Jeff Kaufman's Writing

Gelatinizing Starch in an Emergency

24 Jun 2026 08:00:00 EST

I've been thinking more about disaster preparedness recently, and an important piece of that is food. And for many foods, before you can eat them you need to cook them. For example, a large fraction of the calories in our house are dry rice and pasta. These have ungelatinized starch, where the nutrition is bound up very tightly in starch granules. We mostly can't digest these as-is, but heat and water gelatinize the starch and make it bioavailable. [1]

This means you need some way to heat things up. If the power grid is working then you have lots of options, including immersion heaters and boiling water with a kettle, but in many disasters this is not an option. What else can you do?

Various options with fuel and batteries are possible, but a solar oven is hard to beat. Most would have what they'd need to put one together from tinfoil, cling wrap, tape, and cardboard. It would be worth putting some effort into identifying the best DIY-from-on-hand designs and distributing them. A simple one probably can't get water to the boil, but luckily you don't need that for starch: 180F is enough, and even 160F gets your pasta about halfway bioavailable. You do want to pre-soak, though: as starch gels it becomes less permeable to water, and without stirring or boil to disturb the surface gelling it may not fully hydrate.

The main limitation of a solar oven is, of course, lack of sun. When I look at a place like Boston, though, you should be able to cook a pot of pasta on ~60% of days (~30% in the winter). If you have a mix of foods, including starches that can be eaten without heating, like crackers, cereal, and ramen, you can probably just eat those when there's not enough sun for the cooker. Alternatively a propane stove does pretty well: if you pre-soak, bring to the boil, turn the stove off, and move to an insulated container, you should be able to cook several hundred pounds of pasta with a standard 20lb propane tank.

[1] Now, our stomach isn't the only way to get nutrition out of food: it can also be fermented in our guts. The best case is that you grind it and soak it well before eating. The gut bacteria would need time to adapt to the challenge (replicating to handle the newly abundant food) and while they're adapting it wouldn't be easy gastricly. You'd want to ramp up to this, using it to stretch more bioavailable food so it would last longer while giving our intestines the time to adjust. Then the caloric content is also lower: half to the bacteria doing the fermenting (or lost as uncomfortable gas) and half to you in now-digestible form.

On the uncomfortable gas, it looks like it should be worst initially, and then go down as bacteria propagate that can consume the gas. Whether the gas level (and other gastric symptoms) is tolerable is probably a good gauge for whether the ramp is too steep. It won't drop to nothing, though, and how well it works depends in part on your gut microbiome: if you don't have the appropriate bacteria initially it may never adjust well.

Overall, this really doesn't sound like a good time. Then add in that there's some risk of food poisoning from raw grains that are expected to be boiled before eating, and it seems best to find another way.

Comment via: facebook, lesswrong, mastodon, bluesky

Running An Air Purifier on Batteries

03 Jun 2026 08:00:00 EST

Running an air purifier on a battery could be really useful in an emergency that combined a biological or nuclear threat with a power outage. Getting one that can run on 12V DC and attaching it to a LiFePO4 battery is about $188 (plus $164 for the purifier) for something that will give you 141 CFM for over a week.

I've been thinking about DIY biohardening, primarily to reduce risks from environment-to-human threats, and a lot of what's out there assumes the power grid stays up. This doesn't seem like a good assumption: even if society does a fantastic job protecting essential workers and prioritizing keeping the grid up, I expect many more outages than we have today, and longer ones. If an outage means you lose positive pressure and get sick, that's really very bad!

If I needed to build a DIY cleanroom today, I'd start with my AirFanta 3Pro. While it being HEPA is overkill for cleaning the air that's already in a space, it's great if your goal is to clean air as it enters a space.

The simplest option is to buy a portable power supply. I have the 1,056 Wh Anker SOLIX c1000 and at $450 on Amazon it's comes to $0.43 / Wh. If I trust AliExpress, I could maybe get it for $322 ($0.31 / Wh). These look to be pretty typical for portable power supplies, and I like that the SOLIX supports solar charging.

Another option would be deep cycle AGM lead-acid batteries. This is what I went with in 2018. Doing some reading now, though, it seems like they're rarely worth it anymore. A 100Ah AGM, which you should really only take 50 Ah of, is $160, and a 100Ah LiFePO4, which can be discharged down to 80-100%, is $147. Plus the LiFePO4 is less than half the weight: 24lb vs 57lb.

Unlike the portable power supply, version, this requires assembling a few components:

A coulomb counter shunt, which tells you how much power you've drawn so you know how much is available and whether you're almost out. ($16.19)
A fuse holder and fuses, so a short circuit doesn't start a fire or destroy your battery. ($1.70)
Connectors, so you can easily connect and disconnect without worrying about messing up polarity and destroying something. ($4.66)
Charger, so you can bring the battery back up to full when you have access to power again. ($18.99)

I already had all of this from my earlier inverter project, except for the fuse (integrated into the inverter) and connector to the AirFanta (which takes a 5.5mm x 2.5mm center-positive barrel jack). Hooking it all up, I can run my AirFanta off grid:

If I didn't already have most of this, I'd have been spending $188 for 1280 Wh, or $0.15 / Wh. This is much better than the portable power supply, it also provides much less: I can only use it to power things with 12V DC.

Now, you might imagine someone would sell a box that wraps a battery and provides these extras so you don't need to DIY anything, but as far as I can tell this doesn't quite exist. People sell "battery box power centers" for use on boats, but they don't measure how much power you've drawn. With a modern LiFePO4 battery this is a big issue, because you can't really estimate power from voltage. These boxes also don't provide charging: on a boat that's not a feature you're looking for. So I think full featured portable power supplies and DIY setups are the two main options.

Personally, I'm glad to have both systems:

The Anker SOLIX portable power supply is much more flexible: it powers things over AC, provides USB ports, charges very quickly from the wall if power is available, and can be recharged by solar.
The DIY 12v system is simpler, less likely to break, modular and easy to fix, and cheaper. If I want to go bigger, I can expand my total capacity just by buying additional batteries at $0.11 / Wh.

I can also move power between the two systems with relatively low losses, to take advantage of flexibility or capacity as needed.

I'd really like to know how much power this would draw and how long I could run it for, but without actually building something and taking measurements all I can do is estimate. A big question is whether it could get to useful levels of pressurization: I don't think it would get anywhere close to +75 Pa, but maybe +10 Pa would still be possible and good enough if we can avoid wind by pressurizing something inside an existing building? For now I'll set all that aside and look just at the case that's easy for me to work with: running the air purifier as it's designed to be operated.

So: how long can I run the AirFanta for? What setting should I use if I want to maximize my clean air delivery rate (CADR)?

The manufacturer gives power and throughput numbers, but I expect slightly lower power usage from running it directly on DC. They report 33.2W on the highest setting while I measured 29.2W, so this looks like a factor of 14%, just around where you'd expect. Scaling down by that factor, and calculating CFM per Watt, I get:

Setting	Power (W)	CFM	CFM/W
1	1.93	57	30
2	4.12	141	34
3	9.74	247	25
4	16.58	321	19
5	24.04	374	16
6	29.12	413	14

You can see that setting 2 is the most efficient but also produces less air: if you have unlimited purifiers you should run them all on 2, but if you need more output you might need to run them higher to get sufficient CADR.

We can also estimate the runtime we'd get at different speeds. I'll model the 12v DIY system as a 100Ah LiFePO4 12.8v cell (1,280 Wh) while the Anker C1000 is 1,056 Wh. [1] I'm estimating that the C1000 loses 2.5W just by being on, an additional 7W if it needs to run the inverter, loses 7% on DC-DC conversion (12V port) and 14% on DC-AC conversion (AC outlets). So I'll model the 12V DIY system, the C1000 via the 12V port, and the C1000 via the AC ports (where we then lose another 14% on AC-DC conversion):

Setting	12 DIY	C1000 DC	C1000 AC
1	663	231	87
2	310	152	70
3	131	81	47
4	77	52	33
5	53	37	25
6	44	31	22

The effect of overhead on runtime is substantial, especially at low draw. On setting #2, producing 141 CFM, the DIY system should be able to run for just under thirteen days, the C1000 with DC for just over six, and the C1000 with AC for a little less than three. At higher draw this is less of a concern, since if the fan needs 29W losing 2.5W (or even 9.5W) to overhead matters less.

This pushes the analysis much more in the direction of the DIY system, especially if lower current is enough.

[1] Because the LiFePO4 cell has charge limiting circuitry built in, it's ok to run it to 0%: it will just shut off. While you shouldn't store it fully discharged, in this case I'm imagining we recharge it promptly. This means we get the full capacity from both batteries.

Comment via: facebook, lesswrong, mastodon, bluesky

Somerville Porchfest 2026

09 May 2026 08:00:00 EST

This afternoon Cecilia and I played for Somerville Porchfest, with Harris calling and Danner running sound. There was rain, but not enough keep us from playing, or to keep folks from dancing:

We were originally planning to be on Morrison Ave, where we've been for years. Two weeks out, though, I learned that it wouldn't be possible to close Morrison this year. [1] After lots of scrambling, talking to neighbors and the city, and some help from Lance Davis, we were able to get permission to close the dead-end section of Highland Rd instead:

This meant we didn't have my usual porch roof, and while dancers are reasonably water resistant my gear is not. Seeing rain a few days out I got a cheap canopy:

It was big enough for the two of us and our monitors, but not Harris. And he wouldn't have fit anyway, with his crowd-observation-ladder:

Being away from the house also meant we couldn't easily plug in for power. We probably could have made extension cord work, but Danner and I both have batteries. This worked well, and none of the batteries were below 70% at the end of the 2hr set.

I'm very glad we had a dedicated sound person this time: running sound for myself is never great, since I can't hear what we sound like, but this was a much more complex setup than we've done in the past. In addition to the usual mains and monitors we also ran a pair of delays halfway down the street and a subwoofer. Danner was fantastic, and I'm grateful for BIDA for providing gear and funding a sound person. Here's hoping a lot of people who gave contra a try will come out to one of the regular dances!

To keep the water off the equipment outside the canopy we used trash bags for the speakers and some 18x24 pieces of corrugated plastic for the mixer and batteries. Everything seems to have done ok!

youtube

(If this looks like fun, TryContra lets you find nearby dances. And if you're not sold, Ben Kuhn has a great explanation of what makes contra dancing so wonderful.)

I tried to get neighbor friends to park up one side of the street and move cars just before, but this wasn't enough to get that side fully cleared out. It got us about five spaces, which was enough for more audience space, but the contra lines were limited to the ~18ft between the parked cars. This is enough for two lines, but at one point I counted four squished in there. Harris also had to abandon teaching one of his dances and switch to one that needed less space.

Last year Harris tried a format where he alternated between dances for anyone and ones for people who already know contra, and he used it again this year. It continues to work really well: the new dancers aren't up for dancing every dance and the experienced dancers get to do some more complex (but still not very complex!) material.

While Kingfisher is a bit weird as a contra dance band, I think this mostly translates pretty well to this kind of environment. Having drums and bass, even simple multitasked ones, seems to make fiddle-driven music more accessible to the general public:

youtube

With so much gear to get out and back it was incredibly helpful that Al came early and volunteered to help set up, and of course we used the wagon.

The older kids made and sold lemonade. Nora held signs:

While people were overall super respectful (and joyful!) they did leave a lot left behind, mostly alcohol-related. I took the wagon around picking things up, and it was nearly full by the end:

Just as I was finishing the city workers came by to pick up the temporary street containers, which were also overflowing:

One of the workers helped me dump the wagon into their truck; very helpful!

youtube

It was on the cold side for playing, and fingerless gloves were very important. Probably not a bad temperature for dancing, though!

Overall, it went really well, I'm glad Somerville hosts a Porchfest, and I'm glad I got to play for it and introduce a lot of folks to contra dancing.

Thinking about what to do differently next time, I think the big one is reaching out to the organizers ahead of time to figure out about closing streets. Both this year and last year there was a last-minute scramble for permission to close the street, and I don't think anyone prefers it that way! I'm going to plan to do this way early: probably in a few weeks once the organizers have had time to rest a bit.

[1] After all this, police ended up closing Morrison anyway. I asked about it, and they said it was a safety issue since people gather in the street.

Comment via: facebook, lesswrong, mastodon, bluesky

Introducing faruvc.org

12 Nov 2025 08:00:00 EST

I wanted to link an explanation of how far-UVC works, why you might want to use it to clean indoor air, and what we know about its safety. I didn't find anything I liked, so I made something: faruvc.org.

Let me know if you have ideas for making it better! My goal is to have something anyone can understand, without simplifying so much that it's misleading.

At some point I'd like to include an illustration showing far-UVC in use in an occupied space, but I don't have one I like right now.

(While I'm an employee of SecureBio, this is a personal project.)

Comment via: facebook, lesswrong, mastodon, bluesky, substack

Polyethylene Glycol is not Propylene Glycol

21 Jul 2025 08:00:00 EST

Epistemic status: making things up

About six months ago I got interested in glycol vapors for reducing infections in crowded indoor spaces. I decided to give it a try, but it went poorly.

I was going to order some propylene glycol, but noticed that I already had some polyethylene glycol on hand. I figured that since people recommend propylene glycol, dipropylene glycol, and triethylene glycol for infection control, since "polyethylene" is just "propylene" with more syllables (how different could it be?) that it would work even better. In retrospect this was a very bad idea.

I had heard that you could use a regular fog machine, but these are designed to vaporize a liquid, and the polyethylene glycol I had ("Miralax") is a powder. Instead I decided to enlist one of my kids to slowly pour it into a box fan. This didn't work at all: instead of making a pathogen-reducing vapor it made a mess. To my surprise the powder settled quickly, and it was clear this hadn't worked. It really got everywhere (I can feel some under my feet right now) and it has been a pain to clean up. Worse, our cats seem to have eaten some, which has been an even larger pain to clean up.

While I still think glycol vapors are potentially a valuable tool for limiting the spread of disease, I'm going to wait for official recommendations before experimenting more. And avoid substitutions.

Comment via: facebook, lesswrong, mastodon, bluesky

Masking on the Subway

05 Jul 2025 08:00:00 EST

EDIT: this post is based on a misreading of a graph; see Subway Particle Levels Aren't That High.

Back when I was still masking on the subway for covid (to avoid missing things) I also did some air quality measuring. I found that the subway and stations had the worst air quality of my whole day by far, over 1k ug/m3, and concluded:

Based on these readings, it would be safe from a covid perspective to remove my mask in the subway station, but given the high level of particulate pollution I might as well leave it on.

When I stopped masking in general, though, I also stopped masking on the subway.

A few weeks ago I was hanging out with someone who works in air quality, and they said subways had the worst air quality they'd measured anywhere outside of a coal mine. Apparently the braking system releases lots of tiny iron particles, which are bad for your lungs like any tiny particles. This reminded me that I'd looked at this earlier, and since I spend ~3hr in the system weekly (platform + train) it seemed worth going back to masking. I've now been masking for a week, and am planning to keep it up.

This is an ElastoMaskPro reusable N95 I got for elastomeric fitting. Very easy to breath through, which helps make up for how my beard makes it hard to get a tight seal.

At $30 (vs $0.60 for my favorite disposable) a reusable one comes out ahead after five weeks if I follow the guidance of using single-use ones only once. Now, when I used disposables I would reuse them many times, but the efficacy likely dropped off a bunch: the fit is worse because the elastic stretches, and they get beat up a bit in my backpack. Likely still cheaper to use the reusable one, given how long it should last, but with how I'd use them most of the gains are in efficacy and not cost.

On the other hand, if I wanted to be able to talk to people I'd go with the disposable: the ElastoMaskPro is worse for intelligibility than all the respirators I tested a few months ago.

Comment via: facebook, lesswrong, mastodon, bluesky, substack