Posted By Dana1 on 15 Jun 2017 10:45 PM


I wouldn't exactly call it a mini-split fervor, only trying to counter some broad brush statements. After 7 pages this thread has drifted into some side alleys, to be sure.

I initially got sucked down mini-split alley by anon's assertion that...

................ "Mini splits have pretty much destroyed the geothermal market."

That's an assertion I find utterly preposterous! Even if SOME people are indeed buying mini-split and satisfied with them (even in substantially-sized houses), it's not really what's putting pressure on the GSHP market.

Making broad assertions about where the different technologies are appropriate or not invites a rebuttal when existence proofs to the contrary abound, but it wasn't my intent to try to take over the thread or sell you on a mini-split solution. Phrases like "... at best a retrofit..." or "...the only refigeration/compressor technology available that..." invite the counterexamples, as does docjensers' bizarre assertion that " Mini splits will destroy grid reliability..." .

I'm not trying to convince you that these are better than a GSHP solution for you, only that the GSHP chauvinism expressed isn't necessarily well founded, and mis-statements of fact about other solutions aren't really warranted.

So I moved this to a new thread here since the subject got us off topic.

So lets discuss this, and I ask everyone to chime in here.
"as does docjensers' bizarre assertion that " Mini splits will destroy grid reliability..." ."


Let me throw out a couple math numbers here:

I live in New York, we have cold winters. We need to electrify the heating sector to get rid of carbon emissions, and the total heating load is around 100 GW (Giga Watts). So if I satisfy the peak load with a COP of 4 for a GSHP or a COP of 1.5-2.0 during extreme colds, to satisfy my heating load I would need about 25 extra GW of generating capacity if I use GSHPs, or 50-75 GW if I use ASHPs during peak times.

No, to put things in perspective, the entire generating capacity in NYS is 38 GW, including reserves, and our current base load during the winter is around 17 GW, and 25 GW being the all time winter peak.

So here is the question: Where do I get the extra 25 GW of electric generating capacity from to run ASHPs during winter peak hours, which occurs in the morning without the sun shining for our future solar panels? Add to that the load from future electric car charging, and you have the perfect storm for grid instability. If you can explain to me how to satisfy the enormous electrical peak load coming from air source heat pumps in the peak of the winter, under peak load conditions, you indeed can call my assertions bizarre. I encourage everyone to chime in here .....

Try throwing out a couple of references to where those numbers come from, to start with.

It's only a "perfect storm for grid instability" if one presumes that the grid operators, the state regulators, and utility managers are all stupid, and that the grid management and electricity market revisions already under way simply aren't happening. This is a fictitious straw-man argument about a world that doesn't (and won't) exist. There are several dubious underlying assumptions about what the grid resources will be like by the time 100% of the space heating is supplied by heat pumps, starting with the notion that the total generation capacity will still be 38GW. It's not going to go to 100% overnight, or even in 20 years, and it's not happening in a vacuum- grid operations & resources are already evolving rapidly. The way the straw-man argument is framed presumes crude 19th century type grid management practices which are already long since dead & buried. (The NYISO would already be having multiple severe summertime peak blackouts every year were that the case.) Dispatchable demand response works, and dispatchable grid storage is able to take it even further. PV +grid storage in the form of distributed batteries is already cost competitive with peak power pricing today, and will only be cheaper (a LOT cheaper) in the years to come.

By the time 100% of the space heating is supplied by heat pumps, a number of things will have happened.

A: A large fraction of the transportation sector will also have been electrified, which has two significant consequences to the grid and grid management:

------1> The total generation capacity on the NYISO will have to expand significantly from current levels to supply the additional transportation load, as noted.

------2> A large supply of used automotive batteries batteries will be available on the cheap to be repurposed as grid storage resources (on either side of the ratepayer's meter), as is currently being done in Europe with Nissan Leaf batteries.

B: The efficiency of buildings new & old are economically upgradable (and will upgraded), lowering the statewide peak space heating load by many GW.

C: Under the rapidly evolving electricity business landscape under the NY Reforming Energy Vision (https://rev.ny.gov/ ), grid storage and demand response programs are becoming more economic, not less. As the space heating grid load expands, so can the peak management resources. Neither happens overnight. It's highly likely that under the REV, in addition to "smart charging" demand response, car to grid power will be remunerated favorably during those peaks.

D: Some amount of grid storage or smarter load management will likely be necessary just to manage the 2.4 GW offshore wind NY is already committed to, with more to come as it becomes cheaper than combined cycle natural gas. Offshore wind is already competitive with existing incumbent generators in Germany & Denmark, and will likely be competitive with NY/NE wholesale pricing before 2025, maybe even by 2020.

E: The cost of offshore wind has been cut in half in just the past 2 years, and this party is just getting started in the US. The 2.4GW (operating at a probable capacity factor in the mid to high 40s, or even low 50s) is just the tip of the iceberg. Offshore winds are significantly higher in winter in this region than in summer, and the wintertime capacity factors are likely to be well over 50%, higher than most combined cycle gas plants in NY. By the time 100% of space heating is heat pumps, that 2.4GW number is likely have moved up an order of magnitude, as the cost of that power continues to drop. During Polar Vortex events, the large temperature difference between the land and ocean will drive a significant near-shore wind, even in the unlikely event that there are no other weather patterns driving the wind, guaranteeing that at least a decent fraction of that nameplate capacity is available.

The NY REV is targeting a 23% reduction in building energy use by 2030.

The NY REV is targeting 50% of all grid power to be from renewables by 2030.

There's no way heat pumps will account for even 50% of the (now lower) heat load by then, but a substantial fraction of the flexible resources necessary to support 100% heat pump space heating will already be in place by then.

Somewhere near the bottom line, the "extra" 25GW of peaking capacity can/will simply be purchased or rented for less money than the installed price difference of GSHP/ASHP (in many, or even most cases) in the form of building efficiency, demand response, increased wind generation capacity, and grid storage. The additional grid resources will also be more flexible & valuable than simply the higher COP of GSHP too, since the latter only has a very high value less than 10% of the time, maybe even less than 1% of the time by the time heat pumps.
.

I'll be happy to provide references, please be specific to what you wold like to have references for?

No grid operators, state regulators and utility managers are not stupid. But some of them underestimate the negative impact of ccASHP peak demand on the grid, partially because manufacturers and lobby groups are not forcing about the efficiency of ccASHP during winter peak demand times.

Right now, if ccASHP would have 10-12% market share, and they add about 8 GW to the peak load in the winter, NYS would switch from a summer peak to a winter peak. GSHP having about 250% better efficiency during peak demand scenarios, not so much. Not a fictitious straw man argument (not sure why you get down to the name calling level) but simple math. What regulators forget is the huge amount of demand response needed and the duration (whole night?), not just couple peak hours.

Right now NYS suffers from a growing disparity between decreasing grid utilization, and an increase in peak demand. Grid utilization went from 58% down to 52% in a 10 year period, meaning higher rates and more power generation at idle, while peak demand continues to increase.

Yes, grid storage might be cost effective compared to peak summer demand prices, on a level of MW, but not gigawatts! It is simply inefficient for a grid, and very costly, to cover huge demand peaks occurring only a few hours per year,
The best demand response is the demand you do not create. The follow up costs for the grid are enormous. And last time I checked the costs of the grid are paid by the ratepayer.

A) 1) Indeed, a vast majority of the transportation sector will be electrified, and you argue that we need a much larger generating capacity anyway. But both the heating and the transportation sector will be additive loads, making it more challenging.
A) 2) Used car batteries might help to respond to the demand (Nissan leaf batteries being bad examples, since they are being replaced since they faded too much too quickly), but newer batteries likely hold up longer. But again, demand response will have enormous costs.

B) Yes, buildings can be upgraded, but we all know that the older building stock is sometimes difficult to upgrade, and also has enormous costs associated with. So even if we can shed of 30 GW of peak load with upgrades, should the remaining 60-70 GW of peak load being satisfied at a COP of 4 or 1.5? That assumption of building up-grades was already assumed in my 25 GW additional power needed for ccASHPs versus GSHP I posted above.

C) Yes, I am very much aware of the REV. Yes, car to grid has many opportunities. But again, the large load is the issue, and the longer duration during the night. We are not just talking a couple hours here, and we are talking many GWs. And those cars usually charge at night, and have 50% higher energy usage during the winter, and lower capacity. It is all part of the perfect storm.

D) Yes, unreliable wind and solar will replace reliable fossil fuel and nuclear, one more reason why peak demand will be a much bigger issue in the future, because it forces you to keep that combined cycle natural gas plant at idle 99% of the time, just in case the wind does not blow on the coldest night of the year.

E) Yes, you can have 2.4 GW at 50% on average, but it can be 0% during a peak time. You cannot argue with average anymore, this is peak, and grid stability is about peak loads!. Even if it is 10 times as much, 24 GW, there is still a one in 5 years event where it is close to zero. Now what do you do? New transmission lines to get it from elsewhere? At what costs? We were hardly able to cope with the current winter demand during polar vortex time in 2014 and 2015.

All your points miss the fact that less than 1% of the cars are electric, and less than 2-3% of the heating sector energy use is done by heat pumps. While electric cars and trucks use about 50% more load on average in the winter, the difference between average load heating demand and peak load heating demand can be 250%. Combine that now with a reduction of COP of at least 50% by ccASHPs during peak, and the size of the challenge becomes more obvious.

Specifically in NYS, state regulators still dream about a reduction in demand (more LED light balls). Not sure if you are right about your notion that heat pumps will not account for 50% or more of the heat load by 2030, they have a target of 40% emission reduction as a state mandate by 2030, and they cannot get there without banning fossil fuel in the heating sector. Especially in NYS it is economically thesible, due to a large portion still using oil or propane.

Put what you are saying in front of utilities (and yes, they are smart) let them run the numbers, and they let you know the costs for grid upgrade and power generation, demand response and grid storage will be much more expensive than the price difference between GSHP and ccASHPs.

So with ccASHPs, the appearance of a bargain is simply an illusion!

Yes, in 2017 less than 1% of the cars are EVs, and less than 3% of homes are heated with heat pumps. So what? By the time 50% of homes are heated with heat pumps it's more than likely that 50% of new cars will be EVs, and the availability of cheap grid capacity in the form of second-life EV batteries will be large, and vehicle-to-grid generation capacity will be remunerated.

The heating and EV loads to the grid are only additive in terms of total energy, not peak power. To keep from blowing up the neighborhood transformers with multiple EVs charging, smart charging is FAR cheaper than upgrading the local grid capacity to manage the potentially huge load. Cars don't need to be charged all at the same time, or during a peak grid load. They can be charged any time, as long as it doesn't interfere with their use, and the smart charger for managing that has long since been invented.

Those smart chargers are also a ready-made demand response resource, as a load that can be interrupted during peak grid loads without disrupting the usefulness of the EV. They are also a ready made load dump for excess variable renewable power that enables high penetrations of rooftop PV and wind without excessive curtailment. That same load dump characteristic can also be used for ancillary grid services such as frequency & voltage control. But for not a lot of additional money EVs can also become distributed peak load generation sources. Making that all happen is a matter of adjusting the regulations so that EV owners can be remunerated for those demand response, frequency/voltage control, and peaking power services.

The smart-charging solutions that solve the grid congestion & throughput problem for charging EVs also solves the generation & grid capacity issues for heat pumps. Vehicle-to-grid and stationary battery-to-grid adds some cost, but it's not super-expensive now, and will be a lot cheaper in less than a decade. PV + batteries have already won contracts in technology neutral open bidding for peaking resources in Arizona on cost alone. Both PV & battery technologies are on double-digit percentage cost learning curves, and PV (even distributed PV) will have a levelized cost of energy lower than the NYISO average wholesale power price by 2030. By the time EVs are populous enough to be a problem, the will be populous enough to be a big part of the solution, if not THE solution. (There are many other grid storage and peak management technologies being deployed.)

A recent EV-2-grid experiment being done in CA came up in today's GTM blog: https://www.greentechmedia.com/articles/read/bmw-and-pge-prove-electric-vehicles-can-be-a-valuable-grid-resource

Yes, more generation is going to be needed to support the EV fleet, but it need not be large central power stations or wind farms feeding the grid. In fact it's better if it's small scale and distributed, to avoid having to increase the capacity of all the wires & transformers. PV is cheap and scalable, and only getting cheaper, but it won't be the total solution either.

Then nature of offshore wind in this region is such that it's literally never zero during space heating peaks, and only rarely during air conditioning peaks. Large differences in ocean & land mass temperatures such as those seen during the 99% design conditions guarantee some amount of wind within a few 10s of miles of shore, so an offshore wind farm would be able to safely bid some fraction of it's output into capacity markets. ERCOT (Texas) doesn't have capacity markets, but analysis from 2 years ago demonstrated that near-shore wind (wet foundation or dry) on the Gulf coast was cost competitive with imported wind from west Texas on energy pricing alone despite having a conspicuously lower annual capacity factor, due to that guranteed temperature differential driven wind capacity during peak air conditioning load periods. In NY/NE that effect occurs more dramatically in winter than in summer, but its there. And average offshore wind speeds in this region are much higher in winter than summer, which is a "perfect storm" for being able to make up for the diminished average wintertime output of PV and lower EV efficiency with wind + distributed storage (EVs or stationary.)

There are no one size fits all silver bullet solutions, but there are many tools in the kit that are available and affordable now, and will only become more so as deployment & manufacturing volumes increase. The utilities and all REV stakeholders are aware of it, and the specific solutions will evolve as different price thresholds are crossed, regulations updated, and as more variable output resources go onto the grid. This is a soluble problem.

I'm sure GSHP will be part of the solution too, particularly in existing buildings where envelope upgrades are more expensive than paying for the higher efficiency. Mini-splits and ducted ASHP will be too, and have a significant efficiency advantage over pre-existing resistance heating, but much less of an operational cost advantage over oil or propane heating than they used to, and that will slow deployment.

At recent years' lower oil pricing the incentive for replacing oil-burners for ASHP has pretty much gone away, and in some places propane too. While fossil liquids price volatility is a given, the competition from EVs and heat pumps puts downward pressure on oil pricing, as does the low price of natural gas. This puts a natural economic delay in the ramp-up of heat pump load on the grid, but that can turn around quickly if world oil supplies get tight. But the past 2-3 years have completely undershot nearly all analysts projections for oil demand increase. Some of that has been slower than anticipated economic growth in China, but Chinese policy support for EVs is also part of it. The anticipated large growth in oil demand in India is now not going to happen, since under the new administation there all light trucks and cars sold in India are required to be EV (or at least zero emissions) by 2030. With no demand drivers, shrinking demand in the developed world and much slower than previousy anticipated demand growth in emerging markets, the return to $100/bbl oil isn't likely to happen soon, barring a major war breaking out in oil-country, and #2 oil & propane pricing may even show further down-price volatility. Unless (as has been proposed by draft legislation in MA) there is a major push to de-fossilize space heating by 2030, there is far less drive for ramping up heat pump deployment than there was a few years ago.

With enough economic incentives to smooth demand, people will install thermal storage with to-water heat pumps. And/or maintain supplemental NG or propane heat.

There are lots of commercially available solutions to grid capacity, grid stability and grid peak power demand problems. In the evolution to "electrification of everything" those will be deployed. While thermal storage is cheap per kwh, it's has greater value to the grid as a load dump for excess generation and for frequency control than it does for the total energy capacity grid storage. Cheap demand response using electric hot water heaters has decades of real-world use & data behind it, and while similar things COULD be done with hydronic heat and larger buffer tanks, the additional capital cost of the heating system isn't likely to be the best capital expenditure in terms of keeping the grid stable & reliable. A $2000 tank + $5000 of heat pump could buy a LOT more demand response & load dump and offset peak power if the $7K were spent on $70 wi-fi controllers for 1000 electric water heaters (or even a wall mounted Tesla battery + inverter.)

The total bulk energy problem in the electrification of everything world is solved cost effectively with wind power and distributed PV. The peak power, and grid capacity problem can be solved by distributed PV and distributed dispatchable storage, since that is (in most cases) cheaper than upgrading the existing capacity of the transmission & distribution grids to handle the capacity requirements if the electrification of everything world. The capital cost of adding 50GW of peaking capacity from batteries (stationary or automotive) is far cheaper than the cost of 50GW of new centralized power generation plus the expanded grid capacity to get it all to the load even at today's prices for PV & batteries. (Which is how Con Edison is saving the ratepayer over $1 billion on not having to rebuild & upgrade the Brooklyn Queens substation: http://www.utilitydive.com/news/the-non-wire-alternative-coneds-brooklyn-queens-pilot-rejects-traditional/423525/ ). So even though the distributed PV won't be putting out during peak heating loads, the amount of battery capacity necessary to keep the grid stable with a mostly renewables grid when a decent slice coming from distributed PV would have peaking capacity equal to or larger than the distributed PV itself, and enough to cover the peak heating grid load problem.

The amount of readily exploitable area for distributed PV (without covering vacant acreage) is huge. NREL did a technical study last year that mapped the available existing rooftop area suitable for solar, which determined that rooftop solar at current PV efficiency could cover something on the order of HALF of all electricity use in some states at current consumption levels. Their estimate for NY was that it would be about 25% of all electricity sales, at a DC rating of about 30 GW. See Table 3, p26 ( p38 in PDF pagination):

http://www.nrel.gov/docs/fy16osti/65298.pdf

That's just existing suitable rooftop, not solar farms, not community solar projects, and not solar-covered parking lots, just rootops. Clearly not all of that will get built, but as PV becomes ever cheaper (it's cheaper than grid-retail in NY right now, even without subsidy), by the time even even half of that rooftop PV gets deployed it becomes a severed distribution grid management problem (worse than what Oahu is seeing today), but a problem that can be cost effectively solved by smart water heaters, electric vehicles & smart chargers, and dispatchable stationary batteries. If the REV regulators and NYISO grid operator do their jobs, the grid mix will balance the higher winter production wind power against diminished PV output with a minimum amount of he peak power available from the EVs & batteries will likely be more than sufficient to solve the heat pump (all types) peak load problems (both winter & summer), to minimize the amount of bulk storage (pumped hydro & battery etc) costs. These are all moving targets, and the order of what/how-much gets deployed first & where will continue to evolve, but a lot of smart eyeballs are already on this problem inside the organizations charged with delivering the stable end result.

A perhaps larger scale problem is the amount of very long term seasonal storage needed, which will vary from local climate to local climate. A recent bit of analysis what it takes to get to 80% renewables had this graphic comparing the mix required to minimize the seasonal storage requirements/costs for Germany vs. California:

https://climatepolicyinitiative.org/wp-content/uploads/2017/04/Figure-3.x-Box.png

Clearly NY will have different optimal mix than either, but it's still a big problem clearing the hurdle to 100% renewables in the electrification-of-everything world with lots of possible solution, most of which are still under development. There are numerous aspiiational proposals out there, all of which have flaws, but it's worth considering the longer term picture too. A discussion of one widely discussed 100% renewables energy piece was discussed in this weeks GTM Energy Gang podcast:

https://soundcloud.com/the-energy-gang/the-bitter-fight-over-100-renewables

and...

https://soundcloud.com/theinterchangepodcast/the-changing-market-rules-for-energy-storage

See also: https://www.greentechmedia.com/articles/read/mind-the-storage-gap-how-much-flexibility-do-we-need-for-a-high-renewables

if the $7K were spent on $70 wi-fi controllers for 1000 electric water heaters (or even a wall mounted Tesla battery + inverter.)

Such a 90% discount certainly helps :-).

Does space heating/cooling water thermal storage make more or less sense than batteries and a conventional air-air heat pump? Let's see, 5000 gallons can store > 200 kwh which would require ~$30K (COP=3) in Telsa batteries. Compare to perhaps +$15K for 4x larger, to-water heat pumps and water tanks. Then consider the limited lifetime of batteries (perhaps $50K in batteries over 20 years). Thermal storage wins at about 1/3 the price ($.01/kwh).

I was just visiting a hot country with $.45/kwh electricity. Makes PV solar plus storage for AC a not so theoretical question.

Of course comparing Tesla wall-mount batteries at 2016 or 2017 pricing isn't very relevant to the discussion of what happens when 50% or 100% of space heating in NY is from heat pumps, but the dispatchability of those batteries is more valuable to the grid than thermal storage, and is economic if allowed to participate in capacity and LMP markets. Thermal storage is only dispatchable to the house behind the meter, and only for thermal loads. Batteries are dispatchable to the grid for any use, which makes energy stored in batteries more valuable than energy stored in tanks, or as ice, especially if it's under the grid operator's control.

Batteries of all types are on a steep cost learning curve, and by the time most homes & buildings are on heat pumps and most transportation is electrified, the surplus of second-life transportation batteries will be dirt cheap & plentiful- it's already a business in Europe. Vehicle-to-grid is coming to a grid near you in the next 15 years too.

Thermal storage for AC utilizing ice takes advantage of the heat of fusion of water to store a lot more energy per lb or per cubic foot, and thus takes less space & weight than tanks. This is already a commercially available product (for large or small scale AC) in the US, and valuable where time of use rates or demand charges apply, as well as expensive electricity island grids. Since it s agnostic of the source of the power, behind the meter PV would be fine. With 45 cent electricity it becomes a no-brainer type investment in a cooling dominated climate. The company that recently released the first residential scale ice storage cooling product is Ice Energy:

https://www.ice-energy.com/

https://www.ice-energy.com/residential/

http://www.puretemp.com/stories/phase-change-matters-newsletter-oct-14-2016

http://www.pennenergy.com/articles/pennenergy/2017/06/energy-storage-genbright-and-ice-energy-partner-to-reduce-peak-electricity-demand-on-nantucket.html

https://www.ice-energy.com/wp-content/uploads/2016/03/ICE-BEAR-20-Product-Sheet.pdf


https://www.ice-energy.com/ice-energy-horizon-solar-power-complete-solar-plus-ice-storage-system-palm-springs-cultural-center/

Ice takes less space, but it also requires even cooler temperatures than water storage. So more reduction in heat pump COP.

The COP of the heat pump for ice storage may be slightly lower, but it's not particularly relevant to the grid operator, since the heat pump is completely off line during peak hours, when only the pumps & blowers are engaged. It's designed to be a lowest operating cost solution, not a maximal thermal efficiency solution. The Ice Energy units use a separate heat pump for making the ice during off-peak (or max PV output) hours, but switches over to cooling with the ice during peak (or post sunset) hours. They have designed in enough smarts to make it very flexible, and can even put it under the grid-operator's control if desired. It takes a heluva hit in COP to overwhelm the economics of 6 cent off peak power for making the brick of ice if cooling during peak power periods costs 12 cents, or 10-15 cent (levelized-cost) PV power against 45 cent island grid power after dark.

Since commercial thermal storage units will often need to be sited on roofs, total mass & volume matters. The 144 BTU /lb phase change storage at 32F is the same amount of thermal storage you'd get out of chilling same pound of water from 46.4F down to 32F, at only a marginally higher COP. Chilled water much warmer than 45F isn't very useful for cooling except in radiant cooling applications. The temperature window of storage in the tank isn't nearly as wide as 32F-45F, so the mass is going to double or triple (or more) to do it all with liquid water.

That company is all over the various regulations and incentives for maximizing the value of PV + ice storage in different states utilities, and there are a number of well integrated installations already out there, eg:

https://pv-magazine-usa.com/press-releases/ice-energy-and-horizon-solar-power-complete-solar-plus-ice-storage-system-at-palm-springs-cultural-center/

http://www.utilitydive.com/news/ice-ice-energy-the-hot-market-for-cooled-liquid-energy-storage/408356/

The value the avoided cost of building more transmission grid capacity to deliver rising peak power is high enough that the utility is essentially giving them away on Nantucket. They are taking some state subsidy support but rate-basing the rest , at a net savings to the ratepayers relative to the otherwise necessary fatter extension cord to the mainland:

http://www.nantucket-ma.gov/997/Beat-the-Peak-Initiative

(Tesla is getting in on some of that action too.)

So I read an digested all the arguments ( I think I did) how easy and cheap it is to cover the storage issue with $70 wifi enabled controllers for electric tanks, hot water tanks providing thermal storage at $0.01 per kwh (still cannot follow the math, maybe you can enlighten me!) and how phase change ice storage could be utilized.

So lets get back to the original question:

I have a thermal peak heating load of around 100 GW in NYS for residential and commercial buildings in the peak heating days, typically occurring around 8 am in the morning.

To satisfy this with an ASHP at a COP of 1.5, we are looking at roughly 65GW, and with a GSHP at a COP of 4.0 we are looking at 25GW of generating capacity needs. WE have around 33 GW (plus a 5 GW reserve, partially coming from Solar) to satisfy the peak load.

Polar Vortex recently reiterated the the fact that that peak might decrease during the day to 80%, but also might continue for a week.
So how do I store enough energy to bridge that heating peak (lets say 80% of that peak for 72 hours) with a 1.5 COP ccASHP? So I need about 7.8 KW for that ccASHP to make about 40,000 BTU/h at a COP of 1.5. Assuming a cloud/snow cover and the wind not cooperating.
7.8 KW x 72 hours = 561.6 KW/h needed to bridge the 72 hours. For each household.
A GSHP would be 210 KW/h storage. Even that sounds not achievable right now.

Add to that the other electric loads, plus electric car charging (they do not just sit there and have car to grid battery storage, those batteries need to be charged to drive around the next day), and it becomes obvious that peak demand in the heating sector becomes a huge issue when to heating sector is to become electrified, which certainly must/will happen.

Again, getting peak power from batteries is really CHEAP compared to building peak power generators, and peak power from demand response is cheaper still. By the time 100% of the space heating load is being handled by heat pumps, the amount of grid storage & transmission capacity necessary to keep the grid stable will already be there.

Polar Vortex events are literally never wind & solar free for a week, not even theoretically! ( Stop with the straw-man-in-clown-suit arguments- my sides are already aching! :-) )

Ice storage for managing air conditioning peaks with hooks for putting it under utility or grid operator control is already a commercial product, taking far less space and thermal mass to get the peak relief than a chilled water (but not frozen) solution.

Ice storage isn't a common space heating storage solution,but there are existence proofs using water-to-water very low-temp hybrid thermal solar solutions using large low pressure uninsulated buried tanks, counting on the heat of fusion to supply more "apparent capacity" to the tank, and high solar conversion efficiency at very low outdoor temperatures. Water to water heat pump efficiency is pretty good at 32F, and delivering 40F+ glycol loop temps out of flat plate collectors has reasonable collector efficiency even at 0F. The claim is that it's cheaper and more effective than standard GSHP on a lifecycle basis, but third party verification & modeling doesn't exist. It's also hard to do these as a retrofit. eg: http://thermalreviews.com/PRODUCTIMAGES/20152/210_SNwo1rP.jpg (I don't really see this beating ASHP + more utility scale wind on a lifecycle cost basis, but maybe.)

Electric vehicle loads are NOT additive to space heating loads, since they don't have to be charged at any PARTICULAR time, only IN time. (Electric vehicles can be employed to provide peak power capacity picture under the revised REV.) The EV load is only additive to total-energy load, not the peak-power load. I'm trusting the folks at the NY-ISO to flag when more generation & storage is needed to satisfy the bulk & peak energy loads as NY evolves toward the "electrification of everything" scenario. Not all grid storage needs to be lithium ion battery (and it won't be), but with an electrified EV fleet employed as peakers a good fraction can be, and the pricing of second-life used car batteries will still be dirt cheap, even at storage capacity numbers as large as the outlandish figures used in the straw-man argument with the 72 hours of darkness with no wind.

The total addition of generation capacity needed to support EVs is smaller than most people think. The transportation sector represents an outsized carbon footprint relative to the actual energy needed due to the abyssmal 15-20% as-used thermal efficiency of the internal combustion engine. When cars electrified the efficiency are more than tripled, roughly quadrupled from the power generator to where the rubber meets the road, even with battery round trip losses included.

You give general arguments without addressing the specifics....

Sure 100% heat pumps will take a while, but what about 50% in the next 15 years? NYS is going into a winter peak scenario with less than 10% load provided by ASHPs, and beyond that does not have the power generation capacity installed.

What is the winter energy production of a solar roof, covered by the occasional snow fall? For a 72 hour cold span? Specifically I did ask how to span 72 hours of peak demand use with ASHPs where 561 KWH of storage would be needed. Last time I checked a tesla wall battery has 10 kwh storage capacity and is about $4000. And how do I charge this battery up again?

Again, the time we see heat pump run time in the winter (sometimes 24 hours/day) and the amount of energy needed especially during the night makes ASHPs not even remotely practical.

http://thermalreviews.com/PRODUCTIMAGES/20152/210_SNwo1rP.jpg example you give is a solar thermal storage solution involving a 2500 gallon tank and also involving a geothermal heatpump with a propane backup. Why not simply putting in solely a geothermal system and be done with it...at half the price. Ice storage for heating is simply unpractical. How do you melt the ice, before you freeze it again?

Yes, EV becomes additive for peak since the car has to be charged sometimes during the 12 hours over night, otherwise the owner will not be happy next morning. And the heatpump running all night to satisfy the heat load of the house during a cold winter night.

The problem might be that you are looking at peak for a couple hours, while I look at it for a 12hour, 24 hours or 72 hour period. We simply had periods here where it was 72 hours below design conditions without sunshine (and yes, minimal or no wind). Keep in mind that the rest of the energy load does not go away. Lets break this down more simple:

Assume a single 12 hours night, between -5 and -13F, no wind that night and obviously no sunshine....

7.8 KW x 12 hours= 93.6 KWH storage for the ASHP heatpump
Need to charge my car 80 miles at 450 watt-hours/mile = 36 KWh during those 12 hours
Plus the rest of the house uses 1 kw/h = 12 KWH during the night.

141.6 KWH for 12 hours needed. Where do I get it from?

How Do I store? Or is it simply cheaper (and more practical) to cut that whole usage in half by using a GSHP?

Posted By Dana1 on 05 Jul 2017 10:54 PM

...Electric vehicle loads are NOT additive to space heating loads, since they don't have to be charged at any PARTICULAR time, only IN time....

How is the government going to convince Americans to accept charging their electric cars "IN time"?  Waiting hours for an electric vehicle to charge, from the time it is plugged in, is bad enough.  Now people will also need to wait until an acceptable time for power to be supplied to the charger??  What if someone wants to go out at an unscheduled time for a long drive (everyone doesn't live 5 minutes from WalMart)?

Whats next, permitting only partial charges to allow drive time only to work?

Like you, I'm more than just a bit skeptical about the economics of the thermal battery/ ice storage tank folks, but threw it out there simply as an existence proof that ice storage is being done in space heating apps, not just cooling apps. They claim it's cheaper and more efficient than drilling for it and using standard GSHP + PV, but I've never seen them show the math.

The battery cost learning curve is well into double-digits (faster than the PV panel learning curve) and the doubling rate is fast, as manufacturers are scrambling to build giga-factories to compete for the automotive market. What costs $4K today will be $2K by 2020, and under $1K by 2025. I believe Tesla a white paper indicating lithium ion batteries would retail for under $100/kwh by 2020. Their current cost for the raw battery is estimated to already be below that point, and the notion that it could retail at that rate in just a few years is credible. There are literally dozens of VERY large battery manufacturing plants being built right now. Tesla's facility in Reno is just the first- even Tesla has announce 4 more, and that's just one company.

Sonnenbatterie just released a series of 8-12kwh residential models in Australia that they think they can make money on just leasing it to existing PV homes for AUD$30-$50- per month, which is way cheaper than grid retail there : http://reneweconomy.com.au/sonnen-battery-storage-plan-to-take-utilities-out-of-business-90595/ In Australia rooftop PV sizing is smaller than in the US, but 23% of all single family homes already have rooftop PV, and the poor to zero remuneration for exporting to the grid makes batteries for self-consumption good deal, even bankable through normal lending channels. (Coming soon, to a utility near you...)

An EV battery is good for about 6-7years before slower acceleration becomes an issue, but can go another 20 as a grid battery, used at lower discharge rates. The availability of low-cost used batteries suitable for grid storage applications will go up as the (much dreaded) electrification of the EV fleet begins in earnest.

The 561 kwh of storage per house is too ridiculously overstated to take seriously. Large temperature differences between the land mass and the ocean quite literally guarantees off shore and near shore wind capacity. Reality is probably on the order of 10% of that number, which is roughly one used mid-sized Tesla P65 battery or at most 25%. But even if it's the 94kwh number used it's not a deal breaker by 2030. By that time new rooftop PV will probably not be allowed in many NY neighborhoods unless the installation includes a battery sufficient for self-consumption (as is the case in some neighborhoods in Oahu right now), new grid batteries will likely be on the order of $50-75/kwh, even if the learning curve backs off a bit from recent years. And used EV batteries repackage for grid storage will be cheaper still.

Whether the battery employed is stationary or still in a car doesn't much matter, and a two EV family with a 10 kwh stationary battery for their solar could opt out of the grid or even sell power from their car or home battery to the grid during peak hours, given the right price signals. The interaction with the utilities in 2030 as the REV enters it's teen years won't look anything like what you have today.

The average car in the US travels less than half the 80 miles per day used in the latest straw man: http://newsroom.aaa.com/2015/04/new-study-reveals-much-motorists-drive/

That reduces the average to 13-14kwh per car for overnight charging (if that even made sense) down from 36, using your numbers. Even at 80 miles/day, charging on a smart charger at work (during daylight hours, when the regional PV is putting something up, and the heat load is less) would be smarter than trying to make every house a self-contained nano-utility. With any sort of grid-smarts the heat pump would be the priority load for the home during peak heating hours, if only to limit the strain on the distribution grid.

The cost of upgrading the grid to handle the capacity of charging all EVs overnight at the same time (even without the heat pump loads) would be gia-normous, and you can pretty much bet the planners have this already figured out in better than just a crayon-on-napkin math. With large PV penetration the amount of mid-day excess PV power above the minimum load is a grid management problem, but a problem easily solved by encouraging people to plug in the EVs at work or at the store rather than at home, and using smarter chargers under grid operator or utility control. The local grid congestion/capacity problem and the mid-day excess problem makes me believe the cheaper solution would be to charge more intelligently, and it's already being done, no new technologies need to be developed to make that happen. The high PV penetration with mid-day local backfeeding problem will also be solved by distributed batteries (on either side of the meter), which will enhance local grid capacity, including peak load delivery with less strain on the grid.

Getting to 80% renewable energy (all uses) by 2050 is now widely viewed as viable reliable and cost effective (cheaper than business as usual) by most people who have made a career out of analyzing it, but there is currently a food-fight going on in those rarified circles about the remaining 20%. The speed at which different technologies roll out relative to others makes a difference in cost, and if GSHP becomes cheaper than the grid & storage needed to support the lower efficiency of ASHP under the scenario that actually unfolds it will be promoted, but it's by no means a slam dunk. And the notion that relying ASHP will bring down the grid or create grid instability just isn't supportable given the rapid evolution of the grid technology, including EVs, smart chargers, and cheaper-every-year grid storage.

Even this bit of analysis from less than two years ago seems quaintly out of date, given the street price of grid batteries today:

http://rameznaam.com/2015/10/14/how-cheap-can-energy-storage-get/

The flow battery party is just getting started, for larger longer time scale applications, and may be strangled in the crib is Li ion keeps trending the way it has. Like PV, the installation rates and learning curves for Li ion batteries keeps on beating even the most optimistic expectations.

Posted By geome on 06 Jul 2017 09:10 PM

Posted By Dana1 on 05 Jul 2017 10:54 PM

...Electric vehicle loads are NOT additive to space heating loads, since they don't have to be charged at any PARTICULAR time, only IN time....

How is the government going to convince Americans to accept charging their electric cars "IN time"?  Waiting hours for an electric vehicle to charge, from the time it is plugged in, is bad enough.  Now people will also need to wait until an acceptable time for power to be supplied to the charger??  What if someone wants to go out at an unscheduled time for a long drive (everyone doesn't live 5 minutes from WalMart)?

Whats next, permitting only partial charges to allow drive time only to work?

Most cars are parked 95% of the time. WTF do you care WHEN it gets charged, as long as it's charged when you need to use it? If it's valuable to the utility (and all ratepayers) to charge at one time vs. another, when the rules are adjusted to where car charging can participate in wholesale and ancillary services markets, low efficiency peaking generators will be history, and everybody will pay less.

It's not the "...the government..." convincing people, it's the price signal and remuneration by the utility &/or grid operator to use the smart car charger as a grid resource, charging more heavily when there is excess power going onto the grid (largely a mid-day PV output problem when PV deployment gets high enough.) Like most people, if had an app or could program a setting on the car that allowed me to be PAID to charge the car to soak up the excess power in the middle of the day, I'd do it then rather than at the air-conditioning peak hours where I had to pay a premimum to charge up. It's only worth paying the premium if you don't plan ahead or something comes up, but in an "electrification of everything" scenario fast-chargers (aka "distributed grid batteries") will be available in a pinch, for a price. And that price need not be very high, if the fast-charger station is getting paid on the back end for providing grid services too.

By the time there is any significant market pentration of EVs in NY, New York's REV will make that happen (maybe even sooner than in California.)

Posted By Dana1 on 06 Jul 2017 09:32 PM


Most cars are parked 95% of the time. WTF do you care WHEN it gets charged, as long as it's charged when you need to use it?..

1) Are you saying that charging stations will be available 95% of the time?  How much will it cost to have chargers everywhere they are "needed"?  What will be the cost of repairs, replacements, and upgrades to the chargers?  Will this be added to the cost of electricity?

2) How will the government, regulators, (etc., however you want to refer to them) know when individual people "need" their car charged (outside of their normal routine.)  "Sorry everyone, you can't use 'your' car for unexpected trips, emergencies, or whenever you feel like it."  "We only have enough electricity available to get you to work and back."  Policies like this won't work.  People will demand flexibility to go where they want, when they want.  People will do their best to keep their batteries topped off at every opportunity.  What if people don't have the money to pay the premium price for peak electricity?  Still believe governments won't be involved?

3) What's up with the "WTF" comment?  You used to be more eloquent.

1>In any variation of an "electrification of everything" scenario charging stations would be ubiquitous, and the hardware pretty cheap (it already is, even at very modest production volumes). Fast DC charging stations would also be common, but only what the market requires. Most commuters would plug in at home AND wprk if they are being remunerated for allowing smart charging of their car as a grid asset, and would only rarely have to use a third party fast charger. Most people who own EVs now with a range of over 100 miles tend to plug into a dumb charger at home, or one with a simple timer to take advantage of off-peak rates, and only rarely are on the hunt for a public charging station. With 6-8 homes or 5-10EVs on a residential neighborhood transformer there will be a capacity problem unless at least some automated intelligence is added. Capitalization & maintenance of smarter home chargers will probably be shared by the homeowner and utility (which may be allowed to rate-base their costs where they can demonstrate a benefit for all ratepayers for the ancillary services provided), but there are numerous mechanisms by which public or semi-public charging can be financed. Right now some retailers already pay for both the charger and the power as a means of attracting customers and keeping them in shopping mode, and that may become ubiquitous as well. The cost is pretty cheap.

2>On heavily EV loaded experimental neighborhoods there is already technology whereby the EV user can already program a "be fully charged by..." time, and smarts dedicated to the neighborhood adjusts the individual charge scheduling priority. Of course if EVERYBODY in the neighborhood programmed for a full charge at an indentical time there is a capacity conflict, but the system can easily detect the conflict and notify the last-plugged whether that request would be satisfied. Trading schemes for those willing to pay a premium for priorty charging could surely be accommodated and fast DC charging stations could cover contingencies. If everybody tops off their fossil burner at the same time they wait in line too, pretty common at the beginning of holiday weekends. But fossil burners don't have the convenience of just plugging in at home and using an app (on their phone or in their car) manage the charging schedule based on what's moste important to them, such as, charge ASAP vs. charge as cheaply as possible by 8AM vs. charge only when the LMP is negative so that you get paid to charge the car. These are only a few of many possible charging options easily managed by smart chargers and semi-smart distribution grids.

Governments are only involved insofar as regulators construct and manage electricity markets, but it's the market signals that will determine who charges where, when, and for how much money. The technology necssary to do this stuff has already been invented, but the 19th century regulated monopoly utility model & regulations still in effect in some states is a hurdle to clear (for those states). New York is not one of those states. In New York the REV has tossed the whole electricity market can o' worms in the air, and the speed of evolution of the market rules is breathtaking. It's not fully there yet, but it surely will be by the time EVs and heat pumps are the go-to technologies. Electricity markets have quirks, and badly designed market features can cause serious issues (ala Enron messing with the CAISO spot market), but the view from 40,000 feet up is that not really VERY different from the liquid fuels, which are subject to all sorts of regulations, taxation, etc. Does the government tell you when to gas up, or how much to pay now? In some ways, yeah, kinda... not really. In a more flexible & peer-to-peer featured lectricity energy market the options for car charging will be far greater than what we currently have for fueling up our gas burners.

The regulators also have the responsibility for making sure granny doesn't freeze do death or go broke under the revised utility rules now, and that will continue to be the case as the transportation fleet and space heating electrifies, but it's not a super-hard problem to solve, and by 2030 PV will be the cheapest form of energy of any type. Even with increasing demand from heat pumps and EVs there is going to be consistently downward pressure on electricity pricing. High wind-power states are already experiencing depressed wholesale electricity pricing (one of the things that is making nukes & coal uneconomic), but before 2030 solar is going to doubling or tripling down on that market pricing effect. Premium pricing in 2040 might still be cheaper than residential retail in 2017.

3> Temporary insanity- won't make a habit of it. :-)

It's hard to tell just how fast batteries are going to get ultra-cheap, but they no doubt will. Right now a 200 miles/charge battery is a big expense, a major fraction of the cost of a Bolt or Model 3, but by 2030 a 400 miles per charge battery will be cheaper than today's Model 3 battery. The only real question is how much cheaper. At longer ranges anxiety about when to charge, or whether you need a FULL charge right now fades away.

How often do you have to fuel-up your fossil burner? Me, about once every 500 miles, mi esposa about once every 300 miles, less than once per week for both of us, but close to once per week for me. If it were automated and I could do it at home or work taking advantage of power-glut discounting I probably would.

If you could get paid even a small amount to pour a couple of gallons in the tank overnight whenever wind power was producing excess fuel, or during the middle of the day when PV was overproducing, would you let a utility manage that automatically? If not you, would your penny pinching neighbor do it?

I'm betting lots of folks would opt to charge up when electricity was cheap, free, or paid even a penny or two per kwh to top up.

If you never had to go to a gas station as long as t put the nozzle in when you got home or work and set a timer as to when you needed it fully fueled, would you? Most people don't find side trips to the gas station very enthralling. YMMV :-)

When planning for a longer trip, do you try to top off at a local cheaper spot ahead of time rather than paying the premium at a turnpike pit-stop? Does turnpike pit-stop pricing stop you from taking the trip?

NREL has published a case study for EV charging infrastructure in the early phase of the transition (no smart charging, no smart grid) using Massachusetts' current ramp-up goals, found here:

http://www.nrel.gov/docs/fy17osti/67436.pdf

This is an assumed ramp up to 300,000 plug in cars (including plug in hybrids) by 2025, in a state where there were a few more than 3 million cars registered (2010 data), so it's clearly not an "electrification of everything" scenario. But it demonstrates that infrastructure planning isn't being ignored as the market develops. The grid will have to become smarter to accommodate 3 million cars in MA, but there's time to plan for that.