6 ms·

I do that already, but just for home sanitary water (sorry, do not know the name in English, water for shower etc) with the secondary resistive heating of an "o

by kkfx 4y ago

I do that already, but just for home sanitary water (sorry, do not know the name in English, water for shower etc) with the secondary resistive heating of an "old" (not on sale anymore) Daikin/Rotex M2O EKHHP 300l v3 but I can't do the same for heating: too much water is needed. Actually I'm looking to expand the system with just a set of insulated tanks (found some on sale 500l/~1kWh/day of dispersion at 60℃ inside, 5℃ outside) and a bigger external heat pump + wood stove (for emergency) to heat the water. It's just an idea because: - nothing pre-built seems to exists, apparently nobody is interested enough in try this way to save energy bills; - it not really cheap and DIY style there are many things who can break making the project more an expensive toy than an effective solution. IF I found something that seems good enough next spring I plan to expand, but it's still a relevant and generally interested issue the fact that nearly nobody talk about the dichotomy between p.v.-powered device needs to maximize self-consumption vs non-p.v. powered homes needs to reduce the total consumed energy: with a p.v. system anything should be designed to run all together at full power as much as possible, so automation to power on and off things quickly and things made to march at very different speed depending on the available power. It's far better for instance in that sense a classic resistive boiler who heat water less efficiently than a heat-pump BUT with a constant load and perhaps not a single resistance but few, to being able to "sum or reduce" the power depending on how much you can get from Sun. While if you do not have a p.v. it's far better to save as much energy as possible... Some patterns might be common, like running appliance at a dynamic point in time, one depending on the Sun, the other depending on current grid load (and so price). But many others are different. A stupid example my washing machine, with it's eco-mode actually consume less kWh, but consume MORE from the grid because it have spike loads my p.v. inverter can't satisfy quickly enough from p.v. so they goes on the grid. On contrary running an "expensive" high temp washing program my self-consumption is normally 99.99999% Surely, so far most homes do NOT have p.v. so the biggest market for any appliance is the one who do not count p.v. but is not really expensive counting it. Also it's absolutely not expensive ad ModBUS controls on anything who consume more than 1kW. It does not really change the price. It's absurd not offer such option in most appliance with a significant overprice (let's say 40€ for something that to the OEM cost just 1€).

mindslight 4y ago
> nobody talk about the dichotomy between p.v.-powered device needs to maximize self-consumption I think there are two reasons. First, it's easy to use the grid as a sink, and add in local batteries if you're still concerned. Like here in the US, solar companies size people's PV systems based on their yearly consumption, totally ignoring what time of day consumption takes place, peaks, etc. We're basically coasting on the "net metering" mindset, for better and for worse. Second, control solutions and energy storage solutions are generally quite bespoke (custom). I agree about the need for load management and device communication in general, but we're still (unfortunately) at the point of DIY solutions rather than some standardized communication. Which is probably a good thing, because the whole "smart" cloud crap that companies leap at is dead end technology anyway. In the context of a northern climate with significant heating load, I don't think resistive heating ever makes sense, unless you somehow have access to free solar panels and free land. The tradeoff is going to be the maintenance burden of a heat pump versus how little heat you need, and if you're talking about storage then you seemingly need a bunch of heat. I've been thinking about this problem due to wanting to go solar (with a utility that won't net meter) and also wanting to have a back up generator (but not wanting to get some oversized 12kW monstrosity to power a handful of things in an outage, just in case they happen to all turn on at once). My eventual solution will probably be something like a relatively large capacity battery inverter (eg SMA Sunny Island) tied to a relatively small amount of storage batteries. I'll schedule some loads (eg dehumidifer) to operate only during the day. If I get to the point where I'm significantly overproducing electricity, then I will think about increasing the battery storage.
- kkfx 4y ago
  The issue of consider the grid as a kind of battery is that IF you are paid for the energy you inject you are paid FAR LESS than the energy you grab with a so big delta that's far less than the "efficiency loss" of a battery... When I've made my actual p.v. (here, France, we are allowed to build p.v. systems as generic citizens because before connect any system you need a certification from a third party body) I choose 5kWp because I thought it's enough. Now I understand that to really maximize my self-consumption I need MUCH more peak power because I need to been able to run on p.v. anything at once as much as possible because sunny hours are not that long. It's the same for the actual heating (water-water heat pump) I have a small reservoir because the system is designed to run when I need heat. It should instead be designed to run full power but really MUCH in few hours per day etc. Now, I'm a sysadmin not a p.v. technician not working for anyone in the industry so I find curios that I personally realize that and no one who damn work for perhaps a decades in the industry do not realize the same... About resistive heating: yes it's far from being interesting but it is for a thing: it's extremely cheap and run in any external temp condition. air-water heat pumps do not, water-water (geo or from hot water reservoirs) heat pumps are still dependent on the "external side" temp. A small batch of resistence run by relais piloted via a cheap GPIO board in sequence following the maximum p.v. output possible it's a cheap way to maximize self-consumption. A heat pump can't vary much the compressor absorption witch means that if you need a big one to heat quickly a large mass of water you also need to power it. If you have few smaller you might fire up depending on the available power but it's hyper-costly... IMO having a big heat-pump for when the Sun shine well, a small one for when you can't run the big and you are on-grid/battery and resistive for the rest is the cheapest solution to use at best any kWh you can produce, no proof for that, just an idea, but so far that's the result of my limited experience...

by kkfx 4y ago

mindslight 4y ago

> nobody talk about the dichotomy between p.v.-powered device needs to maximize self-consumption I think there are two reasons. First, it's easy to use the grid as a sink, and add in local batteries if you're still concerned. Like here in the US, solar companies size people's PV systems based on their yearly consumption, totally ignoring what time of day consumption takes place, peaks, etc. We're basically coasting on the "net metering" mindset, for better and for worse. Second, control solutions and energy storage solutions are generally quite bespoke (custom). I agree about the need for load management and device communication in general, but we're still (unfortunately) at the point of DIY solutions rather than some standardized communication. Which is probably a good thing, because the whole "smart" cloud crap that companies leap at is dead end technology anyway. In the context of a northern climate with significant heating load, I don't think resistive heating ever makes sense, unless you somehow have access to free solar panels and free land. The tradeoff is going to be the maintenance burden of a heat pump versus how little heat you need, and if you're talking about storage then you seemingly need a bunch of heat. I've been thinking about this problem due to wanting to go solar (with a utility that won't net meter) and also wanting to have a back up generator (but not wanting to get some oversized 12kW monstrosity to power a handful of things in an outage, just in case they happen to all turn on at once). My eventual solution will probably be something like a relatively large capacity battery inverter (eg SMA Sunny Island) tied to a relatively small amount of storage batteries. I'll schedule some loads (eg dehumidifer) to operate only during the day. If I get to the point where I'm significantly overproducing electricity, then I will think about increasing the battery storage.

kkfx 4y ago
The issue of consider the grid as a kind of battery is that IF you are paid for the energy you inject you are paid FAR LESS than the energy you grab with a so big delta that's far less than the "efficiency loss" of a battery... When I've made my actual p.v. (here, France, we are allowed to build p.v. systems as generic citizens because before connect any system you need a certification from a third party body) I choose 5kWp because I thought it's enough. Now I understand that to really maximize my self-consumption I need MUCH more peak power because I need to been able to run on p.v. anything at once as much as possible because sunny hours are not that long. It's the same for the actual heating (water-water heat pump) I have a small reservoir because the system is designed to run when I need heat. It should instead be designed to run full power but really MUCH in few hours per day etc. Now, I'm a sysadmin not a p.v. technician not working for anyone in the industry so I find curios that I personally realize that and no one who damn work for perhaps a decades in the industry do not realize the same... About resistive heating: yes it's far from being interesting but it is for a thing: it's extremely cheap and run in any external temp condition. air-water heat pumps do not, water-water (geo or from hot water reservoirs) heat pumps are still dependent on the "external side" temp. A small batch of resistence run by relais piloted via a cheap GPIO board in sequence following the maximum p.v. output possible it's a cheap way to maximize self-consumption. A heat pump can't vary much the compressor absorption witch means that if you need a big one to heat quickly a large mass of water you also need to power it. If you have few smaller you might fire up depending on the available power but it's hyper-costly... IMO having a big heat-pump for when the Sun shine well, a small one for when you can't run the big and you are on-grid/battery and resistive for the rest is the cheapest solution to use at best any kWh you can produce, no proof for that, just an idea, but so far that's the result of my limited experience...