The Homemade Heat Pump Manifesto

[AC_Hacker]

Originally Posted by bigsmile

...why the refrigerant out temperature on the low-side is even higher than the water that heats it?

You have asked a heck of a question there. In fact it sent me back down to the basement to check my readings and yes, I get the same kind of readings again.

I'm afraid I'll have to defer to expert opinion on that one...

It does appear that I may be violating the laws of Thermodynamics.

As I have said before, 'I am not an HVAC technician, in fact I am barely even a hacker'.

Originally Posted by bigsmile

Same question for the high-side. It also looks to me to be too much super-heating and sub-cooling going on, which will increase the lift and thus reduce the efficiency.

I'm in the process of getting that info right now...

Originally Posted by bigsmile

Perhaps a TXV can help?

Yes, tweaking the cap tube can also help, but it's not so easy as turning a screw. I really need a TXV on my heat pump (NOTE TO ALL: TXV is good), if for no other reason than to learn about refrigeration and modifications... I even have one but I think it's for a slightly larger unit than I have. I need to see if it will work.

Originally Posted by bigsmile

According to what I read, txv can control super-heating. And am I right in thinking that super-cooling can be controled by amount of refrigerant charged into the system?

That's what I understand, also.

Originally Posted by bigsmile

Once again, I'm really new in this and not sure about what I said.

It is possible that your modesty has too much subcooling.

I shall return...

-AC_Hacker

[AC_Hacker]

Another find:

I would never recommend that anyone ever experiment with and much less actually use R-290, due to the hazard involved with working with an extremely flammable gas, added to that is the fact that it is against the law in many parts of the world, but I did come across this interesting paper called Propane as R22-Replacement in Commercial Appliances.

For educational purposes only, of course.

...and here is a P-T chart for R-290.

Again, for educational purposes only.

Regards,

-AC_Hacker

[bigsmile]

I really mean it when I said I'm not sure. I read about refrigeration a little, but it's all theory. No one really told me much about consequences of these theories, and I have to figure it out myself. I really think about the refrigeration process a lot recently and I think I have figured a few things out. For example, am I right in thinking that if you oversize the exchanges, leaving the compressor undersized, you are less likely to run into trouble (other than a little higher pressure drop at the exchanges perhaps), but if you have the one or both of the exchanges undersized, that will cause trouble? This is the kind of thing that I think is correct, but never get a confirmation from someone who really knows it.

The reason that I have to figure these things out myself is because I didn't get answer when I asked these questions on "DIY" forums. The people there pretty much tell you "DIY? No way", citing insurmountable "technical complexity" that can only be mastered by professionals. They never bother to discuss in much detail about these technical difficulties, like sizing the components of a system.

This kind of negativity, and in some cases, to some degree, hostility toward DIY is not restricted to AC. I met similar atmosphere in a solar energy forum.

[AC_Hacker]

Originally Posted by bigsmile

The reason that I have to figure these things out myself is because I didn't get answer when I asked these questions on "DIY" forums. The people there pretty much tell you "DIY? No way", citing insurmountable "technical complexity" that can only be mastered by professionals. They never bother to discuss in much detail about these technical difficulties, like sizing the components of a system.

This kind of negativity, and in some cases, to some degree, hostility toward DIY is not restricted to AC. I met similar atmosphere in a solar energy forum.

You are not alone in having experienced this. I did too. And so have some of the others who post here.

That is why I'm doing this. I don't have all the answers, either. But I'm learning, and I'm having fun doing it.

Originally Posted by bigsmile

For example, am I right in thinking that if you oversize the exchanges, leaving the compressor undersized, you are less likely to run into trouble (other than a little higher pressure drop at the exchanges perhaps)

Well, my thinking is less along the lines of getting into trouble, and more about efficiency. I've noticed that the minisplits (which are highly efficient) have huge heat exchangers.

I can imagine that there may be a problem with imposing such a large load on a compressor that it can't cool itself properly.

Originally Posted by bigsmile

...if you have the one or both of the exchanges undersized, that will cause trouble? This is the kind of thing that I think is correct, but never get a confirmation from someone who really knows it.

I don't know the answer yet...

It might be time for you to fire up a tourch and build a unit. You already know more than enough to start.

It gets lonely out here in cyberspace, trying to figure this stuff out all by myself... I need help.

Regards,

-AC_Hacker

[bigsmile]

Originally Posted by AC_Hacker

Well, my thinking is less along the lines of getting into trouble, and more about efficiency. I've noticed that the minisplits (which are highly efficient) have huge heat exchangers.

I can imagine that there may be a problem with imposing such a large load on a compressor that it can't cool itself properly.

I don't know the answer yet...

My thinking is that if both exchangers are oversized, then both can handle the capacity that the compressor runs at, thus, they may not be exactly the same size but neither will be tested to find out who's the boss. But if the compressor is oversized, that means eventrally, one will have to be way too super-heated of sub-cooled to catch up with the game. If the evaporator is undersized, it may freeze up. If the condensor is undersized, it may not be able to cool down the compressor effectively (actually, I just learned this from reading your post).

Originally Posted by AC_Hacker

It might be time for you to fire up a tourch and build a unit. You already know more than enough to start.

It gets lonely out here in cyberspace, trying to figure this stuff out all by myself... I need help.

Regards,

-AC_Hacker

Now I have a problem, I don't have a refrigerant recovery machine, and it looks expensive. DIY is possible but may be too time consuming for the moment. Plus it's not really a pressing matter for me (my current HVAC system is inefficient, but works OK), so I may not start to work very soon.

[AC_Hacker]

Fred,

Here's the rest of the response to your last post...

[NOTE: bigsmile notice some errors in my original post, which have been corrected (I hope)]

Originally Posted by Fred_Fredowski

Have you measured the dT in the water going through each heat exchanger? I thought that you mentioned that the sump pump was pumping about 2 gpm through the earth loop, so with a 5000 BTU/hr compressor I would expect about 5 degrees of dT across the low side heat exchanger. Is this what you see?

I did a very careful flow rate test, I stuck an empty gallon jug under the hose that had the water flowing through the low-side (ground loop side) of the heat pump, and timed it with my wrist watch. It took 25 seconds to completely fill the jug, so my flow rate is actually 2.4 gpm.

[* I did notice on the McQuay PDF that the suggested flow rate for ground loops is about 3 gmp/Ton. Since my little heat pump produces under a Ton, I should be able to reduce the flow rate by about half *]

I ran another heat pump test today and logged three periods at 10 minute intervals, with the first test being at 10 minutes from start-up.

Originally Posted by Fred_Fredowski

Have you measured superheat (SH) and subcooling? I think that if the SH is high, you might be able to resize the cap tube to reduce the pressure drop. This should increase the low side pressure and increase the COP. If the SH is not high and the water flow is up around 2 gpm then I would look at the low side heat exchanger and see if that is working efficiently. Many new heat pumps use a large outdoor coil/fan which it allows them to operate with lower lift even using air as the coolant. I would think that a water cooled unit like yours could work even better provided that the heat exchanger is efficient enough.

So if I understand, to find out what the level of superheat is, I focus on the low-side (AKA: suction side) of the refrigeration circuit. First I measure the low-side refrigerant pressure by using my manifold gauge. In the case of my first test, I saw that the low-side pressure was 48 psi. This measurement is then referenced to a PT chart that describes the temperature of vaporization for a given pressure. So for instance, on the PT chart for R-290, that I recently posted, if I look in the chart for the pressure of 48 psi...

...I see that the temperature at which vaporization occurs is 26.6.

So I subtract the measured temperature from the derived temperature like this:

superheat = T(lo-side rfg. out) - T(lo-side PT chart)
superheat = 53.1 - 26.6 - = 26.5

So it looks like the super heat is 26.5

And going even further out on a limb, to calculate the subcool, I focus on the high-side of the refrigeration circuit, so I look at the high-side pressure gauge and I see that it reads 157 psi. I take this value and look again to the R-290 PT chart, and I see that there is no entry for 157 psig...

...so I do a little red-neck interpolating and estimate that since the pressure entry fall about halfway between 155 and 160, that the corresponding temperature value will fall about halfway between 92.01 and 94.19, so without further hesitation, I 'eyeball it' and write down 93 degrees. Next I measure the actual temperature of the refrigerant that is leaving the High-side HX at a point on the refrigerant line, a few inches from the HX refrigerant out port. The measurement I get is 77.5. So I subtract the measured temperature from the derived temperature like so:

subcool = T(hi-side PT chart) - T(hi-side rfg. out)
subcool = 93 - 77.5 = 15.5

So the subcool is 15.5

Does this agree with the method you are familiar with?

Here are the test results...

TEST #1 @ 10 MINUTES:

T(lo-side water in) = 62.4
T(lo-side water out) = 56.4
dT(lo-side water) = 6

T(lo-side rfg. in) = 30.4
T(lo-side rfg. out) = 62.7
dT(lo-side rfg.) = 32.3

T(hi-side water in) = 71.9
T(hi-side water out) = 73.2
dT(hi-side water) = 1.3

T(hi-side rfg. in) = 90.5
T(hi-side rfg. out) = 75.8
dT(hi-side rfg.) = 14.7

P(lo-side) = 48
T(lo-side PT chart) = 26.6
T(lo-side rfg. out) = 53.1
superheat = T(lo-side PT chart) - T(lo-side rfg. out) = 53.1 - 26.6 = 26.5

P(hi-side) = 157
T(hi-side PT chart) = 93
T(hi-side rfg. out) = 77.5
subcool = T(hi-side PT chart) - T(hi-side rfg. out) = 93 - 77.5 = 15.5

TEST #2 @ 20 MINUTES:

T(lo-side water in) = 59
T(lo-side water out) = 56
dT(lo-side water) = 3

T(lo-side rfg. in) = 28.6
T(lo-side rfg. out) = 62.7
dT(lo-side rfg.) = 34.1

T(hi-side water in) = 81.1
T(hi-side water out) = 81.7
dT(hi-side water) = 0.6

T(hi-side rfg. in) = 97
T(hi-side rfg. out) = 80.3
dT(hi-side rfg.) = 16.7

P(lo-side) = 44
T(lo-side PT chart) = 31.06
T(lo-side rfg. out) = 52.7
superheat = T(lo-side rfg. out) - T(lo-side PT chart) = 52.7 - 31.06 = 21.1

P(hi-side) = 187
T(hi-side PT chart) = 105
T(hi-side rfg out) = 89.1
subcool = T(hi-side PT chart) - T(hi-side rfg. out) = 105 - 89.1 = 15.9

TEST #3 @ 30 MINUTES:

T(lo-side water in) = 62.4
T(lo-side water out) = 57.5
dT(lo-side water) = 4.9

T(lo-side rfg. in) = 28.1
T(lo-side rfg. out) = 64.2
dT(lo-side rfg.) = 36.1

T(hi-side water in) = 87.4
T(hi-side water out) = 89.3
dT(hi-side water) = 1.9

T(hi-side rfg. in) = 109.8
T(hi-side rfg. out) = 85.6
dT(hi-side rfg.) = 24.2

P(lo-side) = 40
T(lo-side PT chart) = 18.9
T(lo-side rfg. out) = 52.5
superheat = T(lo-side rfg. out) - T(lo-side PT chart) = - 52.518.9 = -33.6

P(hi-side) = 210
T(hi-side PT chart) = 113.8
T(hi-side rfg. out) = 97
subcool = T(hi-side PT chart) - T(hi-side rfg. out) = 113.8 - 97 = 16.8

Regards,

-AC_Hacker

[bigsmile]

AC_Hacker,

I think on the low side it should be super heat, and thus it's sub-cool on the high side. There seems to be a little too much super heat. My theory is that the cap tube you use were for the air to air AC, which was designed for a higher lift. When you use it on your gshp, it kept this higher lift, and thus will resulted in higher super-heating or sub-cool depending on the charge.

By the way, I think I was wrong in my previous post to say that super-heat or sub-cool on one coil is for "catching up with the game". It should because one or both exchanger cannot handle the capacity of the compressor, so this capacity has to be restricted by the metering device. This will increase the pressure ratio, and thus increase the temperature difference between the two coils. Depending on the charge, one or both have to have too much super-heat or sub-cool.

[AC_Hacker]

I just came across a thread on Extreme Overclocking that follows a project on autocascading chilling of CPUs.

Near the end of the thread was this photo of the main control panel:

...I think I've just gotten a glimpse of the future.

Hackers can be craftsmen too.

-AC_Hacker

[AC_Hacker]

Originally Posted by bigsmile

I think on the low side it should be super heat, and thus it's sub-cool on the high side.

I went back and checked my proceedure document, and you are correct, thanks for bringing it to my attention.

Originally Posted by bigsmile

There seems to be a little too much super heat. My theory is that the cap tube you use were for the air to air AC, which was designed for a higher lift. When you use it on your gshp, it kept this higher lift, and thus will resulted in higher super-heating or sub-cool depending on the charge.

Yeah, I suspect that the exchangers are about 2x oversized. Today, I'm going to reduce the refrigerant pressure and run another test, I'll let you know how it goes.

Regards,

-AC_Hacker

[bigsmile]

I wonder if you are talking about reducing super-heat. If that's the case, you should increase the charge. If you reduce the charge, the pressure of both sides will reduce, so the saturation temperature on both sides will decrease. This means the low side refrigerant in temperature will be even lower below the ground loop water temperature. This means even higher super-heat.

So, basically, if you reduce the charge, there will be higher super-heat but lower sub-cool. And if you increase the charge, you will reduce super-heat but increase sub-cool. To reduce both super-heat and sub-cool, it can be achieved by increasing the flow rate through the cap tube and adjusting the refrigerant pressure at the same time. This actually means that by using ground loop, your heat pump can work at a higher capacity (higher flow rate) then it used to.

Again I need to add that I have zero hand on experience working with refrigeration, and all what I said is from what I think should be the way it works. So, I'm not really sure about it.

Originally Posted by AC_Hacker

I went back and checked my proceedure document, and you are correct, thanks for bringing it to my attention.



Yeah, I suspect that the exchangers are about 2x oversized. Today, I'm going to reduce the refrigerant pressure and run another test, I'll let you know how it goes.

Regards,

-AC_Hacker