One problem with a timer and tied to a heat pump water heater could be controls. The timer will basically cut power to the unit. Every time it powers up, the GE Geospring goes through startup tests sequence. A couple minutes of heat pump run followed by up to 20 minutes running on straight resistance heating and then finished off with heat pump to setpoint temperature. It may also reset to factory default of 120* setpoint every time. In this scenario, you'd lose the benefits of a heat pump water heater.

Just one more thing to note when looking at the GE HPWH as the finishing tank, you have to be careful with the finishing tank entering water temperature.  Talking to support at GE if your doing any recirculation or preheating (via DSH and buffer tank) and the entering water gets higher than the tank setpoint, the HPWH controller will go into error mode because it thinks there is a problem with its temperature sensors.  To prevent this you have to either be certain you will use the water faster than your DSH alone can heat it from ground temp to near the finishing tank setpoint, OR to be safe, set the GE setpoint higher than the high limit on your DSH circulation pump, that's probably 125 or 130 deg F. 

Posted By jokin on 16 May 2012 01:22 PM


Since we are heating dominated, I came up with the following for effective “heating season” COP for a HP water heater……

Effective Overall HPWH COP = HPWH Rated COP / (1 + (HPWH Rated COP / Space Heat COP) )

Example… Hot Water: GE Model HPWH COP 2.35 has one of the highest published/rated COP.

Space: ClimateMaster Heat Pump has rated heating COP of 3.5 (adjusted to include pumping penalty)
Effective HPWH COP = 2.35 / (1+ (2.35 / 3.5 ) ) = ~ 1.4

If the above is correct, assuming a heating dominated climate and ignoring cooling side benefit and tank loses… savings over a standard electric water heater is less than 30% (1 - ( 1 / Effective HPWH COP ) ). 
-----------------------------------------------------------------------------------------------------------------------

http://welserver.com/WEL0267/ I think your number are a bit off since you account too much for pumping penalty. Plus the COP of 2.35 is there at the end of the tank heat, and but it starts heating at a lower tank temp. I have added a link to WEL 267 which now is in A/C mode and mkaes DHW via the hydronic portion of a WF Synergy 3D. I added the COP to the second graph from the top (blue line), so next time it makes domestic hot water we see the live COP. It includes the circulation pumps for the loopfield. One thing which might throw it off a bit is the EWT being already in the mid 40s. Keep also in mind that it goes through a secondary heat exchanger, so the leaving load temp is higher than the DHW temp.

Don't confuse COP with EF (Energy Factor). The GE's EF is now 2.4 (up from 2.35, a minor quibble). More important is the distinction between COP and EF. EF is the net result of COP and standby loss. Typical EF for a good conventional 50 gallon ERWH is around .93, in other words, standby loss amounts to around 7%.

My conjecture (educated guess) is that the GE HPWH has tank insulation as good as the rest of tank heaters but suffers a bit of conductive loss owing to various stuctural connections and penetrations related to the compressor section mounted on top of the tank. If for arguments sake the thermal losses amount to 10%, then the COP would be 10% higher than the EF, so perhaps around 2.64.

All of these numbers arise from the DOE WHAM test protocal which requires, amoung other things, 135*F exit water, 58*F inlet water, 67.5* ambient air, 64.3 gallons water drawn in 6 equal increments (10.7 gallons each) spaced an hour apart. Since HPWH interact much more with the ambient air, any actual use deviations from the test protocol will have significant effect. In particular, lowering tank setpoint really helps - bringing it down to 120 ups COP by another 10% or so. OTOH, cooler ambient temp significantly drops COP.

Draws significantly larger than 10 gallons will trigger operation of resistance elements if the HPWH is in Hybrid mode - we start all ours out in pure heat pump or e-mode. In that mode, resistance elements only operate if ambient drops below 45*F, but recovery time is longer.

80 gal GE HPWH is due out in fall - I'm curious as to price and whether it'll have a higher capacity compressor section.

BTW I tried one of the competing HPWHs - the 60 gallon unit marketed under various brands (Whirlpool, Kenmore, State, AO Smith), but I WON'T do that again. It seems to work well enough but is much noisier than the GE unit.

Posted By jokin on 16 May 2012 10:50 PM
Just one more thing to note when looking at the GE HPWH as the finishing tank, you have to be careful with the finishing tank entering water temperature.  Talking to support at GE if your doing any recirculation or preheating (via DSH and buffer tank) and the entering water gets higher than the tank setpoint, the HPWH controller will go into error mode because it thinks there is a problem with its temperature sensors.  To prevent this you have to either be certain you will use the water faster than your DSH alone can heat it from ground temp to near the finishing tank setpoint, OR to be safe, set the GE setpoint higher than the high limit on your DSH circulation pump, that's probably 125 or 130 deg F. 

I guess that takes SDHW out of the picture. Solar messes up controls for some tankless heaters as well. BTW, I believe the GE HPWH is a product completely designed and built in China. I remember seeing a paper on it a few years ago which was basically an RFP to get different designs. This is where the industry has gone.

http://welserver.com/WEL0267/

It made the DHW with a COP around 4. Keep in mind that the EWT is already around 45,and the entering load is anywhere between 108F and 118F (setpoints). It actually went down to 95 due to a couple long shower. Again, it is all about the lift and using efficient loop pumps. Temp sensor are calibrated, so is the amp sensor. It is calculated with the heat extracted plus the compressor heat minus the power factor. It includes the wilo stratos loop pump drawing about 100 watts.

Today it started making hot water at a COP of 3.8 and went down to 3.4, with an average COP around 3.5. The difference appears to be due to a reduction in EWT right after start up, and then an increase in load temperature once the tank gets warmer. Calibrated all the sensors again with a calibrated digital needle thermometer via P/T ports, so everything is within 1/10 of a degree F and thus within 5% accuracy. This does not include the standby heatloss of the tank, just the COP by what the water is made by the system.

docjenser -  
just to clarify my calculations were for winter operation of a separate air source heat pump water heater assuming my personal home's space heating COP with open loop geo system and large pump penalty (well pump), as well as assuming  that the COP of the ASHPWH was equal to the the posted GE EF rating (which as "engineer" explained is not the same).   Some other test have shown COP as high as 2.75 when using a 120 setpoint and ensuring a 65 or warmer ambient space temperature.   

As for entering water temperature being lower, do you mean the entering domestic water ?  Unless I'm okay with large temperature variations in my hot water, or create much hotter water than I need, won't my water heater typicall working on (heating) water that is within 10 or 15 degress of the setpoint. 


Side note for what its worth...   During summer months when I am using air condition often, another way to get more hot water capacity with my ASHPWH (until I get the circ pump to the bufffer tank installed) is locking the unit into "eheat mode" (HP only mode) and then driving the setpoint up, until you find the point at which you don't run out of hot water.  The additional ASHPW run time adds little or no cost as it is mostly or entirely offset by the reduced runtime of my space cooling WSHP.

The value of most of the appliances we discuss is geographically influenced.

"The additional ASHPW run time adds little or no cost as it is mostly or entirely offset by the reduced runtime of my space cooling WSHP"

Color me skeptical - increasing the setpoint of the ASHP drives down its efficiency (and, perhaps, its durability).

Considered another way, meeting cooling load with the ASHP instead of WSHP is MUCH less efficient - think about it - ASHP is rejecting heat into 120...130...140...150 degree water, however high you set it. The WSHP is rejecting heat into the ground at a much lower temperature. ASHP operation will offset some, maybe 1/3-1/2 of WSHP operaion cost.

Okay fair point, color me optimistic on that one! :)    I stand corrected.   There will be a penalty, the compressor work set by the "lift" (leaving to leaving temperatures) is going to be more with an airsource heat pump water heater... it's not a straight up comparison though,  because you are using different types of refrigerant and compressors, but I think it is safe to say the HPWH will be more energy per unit of heat extracted from your space. 

Maybe a more accurate point would be as follows:  If you are normally okay with your hot water set at some temperature below the max setting on your unit (140 for my GE model), during the summer (air conditioning) months, raising your setpoint (when in "heat pump only" mode) is a cheap, an easy way to increase your hot water capacity.

By the way....

1.  I think it was in this thread that someone mentioned the GE unit production being in China.  I don't think that is the case anymore, as of early this year production shifted to a new/refurbished plant in Lexington, KY. 

2.  Some have mentioned that the GE units switch back to default temperatures and runs standard elec heat for 20 minutes after power is restored, which means it wouldn't work well with a timer.   Twice now the power has gone out, and it seems like my unit (GE HPWH unit built in September of 2011 doesn't do this).  The heating mode and setpoint remain correct as soon as the power is restored.  Maybe the power has to be off for more than a few hours??    Maybe there are different versions of operating software ?

I have not noticed a 20 minute electric element operation interval or loss of setpoint upon power out

As far as start up sequence & 20 min resistance, that was something I saw, I beleive, in the GE manual? Perhaps if it does exist it is an intial startup only, or it may have been a different model or year.

Posted By jokin on 30 May 2012 12:51 PM
Maybe a more accurate point would be as follows:  If you are normally okay with your hot water set at some temperature below the max setting on your unit (140 for my GE model), during the summer (air conditioning) months, raising your setpoint (when in "heat pump only" mode) is a cheap, an easy way to increase your hot water capacity.

Increasing your setpoint will increase your water capacity, I am not sure if the words cheap and easy are correct, since your efficiency and stress on the compressor will decrease with a higher setpoint (load) temperature. This is in contrast to the desuperheater function or output, where the higher temperature does not come with a penalty.

Some follow up info. I ended up with a GeoSpring 50 gallon unit purchased in June 2012. This was the Sears floor model and was on sale for $700. Add on $350 utility rebate, so net cost of $350. On Thanksgiving, showing off my set-up to guests, I note the blinking lights on the display. Heat Pump Failure, sealed system requires service. I was in e-heat mode (heat pump only, no elements) up until then, but it failed to electric heating. In hindsight, I did note that it was running for a large percentage of the time. Called Sears, they sent a new evap coil and dryer. Installed a couple weeks later. Attached photo shows my energy spike in the month of November. Ouch!!!!! An extra $200 for the heat pump trying to compress an open system for heat.

It is very hard to imagine a failed heat pump water heater responsible for an extra 40-50 kwh / day. The compressor on a GE HPWH draws 500-600 Watts, likely less if refrigerant charge low, and I gotta figure there is some sort of loss of charge compressor interlock.

The delta between compressor vs resistance operation should be no more than 10 kwh / day

Curt,

With your pointing out the wattage of the compressor, it does seem unlikely that it was the compressor responsible for the full cost increase. I am unable to guess at what else could be responsible for such a large change and without electric resistance heat and normal household operations, maybe it is the more obvious: a meter read error. It might make more sense if November was high by 1000 kWh and December was low by 1000 kWh. That could make much more sense with the HPWH faulting to the 4500 watt elements for large portions of both months. I guess I will see how January does.