[Solar air conditioning, ice storage] Sources for DC refrigeration compressors?

[Photon939]

I would like to install solar panels on my house at some point but I want to avoid the expense and bad deals for grid-tied solar power systems.

I work with HVAC equipment regularly for my job and have experience with custom refrigeration setups, brazing and refrigerant handling. I have found small compressors available for the camper/RV market but they aren't particularly cheap considering their small cooling capacity and poor efficiency.

Example: http://www.ebay.com/itm/Rotary-DC-Tiny-Compressor-for-for-Refrigerator-Air-Conditioner-R134A-150-300W-/181433530201

They are ideal as they directly run on DC (brushless motor) with speed control (capacity modulation based on available solar power)

The next step up is to use a regular variable frequency drive for AC induction motors with a normal compressor. but most AC compressors are not designed to be speed or capacity controlled and may not lubricate properly running at lower speeds. I would also need to keep the input voltages happy to the VFD and limit power output based on available sun.

Ideally I could use peak sun to generate ice to provide cooling at night. This technique is used in some large buildings to reduce peak energy demands and get cheaper power. In my case it would be making better use of solar generated power when I'm not likely to be at home anyway.

[Someone]

You could probably do better than water as the storage fluid in the tank if you're going to build a double circuit system, and consider reversing the heat pump for heat storage in cooler months (nothing quite like a heated bathroom in winter). Ask around some farmers or agricultural suppliers, variable speed drives are used extensively for water pumping and this moved to solar years ago (you don't see new windmills anymore).

[Photon939]

You could probably do better than water as the storage fluid in the tank if you're going to build a double circuit system, and consider reversing the heat pump for heat storage in cooler months (nothing quite like a heated bathroom in winter). Ask around some farmers or agricultural suppliers, variable speed drives are used extensively for water pumping and this moved to solar years ago (you don't see new windmills anymore).

I have a Teco N3-205-C 5HP VFD that I can try to use for this purpose - I guess I can just keep an eye out for smaller 3 phase compressors on ebay.

I do plan on installing a reversing valve for reverse cycle operation in the winter but the reason for water is that the phase change from water to ice allows for a much greater energy storage capacity for a given volume of water (144 btu/lb) vs just 1 btu/°F/lb for sensible heating/cooling. Refrigeration COP drops with higher temperature delta.

According to the label it's looking for a minimum of 200vac so I would need somewhere between 200-282vdc to keep it happy.
A quick glance at some solar panels it seems most have a VMP around 18v which puts me at about 12 panels to achieve that voltage. Should be fairly doable. I'm more concerned about DC arcing once you get up to those voltages though. Circuit breakers usually can't handle much voltage on DC so I may have to get either special ones or high voltage fuses to protect the system.

[Someone]

I do plan on installing a reversing valve for reverse cycle operation in the winter but the reason for water is that the phase change from water to ice allows for a much greater energy storage capacity for a given volume of water (144 btu/lb) vs just 1 btu/°F/lb for sensible heating/cooling. Refrigeration COP drops with higher temperature delta.

Its an interesting optimisation problem, you can get more storage in the same volume by switching to propylene glycol antifreeze as the additional range has greater energy than the enthalpy of fusion of water, so your COP would need to plummet for the water to be able to store more energy. Cost/Environmental restrictions of course favour water

It looks like most agricultural applications are using DC pumps and not worrying about the speed they run at, so you can scratch that idea. Sad to say but with wanting to run the compressor at a constant speed (a good thing for COP) you're probably headed for a battery or capacitor storage between the panels and the inverter. Masterflux don't offer much detail on their inverters but that could make a neat 12/24/48V system with a MPPT and a capacitor bank offering a minimum on time of 5 minutes.

[Photon939]

It looks like most agricultural applications are using DC pumps and not worrying about the speed they run at, so you can scratch that idea. Sad to say but with wanting to run the compressor at a constant speed (a good thing for COP) you're probably headed for a battery or capacitor storage between the panels and the inverter. Masterflux don't offer much detail on their inverters but that could make a neat 12/24/48V system with a MPPT and a capacitor bank offering a minimum on time of 5 minutes.

Perhaps I should try cutting open a standard reciprocating compressor and see how fast it needs to run before the oil slinger starts working. I could also possibly use a bank of standard electrolytic capacitors to get a bit of energy storage, then program the MCU to do short bursts of higher speeds to make sure it stays lubricated.

On the refrigeration side I could use some solenoid valves with capillary tubing of different lengths on each for capacity control and keeping head pressure up when the compressor is running at lower speed.

So the refrigerant circuit would look something like this:

Code: [Select]

Compressor - Condenser < Solenoid valve 1 - Longer captube  > Evaporator (refrigerant to glycol heat exchanger) - Compressor
                         Solenoid valve 2 - Shorter captube         

[johansen]

not all compressors use oil slinger. some of them have a capillary tube set up to let oil flow directly into the cylinder, using the pressure difference across the compressor to push oil through the tube.


anyhow i think you might have better luck winding a 3 phase, 4 pole coil (easier to wind than a 2 pole motor) and using a 3 phase vfd.

unfortunatly you'll have to weld the compressor back together.

or you might be able to find a small 3 phase compressor from a window mounted air conditioner.. but that might be hard to find and or expensive.

[Photon939]

not all compressors use oil slinger. some of them have a capillary tube set up to let oil flow directly into the cylinder, using the pressure difference across the compressor to push oil through the tube.


anyhow i think you might have better luck winding a 3 phase, 4 pole coil (easier to wind than a 2 pole motor) and using a 3 phase vfd.

unfortunatly you'll have to weld the compressor back together.

or you might be able to find a small 3 phase compressor from a window mounted air conditioner.. but that might be hard to find and or expensive.

So far I found this on ebay http://www.ebay.com/itm/DANFOSS-COMPRESSOR-220V-3-PHASE-SU21G-R134a-8161-NEW-/180862228354

I have not found any scroll or rotary compressors smaller than around 2 tons which is a bit too large. This Danfoss recip seems about the right size. (and for a decent price)

Also going back to my previous idea, I could maybe use a low speed PWM with a liquid line solenoid valve instead of having multiple valves for capacity control.

[NiHaoMike]

You can drive a "single phase" motor as an asymmetrical 2 phase motor. You'll have to measure the voltages and currents on the windings (with the original run capacitor) at full design load to get the correct ratio to drive the motor with. You can do some fancy trickery with FFT as done in the Shannon Liu Quadrature Drive, but simple V/Hz scaling might work well enough.

For an energy storage device to work around issues at low input, a common deep cycle battery will work nicely. Use the top 10% of the SOC range and it will last years. For even longer lifetime, get a few battery cells used for hybrid or electric cars, although finding just a few for a decent price might prove a challenge. (Or take the whole pack and look forward to doing even bigger alternative energy projects!)

[Photon939]

Modern refrigerators are VFD based, so theoretically they should run on DC. If not, remove the PFC module.

I haven't seen any refrigerators in the US yet that have a variable speed compressor, got any examples? Mini-split air conditioners are pretty much the only place I've seen inverter based compressor capacity control. Well, that and commercial >100 ton chillers.

Most capacity control which is now becoming commonplace for air conditioning or refrigeration is based on cylinder unloading (reciprocating) or partial scroll bypass valves for scroll compressors. My parents got a Trane system installed a couple years ago which has a 2 stage scroll compressor, 1.5 and 3 tons of capacity respectively.


I was able to do some preliminary testing with some equipment I had laying around. (VFD connected to mains) They are all single phase 240v devices.

Neslab CFT-33 laboratory chiller
Capacity: 950 watts or ~3200btu/hr
Compressor: CSIR type with current based start relay which engages a start capacitor
Pump: positive displacement 1/3hp, has centrifugal start switch, RSIR type
Reservoir: Plastic, 1.5" pipe cap access
Notes: compressor does start and run with sufficient power but due to the start relay the compressor is stalled until approx 30hz and ~7A of current which engages the start capacitor.

Unknown manufacturer mini chiller
Capacity: approx 100 watts or ~315btu/hr
Compressor: RSIR type with PTC thermistor starting device
Pump: positive displacement, 1/4hp, has run capacitor, PSC type
Reservoir: Stainles steel, large panel access
Notes: compressor starts easily at low speed but once started must cool down for several minutes before the PTC device will allow starting again

[Photon939]

You can drive a "single phase" motor as an asymmetrical 2 phase motor. You'll have to measure the voltages and currents on the windings (with the original run capacitor) at full design load to get the correct ratio to drive the motor with. You can do some fancy trickery with FFT as done in the Shannon Liu Quadrature Drive, but simple V/Hz scaling might work well enough.

For an energy storage device to work around issues at low input, a common deep cycle battery will work nicely. Use the top 10% of the SOC range and it will last years. For even longer lifetime, get a few battery cells used for hybrid or electric cars, although finding just a few for a decent price might prove a challenge. (Or take the whole pack and look forward to doing even bigger alternative energy projects!)

I have considered this as well but the commercial VFD I am using does not have any provisions for driving phases at different voltages. I don't want to delve into the mess of making my own VFD for this project. (or buying one)

Small battery storage for starting power (and other uses) would be nice but I would like to avoid having a 250v string of batteries if possible. Ideally there would be no voltage regulation on the output of the solar panels to the VFD; I should be able to set up the panels to keep the open circuit voltage below the maximum input voltage.

[coppice]

Modern refrigerators are VFD based, so theoretically they should run on DC. If not, remove the PFC module.

Only a small number of high end fridges are VFD at this time, although its certain to grow. VFD is far  more common in air cons.

[Photon939]

Modern refrigerators are VFD based, so theoretically they should run on DC. If not, remove the PFC module.

Only a small number of high end fridges are VFD at this time, although its certain to grow. VFD is far  more common in air cons.

Really? Well, seems I'm too optimistic.

I'd be interested in reading any documentation you have on a VFD refrigerator system if you have any

[Seekonk]

I can see the wives as everyone starts tearing into that new refrigerator.  I do store cold in my fridge.  I live 5 months off grid each year running fridge on solar panels only during the day and operating on the edge of freezing.  Car battery (from vehicle not taken) is used only for start surge current, the rest of the power is real time.  The secret is a large mass of liquid to get through the night.  It is surprisingly successful.  Excess power makes hot water for showers etc. Haven't had to turn on the propane water heater in years.  Everything operates at power point.  Amazing what you can live on if a little thought is put into it.

[Photon939]

Just to add a quick update to the thread, I contacted a local solar installer and quoted me about $13K all said and done (after tax incentives) to install a 6.5KW grid tie system.

Since PA has net metering laws I can generate excess during the day and as long as I consume it all at night, get paid the full rate including distribution etc for the excess power generated during the day.

Currently looking into ways to get the cost of the system down. There is a local company in NJ selling panels at great prices ($0.60/watt) and will see if they will install them instead.

[grifftech]

use a dc motor and an automotive compressor.

[grifftech]

 

use a dc motor and an automotive compressor.

  I forgot to add a belt and something to mount the motor and compressor to.

[johansen]

Just to add a quick update to the thread, I contacted a local solar installer and quoted me about $13K all said and done (after tax incentives) to install a 6.5KW grid tie system.

Since PA has net metering laws I can generate excess during the day and as long as I consume it all at night, get paid the full rate including distribution etc for the excess power generated during the day.

Currently looking into ways to get the cost of the system down. There is a local company in NJ selling panels at great prices ($0.60/watt) and will see if they will install them instead.

if i buy about 3 to 5 grand worth of panels, can get them for 35 to 50 cents a watt, but would have to pay for shipping. if that price includes the shipping then 60 cents a watt might be fair game, depending on how many panels you buy.

But when things are too good to be true, they usually are.

can your grid tie offer recharge batteries and operate without the grid? if not, i wouldn't bother.

the reason is simple, if the utility plus solar plus grid tie = pay off within a dozen years.. something is too good to be true, why wouldn't the utility buy their own solar panels and inverters?

ah, the tax benefits.. yes. well, screw someone else's kids with an immoral tax, deception and debt backed monetary system. anyhow, if none of that is involved, whats stopping you from getting off the grid?

[johansen]

use a dc motor and an automotive compressor.

bearing and seal losses.

just cut off the top of a standard single phase compressor and rewind the motor for 3 phase. (low voltage, if warranted)

several of the compressors i have torn apart had no epoxy or anything in the windings. you should be able to fit more copper in it than you pulled out.