Really low powered heat pump? (Battery even maybe?)

Nazo

2[H]4U
Joined
Apr 2, 2002
Messages
3,672
EDIT: Forget it. I'm wasting my breath. I'm unsubscribing from the thread anyway.
 
Last edited:
The truth is that there aren't many types of heat pumps. The compressor/refrigerant covers everything of significant size, from your fridge to your house to your car A/C. The second is peltiers, with which you are already well-acquainted. The third is a fuel-fed saltwater process of some sort that I don't entirely understand and which is only used on a *really* big scale, like university campuses. I suppose swamp (evaporative) coolers are a fourth.

So you're pretty much restricted to Rankine cycle (refrigerant-based) or peltier/thermoelectric, AFAIK. Peltiers are power-hungry beasts--peltier-based CPU coolers take a LOT of power, way more than 10W. Rankine cycle-based systems are far more efficient--they can remove 3 units of heat with 1 unit of input power. So if your computer is using 250-300W, you need at least 80-100W of heat pump. That's somewhere in the range of a chest freezer, if memory serves.

So if you're really looking for some sort of below-ambient cooler, you can either go peltier, or you can rip open a mini fridge and DIY :)
 
What is the emergency that would require a few degrees off your system? I am not sure what your application is but I will ramble regardless.

Do you watercool? That effectively takes a few degrees off, as seen from your CPU's perspective.

Peltiers should be more than 10% efficient. It depends on your delta T but they are way better than 10%. As mentioned one of the most efficient ways (besides radiator/passive/water towers) is a phase change system like in your window. I've seen some ghetto mods where people pipe it straight into their system from their AC. Phase change are relatively efficient compared to peltiers but their efficiency can drop quickly with delta T as well.

Also some companies (Hoffman) make industrial cabinet AC systems that are meant to recirculate the air inside the cabinet without condensation. The idea here are control cabinets in factories.

The mini fridge idea posted above is likely the most economical if you're into DIY. There's threads on directly cooling your CPU and/or GPU with phase change units that pipe it directly to just your CPU / GPU. These go sub zero.
 
A mini fridge requires quite a lot more power than a couple of lithium-ion batteries. I'm talking small here. Also, two other problems are the fact that they tend to be still relatively expensive given the small scale I'm talking about and quite a lot of them actually use Peltier coolers out of sheer cheapness I guess. (Though in a really small, really well insulated compartment it can become less about the efficiency of the actual cooling method and more about the efficiency of the insulation potentially.)

And this isn't really for the PC. I was just saying the ideas could sort of be adapted. For instance, I could hook up this little Peltier cooler on the 5V line and it would still cool the system a tiny bit for all that it would use approximately 12 watts or so, give or take (guesstimate. I don't have measurements at that.) As a temporary use thing I could almost see it working if I, for example, had the hot side attached to the casing and the cold side attached to a heatsink next to the intake, but ultimately it wouldn't work long before too much of that would come back to the system essentially. So it would have to have a heatsink on the outside to be effective and that's just a royal pain in the rear to drop it a few degrees, so not really worth it. I was just musing aloud more than anything else I guess you could say.

As for what emergency, try living in the southern part of the US without power for several days straight during the summer. I mean, I have things pretty well insulated, but... Basically I'd just want to cool a really small area like inside a tent or something just enough degrees that I could breath. (Unfortunately if it gets too hot and stuffy my nose just closes up and I can't hardly breath at all.) It would likely even involve sitting right in front of the thing the whole time. I just want to be clear here that I'm talking about cooling a very small area a very small amount -- only just enough to try to strike a balance of breathability really. It's also possible for such a situation kind of in reverse during the winter, though the weather is usually less extreme and that sort of thing is even more unlikely. This is something that only happens exceptionally rarely, but when it does it really really sucks.

There may not really be anything better besides methods that require that you have ice or whatever, but I thought I'd at least ask. I've thought about methods using things like evaporative cooling, but this doesn't work well when humidity is high at all and can actually make things worse in fact as it will just make it feel more stuffy as humidity goes up even higher to some extent. Also, it's one way essentially as you can't really use it for heating unless you heat up the water first somehow (which requires quite a lot of energy -- more so than just heating the air normally -- so is pretty pointless.) Though I will admit that it's a heck of a lot easier to stay warm than to stay cool. Plus it's theoretically possible I could lose running water as far as just plain evaporative cooling goes. Though I'll admit it hasn't happened any of those times before. (So far I've only ever had no running water during a really bad day one winter when somehow the pipes still managed to freeze despite all my efforts to prevent that. Of course, as I already said, evaporative cooling won't work for heating.)


I'll admit I hadn't thought about phase change. Are there active phase change methods? Probably most use too much power for this idea, but the basic methodology itself works. It's too bad I can't use something passive like heat pipes, but they simply seek equilibrium which obviously wouldn't do much good unless I stick multiple heatsinks directly on my skin... I'll see what I can dig up about that sort of thing anyway. Maybe there's a really low power one for mobile systems or something.
 
The inside of a tent, even a small one, is massive and not something you can realistically cool with a few normal lithium batteries.
 
I use a small inverter (1000 watt) hooked up to my car to power essientials if power fails.

If you're worried about weeks you should get one of those small, extremely quiet 2,000 watt generators.... Lithium polymers can't do shit over that time span. I have 2x 16AH 6S lipos for my lawn mower and they could power the smallest AC for an hour. That's $250 in Lipos.
 
Not "weeks." More like "week" and I'm not talking about running a full blown AC. Just something simple. The whole POINT here is to keep things to a small scale and simple which somehow I think isn't coming through clearly here. Besides the fact that lithium ion and stuff like Peltier coolers are considerably cheaper than these things people are bringing up, I already have lithium ion batterys in spades (I did say this in the first post) so they cost me exactly $0.00 at this point to get. (I've been finding them useful in eliminating all alkaline and NiMH or *shudders* NiCd batteries from my life. Actually, you might think they are costly since it can take $12 for a two pack, but over a pretty short period of time they cost less than buying 3+ AA/AAA batteries all the time for various things that I've been using them on including a really nice lantern that needs quite a bit of power -- of note here, one lithium ion is roughly equivalent to three alkaline or NiMH/NiCd batteries in voltage with more capacity than the NiMH by far which in turn is far over NiCd, so it's actually a better cost balance than it seems anyway. Alkaline batteries can sometimes exceed lithium ion, but of course are essentially one time use with charging methods unreliable, potentially unsafe, and general design just being unfriendly towards any true recharge with only around ten cycles total even and considerable decrease long before that.) The goal is to keep this small and simple. I'm not looking to run a really expensive generator to run a house AC, a refrigerator, stove, and etc. I just want to make a really small area breathable.

And Ocellaris, I'm not talking about lowering it by 20 degrees or something. In fact, it may be less trying to lower the total area and more just having some cooler airflow coming in that I can point more or less towards my face if it's hot or warmer flow I could point at my legs or something if it's cold.


I recognize that people won't necessarily agree with me and that's fine, but let's try to stick to the subject instead of attacking my reasons with the belief that they aren't good enough for you. You aren't doing this. I am.



PS. In all modern context, lithium polymer = lithium ion in a polymer casing. Still lithium ion. It's better to stick to the general term instead of specifying when speaking generally instead of one specific sort of battery. Oh, and for the record, my Panasonic NCR18650 batteries provide up to 3400mAh (admittedly this is down to an unusually low voltage of 2.5V, so they're really more like 3000mAh or so down to a more standard 3V) and are very reliable and have a good quality to them. These cost $12 for a two pack or less as you go up just getting them from Amazon without OEM bulk buying prices. These are higher quality batteries than many you can buy (but last longer and have a bit more capacity without requiring a higher voltage charge/output) and you can reduce the costs further with others. Not to mention that a bit more of the cost goes into the metal casing and such compared to li-po's casing. A large battery pack providing 16Ah (presumably with multiple cells since I guess it couldn't run a motor like that at that low of a voltage) should cost quite considerably less than $250 and you're paying more for the fact that it's a proprietary battery than anything else. Again though, in this respect it isn't about cost because I have a bunch of 18650s and a reliable easy way to charge them even with power outages such that I can keep all my electronics going for a very very long time with just the ones I already have. Frankly if I had to choose between a really unpleasant generator (and I've borrowed one once before and nearly had to leave the house because of all the incomplete combustion and noise coming in through my windows and maybe door despite my efforts to seal them thoroughly) and a really high costing set of batteries that can be charged more cleanly and used more frequently outside emergencies too I'd go with the batteries still.
 
I used to have one of those cooler/heater setups for food that used a peltier when I drove a lot for work.

For the small area inside the cooler, it would still take around 30 minutes for it to really get cool enough to keep the food from getting warm.

A single small peltier is going to be useless for use when blowing air across a heatsink to even make a difference in the air temperature coming out the other side.

And unless you exhaust the hot air outside of whatever you are in, it is going to heat up the ambient temp a lot more than the miniscule amount of less hot air it will make.

It would be a whole lot more efficient to use an evaporative cooling tower unless you are in a super humid area.

Basically a reservoir, a fan, a pump, some PVC pipe, and a shower head.

http://www.overclockers.com/forums/showthread.php/321055-Building-your-First-evaporative-cooling-tower!
 
A single small peltier is going to be useless for use when blowing air across a heatsink to even make a difference in the air temperature coming out the other side.
If I stick my nose in front of that air flow it might make a difference to me and being able to actually breath. But then I didn't expect much from Peltier cooling. I was hoping someone knew of something more efficient that actually has ever been downscaled. A full blown air conditioner system can't be realistically run through such small power, but then I don't want to cool a whole room or even a small tent by more than a few degrees maximum. I just want to make enough difference for it to be tolerable. Failing that, my nose sort of closes up when it gets too hot and stuffy (and dry is almost as bad, but if it's actually dry I could rig up something evaporative if it came right down to it and as far as I know it's only ever dry in here when the house's heat has been running a lot) so even just having a slightly cooler stream of air coming from somewhere would be nice. I was just hoping I could find something better than a Peltier cooler to do it. And really, there's no reason there couldn't be, say, a really tiny compressor type thing that was intended for something really tiny such as inside a computer system (I still say they actually DID make such a thing once and I saw it on Newegg and almost bought it all those years ago.)

And unless you exhaust the hot air outside of whatever you are in, it is going to heat up the ambient temp a lot more than the miniscule amount of less hot air it will make.
I assumed this was a given. I'm aware of the principles of heat generation and how pretty much any machinery no matter how efficient will produce heat in its process (and why if you opened a refrigerator door and left it open the room would ultimately get hotter and hotter instead of cooler.) The idea was to have some sort of seal with one side sticking out and the other on the inside. A Peltier cooler admittedly makes this easier. If you want to switch thermal flow you just flip the polarity. For heating it might actually work a lot better with that really high level of inefficiency in fact.

It would be a whole lot more efficient to use an evaporative cooling tower unless you are in a super humid area.
Last time this happened humidity was definitely an issue. (I mentioned this already.) I thought about evaporative cooling -- it's definitely the simplest way to go. But as you reach that point of saturation it starts to get less and less effective and, more importantly, ultimately just makes it feel more stuffy.
 
Your best bet is a small compressor. There are some very small compressors that run off 24v and produce around 165w of cooling. A peltier at peak efficiency produces 1w of heat for almost every watt it moves, and also consumes 1 watt of input power.

Aka it has an efficiency ratio of <1.

We can calculate average amount of energy to heat a 1 cubic meter space 1 degree Celsius using U=cPm&#916;T and this works out to about 1250J. A watt is 1j/s so the rest of the math is easy to do... It's a lot of power to drop a space 5 degrees Celsius. Remember, a 2 cubic meter space is 8x larger than a 1 cubic meter space. These values scale very fast.

Also please remember heat and temperature are very different things. Heat is the total amount if thermal power (please consider "cold" as negative heat) in a particular body, space or system. Temperature is the specific temperature of a single measurable part of that system. In your initial 10w example, a peltier running at 10w can certainly achieve large differences in temperature, but a peltier is essentially a motor. With zero load placed on it (aka no input heat on the cold side) it will chill quickly. With more than 10w of heat input, the system will overload and shut down. Do not confuse work potential and temperature. You can safely drop liquid nitrogen on bare skin .1ml at a time. It is very COLD but has very little work potential.

Also nobody is attacking your reasons... The issue is you're talking in hypothetical that simply aren't possible, sadly. Modify the idea. Put metal channels in socks and run some cool water through, or something. Ideally, you could chill some water with a small 12v compressor (uses around 120w, so battery up) or peltiers and send it through a copper coil with a small 12v pump. Put a fan through it
 
Last edited:
Back
Top