Northern Diver,

Sorry to burst your bubble, but the spec sheet for the compressor you referenced says it is rated at 12000 btu (3500 watts). It can be found in the samsung mini-split model no. aq12a9me or sc12za1x, both of which are rated at 1 ton of refrigeration. For the right price, it would be a good unit to tinker with, but much too small to sufficiently cool or heat a whole house.

Jeff5may,
Ah, yes, on closer inspection of the data sheet I found http://www.valinta.lt/ka/Samsung/Kon...00421421_2.pdf It seems this unit has two compressors.

This brings up the idea of using a two or three compressor unit to improve efficiency.

https://lh6.googleusercontent.com/-c...%2520graph.jpg

https://lh3.googleusercontent.com/-U...%2520Calcs.jpg

Two of these 12000 BTU units could be operated together in the winter and only one in the summer. A suitable control system, Arduino based would be my choice, could alternate the running compressor when only one is needed.

Overall I think this would be more efficient than running a single large compressor, especially if the control system has a bit of intelligence, only kicking in the second compressor if the temperature differential between target and actual is large.

Northern Diver,

Looks like we're on a very similar project.
I spotted those compressors on ebay then managed to win a 5KW Dimplex GSHP for a very good price, unfortunately the couriers managed to drop it and destroyed it :mad:(got a refund but gutted as i'm still in need of some heating before winter).

Anyway i then bought 2x 48A124JV1EG compressors. (arrived today, strangely one of them looks like it's been dropped from a great hight and will have to be replaced.:()
I've already bought 750M of 40mm HDPE which i need to lose under the garden somehow, probably a bit excessive for my heating demand(whatever that may be?) but it was cheap so i'll use as much as i can.

Still collecting parts and ideas at the moment, at least a few months away from building a prototype.

Its suitable heat exchangers that seem most elusive at the moment.

Steve

I'm catching up on my favorite threads, and have a few points and questions.

Concerning flammable refrigerant fire safety, it is a good idea to make your flammable surroundings flameproof. Noncombustible materials like concrete or cinder block work, but most of us have wood houses. I found a FANTASTIC range of products from Flameseal, Inc HERE that are used extensively for log homes. Watch the demo videos and see who the customers are-impressive! Some other companies make paint with intumescent properties and there are probably other coatings like Flameseal. Anyone have any firsthand experience with intumescents?

Electrical wiring code requirements can be confusing. By far the best resource I have found is Mike Holt and his books such as Understanding the National Electric Code, and his online forums HERE. The books can be expensive, but you can find a copy written for the previous NEC cycle (2008 right now, since the 2011 is the current one) cheap on feebay. You can then google the changes made since the previous edition and brush up on those.

So, how dangerous are flammable refrigerants?

One of the main tenets the NEC is built upon is that the 18" up from the floor including deeper pits or depressions (for heavier than air combustible gases like gasoline and propane) and the 18" down from the ceiling (for lighter than air combustible gases like natural gas) are the areas where they have different code requirements concerning electrical ignition sources. However, even in commercial buildings, unless you are dispensing flammable fuels like a gas station, there aren't special requirements in between those zones.

As long as your flammable refrigerant equipment is more than 18" off the floor, it's housing is well ventilated so a leak can't bring the concentration inside the housing above the lower flammability limit of 2%, and the system is small enough that even a massive leak of all of the easily lost charge can't bring the volume of air at floor level up to the lower flammability limit, you are not generating much risk. A 2 gallon gas can in your garage or basement for your lawnmower is a much greater risk.

Since R290 is very soluble in compressor oils, only about 75% of the charge can escape quickly even with a catastrophic leak. The rest will foam out of the oil very slowly over many minutes, allowing time for the initial loss to dissipate. Synthetic refrigerants do this to a much lower degree, but the effect is still significant. If you have ever opened an automotive air conditioning system quickly after evacuating it to atmospheric pressure, you know what I mean.

I read a paper on how much flammable refrigerant is required and how much is ok to use. Half a pound is enough for 1.4-1.5 tons of heat moving capacity. 75% of that amount, makes 3.1 cubic feet of gas(0.5lb*0.75=0.375lb. 1cuft of propane=0.12 lbs, so 0.375lb=3.13cuft of gas. For comparison, 1cuft air=0.08lb). 3.1 cubic feet of gas brings 158 cuft of air to the 2% lower flammability limit. A 10x11' room has 165cuft of air in it's lowest 18", so would theoretically be ok. Air currents and ventilation improve the situation, localized higher concentrations near the unit degrade it. Even if you add a 50% safety cushion, a 3 ton unit could safely live in a basement that is 18x18 having 486cuft of air in its lower 18". Build properly, allow for room size, and work safely-you should be ok. I'm still researching R134a and flammables blends that Perdue University worked on, which mitigates the risk even further.

Another article noted that the incidence of fires from refrigerators didn't change in parts of the world when they switched to flammable refrigerants in them, and that washers, dryers and dishwashers have a much higher risk of fire than fridges.

It always comes down to the beancounters, though. Has anyone actually been denied an insurance claim due to a flammable vapor compression system in their house, whether or not it was the cause of a claim? Has anyone seen any verbage in their policy fine print that discusses flammable refrigerants?

Craig
The MMT

Your reply to me a while back mentioned vodka and welding together, it didn't even register to me at the time. How do you weld with vodka?!?

Craig

I also found some interesting references to the human safety and maximum Coefficient of Performances (COP) of some refrigerants, apparently the Carnot efficiency limits for a single stage:

CO2-COP 2.96 (safe to 30,000ppm)
R410a-COP 4.41 (no safety concerns)
R134a COP 4.60 (no safety concerns)
R290/propane COP 4.74 (flammable between 20,000 and 100,000ppm)
R717/Ammonia COP 4.84 (safe to 300ppm, which is beyond what is endurable for even conditioned persons. Doesn't play nice with copper)

CO2 seems to be the laggard and has extra cost involved with engineering for the the high pressure, but it apparently works extremely well in multistage cooling systems cooling to very low temps and is much more efficient than others there. Not much use for us, though.

R410a requires very little flow for a given amount of heat moved, but its supercritical temp (temperature above which it won't condense to a liquid) is very low, so it is nearly useless at condensing temps above 125F-not good for heat pumps providing hot water.

Ammonia is a completely different ball of wax and there isn't much equipment made that is useful for us to hack.

R134a and the flammables have good COP (especially at higher condensing temps) and many other technical good habits. They are cheap, easy to use and easy to find. On the bad side, R134a has Global Warming Potential, requires more flow to move a given amount of heat and the flammables are, well, flammable.

A blend between R134a and one or more of the flammables minimizes the GWP by using less R134a, minimizes flammability risk by using less flammables, and has technical attributes that are greater than any component alone. They are near-azeotropic, so you don't end up with much temperature glide or problems due to changing the fraction of each component during charging.

Perdue University published some good papers on blends. 45% R290/55% R134a had greater heat moving capacity than R290 alone and looked very promising. Anyone tried any blends?

Craig
The MMT

MMT,

Thank you for this very clear and rational analysis of an undeniable problem with R290. Your findings replace 100 pounds of unreasonable fear with 10 pounds of well-thought caution.

I think that anyone who wishes to proceed with R290 must read, and carefully consider your post.

BTW, I still think that a propane sensor that can control a non-arcing evacuating fan should be standard equipment with such a setup. What's good for a boat is good for a basement.

Sincerely,

-AC_Hacker

MMT,

I wouldn't have believed it if I hadn't seen it with my own eyes... but EcoRenovator 'Vlad' picked up one in his home country of Russia. He put on a demo for me and I am impressed.

The technical specifications call out "Water-Alcohol" as the working fluid (which takes the place of an inert gas). As it turns out, you can pour vodka right into the device and away you go.

It is real and AVAILABLE HERE.

Turns out that it is light weight, flexible (weld & cut), and cheap to operate.

Best,

-AC

MMT,

Could you edit-in or post a link to this material? This is very interesting and useful.

Best,

-AC

Leave it to the Russians to find yet another wonderful use for Vodka! Use your work-related consumables for pleasure!