I bought a house and am thinking chiller.

Volvoguy87

I just bought a 3-family house.  The HVAC system consists of 4 Goodman 95% efficient 2-stage gas fired furnaces.  Each system also had an R22 10-SEER AC condenser and A-Coil.  The condenser and linesets were stolen by copper thieves prior to my purchasing the house.  I don't want to reinstall regular split system components because when the furnaces die, I want to install 1-pipe steam (yes, I'm crazy but that's for another thread).  I doubt the A-coils can be easily reused due to changes in sizing standards for them with the introduction of more efficient equipment and R410A refrigerant, so I'm looking at all new AC equipment.

I would like to look into a chiller.  The reasons for this are that I will be able to reuse the chiller should I switch over to fan coils and ditch the forced air systems.  Also, the 3rd floor of the house has no HVAC right now and I'd like to be able to use my new HVAC equipment to cool it.  I can run plumbing to the 3rd floor far more easily than I can install ductwork.  At this point, I haven't yet done my load calculations, but am just trying to learn about what's out there.

I'll need to know what equipment I'd need to integrate a chiller-based cooling system into my house, using the Goodman furnaces as air handlers.  Of course, I'd also need the controls to make it all work.  The furnaces have variable-speed blowers, and I'd definitely like to make use of that feature.  I don't know which manufacturers are making residential-grade chillers or which ones are superior to others.  Any advice would be appreciated.



Clear as mud?

Dave

Empire_2

Nice

While I'd like to address your scenario, my brain would probably explode.  Q:?  How would you bill each tenant for the Chiller usage?  I personally would stay with your 95% eff equipment.  Nice savings there.  Chilled water coils are available and can be sized for any application.  I guess the cost is probably what you will want to look at.  Either are probably going to cost a bit more than you think.....



Mike T.

Charlie from wmass

It depends

Dave where on the globe are you located? Cost tends to kill the chiller idea on most residential jobs. One pipe steam sounds great but why not two pipe vapor?

If you are with in say 200 miles of Western Ma call me.

Volvoguy87

Reply.

I am in Cincinnati, Ohio.



Why one-pipe steam?

I have 2 degrees in historic preservation, a great love of mechanical systems (the older, the better), and a equally great interest in energy efficiency.  1-pipe steam is relatively simple (simple like an aircraft carrier).  At least, I've come to understand and appreciate 1-pipe steam.  2-pipe is great, but I'm afraid of dealing with a houseful of failing steam traps in 10+ years.  I plan on using TRVs on 1-pipe steam radiators on a parallel-flow system installed in accordance with Dan's books and also venting to contemporary standards.  I wrote my undergraduate thesis on the merits of steam and hot water heating and my graduate thesis on learning energy efficient and sustainable building design and use from historic buildings.  I think I can get 1-pipe steam very efficient, with the best balance of operating cost to maintenance.  Additionally, 1-pipe steam involves less pipe, and may be an easier installation in preexisting construction.



For cooling, I (the landlord) would include the cost in the rent.

For heating, I'm sticking with the existing 95% efficient gas furnaces for now.  I'll go to steam in about 15 years, when the furnaces die.  Until then, why replace something that isn't broken? 



So far as cost is concerned; yes, chillers aren't cheap.  Fortunately, I can get HVAC equipment pretty cheap from the company for which I work.  I can get my hands on a 5-ton "Aqua Products" chiller for about $6,000.  The Aqua Products chillers use Coleman equipment.  I'm also looking into whatever else I can get.  I do have access to York products.  I would like to know which brands are good.



Should I have the chiller outside, or inside with a cooling tower outside?



So many variables.  This is fun

Dave

r. perry

chillers.......

I want to do something along the line of what you write about.  I have a 2 storey 1940 Colonial that I want to add a/c to.  Getting ducting down from the attic to the first floor or up from the basement to the second floor is almost impossible without major "surgery".  My initial idea was to install 2 air handlers, one in the attic and one in the basement and run both off the compressor outside.  Discovered this can not be done on a "traditional" freon system, so I began to investigate the chiller idea which would let me pump cold water/glycol/??? to each air handler, giving me a "zoned" system and no ducting between floors.

Found this company yesterday online: Whaley Products Inc., Burkburnett, Texas. 

They make several chillers beginning at 1.5 ton and up from there.  Wide choice of equipment, which I have to admit, much is beyond my limited experience or understanding, but with a few hours of research I think I could cobble together something that might actually work.

Looking at a 3 Ton SA Series Glycol chiller with integral pump tank.  I will try contacting the Co. soon.  Have no idea how "receptive" they might be to an obvious DIYer.

Harvey Ramer

I just installed a reverse cycle 5 ton chiller in a residential

application. It is sweet! It is also not cost prohibitive.



Contact me if you are interested in my experience with them.

NRT_Rob

there are several options

the Daikin Altherma can do heating as well as cooling and DHW in this app. all depends on water temps/output requirements of course.

Volvoguy87

Further along.

Thanks for the replies everyone!



I still haven't been able to get into the house to do my load calculations but so far as the philosophy of equipment and controls go, here is what I've got thus far:



Option #1, buy a packaged chiller

Option #2, buy a condensing unit and a refrigerant-to-water/glycol exchanger and build my own chiller.

Either system would also require the installation of a tank, supply, and return piping, and circulators.



I can get the equipment from the company for which I work at employee-discount rates.  A packaged chiller looks to be significantly more expensive than a condensing unit and a refrigerant to water/glycol heat exchanger.  The disadvantage of building my own is that I've got to do the math right and size each component properly in order to avoid having an ineffective, inefficient, and/or rapidly wearing system.



A challenge is figuring out how to control the system.  What I am thinking is that the chiller will have to operate based only on the temperature of the water within the storage tank.  Due to the varying loads inherent in having multiple independently-controlled zones, I think it would be best to have a dual-stage chiller in order to avoid short-cycling the compressor (remember, this is likely going to be about a 5-ton system).  If the tank's temperature is X degrees, then the compressor would have to switch to high speed.  Would I be correct to assume that I'd need a single-speed circulator on the loop between the tank and the chiller that would only run when the compressor is on?



I think I should have the furnaces (air handlers for this purpose) each on seperate loops with circulators.  The furnaces have variable speed blowers.  How would I control the circulator on each furnace's loop?  Would it matter if I used a single-speed circulator, or would be an advantage to having a variable speed circulator?  I assume the best thing would be for the circulator to run when the thermostat controlling the furnace/air handler calls for AC.



The fan coils on the 3rd floor will be on their own loop with a circulator.  Since the fan coils will most likely be independently-controlled, how should I pump the water through that loop?  I can use a single-speed circulator which would run all the time, or should I use a variable speed circulator which would vary its speed based on the temperature differential between the supply and returns on that loop (More volume when a greater temperture differential is detected, indicative of fan coils running, and decrease volume when the temperature differential decreases due to fan coils turning off or slowing down.)



So gloriously complicated!

Dave

NRT_Rob

various

doing your own chiller probably voids a warranty too.



you probably want to size your tank to ensure runtime. could avoid the need for multiple stage control that way.



variable speed circulators may not be helpful unless you're doing DT control, but ECM pumps will still save energy. if your electricity isn't dirt cheap and you expect a lot of duty time for a circulator it often makes sense to upgrade.



should be able to pull a signal from whatever is running the 3rd floor fan coil.

Volvoguy87

Control.

I've just not been able to get familiar with the water-side of things, I usually preach steam.  I've read Dan Holohan's new book on classic hydronics, but I don't know how much of it applies to cooling.  I don't even know who makes a good circulator, and yes I'd likely look into ECM motor circulators.  Where could I get my hands on information regarding pipe and circulator sizing?

For my 3rd floor, there will be a few fan coils.  If I had a variable speed circulator on that loop, could I vary its speed based on temperature differential between that loop's supply and return in the mechanical room?



So many variables,

David

Harvey Ramer

Well start with the chiller.

That will determine your distribution system. What kind of compressor setup are you going with? Single stage, 2 stage, varible speed there are many different configerations each with there own merits.

With the proper compressor setup you should not need a storage tank.

You should not use a variable speed pump on your chiller loop unless the compressor is also variable speed.

Start at one end and work your way through.

NRT_Rob

you may not need variable speed

but if the pump is going to run a lot, ECM can still save money.



have to check with MFG on chilled water though, some ECM are not ok with being used in chilled water.

Harvey Ramer

Depends how many BTU's you are moving.

I was going to go with an ECM on the setup I did. By the time I figuered the head loss through the H2O coil, the Chiller heat exchanger and the piping I changed my mind. I had to size for 12 gpm at 45 deg water temp with 30% glycol. If I opted to go ECM it would have required the Magna (grundfos) or something similar. Can't talk pricing but I chose the taco 2400-50 as my primary circ for a very obvious reason.

My requirements were 40' of head at 12 gpm. When I turned the circ on the first time the Lflow switch kicked once the temp got down close to 45. I removed a couple 90's from the pex line (i had put them there to make it look neat) and it worked fine then.

With the circ sized so close and using taco's permanant split capacitor motor I am hard pressed to believe there would have signifigant savings with an ECM circ. I measured the circ's power consumption and it was 396 watts.

I have yet to hear efficientcy ratings on ECMs. Don't get me wrong I love them and there capibility to match a load. I still say though, a properly sized 3-phase induction motor is the most efficient means of making things turn.

NRT_Rob

ECM

uses half the energy of PSC to move the same flow. 200 watts could be quite significant savings, depending on how often this beast will run.

Volvoguy87

Circulators.

I am trying to determine the best way to circulate water through 3 output loops and 1 loop through the chiller.  I am planning on using a tank between the chiller and the output loops in the hope that it may be more forgiving of varying loads.  I've never engineered anything like this before and I want to walk the fine line between too much capacity and too little.  And I absolutely do not want to short-cycle the compressor.  I think a tank would give me some buffer room in case I'm a little off on some of my math.



Each of the 3 output loops will be controlled independently of each other.  I assumed I'd use a small circulator on each loop.  I also assumed I'd use a small circulator on the loop between the tank and the chiller.  If there is a better way of circulating the water, I'll gladly listen.  Of course, I still haven't been able to do my load calculations or even begun to figure out pumping requirements yet.



Regarding 3 phase power, this is a residential application and there is no provision for 3-phase power on-site.  I'm guessing the capacity of the system will be around 5 tons.  I'm not trying to circulate the Pacific Ocean, just a comparative puddle.



Thanks for keeping up with the post,

Dave

Harvey Ramer

Rob, I hate to argue (actually I love it) but

I just checked which size Magna it would have taken to do the job.

MAGNA 40 - 120 F

POWER RANGE (MIN-MAX)

25 - 450 WATT

The specified pump would have to run at the max in a fixed speed mode to meet demand. This constitutes that the Magna would be drawing 450 Watts and the Taco is drawing 396 Watts to do the very same job. The Magna is a DC Induction motor and the Taco is a PSC motor. This would mean that there is signifigant power-loss through the Invertor on the Magna's end. I know it is typically said that you suffer a 6% power loss through such an invertor but quite a few engineers have said that 6% is quite a conservative number and does not hold true when it comes to actual testing.

The ECM's main claim to fame (in my humble opinion) is load matching capability and power savings deriving from such. The PSC motor will drastically drop in efficiency when it's speed is slowed while the ECM's retains most of it's efficiency at reduced speed. Therefore it is my opinion that there is little to be gained by going ECM for a fixed constant load. That is, as long as the motors are properly sized. Especially considering the cost differences between the 2.

I am more interested in 3-phase double squirrel cage induction motors controlled by a VFD. These motors can produce over 90% torque before the shaft even starts turning. I will take my hat off and salute to the first manufacterer to implement this motor on a variable speed compressor.



PS,

Volvo guy. A VFD (viarible frequency drive) will take 1-phase 240V and turn it into

3-phase  at a specified voltage and hertz to optimize motor efficiency. It features a soft start, minimal power loss and optimal performance.

NRT_Rob

maybe you've got the wrong pump?

maybe something is different at the larger sizes I'm not familiar with, but at least on the smaller ECM circs this is readily apparent from the pump curves, and we've verified with on site power monitoring.



you are maxing out the pumps I'm looking at though, maybe that's why, 40 feet of head is pretty extreme. maybe the rules change at the edges of the performance curve on the bigger pumps?

Harvey Ramer

The way I see it is that when both motors are operating at max capacity

there is not a signifigant difference in power consumption. But as soon as the load drops the ECM instantly gains the advantage.

NRT_Rob

I hear what you're saying

but with the smaller ECMs that simply isn't true. with the alpha, for example, you match a 15-58 gallon per gallon at half the watts at max output... 45 watts vs 87 watts on the 15-58 for approximately the same output. ON TOP of that you get modulation savings in multizone savings.



I'm not sure why that seems different with this particular example though. but then, I'm not familiar with that taco pump either. I suppose it could be different if you were in the sweet spot for a PSC motor and on the hairy edge for the ECM, but I can't defend that with rigor.

Harvey Ramer

Rob, I think I know what is a large contributer

to this phenomena. The impellor design. Admittedly my requirements on this particular application are unique. 40' of head at 12 gpm. The 2400-50 has a relatively steep curve. It only goes up to 55 gpm. This tells me that the impellor has a sharper twist design hence giving the motor more leverage. The 40-120 Magna has a flatter curve. 40' of head and up to 120 gpm. This would mean that the impellor has a more relaxed twist hence it would torque the motor more at high head low gpm.