? That's why users can suggest a fix to a broken/inconvenienced product.

What could we do with all this data?

- Identifying the common failure modes of product

- Collect fixes for common product failures

- See if a product's quality has changed or gone down at some point

- Add a simple JSON API so other sites and projects can leverage our data.

What a great idea!

Love this idea. I could imagine this being a great way to alert people to product recalls, or even start class action lawsuits in extremely serious cases.

Absolutely love this! Looking forward to using it.

Throwing in my kudos. This is such a needed kind of site. What is your business model? If you can't spill the beans due to seeking funding, I understand completely.

> - Identifying the common failure modes of product - Collect fixes for common product failures

Are these features basically like iFixit for non-electronics?

> We never hear about broken and worn-out products.

I disagree. When I read reviews (mostly on Amazon) I click on the 1 stars and read those first. Those will mostly all be about how crappy the products are.

That's a different aspect - people write reviews about products that are crappy immediately; however, this is about products which are fine initially, but get broken or worn out later, possibly years later.

And how many of those are competitor's bots trying to devalue the other party's goods?

How does failscout solve that?

Why is it failscouts domain to solve? That should squarely be laid at the feet AMZN, not some 3rd party. Seems like such a strange twist of logic to be a question like this.

It's why reviews, in particular AMZN's, are just not worth the trouble. Too many fake ones, so people have come up with "clever" rules for themselves to only read the negative ones. Somehow, they believe the good ones are gamed, yet the negative ones are not? Another strange twist in logic. Once you believe the system is gameable, then the whole system is suspect.

You got me wrong. Failscout is a reviews platform. How does it solve the "competitor sabotage" problem on its own platform? I never implied failscout should fix Amazon's.

Apologies, yes, I did read that like a suggestion of 3rd party reviews fixing AMZN reviews.

To continue the thread but being on the same page, I still think reviews are not something that humans can handle correctly. The people looking to game the reviews will always try harder than the people trying to stop them. I just don't trust online reviews from random internet people. I might ask people I know personaly and am familiar with and would be willing to trust their assessments. Randos on the internet just can't prove anything, because of course a bot could say/do everything a person would try to prove they are not a bot.

For those that are willing to put faith in reviews, have fun. Don't let my pessimism stop you.

| I just don't trust online reviews from random internet people.

What we need is to be able to give extra weigth to reviews we can (somewhat) trust. This means that we need to start keeping "files" on each other, and we need a tool for looking up authors of messages and traversing chains of trust ... Right?

But first filter out all the legion irrelevant 1-star reviews ("The courier left the box in the rain", "I couldn't return it because I waited too long to unpack it because it was a surprise present - SO MAD!!!!", etc).

Thank you for building failscout.

I had a similar web-app-idea after my the computer on my Kenmore dishwasher broke after 13 months of use.

My goal was to identify products that have a catastrophic failure after an unreasonably short amount of time / pressure the manufacturers into improving quality control.

Tangent:

I also had an idea for people to log instances of items stolen from their luggage with the goal of identifying airports where this is frequent.

I had this idea after flying into Paris and finding my Leatherman multi-tool gone. After some research, I suspect that it wasn't actually stolen, but legally confiscated because it had a knife and France has laws against folding knives.

^ This is the kernel.

Consumers see some failures. Service-people see all failures, over a statistically-valid popupation size.

Be Glassdoor for repair service-professionals. And get the legal shielding / anonymity right.

It consistently amazes me how much undocumented knowledge those folks have. "These model Kenmore switches go bad all the time." "The CV joints on these years leak." If you want the real story, find the person repairing the thing all day, every day.

Appliance manufacturers in the US have done a lot to combat this in recent years.

They bought up many of the US brands, and rotate which ones are garbage so that they can continuously trap unwary customers.

Here’s a handy table, to help you avoid the manufacturer of whatever piece of junk you just sent to the landfill:

http://www.appliance411.com/purchase/make.shtml

I just assume the oval "made in america" sticker with the US flag on it implies the item is from that cartel, and is therefore garbage. (Many made in America appliances do not have it!) After that, I double check the list.

I'm a fan of the exit review. Something that is no longer being used (for any reason) and there is now perspective. I first saw bunnie do it at https://www.bunniestudios.com/blog/?p=242 quoting the first two paragraphs:

> I think it’s time to start a new kind of gadget review: the exit review.

> Gadgets always seem to arrive on the scene with a lot of splash and hype, but rarely do you find an article telling you how the gadget fared in Real Life. The Exit Review is something I’m going to try doing every time I retire a major gadget of mine; the idea of it is to reflect upon how the gadget performed over its duration of service. Of course, reviews like this are all hindsight, so they don’t drive sales — which probably explains why nobody does them, because there’s no money to be made doing them. However, as a design engineer I think there are lessons to be learned through reflection, and as a consumer I believe that apples don’t fall to far from the tree — a good gadget maker will get my business again, and a bad one will never see another dime from me.

i love the idea of the site but i would like to see one change. Instead of adding a product when it breaks, let someone add it when they buy it then send a yearly reminder of the products they have added and if they want to update their status. this would have the benefit of: -tracking % of product failures -finding out if they function longer then they are needed. -finding out which products outlast others before they get to the point failure -potentially notify/warn users of potential issues that come up to before theirs breaks

I think a lot of people dont care enough to upload info just to help product designers, some will do it to help out other consumers, but if you can provide more value to the person with the product you want info on that should get you the most engagement.

I'll definitely add such a feature soon. Even if things don’t break - waiting a minimum of 6+ months or 50—100 minimum uses really makes a review relevant. Recurring reviews could be a good way to track the whole lifecycle of a product.

After every year, the reviewer will receive an automatic reminder like "hey there, did anything break or deteriorate?"

Letting the user set the frequency of reminders could make sense since it will vary between product categories. In addition, if the user has nothing to add, a simple click on a button "Condition unchanged" should be enough.

I like this idea, with the refresher email (bonus points for the interval and options being tuned per product category) starting by asking

- still using, works

- no longer need (did its job), works (could still do the job)

- no longer need (did its job), spent

- still need, replaced (broken or worn/torn)

- didn't end up using it much, works (?)

and prioritizing contact especially around the issue spikes in the timeline for said product

There is an issue that a lot of cheapy electronics aren't even still available after one year. :(

As someone who repairs electronics, who has 2 mechanics as neighbors (farm/small engine and car/truck), I get the sense that I could spend a lot of time on a site like this, but I'm not sure that most things I'd speak to are even worth other people reading.

For instance, sometime after 2010 nearly every small engine switched to plastic parts inside, with some brands and models cutting so many corners that there's no bearings on the main shaft at all.

Short of "find an old one and fix it up" there isn't much you can do to prevent this, or even find an alternate brand. Your choice becomes: garbage, refurbish, commercial; with the price being logarithmic across. Commercial stuff may last a while or be easier to repair for commercial entities, but being 10 times as expensive at minimum really hampers my enthusiasm.

I lost count of how many extension cords, axes, rakes, trash cans, hoses, sprinklers, outdoor lights and cameras, etc etc I have rusting or trashed for breaking at this point.

I'm finding the same with power tools. Being a part-time power tool service tech has really burst my bubble about any new tech in tools, plus it's getting harder and harder to get parts for old ones.

I was shopping for an impact wrench yesterday and came across one that was in my budget and was a brand I know I can get replacement parts for. It looked kind of familiar and when I started reading the reviews I remembered why: several people mentioned that the model has a design flaw that makes it overheat and melts the armature fan to the plastic case. A customer brought one in a few years ago and if it hadn't been covered by warranty it would've cost almost as much as a new one to replace the cooked armature. Even at 4-5x my budget the tools are only slightly more dependable, and they've really transitioned from being tools to unrepairable (epoxy-potted) electronic devices.

Recently had an open neutral event that took out or hampered a handful of electronics in my home. Surge protectors largely did their job, and I learned quite a bit about protection and repair in the process. Luckily I was able to find a replacement power/fuse box for my on demand water heater pretty cheaply, but also fixing it would have been doable (including a temporary workaround).

I also recently (finally) refurbished a vintage fender amp (new plate resistors, new electrolytic capacitors, updated power wiring to be safer and have a standby switch, fixed a factory error with swapping a filter resistor position from the schematic (not sure how much of an impact this had as I haven’t done the full circuit analysis). It’s sounding great now, and I’m ecstatic! Being able to fix things is really gratifying.

I’ve long been interested in the premise of “lifetime reviews” like if I graphed satisfaction/happiness with a particular product vs the duration of ownership. Some things start out high, and rapidly fall off, others start middle and gradually glow as you learn to appreciate their choices or robustness. New fridge is great, but then the ice maker makes a puddle in the container and it goes rapidly down hill. So along with the other suggestions for ongoing updates, something like this where I could periodically track various (or a single) metric over the life of an object would be helpful. Aggregating by brand and years of purchase could help see the reliability trend for given appliances/similar.

Incorporating more repair details similar to ifixit seems helpful.

I’m not sure I see the business model, so maybe it’s open source or Patreon-sequence, or you’ve thought of something I haven’t. Best of luck!

I grew up fixing everything -- darning holes, repairing appliances, fixing furniture etc. It was super annoying as a kid but it seemed we had no choice. It's not like the neighbours could afford to do otherwise either.

As an adult I appreciate it -- often it is easier to fix something than to replace it. But yesterday I chucked out a laundry sorter because the repair cost was higher than simply getting a new one. I found it difficult to do.

Edit: I repaired a sentence with an improper verb tense.

Good feedback. I think it's worth considering that most people aren't able to properly diagnose or troubleshoot something that's broken; especially electronics or mechanical things.

I'll try to add questions that will prompt users to reconsider their submission and give readers an idea if this is a legit failure.

Questions could include:

- Did the item break during or after the warranty period?

- Have you contacted the manufacturer for help?

- Would this problem prevent you from purchasing something else from this brand?

- Have you searched for reports of this issue online? What condition was the item in prior to this particular issue?

> I think it's worth considering that most people aren't able to properly diagnose or troubleshoot something that's broken; especially electronics or mechanical things.

This is quite sad to consider. For a device like a telephone or laptop it’s quite understandable: adding the affordance for many (though not all) repairs would add cost and decrease reliability.

But there’s a kind of learned helplessness in not being able to dismantle and consider something manufactured. I’m a backpacker and the lines between make/modify/improvise/repair are often hard to find. And I think it’s also a kind of stance: the same self confidence and debugging perspective are required to fix a tent and figure out who should be president.

I fix some stuff even if it’s more expensive than buying new one. It’s nice to keep old things going, not to produce waste and often times the new thing is cheap because it’s not ’made the way they used to make them’

Love this concept, I hate bait and switch meaning you buy the thing and it has quality parts then then it gets rave reviews, everyone buys it and give it a solid reputation then they gut it and start cutting corners, making parts plastic to save costs while they milk it for every penny. I’m looking at you Craftsman, Wahl clippers too even Braun stuff ughh

You can blame Walmart and later Amazon for this. The lowest-price-at-all-costs is why I try not to shop at either of those places.

Maybe a feature that, given a picture of a serial number, the revision can be decoded.

Sometimes it's "conscious" on the part of the brand, they deliberately release a substandard v2. But it can also happen organically.

There's a book called "Poorly made in China" [*] which explains a common dynamic here which I found surprising.

Major brands and retailers try to squeeze every last penny out of the outsourced manufacturers. The manufacturers often respond with "quality fade", cost optimising the product without the brand's consent. There is an element of "cat and mouse" to this. And this can happen at every step in the supply chain! So the outsourced manufacturer can also suffer quality fade from their suppliers via a similar dynamic.

Often the brand doesn't have that much insight or input into the product itself because they no longer "make" it. They got it made by giving a Chinese factory an item to clone, or by picking an item from a catalog of samples and suggesting customisations and branding.

If the brand "falls out" with the factory they risk having nothing to sell. Their main leverage is being able to switch suppliers. But when they do this the item doesn't come out the same anymore because the new factory has different tooling or processes.

[*] This is not to say that China can not make excellent products. They can, and they often do! But the dynamics of everybody trying to squeeze the last penny out of everyone else can lead to substandard results.

I absolutely love this! I've been trying to get more into re-use of my existing possessions, and this is helpful for fixing whatever breaks.

I think it would be useful for answers/fixes to have a permanent vs temporary classification, because there are some problems where you can it permanently at a higher input or do a band-aid fix you re-apply later but that takes 1/100th the time. For example, I had a Razer Deathadder mouse where the scroll wheel would double-scroll and continue scrolling after I stopped scrolling. I temporarily fixed it by shooting some compressed air in the scroll wheel's housing, but this had to be repeated maybe 6-10mo later. I could have disassembled my mouse, maybe done something to the PCB or switches inside, and permanently fixed it but that's a lot more work and more outside materials.

Adding this to my bookmark bar, looking forward to consuming and contributing in the future :)

If electronics doesn't work just swap out electrolytic capacitors. I fixed few things (monitors) without knowing what's wrong just by doing that.

Me too. For those interested, bad capacitors can be identified by their rounded top. Replace with a new one with the same voltage and capacity rating as the failed one. Prefer parts with a brand and data sheet over no-name generic ones. The latter may not last as long as the original component.

I have an alternate way of figuring out whether or not products are of good quality and long lasting—-buying used. You can generally tell what is high quality and long lasting because older products that still work are both listed there and priced relatively expensively.

Absolutely love this! Added three products that broke on me recently (two of which I was able to fix).

Feature request; flag bad/nonsense posts. e.g. https://www.failscout.co/details/620aad6e8833e70009e1524f

So theres an entry in here that says "Windows: used for 20 years, daily and everything is broken with it".

This is very much a 90s and aughts trope that's long been dead. I dont think I've had a BSOD in 6+ years now...

BSODs are rare now, but only a few hours ago I got an "update" that broke the start button. Clicking or pressing "win" does nothing now.

I'm sure it's fixable, but it's not a reliable product by any stretch of imagination.

The Open Repair Alliance publishes a repair data standard and datasets for downloading as well as insights into the data. https://openrepair.org/open-data/downloads/

From time to time, citizen sciencey "quests" are held to enrich the data and engage people with repair issues.

The data is used for lobbying for the Right to Repair in Europe.

I expected some kind of SRE platform. An image of a shoe was not what I expected :P

Those boots are failing in a really crappy way, are they counterfeit or something?

No, that's normal for 'better' soles made of PU:

https://stories.hanwag.com/en/hiking-boots-sole-coming-off-w...

I had the exact same style of Timberland Pro "Pit Boss" model boots (Brown instead of Tan) fail in the exact same way. Just threw them away last week! I my case, the boots were bought over 10 years ago and used weekly.

I replaced with them with the exact same model again, because the price was half the competition, they fit me perfectly, they were comfortable to wear, and I thought 10 years service keeping me safe while mowing the lawn on weekends was more than adequate for the money expended. Maybe industrial users should consider something other than the cheapest option available.

In fact, it’s a good example of how design solves something. An upfront photo does a much better job than showing you a searchbox (after all, it’s a database so it should start with a search box, shouldn’t it?) and letting you discover that there are no stacktraces in this workshop.

Here's an existing source of analysis of the quality of a lot of mostly very cheap products, some of which proved to be actually dangerous:

https://www.youtube.com/user/bigclivedotcom/videos

Watch enough of the videos and you'll start to see some common problems. Keeping the cost low is what usually results in poor construction quality. Failing to add protection for Li-Ion batteries. AC powered equipment that is dangerously wired. Poor design. Stupid design. Stuff that just could not possibly work in the first place (car ozone generators with just an LED in them and nothing else.) Stuff that emits toxins (over powered ozone generators). Stuff not designed to be repaired (glued together). And teardowns of equipment that is well designed and built also. Part of identifying common failures is identifying common successes as well so people know what to look for (good and bad) when buying stuff.

Brilliant. Like a StackExchange for mending ordinary stuff. Run with this! Keep it open API and user curated. I often fix things around the house and see an obvious design flaw, and wonder if others might benefit from knowing a fix that works, like "successfully replaced sheared rivet with 2BA bolt/nut" or "PVA glue will stick this back on fine".

It appears that submitted data is not available on some open license, right?

That's definitely the plan. Since I'm not very familiar with licensing, what would you suggest? Creative Commons?

I would suggest CC0 for both images and database - crediting authors of images and so on seems hassle that noone would do anyway, would impact just people strongly caring about copyright.

If less permissive license is preferred and you want attribution requirements etc - then CC-BY-SA on some specific version for images? And ODBL for database?

Note that you need explicit agreement from contributors to have things on specific license, so it is a good idea to resolve licensing quickly.

Warning: I am not a lawyer

I don’t think attribution is a bad idea here, since it would lead people back to the site.

Were it CC0, anyone could just put up a clone of the site content under a different license to steal traffic, possibly even for nefarious purposes (e.g. maybe a pissed off manufacturer wants to take over via SEO shenanigans)

I love this. But what I'm most interested in is the low expectations that I clearly have about the things I buy. I mean - I see flip flop and the comment says "..after 5 years" and I'm thinking "WHAT, five years? I'd be delighted if my flipflop lasted 6 months..."

Clearly I need to re-assess my whole way of thinking about modern products...

If you're in Europe, I can vouch for The Sole Workshop[1]. In their words, the warranty covers you, "as long as the sole has tread and the leather has not worn through. They are resoleable, and the maker, Bill, will carve the support structure to fit your exact foot measurements. They're comfy, and you can run in them, if that's your thing.

The tradeoff is, of course, money.

[1]https://thesoleworkshop.com/collections/handcrafted-original...

Major problems:

1) Botting/Troll Farms - existing review systems suffer from sophisticated and persistent false positive and negative reviews. You may have solved one half of this issue, but it's not hard to buy your competitors product, damage it, and submit a negative review. 2) Corporate Cycling - even presuming somehow you only have honest reviews, if your product is subject to frequent refreshes it is too easy for a company to say "oh we fixed that in latest model". 3) Statistical significance - again presuming you don't have the first two problems, how do you know whether a failure mode is common or not? Everything breaks - its about whether or not it breaks 1 in 100 or 1 in 10, whether it lasts for 5 years or 20. I'm not sure subjective "oh this broke for me" tells you that it is common. I think you also need the instances where nothing broke to calculate true failure rate.

Awesome .. I had pretty much the same idea a few years ago .. the name I came up with was .. brokeipedia ;-)

I love the idea but the crowdsource quality seems perhaps problematic:

> The cables would repeatedly break, and had to be replaced every few months. Eventually the headphones stopped working in one ear.

At what point does self-repair get tagged as the culprit?

Anyway, I know this is a bit critical, presenting it in a problem-solving way.

Are you saying that the headphones broke because the customer did a bad soldering job when replacing the cables? Do you know for certain that the cables aren’t actually replaceable without soldering?

Exactly

Front loader washing machines seem to be nearly unrepairable when the main bearing goes out.

We were quoted $400 labor plus around $800 for a complete new drum assembly when the bearing went out in our 7 year old machine. When asked if we could replace just the bearing, they said labor would be around $600 and $75 for the bearing, but they wouldn't warranty the work unless we replaced the entire assembly.

In theory, I could have replaced the bearing on my own, but getting help to move it to the garage, spending a day or two disassembling the entire machine to replace the bearing and then hoping I could get it all put back together was a much bigger repair than I was willing to do on my own.

I ended up paying $900 to replace the whole machine with a newer, more energy efficient model.

What's your hourly rate by the way? I am wondering if you actually lost money while inquiring about these repairs. If you make 100$ an hour, it's just 9 hours of your work to buy a new one. Basically one working day + a smidge.

That calculation only makes sense if you can actually find 9 hours of work that will pay you in money. If you are a salaried employee your time is not always exchangeable for money quite so easily.

In my career as an engineer I was never able to work extra time and get paid for it in any other currency than time off in lieu.

It took less than an hour to schedule the service call, less time than it took to choose and order a replacement. (And that took 2 visits because the first one had some damage)

I would have technically have lost money if I spent a weekend doing the repair myself.

But if I did the work myself, I would have considered it as entertainment, not lost time. I spent half a day replacing a pump in my dishwasher, and enjoyed doing the work.

This is fantastic. Some random suggestions:

- Searching for a product should be the primary action on the home page, not hidden behind a "browse" button.

- It's weird that the highlight example front and center of your home page isn't even a real listing.

- I'm not sure what the relation between the website and subreddit is. If someone posts something on Reddit, you probably don't have the right to republish it on your site.

- Probably too late for this feedback, but if the purpose of this site is to build a repair catalog then associating it with the word "fail" is probably too negative. You are simply inviting people to complain about crappy things they bought.

Yeah, per my comment below if the intent is to document failures I think this project is a failure - but documenting known fixes is much more useful.

I've been wanting something like this for a long time. Our society talks way too much about success, growth, and positivity. Even reading reviews failure patterns are hard to spot, they get buried by reviews of day old products.

The power controllers for US refrigerator interior lights are designed to fail shortly after the warranty runs out. It costs more than $100 for a replacement.

(What, the light in the fridge needs a power controller? Evidently.) It has two surface-mounted resistors that are too small, under-rated, and burn out. They can be replaced with higher-wattage resistors, which sometimes works.

Of course the better solution would be to toss out the power controller, and jump the input voltage straight to the light. But connectors, voltages, yada yada.

What?

I'm pretty sure the light in my forty year old European fridge has never been replaced and is a perfectly ordinary 230 V incandescent bulb. Even if it were replaced with a LED why would it need any extra electronics?

Note I said "US refrigerator".

But with LED lighting, you need a converter from mains voltage to low DC voltage. A little bitty transformer and a pair of diodes, or even a low-duty-cycle oscillator driving a switching transistor, diode, and capacitor would suffice, but there is a lot more junk on the circuit board in US refrigerators.

I would like a search that show products that lasted 5 years, 7 years, 10 years or something like that.

For most things that is a good endossement if it lasts that long.

I wonder if you could scrape some of these price matching sites, or even integrate with them. That way you could get a lot of product names and run a whitelist of products in stead of relying on people adding 100% correct and unique names.

Would love to see common failure modes before deciding to buy something. Of course this could become an attack vector for the competitors.

Reminded me of Bad Designs: http://baddesigns.com/

Brilliant!

Nit: it's not clear where to put repair procedures if you're reporting something you've already fixed.

> We take your privacy very seriously

I was about to roll my eyes since this is exactly the kind of title used by people who do not care about privacy. Maybe I'm just too jaded but it might be better to leave that title out (the paragraph below it says enough) or at least reword it.

I love this! However, I wonder what your growth/sustainability plan is. One of the first things in a 2 sided information marketplace is to nail the growth and retention model.

Curious to get your ideas on how to solve this :)

How to solve growth? I think it depends on your company, team skills, and space. But you should be thinking about it, otherwise your bucket will be leaky - and 2 sided marketplaces are already really difficult.

Interesting and useful, but I'd also like to know what products has worked exceptionally well. Carrot & stick, you know

So sort of a Consumer Reports for social media?

Sounds like a great idea. Are you planning to have a rss feed at all?

Should also have some way to filter via country/ location.

The "printer" section is going to be a hoot...

I love this!

Me too. Till it gets traction and then corruption.

I hope this is indexable by Google/Duck so it can be found when needed...

Reply on news.ycombinator.com