Ask HN: What can you realistically manufacture in your garage?
Given a two-car garage, what can a small team (2-3 people) manufacture that can be sold for some amount of profit? Imagine access to capital of 20-50k USD at maximum.
Interesting would be items whose manufacture could be automatized to some extent, but this is not necessary.
I am not particularly interested in the legality of this at the moment. But safety considerations could be important.
427 comments
[ 6.1 ms ] story [ 297 ms ] threadLathe, CNC Mill, Drill Press, Bandsaw, Bend Brake, CNC laser/plasma cutter. That'd be the basics of a fully featured metal shop. Buying used and upgrading as you bring in some revenue would keep you under your price target.
I'd love to move to a more "robust" process, but options for materials and widespread access to 3d printing provides a lot of versatility for a single-person business where I want to control the entire product and process end-to-end.
Just need to find your niche.
I wanted the product and knew others in the industry would too. Once I was able to get it in the hands of a few operators to see and use, it started selling "organically" through word of mouth.
helps if you have breadth of knowledge and experience in a variety of fields also. interdisciplinary solutions still have plenty of untapped market potential
other than that there is no magic formula that i know of. its kind of one of those "if you have to ask..." type things.
For example, hang out at small-town airports or boat harbors, or with lawyers and doctors. If you can find some small part that would help them use their boat, then $300-500 for it might not even be an extreme price especially if they see you using it first.
(This is not an argument to go buy a boat and plane).
For example, I designed a set of sprockets to drive my blinds, and the total weight of the sprockets to drive 3 sets of blinds is less than 15 grams of plastic. This goose-shaped figurine is about 35g of plastic: https://www.thingiverse.com/thing:3906053
For something that costs me $0.17 at home, Etsy people want to charge upwards of $4 plus shipping and Shapeways or the like are $20+.
Imagine you had a bunch of arcade machines with a proprietary button to turn each on. You can buy the button, but the manufacturer requires their service people physically install and bind it your arcade. Most operators want multiple buttons cause they always break, but they also don't want to deal with downtime during replacement.
I sell the button for 1/3 of the cost with modifications allowing anyone to install it.
For me, reverse engineering it was easy, the most difficult part was/is sourcing electronics with custom specs.
For a guitar, however... possibly having the body made of wood like a real electric guitar could be a selling point.
I always have ideas for little products (some involve electronics others not) I'd like to build, but how to go from raw idea in my head to working assembly, I'm lost on: choosing motors, control board, mechanical reliability, etc. Maybe I just need to read some ME books, but if they are like math and physics texts, there is gulf between the text how to do build something practical.
Recently I had a small project idea that involved a pretty basic PCB. I was surprised at how easy it was to teach myself board layout (using KiCAD) and several related skills. You can get one(ish)-off PCBs manufactured by PCBWay or OSH Park on the cheap. I haven't finished the project yet (supply chain issues), but I'm pretty sure it'll be successful, and fairly cheap, considering.
There are lots of resources online if you need more external help learning. Two I was pointed to (but haven't really dug into too deeply yet) are All About Circuits (https://www.allaboutcircuits.com/) and Ultimate Electronics (https://ultimateelectronicsbook.com/). For the more physical aspects, YouTube has been a great resource -- for example I needed to solder a 5mm x 5mm QFN IC to a breakout board, and didn't want to spend the money on a high-end rework station. A random video on YouTube taught me how to do it with some flux, solder paste, and a cheap hot air gun I bought on Amazon.
Next I'm going to look into getting a 3D printer, and learn how to design stuff for that. I expect the learning curve there will be higher (at least for me), but I don't see it as out of reach.
I'm in the high power rocketry hobby and a 3d printer was a game changer for me. I have a prusa i3 mk3s+ and can't recommend it enough. For software i originally started with onshape but then went to Fusion360 because that's what everyone else in the hobby uses. Learning curve is steep but there's tons and tons of good material to help you learn. My workflow is Fusion360 then PrusaSlicer then save to sdcard then print. You can really optimize further but i don't sell 3d printed parts so it's good for me.
On the other hand, i do have a big arse laser and i stood up a little website and sell a simple cutting service to my buddies in the rocketry hobby. I like my website, i have no database! I actually use some metadata features of the stripe api for my order database hah.
https://oakclifflaser.com
I have since gotten into 3d printing and have printed molds and poured silicone with good luck. Some of that silicone has been the final part, some were then used as molds for pouring resin in. I've even added glass fiber to some of that resin before pouring to make some pretty sturdy parts.
Maybe some of the parts you are 3d printing now could be done with resin? Bondo makes a product with glass fiber in it, but most of the parts I'm doing are fairly small, and the bondo has long fibers in it, when I make my own fibers, I just cut fiberglass and can make it whatever length fiber I want.
I have looked at SLA printing and "bench/desktop" injections molding setups like MicroMolder, but I've ruled those out cause I don't have enough ventilation or appropriate space for handling resin -- I don't actually do this in a garage, but a spare room.
That fiber reinforced Bondo is some pretty nasty stuff in the fume dept. If you really wanted to go that direction you could vent out a window or use some active carbon filtration I presume. But it starts adding up, depending on the quality you need you might also need a pressure/vacuum vessel as well and a pump.
I've managed without, for my micro volume of personal stuff I do.
[0] tube is measured by the outer diameter; pipe is measured by the inner diameter
https://postalpro.usps.com/mailing/competitivepoboxes
Are you in the US? If you're trying to buy those tubes, are you trying to order very long lengths? UPS apparently has a limit of 108" total length, and a sum of 165" for length + "girth".
As for McMaster overall - I agree with some other sentiments here that they're great, though I do admit the "McMaster tax" (paying 10~200% what you could find the identical part somewhere else for) can be annoying at times, but is worth it when you want a reliable supplier with almost universally good quality products.
edit: Based on your reply to a sibling comment, I looked around, and found this thread that seems to reinforce your experience that McMaster shipping to Canada is...unreliable: https://www.chiefdelphi.com/t/mcmaster-carr-supplying-to-int...
I would try the recommendation there of ensuring you have a business name on the order, or, contact their support - I've had to contact them a couple times and they were generally quite helpful.
Good luck.
Hi xxxx,
We only ship to large businesses and schools in Canada. We can't accept your order. I'm sorry for the inconvenience. You might want to try Fastenal or Motion Canada.
Lauren
If that's a problem to them, I'm guessing they don't have a fully automated solution to that like e.g. DigiKey does.
I'm not sure if this is a super variable experience or if it has changed over time (I first ordered from them in 2020).
https://www.aircraftspruce.ca/
I've built a bike frame with their 4130 tubes.
"Due to the cost and complexity of shipping our products to Canada, we are only able to accept orders from businesses and schools. We’ve canceled your order. If this material is not for personal use, please resubmit your order online using the business or school name."
After that, I added my company's legal name to both shipping and billing address, put "Please note: this order is for business" in order comments the first couple times, and had no problems since.
I wish we had something similar in Canada that has a good selection and isn't a pain to deal with, but as it stands, McMaster is too good to give up on.
ETA: Having read your other comment, it appears they might have restricted their policy since then, but grandfathered existing clients. That would be really sad for Canadians.
https://www.torchandfile.com/
https://www.cyclingabout.com/heaviest-touring-bike-ever-behe...
(I met this guy in Cambridge Mass in the 80's).
I remember this from the 80’s too, I was 8 when I met him and it was the coolest thing ever.
Things could still be made smaller if there were more antennae nearby: hence lower transmit power. But I think the smartphone format is pretty dang near perfect.
I think that steer is something easily replaceable. I want to do lowriders - when the lowest point of butt is lower than a line between two axles. This is an incredible fast form-factor for totally good roads. Thanks HN for a discussion.
But at this point carbon fiber is pretty easy to manufacture. For a bike frame the hard part would be design optimization and stress calculations. It might even be easier once you have the molds. I'm not really sure I understand why some upstart carbon bike company hasn't cleaned up given how inexpensively some of the no-name carbon parts from china are (and how they appear to be at least as good if not better than some of the name brand stuff in some cases).
Print shop type things, especially specialized like vinyl cutting and large banners, could be a good business depending on where you're at. Might be able to buy used or lease equipment too.
Woodworking / furniture shop and / or antique furniture restoration might not be terribly capital intensive.
Plenty of people use gas, but I don't see why given the increased risk and cost.
For gin you need to make or buy neutral spirit first. Buying it is ideal as larger providers can make it cheaper and cleaner than you, but needs a license so you won't do this while you're developing a product.
Then you will likely use a pot still to make your final product. You can use the same equipment to do both processes if you have a modular design.
None of the above is set in stone - gin is a bit like jazz and breaking the rules is common.
The problem with the space isn't really the capital but noise abatement, waste disposal, and inventory. The ideal product would be quiet to make, not use tough chemicals to dispose, and materials you can buy in bulk and small enough not to take up a ton of space in the worksite.
Of course it also benefits remote tech workers who have an LLC or whatever. And handymen, et c.
The key is to fly under the radar and not cause problems.
I've seen some large HOAs that cover lots of homes where they provide free or discounted community services, like a gym, pool, tennis courts, etc. But it seems silly to structure that as a HOA rather than some sort of voluntary community organization. Or just do it more traditionally and charge membership or use fees to use the amenities. If that's not financially feasible, then perhaps there aren't enough people who actually want them, and so maybe they shouldn't exist.
And I get that many homeowners want to control their neighbors in order to keep the neighborhood frozen in some particular state, but it's... pretty gross. I hear the line "protecting home values" trotted out all the time, but that just feels like an excuse for busybodies who want to control others' lives. Things like not allowing a garage-based business that produces lots of noise and pollution should just be covered by regular zoning laws.
I assume they intend to narrow down ideas based on their specific situation.
Nobody is checking what's going on in garages and basements in Detroit and the police in that kind of place would laugh off a call like that. Not having snooty busybody neighbors (an impossibility in most of the "nice neighborhoods" and "good school districts" that HNers generally buy into) seems to matter more than legality in practice. Nobody else cares if you're legal as long as you're reasonable.
"The following occupations are eligible for a Home Occupation Permit subject to restrictions discussed in the next section. If the occupation is followed by an asterisk, the use is also subject to special conditions also discussed below. Eligible home occupations are:
1. General office uses, such as accountant, administrative assistant, answering service, appraiser, architect, attorney, bookkeeper, broker or agent (real estate, insurance, etc.), counselor, consultant, drafting service, engineer, interior decorator, secretarial service, word processing service, and other office uses whose characteristics are substantially similar to those listed.
2. Commission merchant, direct sale product distribution, internet, or mail order business.
3. Dressmaker, tailor, fashion designer.
4. Mobile vehicle glass installation and mobile vehicle detailing.*
5. Pet services, such as pet sitting, pet grooming, pet training, and veterinarian care.*
6. Office for contractor, handyperson, janitorial service, landscape contractor, gardening service.*
7. Artist.
8. Tutoring.*
9. Small equipment, appliance, and computer assembly, repair, or reconstruction.*
10. Healing arts professional, including physician, surgeon, chiropractor, physical therapist, acupuncturist, and somatic practitioner.*
11. Hair stylist, barber, and manicurist.
12. Swimming instructor.*
13. Cottage food operation as defined in section 113758 of the California Health and Safety Code."
=================
I saw somebody get dinged for this once for running a Twitch stream out of their garage and accepting donations, one of the only times I've seen the modern "contractor" trend work out in anyone's favor:
- https://nitter.net/happyf333tz/status/1036846647945261056
- https://nitter.net/happyf333tz/status/1040074413599678465
[0] https://casetext.com/statute/california-codes/california-hea...
there's lots of tutorials on youtube, seems like a gratifying hobby with a potential for profit if you take it seriously as a day job
If it were me, I'd also be looking at scenarios that involve any kind of "thing" that I could acquire cheaply and re-purpose somehow. Making lamps out of old wine bottles, that sort of thing.
Robots, unmanned vehicles of various sorts, all sorts of small electronic gadgets, probably some auto accessories... really, the range of things you could (at least hypothetically) manufacture in a space that size is huge.
Now whether or not you could manufacture the thing at scale may be a different question. You could probably easily accommodate doing something the size of a small home appliance (think: washing machine size) if you only had to do one at a time. But doing that at scale might well require more space. So is the intent to stay in the garage and run an enduring business there, or just to ship a prototype, prove the model, and then maybe expand? Or is this purely an academic thought experiment?
The Buster Beagle was a real game changer in this space, though if your parts are really small there are other even cheaper options.
The goal is to, as you say, prove the prototype and then get a larger dedicated space. The product I have is not super niche and could theoretically grow a lot, but we are planning to be pretty adaptable by focusing a lot on COTS components, the kind that you can go onto Alibaba and find 5 factories for whatever you need.
My understanding is that OP is setting aside legal issues such as zoning, etc for the moment and focusing on feasible business ideas.
However, I could be wrong.
This seems like the sort of work that was spun off from some kind of existing business relationship.
I'm really into analog photography and I can think of several products that people (including myself) would pay for but that it's just not economically feasible for a big company to hire a bunch of people and pay them wages, health insurance, and 401ks to make, on top of the actual cost of the product. I'm looking to gradually put together a workshop to try and make some of them, and if I do them, I might as well sell them. Even if it's $100 out the door for a $5 piece of metal, there's actually a market for that in terms of hobby income, it's just not a market that will sustain full-time employees and mass production.
Between 3d printing, stamping, a CNC mill and lathe, casting, a laser cutter, and a vacuum oven, you can really do a tremendous amount of stuff in your garage, especially if you are willing to leverage these tools together. 3D print a part and then cast it in a durable metal, machine it to clean it up. CNC mill yourself a stamping die. Use the vacuum oven to cure things, or dehydrate your filament, etc. Like on paper that's pretty much a tool-and-die shop, given sufficient effort you can make things that will let you make whatever else you want - much like chemistry you're never more than a couple tools away from the thing you want, you just have to make the thing that will let you make the thing you want.
Optical lithography is not that hard either as long as you're not working at semiconductor scale. There's that kid that is making chips in his garage over in the UK or something. But you can use that as a manufacturing technique for other stuff. Or use resist etching like for PCBs.
In a lot of cases, really the only limit is when you bump into something that's restricted or too hazardous to keep around even if it's unrestricted. Like boy, mercury intensifier works great but... I like my nervous system the way it is. Even selenium intensifier or pyrocat developer are pretty yikes in terms of the MSDS, very much a "better have a fume hood" thing, or do it outside (in a daylight tank).
Incidentally, but, my most insane "I'd love to do that in my garage" is custom lenses. I know the accuracy is probably just not there compared to what you'd realistically need for good results, but it really seems like single-point diamond turning should be something that is achievable with a high-end setup (say $25k) in this era of solid CNC mill or lathe setups for half that price. Maybe it's something you could build out of a CNC setup but again, is it accurate enough to make it worth it (not sure of the tolerances required, at least 1/1000th, probably 1/10,000 is better, 1/100k or finer should do it, which I guess isn't too far out of what you can do with a lathe, it's all just end work and you have to be precisely optically centered and aligned). Coating is one where you'd need the vacuum oven for sure, assuming it wasn't too toxic (iirc coatings are fluoride based). Growing optics-grade fluorite or calcite crystals also might be possible for lens blanks (although again, maybe too nasty) - or glass casting too. You'd need an optical bench too of course.
There is definitely a market for that I think - all-fluorite lenses are excellent for wide-spectrum photography (UV to IR in the same lens without focus shift - see the Coastal Optics lenses f.ex) and people (companies) pay big bucks for those, like $50k is entry-level for something in that class if you go out and buy it new. And with single-point you can make aspheric lenses as easily as spheric, so you could do all-aspheric designs that aren't commercially viable for mass-market lenses... as well as super-high-quality repros of classic lenses that are obscure or just classics. People would pay for a neo-retro Hypergon or modern takes on sonnar/h...
https://pinnermachineshop.com/
https://skgrimes.com/
Bikepacking bags (and to some extent ultra-light hiking) are good examples that come to mind where people started things as hobby businesses in their garage and they quickly expanded because of demand.
A trivial electronic gadget to turn an oil cooler pump on and off is a good start. Electronic assembly isn't noisy or stinky. At most you might need to bend sheet metal for a case.
Guitar players love audio processing gadgets they can turn on and off with their toes. Make something that can be adjusted to give them a unique sound. Distortion is easy to layer on.
> ...not particularly interested in the legality of this...
And potentially legal, depending on locale.
I get better product now than I used to. Mostly it’s about genetics anyway.
The number one bit of advice I give to new gardeners (my garden is pretty well developed) is ‘less is more’. Basically leave the plant alone as much as you can. Water deeply instead of frequently and use many seeds to start with whittling down to the best genetics.
For example for my 5 pumpkin plants I planted around 50 seeds from last year. Waited a couple weeks and then pulled the smaller plants. Do the same again a couple weeks later. Then when transplanting cull once more. Plant three in every hole. Two weeks later cull until there’s one plant in each spot.
Then for advanced mode you use the seeds from this selection process next year, and they get more and more adapted to your micro climate.
>> for my 5 pumpkin plants
Is this some sort of code-speak, or marijuana variety, or are they really just talking about pumpkins?
I deffo could have made that more clear though!!
There will always be new customers
https://www.youtube.com/watch?v=WyXSNzywqvw
The only problem with living things though is then you have to care for them. Even if you don't feel like it that day. And if they don't sell you still have to water them or throw them out.
There are people who breed specialty plants also and make a (modest) living from their backyard. I know a retired guy who does this and sells his developed cultivars (ships once a year) on Facebook marketplace. He does like 20-30K in sales a year from his yard but it took him years of growing to start developing the cultivars.
Not spreading plant diseases around ends up being a big deal.
However, the inspection and license is not very hard or expensive.
(My dad grows and sells seedlings of a particular and somewhat obscure native fruit as a side business.)
You'll never outcompete a large market, but for people making < 1000 of a thing, there's not a lot of options.
The margins are really high.