Another DIY HP Waterheater (pics and info)
 

First new thread from a lurker, this is mostly a documentary to spread what I have learned. I have been slowly poking at a heat pump waterheater project for years now. Iv lost count of how many alterations have happened along the way. The quick and dirty info of its current state: 12k r-410A heatpump window unit. Swapped to r-134a for better working pressures. TXV evap control, series/parallel piped BPHE r-410a rated, small 4-8gpm taco pump, chinese heatpump waterheater controller, and some fancy pipe work.

My design goal is a outdoor remote piped unit (funny to think cold air is a byproduct!), at least a 2.0 COP, maximum water temp 130*F (115-120* desired), and totally automatic.
The current iteration is achieving about 2.5COP at 115*F water temp at 80*F ambient, moving approx 5800btu. I literally just charged it with 2lb of gas and let it run. There may be some improvement to be found in a proper subcooling adjusted charge as it felt like only the desuperheater exchanger was doing all the work.

(Please be patient with links, my server is underpowered for what I demand of it)
*invalid url*
*invalid url*
I have updated my site and images have moved. Please use this link

I understand the cautions of using BPHE in a potable water circuit, but they saved a lot of space, and just one could handle the full load. They are "wired" in series on the gas side as a desuperheat/condenser, so even if one fouls, it will still operate fine. The BPHE are piped in parallel on the water circuit, counter flow of the gas circuit. I have also included a fine mesh strainer on the inlet of the unit. I had trouble deciding to stay with hot gas defrost as it does show some loss of hot gas temp when running, and I'm not sure how low an ambient this unit will still be efficient. The controller has inputs for safety disables, so I have included a high and low pressure cutout, as well as a water pressure cutout (no sense in it running if the waters off! Also prevents dry run, and shut down on plumbing failure). The filter dryer acts as a receiver for the txv in this arrangement, convenient! I decided to use the original evaporator in the recent variant for a few reasons. This gave me a 3 speed lower wattage motor, a much quieter blower, separates the condensate from the components, allows me to seal up the component side and insulate the box, instead of every component by itself. The unit is a 12k unit, and I'm only moving about 6K of heat, so the evap should be capable of lower ambient temps. It has significantly reduced unit noise, and also offers the ability to duct the cooled air. Im seeing about a 25-30*F delta coming out of the cold side on full speed.

Here is the growing gallery of photos of *invalid url*
Please excuse the lack of any naming scheme. I know what is in my head lol.

I have updated my site and images have moved. Please use this link

The previous phase of this project, you will see in the gallery I was attempting to use the old condenser as the evap. Trial and error showed some need to change the coils circuitry, which resulted in some very messy brazing, mistakes, unbalanced circuits, and other things that made things not very nice. The fan was also loud, single speed, and some 150watts by itself. It also *invalid url* drew the air around all the components[/URL], wasting significant heat. Also resulted in a lot of *invalid url* un-tamed piping jumbled into one corner [/URL]around the compressor. And further, all the components were sitting in a pan of condensate. While this arrangement worked ok, after a couple years of fooling with it, I decided to go the route of the original evap. I work on PTAC and windowshakers these days, so getting another of the same model was a matter of going out back to the scrap pile lol. The first version is a mess of a learning curve and years of fiddling. The newest version was put together in 2 nights, and tested in an afternoon. Yay experience!

I will update this thread as things advance, and feel free to ask anything you may have questions about. The spreadsheet pictured is just something I slapped together, ment for limited data calcs. But if there is interest, Ill buff it up and post a link. You feed it time, temp, amps, volts, and water capacity, it does the rest. Iv spent enough time fooling with this thing to learn a thing or two. Just don't expect an expert reply lol.

*Saved post*

I forgot to mention, the goal of this unit is to store heated water in a un-powered electric 50gal waterheater. I may leave the elements in, and use a outdoor ambient thermostat to transfer from HP to Elect in <1.0 COP operating temperatures.
Also to note: Switching to r-134a on a r-410a compressor proved fairly significant capacity loss from the compressor. I can only assume the 410CP uses a smaller displacement pump than a 134CP, likely due to the very different pressure and mass flow rates of the 2 working gases. There is probably a slight performance hit as the motor is slightly over-sized now, but that also offers a bit extra overhead to work with.

Looks like an awesome monster machine to me! It looks as if you've done lots of mods to the original design. I have some questions:

So how does your hot gas defrost perform? There have been a couple members interested in the process, not usually used in consumer machines.

Since you live in Florida, how come you don't use this for air conditioning as well?

The defrost is time/temp based. If the controller sees the evaporator below 25*F for 90minutes, it will enter defrost. At this point, the controller turns off the evap fan, and energizes the reversing valve. It will stay in this mode until evap temperature reads 45*F, then return to normal operation. I think the max defrost time is 5 minutes (which is a very long time!). This is nearly identical to defrost cycles on your typical household heatpump, with the exception the controller exposes all the time and temperature variables.

I considered using it for a/c, but tell me if my logic is sound: In spring/summer/fall we often have outdoor temps in the 70-95 range. It will almost always be warmer outside than inside on these seasons, which are in my favor for higher evap temps. The humidity load will also do wonders. The 2-3 months we call "winter" will take a hit on efficiency, but not sure if it would be cost effective to "cascade" the unit via the whole house heatpump. That, and space inside is limited, placing the unit in a living space (or ducted). What Could work is duct the air inside during the summer, supplying conditioned fresh air to the house (hmmmm )!

Edit: I re-read your question, and it looks like your asking how well it works, not How it works. Honestly, it hasn't been cold enough to get the evap below 45*F lol. I will say that forcing defrost, the coil goes from 45 to 130*F in about 45seconds, burns off the condensate in a hurry. Once the unit has stabilized, the compressor holds a Lot of heat, and is immediately available to burn off the evap. The heat load of the water in the BPHE also offer a huge supply of instant heat to the system. Do note the capillary bypass and check valve around the TXV for this to work. Don't forget this vital part! (wana know why I have a highlimit? LOL)

Why R134a instead of R290 or R433b/ES22a? What you essentially did was a double "Davuluri Treatment". Your max temp is going to be limited by the discharge temperature limit of the compressor, for which R290 (or R433b, which is mostly R290) would do better than R134a as it doesn't superheat as much on compression. (Counterintuitively, that's actually more of a problem at low ambient temperature since the compression ratio goes up.) The reduced suction pressure avoids compressor overload at high ambient, but just going from R410a to R290 will do it for any reasonable ambient temperature.

You can get a good boost in efficiency by swapping the PSC fan and pump for ECM. As a bonus, a variable speed pump lets you take advantage of stratification by drawing cold water from the bottom and putting in hot water at the top of the tank.

BTW, for best efficiency, operate at 105F or so most of the time and use remote control (with a timer) to boost it up to 140F for washing dishes.

Mik
My dishwasher has a built in heater that boosts water temp up if it is not 120 F was used to "old school" where you ran the tap closest to the dishwasher for a few minutes to "get the water hot" before running the dishwasher.

Today, I actually lowered the water temp to 110 F, but I notice that it does take a while to get water hot in the shower (wasting water).

Such are the trade offs . . . .


Steve

The amount of cold water sitting in the lines is the same regardless of the temperature in the tank.
Also, if you plumbed the unit into the hot and cold lines under the bathroom sink, it can also operate as a hot water recirculation setup. For those in cooling dominated climates, the summer efficiency gain from having the unit also operate as an air conditioner outweighs the loss during the winter. (And even then, if dehumidification is needed during the winter, it can still be a net gain.)

Mike,

I am confused. Is "the unit" mentioned twice above, a dishwasher? And if so, how can it act s a dehumidifier in my hot/humid summer?

Lastly, how can it be used as a hot water recirculation setup?

Am confused . . . .

Steve

The unit is the heat pump. As for how it can work as a hot water recirculation setup, if you have it connected to the hot and cold lines under the sink, you can run the pump to move the cold water in the hot line to the cold (needs a reversible pump, thus not practical with off the shelf parts) or just run it to put hot water in the lines.

The reason for r-134 is because it's readily available, I already have 30lb as well. I will never run above 130*F, 115*F makes for good shower times and temps. Also the higher super heating I think is in my favor, since heat is my goal. Not too concerned about overloading the comp as its already running at half it's intended capacity. Dishes and cloths are done in cold water, and I don't have a dishwasher. The controller does have a timer function, which I may use to prevent nighttime running when the ambient is lower. I'm not seeking absolute returns, so that saves me a Lot of money on ecm motors. The evap blower wouldn't be interchangeable anyways. Don't go by where the lines cross for cop vs. Temp, it was basically intended to show relation and returns at x temp. As for possible winter dehumid... my winter relative humid is usually below 20% and my sinus crys. Lol

Sent from my Samsung S4°

I should also note that, I have so much condenser surface area, the liquid leaving temp has a almost 0* delta to water inlet temp. Something on the order of 2 tons capacity, pushing 1/2 a ton of heat lol. Once I correct the charge, Ill recheck my vapor temperature. The last iteration of this system was 150-160*F vapor, but I could be completely mistaken. I recall seeing about 250psi vapor/discharge the other day, which relates to 146*F (at 130*ish water temp). The controller Im using also has a discharge vapor temp sensor that will shut it down if it goes beyond a set variable (Think i have it set to 180*, since thats well above expected conditions). Previously, I had also experimented with water flow rates, with surprising results. I had to dial it all the way back to about 0.4gpm before I started seeing liquid temps more than 5*F Delta. The taco pump Im using now is for sure overkill at the moment, but once I factor in 20-30ft worth of pipe and fitting restrictions, I suspect ill be down in the 1-3gpm range. As for the suggestion of using it as a heat loop, that is a great idea. The unfortunate part, my heater is in a optimal position... right in the middle of all the fixtures. Would need to run 2 loops. And with a setup like this, I would want a dedicated feed or return instead of using the cold line (because then your bleeding hot water to get cold water! lol).

Couple other specs, 80*F Ambient runs about 50-55psi low side. Highside vapor pressure varies a lot by water temp. 80* water is only some 110psi, 115* water is about 150-175, and 130-135* water is about 250*. I will make more accurate notes next time. Total system current never exceeds 3.25amps, comp only pulls about 2.5, well below the LRA 6something its rated for. Lots of overhead. The previous system had an accidental excursion, pushing the water temp all the way to 195*F ! before I found it and turned it off. Wish I could have ampped that lol. Hi-side was around 475psi (200*F), just barely under critical temp (213*F) lol.

It will not excessively heat the cold water line. If you use a reversible pump, the control circuit stops the pump as soon as it senses that the hot line has become hot. (That's how most hot water recirculating systems work.) If you just run the unit, the water coming out the bottom of the tank would be warm at most until the tank is full, at which point it should turn off as there's no point in continuing to heat a full tank.

Trouble is, this unit only adds a couple/few degree rise to the water, requiring multiple passes. Dialing back the flow would overcome this, but that sounds like quite a balancing act (perhaps only suited to a such mentioned ECM pump with temperature feedback). Not sure how that would effect my COP as well, since a depleted tank has a much lower condensing temperature, than trying to maintain 115-120*f outlet temperatures. Something to mull over, maybe up for experimentation.

Still need to decide an optimal way to tap the tank. I dont Want to tap the tank drain as thats a sure source of sediment. I could pull the water from the inlet dip tube, but there will be conductive heat gain on its way up. I considered removing the lower element and install a short dip tube to the bottom, or possibly a coax'd tube in through the drain with a slight rise to get it off the bottom. Im going to push the heated water in through a Tee with the T&P valve in the top. That will ensure that cold water cant bypass the tank through the heatpump. The budget for this unit is about at its limit for feasible returns, so the ECM pump is out of the picture for this round... (and I got the taco for 25$ lol)

I'm not quite clear on the adapting strategy of these ecm circulators. Are they constant head/constant flow, or are they temperature feedback? Building a simple pwm dc drive with temp feedback for a DC pump wouldn't be too difficult. Not sure I want to try to venture into driving a split capacitor ac motor with a scr drive. But I guess that will depend on what kind of returns I get out of multipass vs constant control output.

I think you have a very interesting water heater going on here.

Thanks for sharing all the charts and photos, very useful. I wish more people were as good at describing and documenting their efforts as you are.

I don't think we've had anyone substitute R134a for R410a before. Interesting swap and an interesting reason for doing so.

When I first looked at your machine, I had to blink a few times, I thought I was looking at an old VW engine... but then I dialed it right in!

As I recall, one of the comments was a question about why didn't you consider air conditioning? Well... isn't the output from this beast cool air? I think that just a bit of ducting and you would have killed two birds with one stone.

My question is tube length... Most of the commercial units I have examined have longer copper loops that would seem necessary, but they are that way to reduce vibration strain by spreading the strain out over a longer length. And then fair-sized chunks of rubber are put on those loops as vibration dampers. I believe that jeff5may ran into the problem of a too-short loop giving out due to vibration strain.

BTW, with reliably warm Florida air temperatures, and the modest level to which you are heating your water, I think that COP 3.5 is not too much to expect from a Heat Pump Water Heater.

If you get the urge to do another one, you might try a R22 compressor and filtered BBQ gas... I really like that combination. Your skill set is such that anything you build now will go together fast.

But, all-in-all, this is a great effort. You have tried quite a few interesting ideas... and you have them working. That's a good thing, right there. No doubt.

Congratulations!

-AC

In my hurry to reply before bed, I forgot to explain some of the bigger deciding factors of choosing r-134a. Using a constant of 54* evaporating, 150* condensing temperatures.
R-410a pressures would have been pretty extreme, 155/610psi. R-22 a bit more acceptable 90/370, but not oil compatible. R-290 was considered for some time, given 85/327, but seeing as r-134a was 51/260 which are much more manageable, and available in bulk right out of the bottle. It also works with the existing POE, and mixing a little PAG in wont hurt it as well (using automotive charging equip that has PAG in it, as well as the use of dyed oil for leak finding). I didnt bother considering R-12, since its becoming so expensive. Something you may get a kick out of, is look up the blends in Freeze-12 (a "dropin" replacement that we used. I feel very ripped off). And speaking of which, AC-Hacker, you will like this link. The epa did something useful! Composition of Refrigerant Blends | Alternatives / SNAP | US EPA And to all you r-290 supporters, you may be entertained knowing I dissected a hotel minifridge to discover the compressor was rated for r-290, r-134a, and blends (USA Model)

R22 works fine with POE oil. I believe the issue with PAG oil is its low breakdown voltage, so it's best to avoid any of it in a high voltage compressor.

Iv heard that mentioned before concerning hermetic hybrid car compressors. They strongly suggest not charging prius without purging your equip, for example. Fortunately I only have traces of pag residue from my equip. Maybe I was thinking r 22 wasn't a good mix because it's normally used with mineral, and just simplified it in my head to prevent contamination.

This list is rather dated. As you may or may not know, there are oodles of "new and natural" refrigerant blends on the market now. The OEM suppliers have now crossed the line of including "highly flammable" hydrocarbon refrigerants in the circuits of small appliances. The powers that be have been granted permission to do so below a certain mass of charge, and the mini-split and window-shaker market falls within the boundaries.

This trend has yet to penetrate the USA market, mainly confined to European and Asian markets. I believe it is just a matter of time until hydrocarbon refrigerants start appearing in refrigerators, dehumidifiers, and portable A/C units in the Americas. Your experience with the mini-fridge is living proof that at least a FEW of these units are being sold here now. But I can't help but believe that the r134 and r410 units are on borrowed time.

For some reason, most US HVAC techs freak out about the "danger" of hydrocarbon refrigerants, yet have little concern about the even more flammable Lithium batteries everywhere. I have a clue it has to do with the pinkwashing (their push for R410a as the future refrigerant) by the fluorocarbon industry...

Personally, flammability doesn't bother me. If you vac out the system, it's not flammable without air. So what might happen while brazing on the system, a surprise candle? As long as there isn't any saturated oil anyways. Anyways, topic has wandered a ways from the threads intentions.

I believe you made a good choice going with R134a as the working refrigerant in your rig. It has been done before. I am riveted by your dialog concerning your selection process. I know R134a has a very wide "acceptable" pressure range, making it useful in both refrigeration and air conditioning. But I did not know that you could push R22 or R290 that high without destroying something.

Obviously, R410a was designed to work at high pressure levels at moderate temps, not so much at low or high extremes. Can you elaborate on the envelopes of these refrigerants? Assuming a system (as yours) originally designed for R410a, where the compressor and plumbing will handle the high pressure, how hard can you actually push these alternative gases before they won't work? What goes wrong when a limit is exceeded?

The reason I ask is because so many of these R410a systems are hitting the scrapyards. I would love to experiment with propane or propylene in these units, but without an understanding of the temps/pressures that are achievable at safe margins, I am apprehensive.

It has to stay within the high pressure limit and the discharge temperature limit. A R410a compressor is going to be rated for a very high pressure (400PSI is considered a "normal" high side pressure) and the recommended discharge temperature for any common compressor is less than 190F. In practice, because R290 and mixtures based on it don't superheat as much on compression, 140F and above condensing temperatures are easily achieved.

Now, Im absolutely not an expert on this stuff, and should be fully verified before you use it to make any decisions. There are several things that must be considered when selecting a gas, type of compressor, oil viscosity and type, and worst case scenario.

I have a chart that lists boil at atmo and whats known as Critical Temp.
Crit.Temp is a very important thing to know if you are trying to make heat. Wikipedia explains it very well with "Above the critical temperature, a liquid cannot be formed by an increase in pressure, even though a solid may be formed under sufficient pressure." That said, you dont ever want to go near this region. To name a few, R-12 233*F, R-22 204*F, R-134 213*F, R-290 206*F, R-410 161*F. So as youll notice, 410 is by far the lowest, but 161*F equates to 688psi! Not a good gas for heating beyond say.. 110*F. (and the reason why desuperheater reclaimers dont work as well these days)

You also need to consider your Evaporating temperatures, and your condensing temperature. This can be a hard number to come up with sometimes as there are tons of variables. But for example, your making heat. You want to produce 115*F. Lets consider 2-3 common goals of 5, 10, and 15* of delta temperature. It would not be unreasonable to have 120, 125, or 130* condensing gas for this condition if your condenser is closely sized. Next youll turn to a pressure/temperature chart and find one or more gasses that are within your acceptable pressure range. For medium and high temperature stuff 200-300psi is a good place to be as most pipe work, fittings, and the alike can handle this easily. At this point, you would be eyeballing R-22, R-134, R-290 for example. R-410 is way out of the picture now since you would have to upgrade your stuff to handle 475PSI operating pressures!

Now things can get a little tricky, as the heat moving ability of these gasses vary, as does the low-side. This will effect how many BTU your system moves, given a constant of compressor capacity. The higher the spread of your pressures, the higher the compression ratio will be. The higher the compression ratio, the less "mass flow" the system will have, which means the less heat it can move. But the CR also has a lot to do with how much heat you "make". Things heat up when compressed, so a higher CR would be ideal to make heat. Its a balancing act that might end up being steered by what acceptable, available gases are at your disposal. You must also consider your compressor. Compressors come in all varieties of motor size and pump capacities, and are usually built to match a purpose and a gas.

A system expecting high heads, and a middle lowside will have more mass flow, and a lot of pressure to push it against. This will have a significantly larger motor section than a system with medium head, and a low suction, since it wont be moving much gas at all into a so-so head pressure. Thus the designations of LBP, MBP, and HBP (low, medium, high back pressure). LBP being the weakest, and HBP being the strongest motors. The pump capacity has a lot to do with expected pressure, and the intended mass flow. R-410 has a higher pressure, but less mass flow, so it will have a much smaller pump than a r-22 system with lower pressures, but more flow, even if it has the same size electric motor! Thats why if you use a 410 compressor with 22, 134, or 290, you might expect to get half as many BTU out of your unit (with consideration you changed your metering device to match your gas), but at a reduced running amperage as well. While it wont be as efficient at the given job, youll have a hard time overloading it. If you tried to use a 22 compressor with 410, it would be pulling very high currents, and likely have a very short life as its trying to pull off almost twice the work it was made for.

The same can be said about using a fridge/freezer compressor to build a baby window a/c. The motor is too small for the given load, unless you reduced the capacity with a lower CR gas. But wait, more to consider? Lets say you have a window a/c compressor, and you want to build a bad arse freezer. Your condensing temperature might only be 95*F, but your evaporating at -20*F. At this level, your mass flow has dropped to only a trickle because so little gas is evaporating. Most compressors rely on suction gas to cool the motor section. Now you have a compressor with a huge, very unmatched motor, with a little kid blowing through a straw trying to cool it down. Again, a very short life. Using a HBP in a MBP is ok. using a MBP in a LBP may be acceptable. You really need to reference your comps spec sheet to see its design limits. But trying to mix extremes will not be in your or the poor comps favor.

Things also to consider, the higher the difference between your evaporator temp and your condenser temp, produces a pretty nasty drop curve in your energy returns. It takes a lot of energy to move heat against its will. You can move a lot of heat a little distance, or a little heat a long distance at the same price. The closer you can keep the difference, the less its going to cost you on your power bill. Bigger coils, and thoughtful of what your trying to make go a long ways. To really simplify it, If you dont NEED 130* for your purpose, but 110* is your limit... maybe 115*F quicker will satisfy your need. This might mean finding a way to deliver your heat in a faster manner. Higher flow fans, larger capacity pumps, what ever it is you have to do to get the heat there faster with a lower delta. Instead of using lots of delta to drive the heat through your medium.

Please know, there is a lot more to this field of knowledge out there. This would be considered "beginners intro". So many things to consider to make a properly efficient system, that wont suddenly let the smoke out. It is often easier to research existing products from large experienced companies that have millions to blow on R&D, and extract knowledge for your own personal gain. Go look up a a/c unit of your desired BTU. Find out what gas it takes, go find a parts list and look up the compressor PN. Go google that PN and find a spec sheet. This will tell you the pump capacity and rated power. Now you have a gas, and a pump spec to build around to make a a/c. If you want to make a heatpump, research that. A Freezer, research that.

I dont have the knowledge to tell you how to do the math to properly size a compressor, or what capacity your compressor will have. Google it, there are dozens of papers out there that are pages long, with math that makes my brain itch, GL!

Thank you for the highly informative explanation! This has thrown dynamite into some of the conceptions I had concerning the limitations of the various refrigerants and compressors, as well as their proper and "fringe" uses. My level of understanding has just been broadened.

That was a very nice and concise little overview. It is hard, with a subject as wide as refrigeration, to get something like this.

In my learning, there wasn't really a basics page. I dove into refrigeration with operational knowledge and nothing else. I didn't know there were so many different compressor configurations, how much gases varied, such and so forth. I had visions of using freezer compressors to make baby a/c units and such. Now I know why my mini fridge has trouble starting on the same weak circuit a small window a/c doesnt (motor size/starting torque) While perhaps a bit off my original topic, it may still help shed some light for future creative minds. I had no idea the post was so long winded until after I submitted it lol

Hi all,
Been lurking here for a while learning lots. Thanks to all of you.

RB, thanks for that explanation. Perhaps you could help me fill in some blanks in my understanding.

"Above the critical temperature, a liquid cannot be formed by an increase in pressure, even though a solid may be formed under sufficient pressure." That said, you dont ever want to go near this region. To name a few, R-12 233*F, R-22 204*F, R-134 213*F, R-290 206*F, R-410 161*F. So as youll notice, 410 is by far the lowest, but 161*F equates to 688psi! Not a good gas for heating beyond say.. 110*F. (and the reason why desuperheater reclaimers dont work as well these days)"

Having your refrigerant becoming supercritical would be a problem if your condenser became overwhelmed and couldn't reject enough heat to condense it before cap tube / txv. Or if it were to exceed the pressure rating of the machine and potentially rupture something. But what else would happen?

If the fluid failed to condense it would whistle through a cap tube as vapour and equalize pressure between the high side and low side (basically turning your evaporator into another condensor). I'm thinking a txv would do the same because the vapour would be superheated entering the evaporator already. So basically pressure in the evaporator would render it incapable of absorbing more heat. No pressure diff = no phase change= no heat pump.

If the pressure of the fluid in the supercritical condition didn't exceed design spec, would it not just sit there blowing heat until the condensor started to condense again? Or would a critical failure result?

I haven't even started trying to think how efficiency would be affected.

I'm a marine engineer not an hvac guy. My experience is mainly with steam, so I could be making some incorrect assumptions or poor analogies.

CO2 systems generally operate in the supercritical region by design, although the pressures involved are high enough to make DIY impractical. At the other extreme, water vapor operates at too low of a pressure (and chemical incompatibility issues) to easily build practical systems with, even though it is a very efficient refrigerant when properly used.

As a general guideline, higher critical temperatures and lower operating pressures (within reason) correlate with higher efficiency.

Consider this. you have 1 lb of liquid. now turn it to gas. Its going to do 1 of 2 things. Its going to take up a lot of space, or its going to make a lot of pressure. I get the feeling you could very quickly experience thermal runaway. I never actually researched what happens over crit temp, I just know its well out of the design range for almost anything Id ever need to do lol. It would probably also displace liquid in your condenser, reducing its capacity. That, and that poor compressor would have to be way out of its discharge tmax lol.

Funny you bring up water, Mike. I was doing some research into heatpipe heat sinks. Thought it was very interesting that many many use water under a vacuum to get it to boil and condense at such low temperatures. Iv started fooling with heatpipes as pre-stage in a HRV tinker toy.

Water has a very high critical temperature and very low (vacuum) operating pressures. The latter presents many problems in making a practical system. Even small pressure drops are significant, so split systems using water vapor are impractical. (In fact, even package units are difficult to engineer. Most usual is a chiller.) Also, the compressor would ordinarily have to be impractically large to get a useful capacity, at least until Brittany Benzaia figured out how to use a switched reluctance motor in a centrifugal compressor that runs at over 100k RPM, thereby getting a high flow rate in a small package. (In contrast, ordinary compressors run at about 3600 RPM or 3000 RPM.)

Not to take bloody forever to finish a project, but things have once again occurred in development. Discovered that the evaporator between a heatpump and straight cool model differed. The heatpump coil, which I was using, had the suction coming off the air downstream of the coil, which was not very good at extracting all the available superheat. The a/c version has the suction on the front of the coil, allowing higher coil capacities as the superheat could be set much lower. So of course, I changed the coil. Fixed my txv hunting issues and increased system capacity. I achieved almost the same btu at 54*F ambient, as I was getting before in the mid 70s! Also got to play with my new gauges to very accurately setup superheat and subcooling. I have also learned the value of a proper vacuum. My previous equipment was old and mal-cared for, and in no way or form could pull an acceptable vac below 4-5000 microns. Fast forward to good cared for equipment, it took forever to get down to 300 microns, as the system had previously uncaptured air and moisture in it. Took a lot of heating of the drier and compressor to get it to stop off gassing moisture. I have also made an attempt at organizing and documenting the experience in a proper photo album, for your viewing pleasure! Still using the same old server that can be a little slow, so give it a moment.

Its been a rough year, so any project development was pretty much halted. I'm taking a more Me Time centric approach to this year as it seemed like everyone else owned more of me than I did, so hopefully things will start progressing again. Losing track of your life, a little hospital time, and a couple major disasters changes your outlook on things, let me tell ya.

Thanks for sharing your results.

Is there any way that you could show with photos what you mean regarding, "The heatpump coil, which I was using, had the suction coming off the air downstream of the coil..."

This is really good info that everybody needs too understand.

Great work!

-AC

When the refrigerant will not condense or it is too hot, the gas builds up on the high side until it does condense. With a cap tube metering device, it passes roughly the same velocity of either gas or liquid. The liquid has much more mass, so more heat flow occurs. Equilibrium results. With gas flowing through the cap tube, much less mass is carried to the low side. The compressor will literally pull a vacuum on the evaporator if it can physically pump the cr. more likely, it will trip out on its thermal overload.

With a txv metered system, the situation is much different. Gas in the liquid line will flow much more quickly through the valve. As the evaporator pressure rises, the valve will close quickly, causing the needle to hammer into the seat. If this happens a lot (flash gas in liquid line), the txv will lose its control mechanism and act as a leaky faucet. Eventually, the compressor will be flooded by a burping evaporator and let out magic smoke. The magic smoke will stay in the sealed system and contaminate everything, causing not only the need for a new compressor, but a decontamination of the entire system. Auto techs call it "the black death".

(There are 2 pictures here. If none/one loads, refresh. My servers slow lol)
AC, the design changes I think affected me mostly because of my use of TXV which was sensing a colder vapor line on the HP version vs the AC version. This unit was originally capillary, so superheat was whatever they dialed it to. Now, this design variant probably isnt a universal thing, more of a per model modification. But, to demonstrate the differences found, heres a couple pictures.
This is the A/C version with the vapor line final passes across the front of the coil. This exposes the warmest possible air to the last passes and makes the most use of the evap. This is an example of a counter current exchange.
http://ryanbonigut.com/index.php?cmd...TFlMGQwYzMzMDQ

This is an example of the heat pump version of the evap. I fully suspect this difference was to improve heat performance. Since flow is reversed, this would make the last pass on the front of the coil instead of the rear, again creating a counter current exchange. The negative to this design is its a co-current exchanger in A/C mode.
http://ryanbonigut.com/index.php?cmd...Y0YTU0ZTM1YjEx

Both coils used the same number of rows, same fin per inch, but the plumbing was quite different. Both coils entered as 2 channels, split half way through, and exit as 4 channels. The differences are the location of the inlet and outlet ports, as well as the paths of the gas through the coils. Again, I found the HP version coil to not be nearly as efficient as an evaporator, as the AC version was. Im not sure if it was an actual coil performance difference, or the fact that I was unable to dial in my TXV properly with the HP coil.

Im not sure how far along everyone is with understanding superheat and subcooling. If Im talking greek to anyone who fancies building something like this, read it, learn it, understand it, or it will never be perfect! The heatpump coil I found having to set very high (20-25*F!) superheats to prevent hunting. The problem with that is it was severely starving the coil, reducing capacity, and I had no way to see if liquid was actually leaving the coil at low superheat settings causing the hunting as it flashed by the bulb.
The A/C evap, I was getting very good results at 5* superheat (with some hunting around 2-3* which is expected), but bumped it up to 10* superheat as my charge is slightly larger than the accumulator is designed for, the evap is a little smaller than Id prefer, as well as this unit may experience low ambient conditions. More of a safety factor at a slight loss of capacity to prevent compressor slugging. Its worth noting that the ambient to vapor line difference was only 5*F, so even at 57* it was doing a pretty good job, and might even be able to operate down in the mid-high 40's without defrost cycles. Really need to relocate my high side port into the liquid line to get a proper set on the subcooling now.

For the sake of anyone new to the topic or learning, couple key facts defined:
*Counter current - When your working fluids/gases flow in opposite directions. Your warmest air enters the opposite side of your cold gas inlet resulting in the exiting gas being almost the same temperature as your inlet air. Best way to get the most energy out of your system.
*Co-Current - Your warmest air meets the coldest gas. The result is both exit the other side as a average between the two. Your only getting a percentage of what the system is capable of producing.
*Vapor line - This is the gas line coming out of the evaporator after all the liquid refrigerant has boiled off. It is the larger fairly cold line.
*Liquid line - This is the condensed "room" temperature ref. flows through. This is a smaller, usually warm line.
*TXV - This is a temperature controlled metering device that controls the flow of liquid ref into a lower pressure vapor.
*Capillary - A cheaper static means of metering the ref. Diameter and length is carefully chosen to get the right flow.
*Superheat - How many degrees above boiling temp the vapor leaving the evap is. Superheat's purpose is to make sure all the ref has boiled off.
*Subcooling - How many degrees below condensing temp of the liquid ref is. This is to make sure there are no bubbles when reaching the metering device.

(There are 2 pictures here. If none/one loads, refresh. My servers slow lol)
This isnt from this specific model (it had a smaller version of this), but relevant for heatpumps/coil discussion. This is what we know as a Weld. Its a spun filter drier, a check valve, and 2 sections of capillary tube. The bottom of the drier is attached to the condenser, the open end of the captube to the evap.
http://ryanbonigut.com/index.php?cmd...WRjMDRjY2RlZWU
http://ryanbonigut.com/index.php?cmd...jRmMjM1ZDZhNzU

In a/c mode, it enters the drier, through the check valve, and through the longer length of cap tube. In heat mode, the flow reverses, entering through the long captube. The checkvalve is now closed forcing the gas to also pass through the shorter captube.

This design allows them to run a lower superheat in a/c mode, and a higher superheat in heat mode. This compensates for coil size, design, and temperatures. This may explain why they can get away with their coil designs, as most other brands dont have this neat little feature (split systems for the most part do have double metering, package are typically single fixed metering).