Hacker Newsnew | past | comments | ask | show | jobs | submit | akumpf's commentslogin

There are a lot of cottage food bakers out there, making/selling artisan bread at farmers' markets and through side shelves in established retail/grocery spaces.

The big issue with home baking is scale -- there are often regulations that limit what you can use (often it MUST be in the kitchen within your primary residence, and it can ONLY use a standard residential oven). And if you can't make many loaves, the prices end up being about double compared to industrial produced bread. Hence the $10-25 per loaf at the farmers' market.

So I've been reverse engineering what's possible, given the cottage food law constraints (specifically in Ohio, but applies to a lot of states).

There are lots of non-standard things that I've found actually work quite well for me: Kneading the dough by slowly rotating it in a dough tub (no stretch-and-folds). 18hr+ room temperature bulk fermentation (using just a few grams of starter). Standard oven with 2 "decks" using baking stones. Using only a 10-15 minute bench rest (microproof) then quickly shaping loaves before going straight in the oven (no pans, parchment, banneton, or couche).

So I've put it together as open research for anyone to use. There's also a dynamic recipe page with my production quantities for various breads over here: https://palmbread.com/open_research/baker_recipes/

If you've ever considered baking bread at scale, please give it a look and let me know what you think. Thanks!


Some context: I have kids and I build functional prototypes as my day job.

There are tons of circuit kits out there, but I’ve been surprised/disappointed with how pretty much all of them are designed. So I’ve built out a new approach that uses a graphical sticker on a breadboard to make it easy, but still uses the raw components to make it work so the maker can gain comfort/experience with real engineering.

I also wanted the kit to do something truly interesting/useful. Not just blinking a light or mimicking something else you can easy find elsewhere. This first kit is a fully generative music box that can play millions of different song patterns.

Would love to hear thoughts from the other maker/engineer parents on here :)


Five years ago I worked on an app called Fiddlewax Pro (later renamed Firo); an advanced chord/drum/keyboard/looper. Unfortunately, as iOS updated, the app eventually stopped working and fell out of fashion.

Firo4 is an attempt to bring an old app (that I truly loved both making and using) back to life.

Direct link to video if you just want to see it in action: https://www.youtube.com/watch?v=OtpXpVq30dY

Since it's a continuation of an old app (that originally sold for $20, but was made Free for its final days), launching Firo4 for Free seems like the right move for now. Still deciding where to take its revenue model, but it definitely feels good to move it out of the archived box and get it running again.

Any suggestions on what kind of pricing models are best for a situation like this?


Direct link to the video here: https://www.youtube.com/watch?v=DEhSXTKpjQE

Also, all the design files are open source and available on Thingiverse: https://www.thingiverse.com/thing:3649542

Let me know if you have any thoughts/questions!


Direct link to video here: https://youtu.be/kWE8AzJY8qc

I'm the dev. Let me know if you have any questions/thoughts!


Direct link to a video of the TuneFast Harp here: https://youtu.be/yGjTQOmBWsM

And here are the files/info if you want to print your own: https://www.thingiverse.com/thing:2709667


In short, Picnic is google docs-style collaboration for graphic design using just a synced filesystem (google drive, dropbox, sync, local shared folder, etc.).

There's a video of it in action at the bottom of the page.


It's essentially google-docs style collaboration (multiple users in a single doc), but doesn't require a dedicated server and has to get around using all the diff/match/patch magic you can apply when working mainly with text (since Sketch uses lots of binary files).

What's particularly cool is making an existing app collaborative by simply creating a plugin and leveraging realtime syncing services. Curious how many other apps could use this same approach. :)


This has been a lot of fun exploring over the past few days. Let me know if you have suggestions/questions about this kind of hybrid approach using 3D printers.

Anything else you'd like to see printed with multiple materials?


1) Thank you for writing this! Some of the techniques you've outlined here, I have thought of so many times... but deemed them too dangerous. Your post has given me the courage to try some of this out on my own. To begin with, I will print some artsy statue with ... uhm, a low infill (like say, 10%?) and routinely fill it with some heavy sand or viscous liquid. I think it'll work well... it'll save me material, but on the other hand it makes 3d printing very hands on huh! I suppose I should pause and fill every other 30 minutes or so.

2) > this approach is ultra powerful, and it's the same thing you see on some of the high-end maker printers (like in the TAZ production docs that describe how to use strong/precise steel parts combined together with 3D printed parts).

Sorry man, what do you mean here? Can you maybe please specifically link to what you are referring to here?


To answer #2:

If you inspect most of the FDM (maker-style, RepRap, etc.) printers, you'll notice that they themselves are constructed of 3D printed parts. But the way they use the parts is a lot different than most end-users (a single purpose item that works stand-alone; like a statue, cup, wall hook, etc.).

Instead, the way the 3D printed parts on the printer are used (in general) is by combining the strengths/attributes of multiple materials. All fasteners are standard metal parts, with mating metal nuts (often embedded or even hot-inserted into the back of a printed part). Precision bushings are press-fit into printed parts to use along steel guide rails.

The TAZ assembly docs show a lot of that thinking, so I thought I'd link over to it in the write-up.

In short, it's the kind of thing that I'd love to see more of because it leverages the best properties of multiple materials instead of being all one or the other. :)


(I can't tell from the post if you are able to print 2 materials, or just embed 1 in another, so I'm not sure if this is what you are looking for)

At one point I wrote a STL converter for medical images (DICOM -> STL), for MRIs and CAT scans: https://github.com/chrisvana/printer

It would be interesting to print a cancer tumor with one material the and the surrounding tissue with another. Some data here: http://www.cancerimagingarchive.net


> It would be interesting to print a cancer tumor with one material the and the surrounding tissue with another. Some data here: http://www.cancerimagingarchive.net

Wow, that's like the first cool reason I have stumbled upon for a dual-extruder printer. :) (a lot of people think they need a printer with dual-extruders -- they don't, and actually the biggest reason they don't need it is 3d printing suddenly gets twice as difficult when you're doing it with dual-extruders)


> 3d printing suddenly gets twice as difficult when you're doing it with dual-extruders

Not really... If you use something like this http://www.thingiverse.com/thing:1680034

In this case, you extrude motor 1 and print. Then to switch, you retract until you're before the Y, and extrude with filament 2. Use a wipe-tower to get a consistent flow, and you have 2 colors (of the same material) with 1 hotend.


My experience would be that it is a lot more difficult.

I've got one of these (the cyclops) and when it works it works well (although even with a wipe tower you get some colour bleed).

https://printedsolid.com/products/e3d-dual-extrusion-chimera...

It is very easy to jam one of the colours, you have to be very careful about retraction distances because if you retract to far the other colour can leak into the cold end part and solidify and then you only have one colour working.


Its the type were the printer prints 1 material and the operator manually adds the others.

I've done this to put captive nuts into a 3d print by adding a M00/M01 into the GCode right before the layer they were covered up in.


for DICOM, don't you need some way to do segmentation to get to the data that you actually want?


Did you mean diffentiating tissue?

Yes, that is true. You can use ISO levels from the images to segment different density regions, which is more or less what a doctor would do, AFAIK.

Usually the entire picture is non-zero, so you have to set a cutoff somewhere in order to get a real blob.


Great post! You should post it to this 3DP facebook group[0] for feedback on slicer settings that could help with this. I think for liquids you might want to use additional perimeters, increase the infill density, and maybe even slightly overextrude to avoid gaps. I know /r/3dprinting would love this post too!

[0] https://www.facebook.com/groups/makerbot/


*Under Pressure = Barometer (which is correlated with various weather patterns and doesn't require other connectivity to function).

Full disclosure: I'm one of the devs and I would greatly appreciate hearing your thoughts about LightNudge. What do you like about it? What is confusing or unappealing about it?


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: