The other one noted if you don't specify the density of plastic for bags, or paper for bags and packing, you get clingfilm thinner than you thought existed, and paper which is almost tissue in its weakness. You don't even get boxes to put the boxes in, if you don't specify boxes to be delivered in boxes. So now wrapping a pallet becomes a nightmare if they don't stack. And if you don't specify how many to stack, and how to pad the stack, they won't do unit height stacking if it costs labour time. Your risk.
Some of this like the casting mistake, or the knob thing, could happen anywhere and you have to be close to final manufacture spec to find out e.g. the metal coating impinges on the knob at the free space you specified, because your test rig didn't have powder coating. Or, that a design feature you need like the light entry holes, is used by the casting engineer as pour points because it looked like you'd specified mould pour points not functional holes.
But other things like "yea, you didn't spec how long to make the tails so we cut the tails as close as we could" is just the cheapening above: if you don't SAY its a 10cm tail for the connector, it will be 2cm, if saving 8cm of cable saves money for them.
I've read some stuff which says the cost of 5 SBC boards with pre-applied SMD is now so low, you might as well order 5 so you get at least 1 which works. That means they will wind up working out your tolerance for failure, and produce goods to meet that: if 1 in 5 is viable, thats what they'll target.
They'll make this amazing Remote Control Car, with good suspension, a battery that lasts half an hour, plenty of power, and just all around amazing. But then it'll break after a day because somebody saved 1/20th of a penny by speccing this impossibly thin wire the thickness of a human hair to hook that powerful battery to the powerful motor and inside the remote.
They could have used actual wire-sized wire and had the most amazing product ever, for roughly zero more cost. (Possibly less, since surely it must cost _more_ to manufacture and solder micron-diameter wiring). It just makes no sense.
I ordered cups and did specify the thickness (based on a reference) of the plastic but didn’t specify how thick the boxes they shipped in should be. Guess what happened!
Now go with Kyosho or Tamiya and you DO know it will be best in class, but at 10x the cost.
There are professional communication/training courses for working with Chinese vendors/colleagues that spell all of this out, because it's not some secret. It's just a very different culture, with high context communication (I'll let you read what the practical implications of that are elsewhere). Want to have your mind blown? Look up what it means when they say "yes", when you're explaining something.
Being a low context person, I have significant and severe communication problems when working with Chinese colleagues/vendors.
It was mostly the same as anywhere else, you go talk to them in person, tour their facilities/processes, and see what else they've built.
I was warned strongly about IP theft and cost cutting, but didn't find that expectation quite met reality. It may have been that our products were mostly un-copyable, and we specified everything precisely, or were just lucky.
Personally, I never really had too many issues sourcing from China because I made sure I was always introduced to a reliable partner first.
And secondly, I told them when deciding on two options, choose the better quality option, regardless of price.
Basically, I didn't tell them to save us much money as possible if that made all the difference.
Knowing which is being spoken or heard is going to be hard.
Is there a term for this? Because I see it in my personal life as well dealing with some low price manual labor that doesn't speak english.
Instructions often get lost in translation, the reply will be "yes" and it doesn't get done. I know they want to sound professional and confident, so saying no or asking questions is a "bad thing".
I've got suppliers who I can send a difficult part to and know that I'm going to get exactly what I expect, faster and cheaper than just about anyone else. It took a few years to get to that point, but these few vendors make it really hard to go with anyone else, much to the chagrin of the sourcing team who rightly recognize it a risk to rely on just a few suppliers.
Once you get to a certain type of supplier you end up running into the problem where their processes are such that they won't do anything without you clearly documenting it. They simply refuse to make any assumptions on your behalf. They can be so frustrating when you are used to the other way of doing it. I simply cannot answer some questions because I'm so used to my other suppliers just doing it correctly and haven't ever asked about it.
I've had a 90% failure rate on what was supposed to be the final prototype before production. Turns out they hand-soldered the batch because the proto run was (obviously) only a dozen units - and some parts were just too tricky to reliably hand-solder.
I understand the logic as fully-automated assembly has a nontrivial startup cost, but a big reason of doing later prototypes is evaluating the manufacturing process as well. If the assembly method used doesn't match what I can expect in production runs, what's the point?
Weirdly enough the batch before this was totally fine. In the end we did get a massive discount on the hand-assembled run and managed to do all the testing with the one prototype we got working with some small rework, but it still cost us quite a lot of time and money. We would've happily paid a significantly higher fee to have them just do it properly - per-prototype cost is pretty much irrelevant during development.
The objects were order of magnitude simpler than in the post (no electronics and no plastic, only metal) so maybe that doesn't compare, but I never had any bad surprise from any supplier, including packaging (which can be quite complex and involve several providers), etc.
Everyone will gladly send you samples (for free!) and prototypes of what you imagine (usually at cost) and if you're explicit about what you want and validate each step before the next, everything goes well.
Eventually I moved on to other things for mostly bureaucratic reasons; selling objects in Europe is an administrative nightmare that's simply not worth the hassle.
But the manufacturing part was not just smooth -- it was the best part of the experience.
(And I never left my town and never even talked to anyone over the phone: the primary means of communication was email.)
Edit: why would anyone downvote this, and so fast? If anyone thinks I'm being insincere, I have proof! ;-)
HN loves downvoting. It's Nerd Reddit. Don't sweat it.
That is very rational. Each 9 in uptime or quality represents expense. The expense of moving to the next level up can't always be "shift left", but instead done at the point in the process where money can be applied.
Lets say you have a process that goes right at a minimum of 20% of the time for cost of 100. The manufacturer can add QA that makes the cost 120. Is it better to trust the manufacturer at the cost of +20? Or is it better to do your own QA for 20 and gain any correct pieces above the 20%?
I think it was one of the many threads off "Bunnie Huang's Essential Guide to Electronics in Shenzhen" because the specific incident I can't find.
Depends on the shop. The one I use for prototyping has been around for at least 15 years with a good reputation.
Years ago, there was an HN article "You Need More Lumens"[1], which in turn led me down a rabbit hole.
I ended up purchasing:
4 standard table lamps from Target,
28 2000-lumen Cree LEDs bulbs[2] and,
4 7-way splitters[3].
[1] https://news.ycombinator.com/item?id=10957614
bigclivedotcom video: https://www.youtube.com/watch?v=ISTB0ThzhOY
The absolute cheapest lumens per dollar COB would be the GVM SD300D, although I highly question the reliability and light quality.
I never got around to putting them on a dusk-to-dawn timer, so they've been burning 24/7 since I purchased them at the end of 2020 (except for the occasional power outage, of course). I paid $20/each for them.
Sample size of 1 (technically 2), but there are definitely products on the market that meet your criteria.
Interestingly, 4 of the 6 that I had running all died in the same ~3mo period, but still I was pretty happy for 4 years of use for $25/ea.
https://www.costco.com/p/-/enbrighten-ultrabrite-hex-lights/...
Getting just lumens is cheap. Getting a full spectrum of light is where costs increase.
This is the reason I chose to go with the specific Cree bulbs (linked in original post) that get a 90+ CRI rating
In my office I have 6 of these, for a total of around 13,000 lumens. It effectively 6x'd my light output for around $150. Worked wonders, especially in the PNW winter.
It rains only once ... but for six months :)
Besides not wanting to waste the money, I doubt the lamp will last 5 years (not 5 years of projected use of XX minutes per day…). 580W converted to heat on a small disk will take its toll.
150lm/w would make it at least a cut above domestic lightbulbs.
200lm/w would make it a premium product.
I do think it's actually quite hard to beat the Brighter lamp on all of: Lumens, $, QoL (ie: Google Home integration + temp control), Form Factor (ie: not looking ugly), CRI.
I personally noticed issues w/ CRI & Form Factor quite a lot with my previous options.
I would get/build such a thing for my mental health, but I worry the LED illumination will be counter-productive.
The only thing to watch out for is that the lamp base you're using can support the high wattage.
Newer LED phosphors are typically 90+ CRI, and I commonly find 93 CRI bulbs available off the shelf.
Personally if I wanted "daylight" replicated by LEDs I would go for a higher quality white grow light that included deep red LEDs. Just be sure you don't get one that is also outputting in the UV range, although most don't.
> Blink and you’ll get a different measurement.
This means your environment is not controlled enough. Also quality control is usually done in terms of statistics. You might want to read something called gauge R&R. That being said, you should be proud of being able to ship a physical product!
As for quality checks, software quality teams pales in comparison to hardware quality teams. Mainly as you said, there’s a lot checks you can do in software. For hardware, bigger companies have to have their vendors qualified. The vendors have to follow their customer guidelines and do outgoing inspection. Then the company has a division to do incoming inspection. There’s a traveler that follows the kit (of parts) and there’s usually subassembly quality checks. Then final full build checks before it leaves the door.
I think the first thing to focus on is the stats portion - do you have appropriate FAI/SPC/OQC with Cpk requirements defined? Gauge R&R plays a much smaller role, especially in something that is relative
Except those with ears and/or eyes.
It turned out that USA / "the west" lost the engineering knowledge to manufacture "stuff" (in this case injection molding and other procedures): Nowadays, we simply create the plans and schematics (e.g. CAD files) and let the Chinese do the building.
We made our first hardware by hand, i believe we did 15 units. I remember my cofounder broke down because he couldn’t take the pressure of receiving fifteen orders and now we had to make FIFTEEN by hand, lol. But we were able to figure out SO MANY issues before mass production. And of course even then many slipped through
For this specific application, the manufacturing method determines the porosity of the material, and therefore the heat transfer.
CNC prototype parts will have better heat transfer than pressure die cast, and the pressure die cast will perform better than pressureless cast parts.
https://en.wikipedia.org/wiki/Design_for_manufacturability
Some are good at designing to factory capabilities, yet others try something slick and get schooled the hard way. =3
I had a similar experience. I quit my job as a SWE at Google and built a hardware product on top of the Raspberry Pi.[0]
I don't really have anything helpful to add, but I relate a lot to all of the gotchas you encountered. Shipping a hardware product made me appreciate software so much more, especially SaaS products where you can ship a fix immediately.
With hardware, it's painfully easy to make a mistake and not realize it for 3-6 months. And by that point, you have this whole manufacturing pipeline you have to halt and unwind. And with overseas manufacturing, tariffs, and shipping costs, you can't even unwind some of this stuff, and you just end up with useless material that you paid 5-6 figures for.
We're definitely spoiled in the software world with the relative ease of fixing bugs.
I'm sure there will be more challenges as well but as long as you keep focusing on the experience you're delivering, I'm sure you'll continue to get past them
One downside is that the active fan cooling design is questionable - the air goes over the top of the LEDs, and there aren't any dedicated exit holes so the air is just squeezing through the very small gap between the glass and the heatsink. There are also blotches of paint that worsen the TIM contact between the PCB and the heatsink. I used a rotary tool to remove those blotches.
What finishing process, thickness, overspray, masking requirements does your drawing package specify?
What happened after this? the factory have to replace the casting mold at their own expense or you have to pay for it?
This is also why so many crowdfunded projects fail, people go into it with no idea of how hard it is to get something to market and waaaay underestimate the time and cost. Years ago for the first project we did we took an absolute worst-case estimate, then doubled the time and cost on that. We came in on time and under budget, but only just.
Hats off to the author for making it through! What a start to the journey!
Previous experience? Or you know someone? To me that always seems like 90% of the battle with manufacturing.
If you are American, and you are manufacturing 500 units, do it in America. Yes, it's more expensive per-unit, but at 500 units you don't need a tiny per-unit cost.
American manufacturing is more flexible, higher agency, and people tend to adapt to underspecified instructions better. The communication loops are stronger.
That said, America has enormous tradeoffs. If you need 10,000 of a part in 4 hours because something weird happened, China can make it happen. America... might take months. But if your batch size is 500, it's better for an American living in America to cut their teeth on American manufacturing and go to China when you need 10,000+ units per run.
It seems like the design was being changed up to the minute these were shipped to customers, so it doesn't seem possible that any testing was carried out on the final design?
This product is about at the point of DVT in development flow, and therefore would be sent to testing about now. But, instead, being sent to backers.
PS, not a hypothetical circumstance for me. I've previously certified a number of luminaires under UL and CB Scheme. I was the technical chair of ANSI C136.37 for several years, and on the working groups of several other standards.
I was trying to ask in as charitable way as possible...
I could search for it if you want to read about that.
Congrats on shipping. I'm living in the EU working California hours (4pm-1am) and will definitely be buying one.
We had another 6'5" customer who was worried about the height but they said it was totally fine even with shoes on.
Pixels dice have also been going on a “fun” journey with the weight of a huge crowdfunding raise. I’ve been following their updates with sympathy: https://www.kickstarter.com/projects/pixels-dice/pixels-the-...
I have an appreciation for very bright lamps, and the project is neat, but that stuck out to me.
I'm always fascinated by people who both feel comfortable ignoring maybe the single most impactful society-determining apparatus but will also say "no one could have seen that coming", where that is whatever they were unaware of because they chose to check out. I find the stance so fascinating because for myself, it would be impossible to not try and understand why the world is the way it is.
Everything is downstream of politics whether people want to recognize that or not, and choosing to ignore it is, in fact, a political choice.
This term evolved into the modern "idiot" which we are familiar with.
After the extermination of Melos they could credibly say they were less responsible for the actions of the polis.
And had a higher chance of deflecting the inevitable revenge on to the non idiotes Athenians.
It's genuinely baffling to me why business owners pay so little attention to the politics that will directly impact their business.
The entire tariffs thing was incredible obvious to me (I am Australian) and I only check in on US politics for 10 min a couple of times a month, any less and it would be zero.
Trump in 1989 talking to Diane Sawyer: "he would impose a 15% to 20% tariff on Japanese imports".
Trump in 2011 in his book "Time to Get Tough: Making America #1 Again": "I want foreign countries to finally start forking over cash in order to have access to our markets. So here’s the deal: any foreign country shipping goods into the United States pays a 20 percent tax. If they want a piece of the American market, they’re going to pay for it. No more free admission into the biggest show in town — and that especially includes China."
Trump at a rally in Vegas in 2011, referring to China: "Listen, you motherfuckers, we’re going to tax you 25%!"
Trump in 2018: If the Europeans are "not going to treat us fairly... then we're going to tax all those beautiful Mercedes-Benzes that are coming in."
Anyone who didn't think tariffs were coming is a fucking moron.
More than half of them went bankrupt.
One guys kept dumping money into a new gym buildout mere weeks before the months-long lockdowns commenced.
He’s now unimaginably successful at YouTube but at least I’m better at predicting the content of tomorrow’s newspapers.
The thing which was easy to predict is that Trump is going to continue his trade war against China. It is also easy to predict that in a trade war companies who manufacture some product in China and sell it in the USA will suffer.
That prediction is enough for one to stay out of that kind of business. But it is not enough to do trades and profit from it.
If you could predict that Trump is going to announce x tarrifs on y tomorrow at z time that is much more likely to lead to succesfull trades. That is hard to predict.
Whoever was holding aggressive poly market positions on “POTUS poops pants at presser” is a millionaire now. We all know he wears diapers and has massive flatulence, but who would have predicted that specific thing?
https://www.morningbrew.com/stories/2025/04/30/amazon-wont-b...
I'd argue it's the other way around. Politics is downstream of everything else. In other words, it's easier to predict the politics of tomorrow based on the culture today than it is to predict the culture of tomorrow based on the politics of today. I'd go as far as to argue that political details are almost irrelevant except in the most extreme cases where political figures change culture (Constantine or Hitler for example). The current political climate is the result of the cultural climate, and if it wasn't, the people in office would have never been elected in the first place.
National politics doesn't teach you any more about how the world works than the politics of your workplace or your school.
https://polymarket.com/event/trump-imposes-40-blanket-tariff...
They didn't do everything perfectly, but this looks like a normal learning curve for product manufacturing. Textile manufacturing would be even steeper.
Multiple sources spread over an area would be more efficient, as each LED could run at lower current and temperature, which increases their efficacy.
The light would also not have to bounce off ceilings/walls multiple times, while some are nice for diffusion, every reflection is lossy.
I suspect you’d also get better reliability.
Notably:
> When we got back, we watched as the laser engraving technician printed out a copy of our alignment template, deleted all of the guide lines from the file, imported the legends into the laser engraving software, and proceeded to try to eyeball the correct label placement.
> Going in, we were annoyed at how far off the legends were. In retrospect, it’s astonishing how good his manual placement was.
Tooling up a production line for even a toothbrush is well over $1.5m to get the first unit off the line. Building these factories is a different skill set, and everybody is bad at it at first.
Note hardware has a 1:6 success rate compared with service companies.
Best of luck, =3
https://www.harborfreight.com/10000-lumen-4-ft-linkable-diam...
Also, if you've ever been in a Walmart or Forever 21 at night, you'll know that constant LED white light is probably not the best thing for your eyes.
whether or not it's worth it depends on user.
https://myopiainstitute.org/imi-whitepaper/imi-the-role-of-l...
lack of light is generally the leading hypothesis for why there is a myopia epidemic actually. from people being indoors most of the time for school or work.
though unfortunately scientists are still researching if it is a specific frequency of light etc... people are missing
You definitely don't look at it directly.
Never done casting let alone worked with Chinese factory to ship hundreds of units but: this sounds potentially intentional at factory's end. It's plausible that these fins didn't pass factory guy's manufacturability gut DRC who would made changes thinking the customer would just give in. IIUC molding people in general don't like corners and narrow channels with no sprues and/or gas escapes. Especially the outer ring of pins appear to be exactly where they like to place sprues.
I wonder how this problem was eventually solved. The final product seem to retain the number, height, draft angles of the fins, but fillet radii appear to have increased(2mm -> 5mm?) and the entire body shows more material shrinkage at the outer edge of the body as well. Was it just no pins and higher defect rate, or something else entirely?
[1] https://www.pcgamer.com/introducing-gameref-the-anti-cheat-h...
After 20 years of system engineering, I just expected this to always be the case. Until my most recent job with a bunch of startups, where people fly by the seat of their pants, there's no communication, documentation, protection or testing, for anything. I am pissed off daily that things don't go wrong, because people now think this is normal, and it goes against everything I've learned from experience. It seems I stumbled onto the corollary of Murphy's Law: when you expect everything to go wrong, nothing does.
One recent run fun issue I had was a pneumatic timer that worked fine in testing in the shop and outside my house
But once in the field the sun heated up the tube enough to trigger the sensor and get stuck in an on state requiring a plug on one end with a hole big enough to let the pressure out but small enough to let pressure trigger the sensor
I recently changed my car's headlamps to Chinese LED which claims to be about 37kLm and I don't know how much it is probably less than that.
Two of those lamps costee me around $24 on Amazon US (pretty sure under $10 in China).
What makes this $800+ ?
It's possible, they exist, many such LEDs are probably manufactured in China ... but the legit ones are probably more expensive, and you may need a more recognizable brand to do some QA, and keep pressure on the factory to not slip quality or inputs.
Consider the cheap screwdriver included with the lamp in this story: unexpectedly, many were more faulty than the cheapest $4 screwdriver you'd find in any hardware store. The more stories you read about manufacturing stuff in China, the more you'll see very strange things. It's not about nationality or anything, it's an extreme kind of optimization. If you didn't catch it already, maybe you didn't really need what you thought you asked for ... they're just checking/optimizing
I don't know much about car headlights, but chatgpt says high beams are typically 25-45 watts, and assuming a generous 200lm/w that gives you 5000-9000 lm.
Roughly speaking, it's expensive because it's 50 lbs & tons of electrical components (that are much higher quality than $24 headlights).
That's all there is to it. Take a look at https://www.youtube.com/watch?v=5CJqAJ2LXw8&t=852.
Is it the ability to change Hue that makes this expensive?
There are a bit over similarities with the article, like miscommunications occurring or needing to specify in exact terms what is required. But I think it is slightly different now, the stories in the book was from a while ago and in shenzhen instead of zhongsan, so I am sure things have changed, and this article is more up to date.
[0] Despite the title, it is not entirely solely about hacking in the security sense. But also the traditional use of the word "hacker".
I rather like the way it releases concentrated light towards you, as the sun does through a window, but not everyone wants harsh shadows. At the same time, the reflector makes the 20.000 lumens feel like a lot more, as the light doesn't have to illuminate the whole room, just where you're sitting.
- Go for a daily walk outdoors after lunch
- Get some vitamin D supplements
PPAP were developed to help tackle this sort of thing in a somewhat uniform way, but vendor diligence can take many other forms, too.
> That was the worst period of my life; I would go to bed literally shaking with stress. In my opinion, Not Cool!
How you get funding for a hardware startup without cursory research into this is staggering.
There's demand for high intensity lamps (and other hardware projects), a lot fail but some succeed, and lessons are learned. And not just by the people starting these projects, a big part of why these projects start in the first place is because the manufacturers make it accessible too.
I recently ran an experiment where I increased prices by 50%, and sales volume actually went up. It’s wild how much copy and creative assets can alter perceived value, allowing you to justify a higher price point for the exact same physical product
The store is still online so I assume it must be. Let me run this by my wife haha.
And i wish this kind of people to fail miserably. Too many times they tried to make me the scapegoat for their bullshit project, already sold on paper but with nothing more than a render to show, now looking for an hardware guy to point the finger at to blame
And they got the design right, insofar as they could see at that point, hence having 500 units built. In a lot of cases you can't have a single unit built as a one-off.
What stood out to me: the factory miscommunications and quality issues compound because you can't iterate as fast as software. Each mistake costs weeks and thousands of dollars.
For anyone considering hardware: if you're not getting deposits or strong signals of purchase intent before tooling up, you're basically gambling. The author's approach of getting commitments first is the right playbook.
Day 1 of product manufacturing class.
Cost is not an issue because anyone willing to pay $1200 for a god damn floor lamp would surely pay $1500 (and have the satisfaction of keeping your fellow neighbors employed).
Anyone that has ever manufactured anything in China—or India for that matter—knows unless you spec every little detail and dimension and tolerance it WILL be ignored and violated.
Whereas in the US & Canada there is a different level of workmanship and culture where you might get a phone call instead of a box of weldments bent 30° in the wrong direction because the drawing didn't say it couldn't. And you probably could have visited the factories a half dozen times before starting production (did you really not plan on building prototypes?)
Maybe even more would be willing to pay if it had a Made in USA stamp.
Not 100% true. With $1200 it has wider market reach than $1500. There are still people spend $1500 for a floor lamp, but 100% it will be fewer if it was priced at $1200.
Also pricing it too high means higher chance a clone will exist, because they can copy literally 1:1 and price it a little lower.
Another thing in China is they move fast. As long as you have enough money, the time to market is insanely short.
How/where did you find your suppliers/factories?
I believe people like you will become very rich (because you’re so compatible with money making), but I have a prejudice that you lack depth in your understanding and intentions.
I almost never feel jealous but while reading your blog I could feel this feeling of freedom of creating that I seek and I couldn’t help but. Congratulations.
Had I ordered a $1200 lamp and received a $0.1 screwdriver with it, I would be livid.
I guess it goes to show in what kind of an inflationary environment we live in.
Such a wildly privileged take.
Unrelated: why offshore this? Why not end-to-end local development and production?
How much power does one of these consume?
A: 580 Watts from the wall
Checking store ... 1300 ... um ... no.
I can buy a professional LED lighting rig, the stuff used to film movies, for cheaper.
I want to support local manufacturing too, but in many cases the choice isn't there. The Trump dictatorship's tariffs are an attempt to "encourage" this manufacturing back to the US, but redeveloping the factories and supply chains will take years - if there's even people willing to commit to it, because despite high import tariffs, it's still cheaper to import things. And even if it wasn't, it'd still be cheaper and easier to go through a 3rd party and back channels, or to a country not affected as much by the import tariffs. Whole manufacturing lines are / have been moved to e.g. Vietnam or India to avoid the tariffs, and that's still cheaper than moving to the US.