The Homemade Heat Pump Manifesto

[Vlad]

I found this easy explanation about TXV function.

HVAC/R Everything You Need To Know About TX Valves

When it comes to TXVs they are very complicated devises but they make your system run more efficient and safer for compressor. Most of them are adjustable so, to change your superheat you don't need to open system again (cap tube case, what a pain) but just adjust it.

Nobody asks you to know all functions and all details about txv. You just need to know how to select one and how to adjust it.

[Ko_deZ]

Hi guys.

I have been very busy for a long time now. Regarding Arduino stuff, I said earlier that I could help out if someone needed it. I have yet to receive any questions, but still AC said that there was little help to be found in this area. I would like to contest that. Please let me know what you need, and I will mock up a circuit and some software that will do what you want. I am pretty sure that I can make anything you need in a very short time. This is what I do for a living at the moment, embedded systems. For others to get ideas, I suggest posting the requirements here. Something like:


Three temperature inputs, T1, T2 and T3
- Readout to serial interface every second
--- T1 is indoor temperature
--- T2 is outdoor temperature
--- T3 is accumulator tank temperature
Three relay outputs, R1, R2 and R3
- R1 goes on 20 seconds before R2, and stops 20 seconds after
--- water pump for ground loop
- R2 goes on when T1 <= 40C and off when T1 >= 50C
--- Used for starting/stopping compressor
--- Never run for more than 30 minutes
--- Minimum pause of 10 minutes if the full 30 minutes are reached (deicing of ASHP?)
- R3 comes on if R1 is running and T3 is not increasing (driving alarm buzzer)

(did not bother to figure out how to format bullet-point lists here, sorry)

This is all just random stuff i thought up right now, you need to figure out what you want. If you want to start a pump for solar collectors at a given temperature, use the solar collector as source for the heat pump, trough solenoid valves want to divert waterflow to a outdoor water HX to get higher input temperature when outdoor temperature is higher than the ground loop etc. All of this is quite easy logic, and I believe I can make software that will be easily extendable and understandable. The circuitry will be very simple indeed. Temperature measurement is a matter of connecting a tri-leg transistor like thingamabob to ground, +5V and one of the arduino inputs. If you need negative temperature, you must add a diode. Output is a resistor, transistor and relay. It is all very simple, I am certain all of you will be able to do it, just let me know what you need.

-Morgan-

[AC_Hacker]

Originally Posted by Ko_deZ

Regarding Arduino stuff, I said earlier that I could help out if someone needed it.

Morgan,

This is really great!

I would say that one of the most basic requirements would be a delayed compressor ON relay, that would protect against intermittent voltage events. So, it would be a delay with a duration of about 2 minutes after current ON, before the compressor relay switches on. I see SSRs for sale cheap on ebay, theoretically one would last for a long time. I am not sure of the efficiency of a SSR, though. Some solid state power devices leak current in the 'off' state, some don't deliver full current in the 'on' state. What is your opinion here? Is there a way to protect electro-mechanical relay's contacts from degradation due to arcing?

Another feature would be in the case of a water-source heat pump. It would need to monitor the temperature of water going from the ground loop into the evaporator HX... If the water gets too close to freezing as it is coming out of the the HX (this would depend on anti-freeze percentage), it will freeze and quite likely damage the HX.

Another would be a way to monitor over-pressure in the compressor, an do an emergency power-down if over-pressure occurred.

Your other ideas look very good, too.

-AC_Hacker

[Ko_deZ]

AC, that should be easy enough to do. Sparkfun carries some current measurement devices, but what kind of pressure sensors would you use? Any other ideas?

Regarding SSR and electro-mechanical. The SSR does have a voltage drop across it. Ususally SSRs have something like 1.5V drop. Pulling 1A will give you 1*1.5=1.5W heat loss. Not a whole lot, but with more current you will get more wattage lost, and potentially a need for cooling. On the other hand, the EMR will not have any of this. If you switch one cycle every hour of every day for a year, you have done 8760 cycles in a year. Take for instance this:
LINK
It has an electrical life of 100000 operations at max rated load, 10 million without load, so somewhere in between if you do not pull full power. That should last the lifetime of the unit. If not, you could always change to a SSR later if it does not work well. It is not expensive to replace. I would go for EMR.

Temperature measurement of the loop water will be easy enough to fix.
Seems like a state machine design is brewing here.

[MN Renovator]

If you are protecting against an intermittent voltage event, can't you just set something simple up on the low voltage side, if you lose power, just cut the thermostat connection for 2 minutes. It would save a ton of hassle, wouldn't it?

[AC_Hacker]

Originally Posted by Ko_deZ

AC, that should be easy enough to do. Sparkfun carries some current measurement devices...

It probably won't be necessary to actually measure current to get the two-minute pause to work correctly, although it could be very useful for performance monitoring.

Originally Posted by MN Renovator

...can't you just set something simple up on the low voltage side, if you lose power, just cut the thermostat connection for 2 minutes. It would save a ton of hassle, wouldn't it?

The compressor will still need a high current power device either SSR or EMR. I think that the way to do it is to make sure that the power source that runs the compressor will also run the Arduino. And have an initial two-minute wait loop whenever the Arduino powers up, before the Arduino proceeds to the other tasks.

Regarding the exact type of pressure sensor (switch?) I'm not so sure. I do know that they are included in commercial equipment. I also don't know what the over pressure limit we would want to set as a limit for R-290. I'm hoping BradC and/or Vlad might have some information for us on this issue.

Also, BradC indicated that in calculating superheat, that there is a way to do it with just temperature measurement. Perhaps there is a way to protect the compressor from over-pressure that only uses temperature measurement, too?

Originally Posted by Ko_deZ

I would go for EMR.

Sounds good to me. Do we need to include a diode or cap to prevent contact arcing?

-AC_Hacker

[BradC]

Originally Posted by AC_Hacker

Regarding the exact type of pressure sensor (switch?) I'm not so sure. I do know that they are included in commercial equipment. I also don't know what the over pressure limit we would want to set as a limit for R-290. I'm hoping BradC and/or Vlad might have some information for us on this issue.

Also, BradC indicated that in calculating superheat, that there is a way to do it with just temperature measurement. Perhaps there is a way to protect the compressor from over-pressure that only uses temperature measurement, too?

Nope. Please use a HPCO on the discharge of the compressor.
Something like these : Search Result Automatic Controls : Detail | Saginomiya Seisakusho, Inc.

They are about $20 list at my local fridge place. Do not make it a software function. It must knock out the compressor.

These are even better, as they must be reset manually : REFRIGERATION HIGH PRESSURE CONTROL : HLP530ME | eBay

Available pretty much everywhere and fully adjustable.

Your HP cut out should be sized for a fault condition only. Because I condense at 26Deg C, I set mine at ~180 PSI, which is about 40 Deg C. On my Air cooled unit, the HPCO is set at 360 PSI, which is about 65 Deg C for R22 (which the unit was designed for).

Pick an appropriate safety margin and aim for that. Use the closest matching fixed switch you can find.

You really should pressure test your high side to about 1.3 times the pressure the HPCO is set at.

For example, mine is set at 40C. On a stinking summers day here, with the system not running and in the direct sun, the ambient is going to rise _above_ 40C. I could see potential pressures above 55C in the system with it sitting idle. So you need to account and test for your worst possible case.

[Vlad]

I do agree with BradC about hardware HPCO and LPCO.

These MUST be specially design switches. They are SAFETY switches. HPCO saves your system from overpressure and rupture (running compressor will blow itself into pieces). LPCO will cut off in case of lost of charge. HPCO must be below designed hi side pressure(you can find it on name tag or somewhere else,it depends how "creative" designers were).

Usually for R22 systems HPCO is just bellow 400 psi. To connect them you just put them (or even some more safety switches) in series with contactor coil.

Also it is always a good idea to add 5 min time delay relay for restart. They are usually 5-10$. Most of modern thermostats have time delay function built-in.

For water coil frost protection you can use remote bulb thermostat. You can use old style or digital. It can be added to the same circuit with other safety switches in series.

Here is really cool digital thermostat with all bells and whistles you can only expect from thermostat. Price is also cool.

A419GBF-1C - Johnson Controls A419GBF-1C - Single Stage Digital Temperature Control (24v, SPDT)

If you want to be more sophisticated you can add extra "software" control. But it must be separate from basic safety switches.

I wanted to bring another issue - plumbing. Because we use water source HP we have to protect water side in case of refrigerant leak into it. In case of refrigerant leak high pressure will blow water circuit. Also if your water circuit is connected (even if you put isolating valve) to a potable water system, special back-flow preventer must be added. This is a part of plumbing code and you have to check your municipality what kind of valve must be added.

[Vlad]

AC, I might missed it. Do you use just a plain water in your ground loop? If so, do you have a reason for using water and not antifreeze?

I am curious, because in case of power failure your water pump will stop pumping immediately compressor will do the same but remaining refrigerant will keep evaporating and freezing your evaporator.

[Vlad]

Originally Posted by BradC

See, in Australia we are lucky. LPG automotive fuel is an unquantifiable blend, but BBQ Propane is over 95% Propane.

In any case, properly measuring the P/T of what is in your bottle will tell you a _lot_ about the purity of what is in there.

I'm investigating the fractional distillation of a Propane/Butane mix. The thing that is bolloxing me at the moment is the US has a habit of adding Methanol to propane. I'm trying to come up with a useful way of separating that, but that means I actually have to pollute Propane with Methanol to try it, as our Propane is dry enough not to warrant that level of pollution.

I take our BBQ propane and circulate it through a filter/drier a couple of times and I have as close to R290 as you can mostly get. No wonder they call Australia "the lucky country".

I was thinking to charge my system with R22 and after switch it to propane. But after reading more about propane use as a refrigerant I will try to charge my system with propane first.

The choice is to get enviro-green-$$$$-propane or get BBQ botle and make your own. In Canada (I guess US is the same) there is no difference between BBQ propane and car fuel propane. If I am wrong please correct me.

What we have is mostly propane some butane some non condensables and some other hydrocarbons mixed as liquid.

To get to R290 as close as possible we can try to separate butane from propane. This two gases have very different boiling points. Butane Boiling point is −0.5 Degree Celsius. Propane Boiling point is -42 Degree Celsius.

In Russia many people use propane/butane mix for cooking especially in country side. Usually propane bottle stays in cold shed and not inside the house. In winter time butane becomes a real pain because you have half of the bottle full (propane is gone) but you can't use it.

If we cool bottle down to -5-10 Degree Celsius and start pulling propane as a vapor (bottle valve up) with recovery unit then propane will start evaporating (and it will help to keep bottle cold) but butane will stay as a liquid. Also this will keep water in original bottle as well.

We will get some non condensables but it is easy to get rid of them by purging and charging as a liquid.

We will not be able to get pure propane like R290 but it will be as close as possible and not that hard.

And yes we need to use filter/dryer during the process.

Please guys let me know what you think about this.