Hacker News new | more | comments | ask | show | jobs | submit login
Fridge 0.1 (joeyh.name)
257 points by edward 9 months ago | hide | past | web | favorite | 70 comments

Hey Joey, fan of your projects. I'm not sure if the fridge you have pulls a vacuum when it is closed, but pulling a vacuum can make a fridge much more efficient. I worked with some of the engineers that designed this -80C freezer for the International Space Station (where power is at a premium):


One of the key efficiency factors was being able to pull and hold a vacuum. And obviously the more efficient the engine that pulls the vacuum the better.

If it's a modern high efficiency unit, it probably already pulls a vacuum. If not, it may be worth physically hacking.

It probably goes without saying but if you're pulling a vacuum in the fridge you have to consider how often you'll open it. If it's for long term storage and you open it a couple of times every month at most it probably makes a lot of sense but if it's your main fridge you open potentially dozens of times every day it might not make sense to make a full vacuum every time.

That being said since the system doesn't use batteries to store energy it might make sense to use any leftover energy to pump the air out of the fridge even if you don't reach a full vacuum.

You could also use excess energy to pull a vacuum in an external tank, then use that stored vacuum to quickly return the fridge to a near vacuum after opening.

Or maybe just build up the vacuum externally throughout the day until a certain time, then use it to pull the air out of the fridge for the night when it won't be opened and won't have power.

Talking about high-efficiency systems like this it's a given that we're testing for the best-case scenarios.

But you do make a really good point. In cruising, people who choose to have a fridge have systems for this.

It is usually to get saltwater really really cold or frozen in some kind of containers: it is popular to use PVC pipe, capped off of course, filled with seawater (or a mix of fresh and sea).

You store your food in the freezer and charge up these pipes, cycling them on to the top of your food in your cooler (not powered) once every 3-4 days.

This: "1 kilowatt of solar panels has managed to cool the fridge down close to standard fridge temperatures."

Is kind of silly to me though... because that's an "incredible" amount of power.

1kW is 'peak' rating, to get the real power a rule of thumb is that they'll run at that peak for 3-5h each day (depending on climate)

Pessimistically that means 3kWh/day, which if you're storing it efficiently (in a battery or as thermal mass like OP) gives you a nominal 125W supply

Yep, and the actual kWH per day with the fridge has so far been in the range of 1-2, and bear in mind that also runs everything else in my house including a rather energy hungry satellite modem.

The link is broken, I think it was renamed https://en.m.wikipedia.org/wiki/GLACIER_(refrigerator)

With the vacuum technique, I imagine you would have to think more carefully about the containers you use (e.g. lids would probably pop off tupperware-style containers).

And unwrapped fruit or veg would dry out really quickly. I wouldn't want this in my regular fridge.

Theoretically, I'd think you could just wrap most of the central compartment(s) with a (permanent/sealed/static) vacuum chamber--like a big thermos.


Temperatures seem more than a bit high to me-- FDA recommendations [1] are that fridge temperatures should remain below 40°F (~4.5°C) for food safety (to inhibit bacterial growth). This fridge is never making it below 5°C, and (from the real-time data linked in the post) it's peaking uncomfortably close to room temperature-- I see one spike all the way up to 16.5°C, and it doesn't look like it recovers quickly at all.

[1] https://www.fda.gov/Food/ResourcesForYou/Consumers/ucm253954...

It's generally sitting around 0.5-1C.

The relevant part from scuttlebutt:

@IBob on temp probe placement, I think I know why it was seeing such sharp swings when outside the water. I had the wire coming down along the inner fridge wall corner, touching it for a foot, and then the probe was pulled out an inch from the corner. So, it was both very close to the walls, and wire's contact would suck more heat away when the wall cooled down.

The spring thermometer I have mounted in there to measure the air temp now avoids this problem, and has been staying just above freezing during long compressor runs despite frost buildup on the walls

So make sure your food doesn't touch the walls of the fridge or it will spoil. Got it.

Why will it spoil if it touches the walls? Wouldn't they be the same temperature as the area inside?

Another resource in this kind of measurement would be the mean kinetic temperature -- often it's not entirely useful knowing that the fridge is 15 C or whatever after you open the door and get something; you really want to have a reasonable measurement of how much the temperature of the contents changed in that time and in the time until the fridge interior returns to the set temperature average.


> (Numbers in this graph are running higher than the actual temps of food in the fridge, for reasons explained in the scuttlebutt thread.)

Some serious over engineering going on here. The fridge should be set to the coldest possible temperature without the food freezing, as close to 1 °C as possible an run whenever there is solar power available. The only software need I see would be a warning if the internal temperature got too warm.

Instead of smarter software, the focus should be on insulation and adding mass to hold the heat. For example, I put a foam insulating board over my 22 cu. Chest Freezer to improve it's efficiency.

The classic, low-tech (and sustainable!) way of adding thermal mass is to embed it in the ground and leverage the thermal mass of the earth around you; seemingly pointed out by comments [1][2] that haven't gotten as much traction as ones discussing high-tech. Then you don't have to cool nearly as much. Talk about over-engineering.

[1] https://news.ycombinator.com/item?id=17107254 [2] https://news.ycombinator.com/item?id=17107439

That's a pretty good option if you already have a basement, but if not, seems way more over-engineered to me. Excavating, building walls, dealing with water leaks...

A complication that my blog post doesn't touch on is that the fridge is sharing the same solar panels that power the rest of my house. I want a good battery charge and a cold fridge, and I want the right tradeoffs to be made in non-ideal situations.

You aren't wrong, but there is some value in a more complex control system. For example, the fridge uses some sort of hysteresis to prevent the compressor from cycling on and off too quickly. Let's say it runs for 15 minutes every hour. You ideally want it to schedule that last 15 minute run at the latest time possible when enough sunlight is still available to run it. This shortens the amount of time that the fridge needs to be able to stay cold without power.

I've thought about trying to add that eventually, but I think I would need a weather forcast with insolation data and also some modeling of when the sun goes behind trees/hills at different times of the year.

Anyone know a good insolation forecast API?

Never used it, but I've heard good things about the darksky API: https://darksky.net/dev

An even further radical simplification would be just plugging the fridge into an outlet. But that seems like it would be missing the point a bit.

He lives off-grid.


He doesn't have an outlet to plug it in to. He gets all of his electricity from the solar panels. His whole house runs on 12 volt DC.

Right, which is 'overengineering' compared to getting a generator and an inverter.

Who wants to run a generator all day? Solar panels, bank of batteries, generator to top up the batteries in the winter when you're not generating electricity from the panels. I live off grid and this is my set up.

During the summer I don't need to run the generator at all and I only have 350w of solar panels. This is mainly because I run almost entirely 12v appliances.

Or moving to a city?

Living off-grid is overengineering compared to the alternative. It's not surprising people would jump through seemingly unnecessary hoops once they've made that decision.

He lives in the woods, where there is no grid. It's not overengineering, just engineering with different goals.

I didn't communicate clearly. What I was trying to say is that "why didn't you just do X" critiques have to be viewed in the context that this person has already decided to live with these particular constraints. Which is why "just plug it into the wall outlet" isn't useful feedback.

My parents live in a village, they have basement, few meters underground (under the house) the average temperature is 0-5C (even in the summer when outdoor temp is 30C). All the food is saved there for years!

Engineering sometimes is so wrong, people before tech had a lot of simple an nice hacks!

My new freezer says that in the event of a power outage, if we don't open the door food will remain frozen for 48 hours. All the walls are unusually thick, which I presume indicates a lot of insulation.

Mr. Hess might want to consider adding additional insulation to the outside of his unit, especially as temperatures rise in summertime.

Yes, I'm pretty sure standard refrigerators have far less insulation than they could have. Probably mostly for weight and cost reasons.

A good quality camping coldbox (e.g. made of fibreglass or thick polyethylene plastic) can keep ice solid for days even in an Australian summer.

On long river and fishing trips, a standard cooler can work with the help of a good shady spot and a towel that has been dipped in the river and draped over it.

Evaporative coolers were popular in Australia before the widespread availability of electricity:


A smaller version is useful for storing insulin on the go:


I don't think they're evaporative. Plus you need to recharge?

Also, "patented crystal technology" what?!

They are, you soak them before use to 'recharge'.

Oh I appreciate you chiming in! I had no idea these existed.

Cool project. He's loaded the fridge with some ballast (17 gallons of water), so while there isn't an electric battery, there is a reserve of another kind.

In cooking, I like to think of large thermal masses as a thermal battery. Thus, a cast iron pan is good for searing steaks than typical pans because it's a beefier battery.

A capacitor is an even better analogy - "charge" is asymptotic, and the function of the cast iron (and the water in the fridge) is to smooth out local fluctuations in heat (rather than to store it long term).

Interesting observation. Is the line between capacitor and battery a matter of intended storage duration?

Nice project. Dependig on the location a ground fridge could be an alternative. Soil temperature is close to the annual mean in a few meters depth. https://www.builditsolar.com/Projects/Cooling/EarthTemperatu...

There is a massive amount of inefficiencies in our power usage.

Resistor based water heaters and space heaters, heat pump / compressor based air conditioners and water and space heaters, fridge and freezer cooling compressors, work air compressors etc all could be used to increase grid stability instead of now decreasing it. Or, if you have your own generation or are off grid, to reduce the electricity trading or storage needs

For my diabetic mother in a disaster, I would pay a few grand for something like this that is reliable — it could even be tiny!

That's pretty fascinating, and might be something I can work with. Thanks!

“the fridge itself becomes the battery -- a cold battery.”

Wildly my first thought from from this is that you theoretically could use a peltier in reverse to generate energy. It wouldn’t be efficient but possible.

At a different level, you can also think about your interactions with the fridge. You could segment the fridge such that it had different insulated sections that you use.

Another hack is to not store things in the fridge that don't need to be.

Bread, eggs, etc.

Maybe curdle the milk!

That looks like a big fridge, maybe then it can be smaller.

Although smaller doesn't necessarily mean more efficient you may want to check the star energy rating.

He has made the fridge "smaller" by putting 17 gallons of water in it (per the scuttlebot link). This is even better than a smaller fridge, because the cold water itself keeps the fridge cool when it is off.

Thanks, that's interesting. I like to read the article before commenting, but I don't usually go a level deeper!

Instead of an electrical fridge, use the solar heat directly to boil the refrigerant, as with a gas-powered fridge (once common, now rare due to widespread electrification). The power loss of solar->PV->motor compressor is quite large while a gas refrigerator has no moving parts: https://home.howstuffworks.com/refrigerator5.htm

I have three old solar panels in my boat for a total of 105 W, and 2 cheap 85Ah marine deep cycle 12 V batteries.

It copes with keeping a number of phones, radios, tablets and a small laptop charged, some moderate use of lights, an occasional run of a webasto diesel air heater, and continously powering a compressor driven fridge box.

I think the batteries are one of the cheapest thing in my system.

Batteries are one of the most expensive parts of an offgrid PV system once you start looking at prices over decades.

My 1.5kw solar panels and electronics cost $2k (self-installed) and will last 20+ years.

To run the fridge on demand from batteries during the night and rainy days, I'd need something approaching the capacity of a powerwall, $6.6k every 10 years.

On a larger scale something like this has been proposed (or maybe even already implemented somewhere?) as an economical way to deal with variable renewable energy (wind, solar); when VRE is plentiful and cheap, giant industrial refrigerated warehouses cool down a few degrees, and conversely when VRE is scarce and expensive, let it heat up a few degrees.

http://www.axiomexergy.com -- they build "ice-batteries" where they create ice at night when it's cool and electricity is off-peak, then use that ice to cool industrial refrigerators (supermarkets, cold storage facilities, etc) during the day.

What is potentially most interesting with this project is that the thermal mass allows a 14 hr down-cycle - could a regular-on-grid fridge be run like this, or close to it, with economic benefit in terms of electricity use?

Or would this simply just cause spikes in electricity usage to compensate for the down-cycle along with increased mechanical wear?!

Any economic benefits would likely come from using cheaper off-peak electricity. Once smart meters and flexible power prices become more common I would expect fridges (and other appliances) with such features to become more common.

From the scuttlebutt thread (on mixing water and glycerol in order to target a freezing point close to 5C or so):

> Glycerol sadly has the tendency to supercool,so once all of your solution melts you will lose all of your seed crystals and then you have to drop down to a really low temperature before it will get back to a solid.

Add sand to the water tanks?

12 volt fridges used for car camping might be much more efficient than a regular fridge if you rely on solar power.

How are the graphs being done? I'd like to look at using them for a project.

Why don’t you use a supercapacitor bank instead of the powerwall?

this has much merit

Somewhere, a team of coders at a top government agency is furiously designing Stuxnet 2.0 to break his fridge, bring him back to the grid life, and raise the nation's GDP again.

Applications are open for YC Summer 2019

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