First of all, thanks for putting out a fascinating and informative podcast. I found this particular issue to be very confusing. After reading it as carefully as I could, I still don't understand exactly what the guests were proposing. Specifically: 1. hwhat exactly is this "household spare capacity"? Are you referring to extra solar power that they don't use themselves? That usually already goes back to the grid and gets used by others, so it doesn't leave anything extra for data centers. Do you mean that households have the capacity to use more electricity than they normally consume? True, but the unused electricity just goes to other customers. The grid is balanced so there is generally no extra electricity going to waste that could go to data centers. 2. How exactly would this heat pump idea work? If people are replacing conventional A/C and/or resistance heaters with heat pumps, that would result in saving electricity,. but resistance heating is primarily used in the South, not all over the country, and not, I believe, very much in California specifically. If people replace oil or gas furnaces with heat pumps, it's great for the environment, but they'll need more electricity, not less, so it doesn't help with this problem. Are you proposing that data center owners purchase heat pumps and give them to homeowners? How would you get them to do that? 3. While I'm very much in favor of electrification, switching from fossil fuel power to electric power requires more electricity, not less, so implying that this could help solve the data center policy is confusing to me. I'd appreciate an explanation for this argument that explained exactly how it would work, because, at the moment, I can't make sense of it.
Couldn't help think of this conversation when I read the following... seems like Base Power's general model could very easily be adapted to directly provide hyperscalers with the capacity they need...
The world is a better place for Ari and Carla. Also, I loved the "a funny thing happened on the way to the podcast" intro. Go, David!!!!
"There should be media about it.....there should be popular media". Yes, there is a story--a clear, simple truth-- about how electrification could make our futures better than the burning Hell of climate change we are facing right now. Too bad mainstream corporate media is AWOL on this and other issues. Like, did you know that there are people protesting the Trump administration and the war in Gaza? So, it is up to independent media to be telling this story. Tag, you're it. Good luck.
I feel like I'm missing something from a resource adequacy standpoint. Data center load is going to be highest in the summer when cooling demand is the highest, but your heat pump program is basically only giving you winter capacity (not sure how exactly it'd end up as accredited capacity since i'm not super in depth on CAISO's capacity accreditation process but the resistance to heat pump swap really only helps you when heating is needed in the first place). Like it'd be really cool if this worked out just because people getting heat pumps rocks but I'm really not clear how this is effectively helping meet data center demand.
IIRC, Doug Lewin said there was a *LOT* of resistance heat in TX. So, when I heard Octopus Energy was entering that market, I was a little disappointed to hear them say they would not be deploying heat pumps.
Recently Denver Water was in the news here for raising it's "tap fees," aka "System Investment" fees. Some reporting showed these to be $20,000 to $50,000 per "unit," across Colorado, with the high end for big houses in the 'burbs. Made me think should we charge these kind of fees for electric service? Or maybe stop charging them for water to homes as part of housing "affordability" campaigns?
If we wanted to apply it to electric, maybe $10/W of average demand. Then if a hyperscaler or whatnot could find "efficiency as a resource," at less than that by all means let them pay to reduce or improve our load profiles and reduce their own system charge. Google says data centers cost $11/W to build so the threat to double that for the electrical service might get someone's attention.
First of all, thanks for putting out a fascinating and informative podcast. I found this particular issue to be very confusing. After reading it as carefully as I could, I still don't understand exactly what the guests were proposing. Specifically: 1. hwhat exactly is this "household spare capacity"? Are you referring to extra solar power that they don't use themselves? That usually already goes back to the grid and gets used by others, so it doesn't leave anything extra for data centers. Do you mean that households have the capacity to use more electricity than they normally consume? True, but the unused electricity just goes to other customers. The grid is balanced so there is generally no extra electricity going to waste that could go to data centers. 2. How exactly would this heat pump idea work? If people are replacing conventional A/C and/or resistance heaters with heat pumps, that would result in saving electricity,. but resistance heating is primarily used in the South, not all over the country, and not, I believe, very much in California specifically. If people replace oil or gas furnaces with heat pumps, it's great for the environment, but they'll need more electricity, not less, so it doesn't help with this problem. Are you proposing that data center owners purchase heat pumps and give them to homeowners? How would you get them to do that? 3. While I'm very much in favor of electrification, switching from fossil fuel power to electric power requires more electricity, not less, so implying that this could help solve the data center policy is confusing to me. I'd appreciate an explanation for this argument that explained exactly how it would work, because, at the moment, I can't make sense of it.
Couldn't help think of this conversation when I read the following... seems like Base Power's general model could very easily be adapted to directly provide hyperscalers with the capacity they need...
https://techcrunch.com/2025/10/08/base-power-raises-1b-to-deploy-home-batteries-everywhere/
The world is a better place for Ari and Carla. Also, I loved the "a funny thing happened on the way to the podcast" intro. Go, David!!!!
"There should be media about it.....there should be popular media". Yes, there is a story--a clear, simple truth-- about how electrification could make our futures better than the burning Hell of climate change we are facing right now. Too bad mainstream corporate media is AWOL on this and other issues. Like, did you know that there are people protesting the Trump administration and the war in Gaza? So, it is up to independent media to be telling this story. Tag, you're it. Good luck.
I feel like I'm missing something from a resource adequacy standpoint. Data center load is going to be highest in the summer when cooling demand is the highest, but your heat pump program is basically only giving you winter capacity (not sure how exactly it'd end up as accredited capacity since i'm not super in depth on CAISO's capacity accreditation process but the resistance to heat pump swap really only helps you when heating is needed in the first place). Like it'd be really cool if this worked out just because people getting heat pumps rocks but I'm really not clear how this is effectively helping meet data center demand.
IIRC, Doug Lewin said there was a *LOT* of resistance heat in TX. So, when I heard Octopus Energy was entering that market, I was a little disappointed to hear them say they would not be deploying heat pumps.
Recently Denver Water was in the news here for raising it's "tap fees," aka "System Investment" fees. Some reporting showed these to be $20,000 to $50,000 per "unit," across Colorado, with the high end for big houses in the 'burbs. Made me think should we charge these kind of fees for electric service? Or maybe stop charging them for water to homes as part of housing "affordability" campaigns?
If we wanted to apply it to electric, maybe $10/W of average demand. Then if a hyperscaler or whatnot could find "efficiency as a resource," at less than that by all means let them pay to reduce or improve our load profiles and reduce their own system charge. Google says data centers cost $11/W to build so the threat to double that for the electrical service might get someone's attention.
I had this *exact* thought when I read how much of USA's grid capacity is there to support HVAC. 300GW!