Looking for EE Partner(s) to start an open source hardware project / company

[snick]

Hello everyone,
new joiner here, and starting with, probably, a very controversial request. (I guess this should be crossposted in the Open source hardware category, but here feels more appropriate).

I want to build open source hardware and software for the residential energy management (inverter, mppt, bms, ev chargers),
and eventually build a company on top of it. 

Disclaimers:
- I have a very rudimental understanding of EE
- My day job is in software. I work on machine learning problems in one of those big tech companies, and have roughly 20 years experience in the field.
- I have a good understanding of the cost ($$) of what I'm proposing here
- I have never design, implemented, certified and sold hardware. I did, although, got trough certification of some real time software in the automotive industry. The bureaucracy aspect is understood.
- I'm here to learn, but please be open minded and nice .  When a newbie comes to me with software questions, I always start with the assumption that maybe they know something that I don't know, despite my experience in the field.
 
Some beliefs / intuitions.

- The user experience when it comes to electrification of homes, energy management, and setup of energy positive homes is suboptimal (it's complicated, confusing, and suboptimal from a performance standpoint)
- I'm seriously worried we are going through a brand siloing. Talking about the European market, you start seeing more and more companies offering 'full solutions' that are not inter-operable with other hardware (Inverter / mptt, bms, energy meter and optimizer with proprietary protocols, etc).
- The digitalisation of the hardware is still years behind. I worked with few protocols (CanOpen, Modbus, some proprietary ones). They are rudimental, hard to use, and prevent a lot of further optimisation (ie.: they operate best effort with no real-time guarantee, they are often buggy because the underneath firmware is not fully implemented, updates are hard to execute, etc). 
- A lot of the products out there, requires a full commitment to a use case from the get go. I really hope this is not going to trigger people, but I feel there is no much of a pathway to modularity. (ie.: if I double my system kWh capacity, I need to replace completely half of my components, rather than just add additional).
- There is a business model : When it comes to this segment of Hardware / Software, being open source allows experts to implement / use it for free (license alike of AGPL), and help iterating and improve it, while normal consumer would still purchase the hardware / service.
- I can talk for hours on the value of having advance control software in a residential system, from local performance to grid optimisation.


If you read up to this point, thank you. You are either think I'm crazy and full of BS, or you think I might have some point, and you want to know what's next.

I'm looking for a partner(s), that is interesting in exploring this one layer deeper.
I know I want to eventually cover the grid tie, and island use cases, battery and ev (v2h), but there needs to be proper sequencing and decide where it makes sense to start.
Ideally we would first brainstorm the entire product line, go extremely wide with the thinking, and then decide where to start first.
I don't have a geo preference, but ideally UK or Europe for a timezone perspective (or at least someone familiar with the grid systems / legislation in 1 of the european + uk countries)
I'm happy to pay for consultancy, but what I'm really after is a partner to work on this together. I don't believe paying my way through this is the way to go. I need someone that is willing to challenge me, while believing in the end goal. If I just throw money at my intuitions, I will get something most likely not useful as an end product.

So, anyone interested ? (open also to general commentary, but please be nice, I'm conscious I'm a bit crazy) 



 

[Bicurico]

How do you expect to make money?

Why do you think you can achieve better solutions than global players?

Why do you want to make it open source?

Doesn't make sense to me.

[Phoenix]

You don't need an EE to work on this project. Start with a raspberry pi and put a hat on with your required CAN RS485 whatever transceiver hardware.

Once you get thousands of sales maybe it will make sense to roll your own hardware.

[snick]

Again, we are discussing about convictions and beliefs (that can be wrong), would love to hear if there is anyone that approach this with 'why not' rather than 'why'

Re Bicurico questions:

How to make money:

- hardware is sold at cost (covers cogs + operations)
- revenue: recurrent from energy savings subscription.
 

Why do you think you can achieve better solutions than global players?

I cannot have better margin than a big player. Full stop. What I can do is serve a niche better than them.

The energy positive home segment is underserved, and the users interested in it are usually prosumer that wants more control. I'm talking about tens of thousands of installations, not millions.

Why do you want to make it open source?

Why not?
On hardware:  The value here is not on the hardware, it's on the software (orchestration part of it). No electronic company have the know-how to run it. If they 'copy' the design, it means it's compatible, which is a win-win.
On Software: the fact that some software is open source doesn't mean there is no room for a paid service. (in fact this is how open source communities payed for themselves).
Ultimately, if someone find this useful, copy it and scale it to millions, I feel the entire planet is better of. so win win even in that case, and If I lost money in the process, I think the ROI is higher that donate to some green charity and I'd probably have a lot of fun in the process.
 

on Phoenix (prototype it with what you have, sell it, if volume, scale)
- that's more or less what I've done so far, for projects for myself and friends. not a raspberry pi, and not a can rs485, but same concept, micro controller, some reverse engineering of SE and SNA products, and of you go.
- I personally think it's hacky while it works, and I'd never buy something like that for myself. But I'd buy an inverter that offers the same capabilities, and gets certified anytime. 

to put it in other words, you can build a Nest thermostat with a 'raspberry over a rs485 over a cheap honeywell' as much as you want, and you would not sell more than 10 of them, but this doesn't exclude the fact that there are ~3M people a year willing to drop 300 USD for the same thing, just packaged as a real product (and benefit for it)
 

 
 

[Phoenix]

Have you heard of SwitchDIN? They are (one of) the leading energy management system controllers in Australia. They work with residential customers, utilities, VPPs etc.

Their residential Droplet device is... A raspberry pi in a branded box. They are working on their own hardware in time but why take resources away from their key offering (the energy management, the software, the AI stuff) for it?

https://www.switchdin.com/

As you seem to be aware, this area is hugely volatile. Do you really want to try and hard lock down your hardware before operating in the area for a while? I mean really operating with many varied customers, not just your lab prototype.

Could you aim to be hardware agnostic?

[nctnico]

As you seem to be aware, this area is hugely volatile.

I agree. I think in the end the grid operators are going to catch up (both transport capacity and storage) which make local energy management devices obsolete. It also looks like a lot of these devices are simply domotics devices which aren't gaining much traction anyway. Even though many companies are pushing domotics systems hard for several decades already. You can't beat using a physical switch where it comes to simplicity. Even car makers are going back to physical switches because the touchscreens don't work for the majority of people.

What is left is a solution looking for a problem. I just looked at the SwitchDIN website which turns out to be a good example of that. There is a lot of blabbering on about green this &that but not even the slightest hint of telling what actual problem they are solving.

[Black Phoenix]

What is left is a solution looking for a problem. I just looked at the SwitchDIN website which turns out to be a good example of that. There is a lot of blabbering on about green this &that but not even the slightest hint of telling what actual problem they are solving.

Not even proper case studies showing equipment installed plus software. I had curiosity to see their hardware solution based on the raspberry pie but nothing...

[snick]

Very interesting points here.

On sitting on top of hardware, rather than developing the full hardware.

- I wish I could just do the software integration part. I've been in conversation with Schneider electric for almost 2 years, and very close to get founding and contract, just for everything to drop as they change their CIO.
- I tried with other companies and I get one of the followings:
1) we don't care about software, we make money on hardware.
2) we don't understand software, we just buy the firmware from company X, and put it on our hardware.
3) we are integrator, and we are making money out of the complexity of the integration, we don't want to run ourself out of our own business.

If you know any other company out there that is willing to integrate their hardware, I'd be happy to write the 'whatever-proprietary-protocol - to - modern stack' and write the software from there, but as it stands right now I have had no luck.
Even if I was to find someone, I'm not sure how much the integration part of the business can scale (there are really a proliferation of protocols out there that is ridiculous).

On SwithDin
I have no idea what SwitchDin is trying to do. It sounds like an optimizer (great!), but it doesn't solve the user experience that I want to target.
'I have a home, I want to make it energy positive, what do I buy to get a modern user experience that just work' ?

On maturity of the infrastructure, and the future trend
I honestly didn't thought about it, nctnico. Or better, I thought about it, but thought the trend was going towards decentralization and fragmentation of big players into smallers.
If you are right, then yes, the target audience just got incredibly small (off grid, micro grids). I'm not sure if that is >10.000 users on same hardware stack. Most likely not viable as a business. Which is ... sad, and a dead end for my idea.



 

[Bicurico]

Photovoltaic installations currently only make sense in two cases:

1) You live somewhere remote and off grid

2) You get subvention from government, in which case I doubt some open source hacked solution is accepted, especially when you want to sell the energy, feeding it back to the grid. If that is not the case, then you will hardly have a ROI.

I am open to be convinced otherwise, but following my own math, at least here in Portugal, there are better opportunities to invest money.

[snick]

I don't want to put another hacked solution in the market.
I want to build a legit product, which happens to be open source. (I'm really not sure why people always equates open source with hacky/bootleg/uncertified).

You are correct in stating that the economical use case requires 1) either off grid as goal or 2) the production side of the system is subsidised.
I would add that, with the cost of energy increasing, the economical case for an highly optimised system is becoming more and more compelling. (20-30% saving on a 200+ euro bill -> 10k savings over 20 years).

If you are saying that instead of spending 10k in this, people would rather invest in S&P at 7% y/y, and make a lot more money, you are again, correct, but I'm not really sure everyone care just about the money aspect (or have the mean to think in that way).

[Faringdon]

Absoluteley superb idea of Snick.
Yes, the hardware must remain non faulty for 10years++ for investing in Inverters, off-grid etc to be  de-risked.
So the hardware needs to be open-source...so if your inverter breaks down after 12 years, then its easy(er) to fix it, or replace that probably little simple bit that failed, rendering the whole system inoperable.
People wont buy inverters now because if it fails after 12 years.......then  in replacing that inverter, you have lost all your "pay-back". You cant get it replaced on garantee as garantees are only 10 yrs max.
Then there's batteries......when you go out of your 10 year garantee...the cust has to replace their own batteries....and battery tech now is pretty lousy, they go non balanced all the time...just look why CD02 stays in business  trying to improve this. So the cust  has to be sure not to get ripped off with lousy replacement batteries....and the cust knows nothing about batteries....open sourcing the whole thing will improve this nightmare for the customers.

[AndyBeez]

[ sorry if I sound a bit Dragon's Den here ] 

I'm thinking four words that describe your mission: Plug And Play Energy?

Is your company a design authority or, for the governance of a consortium of interested parties who do the design?

By open source, how open? Is there licensing involved as only commercial operations would be interested investing their resources.

So what are the BIG consumer problems that your approach will solve?

I suppose you might use smart contracts in your protocols [and a reliable blockchain] - so is your open source solution cloud dependent?

Just how many other parties would need to be incentivised to adopt your open standards when they have already invested copious resources and reputations in their own closed loop business models?

[nctnico]

I don't want to put another hacked solution in the market.
I want to build a legit product, which happens to be open source. (I'm really not sure why people always equates open source with hacky/bootleg/uncertified).

What is your actual goal of open hardware? I think a much better goal is to maximise interoperability by offering free licenses to use the protocol.

I have been involved with projects that are based on open hardware provided by a major research institute. However, what I see is that their hardware and software design just isn't suitable for many applications. So every manufacturer that makes similar equipment comes up with their own hardware and software tweaks (or implementations from scratch) that fit their requirements. The protocol that those devices use is interoperable. So devices from all the different manufacturers will work together.

A long time ago I have been involved with domotics as well (wrote interfaces to several systems). I got the same feeling you have; companies like Schneider are very set in their ways and don't even have the ability to really develop anything themselves. And often domotics systems have very limited or convoluted interfaces. At one of my former employers we made a simple touchscreen to switch lamps on/off and choose combinations of lamps. Smartphones wheren't a thing back then. Super simple interface and speaking to an existing domotics interface from a major supplier with the help of some good old reverse engineering. Right next to it was the official interface. We could not get anything to work through that interface. People from the supplier where flabbergasted by how simple and straightforward our interface is. So yes, I get your frustration while trying to sell something to companies that are very set in their ways.

Bicurico makes a very good point indeed. For most people ROI will count. Ofcourse there is always a percentage that doesn't care and wants to spend money on the latest toys but for the majority you'll have to somehow make sure your solution is mandated OR really does increase ROI. But the latter is hard to achieve. Making a habit of just turning of the light is free.

[snick]

Thank you @Faringdon. Indeed, why I want this to open source is to extend the lifespan of the hardware, and avoid cheap solution. If they can see what's inside without buying it, everyone can have a more informed decision on the quality / value.

AndyBeez, those are great question, and it sounds that you are following along my reasoning.

1- Yes, the company will act as the certification body for the implementation. We certify our own implementation, but if a 3rd party wants to certify their own implementation, they are welcome, and we'd do it for free (life cost of running the certification process).

2a - Hardware: as open as you get the schematics (both components, and the traces), and can ideally start your own production. The only requirement is that the hardware needs to get certified to sell. If you are private, no need to certify.
 
2b - Software on hardware: all protocols are following their own license, (ie.: if we use Matter/Thread, we stick to their license). The firmware should be open source GPL (you can do your own version but you need to release it).

2c - Software on orchestration: the data logging, measurement, etc are open source, the way the model for optimization is trained is open source, but the model train*ED* is not available. I want to make sure that: if there is a new optimizer that is better than our, everyone can point their hardware to it, but nobody can get access (privacy) and reuse (3rd party utilization) the model we trained thanks to the customers that select our service.

3- I want to solve 3 problems: (1) it's easy to configure, setup and operate. (2) is modular and can be expanded over time, as the usecase  changes. (3) because of better operability between the components of the system, it optimize the energy consumption as close as possible to the optimum savings.   I believe (1) is the most important, (3) is the long term win and (2) is a nice to have only if it is practical / add real value to the user.

4- I'm still not sure where to put the Inference part of the optimization software. Right now, the easiest way is going cloud. I believe that's the best way to get this off the ground. Ideally, the intelligence (training and inference) should run locally, so to make the hardware 'smart' without being constantly internet connected. I think the smart contract part comes in only when we move from single homes, to network of homes that do global orchestration. That will require density of penetration, VPP, etc, which is way down the line, but ideally, yeah, that's the way. Someone can only dream.

5- I don't know. And it's hard to tell. Ideally everyone's adopt it, and we are done. Realistically, nobody adopts it, and if you are lucky you do everything yourself, up to a small use case (10.000 homes). From there the benefit should speak for itself, and ideally the industry should catch up. One thing is certain, I rather prefer to have less parties but more geographically close, than many parties, with low local density. (because of the end-game from point 4).


Thanks for indulging my craziness. Love the 'Plug and Play energy', it does hit the 'simply works out of the box' concept. 

[snick]

nctnico, I think we are getting closer here.

Yes, the protocol is open source, with a free licence.
Ideally, every manufacturer can produce their own flavoured hardware, as long as it meets the protocol, and they get certified.


My main concern is the bootstrap problem. Unless there is real hardware out there that is threatening the big player (SMA for instance), they really don't care of implementing your protocol. Their sales lifecycle doesn't include post sell monetisation, and as such, there is no incentives for them to adopt it.

On the other hands, if they are either loosing market share to 'our implementation', or customer are passing on their product because they are not interoperable, there is an incentive to jump onboard.

My approach is to brute force this realignment of incentives (at the cost of spending money) and getting real hardware out.


On the bigger topic of 'why the big player cannot come out with something easy to use' I'm sure there is a lot to be said (see The Innovator Dilemma), but I keep getting surprised, and is too evident to not feel is an amazing opportunity.


 

[ebastler]

Adding your own hardware products (instead of a pure software/controller/interface solution) would give you much more revenue and margin. And it would act as a "dongle" for the open source software, since hardly anyone would make their own power electronics.

But I think you will find that making and selling hardware is much more complex than you think. You mention in your initial post,

- I have a good understanding of the cost ($$) of what I'm proposing here
- I have never design, implemented, certified and sold hardware. I did, although, got trough certification of some real time software in the automotive industry. The bureaucracy aspect is understood.

but I beg to differ. Getting your design certified may be somewhat akin to getting automotive software approved. But you will also need to deal with supply chain, inventory cost, manufacturing infrastructure, sales and physical distribution of your product, warranty, repair, electronic waste recycling schemes. Given the nature of the intended products, you might also want to think about liability insurance.

What was your estimate regarding "the cost of what you are proposing"?

[...] I thought about it, but thought the trend was going towards decentralization and fragmentation of big players into smallers.

Would you have a recent example of any industry which has moved in that direction?

[AndyC_772]

One more random data point...

We have solar panels on the roof and battery storage. The system has been up and running for 2 years now and, now it's correctly installed and set up, it's working very well.

One of the hurdles when it was installed, was getting the necessary permission from the grid operator to export surplus power. This was, as you might expect, a generally unfriendly, bureaucratic and jobsworthy process, which I don't think I'd have been able to complete for myself as a regular householder. The regulatory system is very much geared up towards "get a licensed electrician to...", which is a shame as it's hostile to both consumers and would-be developers of new equipment.

If you do develop something new, think *first* about who will install it and why they'd choose it over existing, known good, certified solutions that their installers already know well.

In terms of modularity and upgrades, the only way I'll be upgrading the solar installation is if I somehow end up with a bigger roof, and that's not going to happen. My batteries are modular already, but the only reason to upgrade them would be economics (ie. storing and using additional power myself rather than exporting it to the grid and buying it back for a higher price layter on), and the numbers just don't work out - even with today's electricity prices.

In fact, we now need *less* battery storage if anything, since we now have an air source heat pump too. Installing that was an absurdly costly and disruptive process that's a topic for a whole new thread on its own, but it does mean we now have a way to make good, efficient use of any surplus electricity.

If there's a missing piece of the puzzle, it's the overall system monitoring. Right now all I really have is the web portal provided by the battery inverter manufacturer - it knows about total power flow to and from the house, grid, solar panels and batteries, and updates every 5 minutes. That's a ridiculously coarse timeline, but worse than that is it involves reporting all this data via the web to a server in China before I can log in and look at it.

What I'd love to see is a separate monitor, with current transformers on all the major circuits, ideally a way to also interrogate the battery inverter for SoC plus DC voltage and current - and, crucially, absolutely no cloud, IoT, account, privacy policy, terms, mobile app or other networked features whatsover, besides being able to plug it into my own private network and browse its built-in web server.

My home, my usage, my data - and no, I don't have a "smart meter" either, because d'uh.

[AndyBeez]

Floating away from the technicalities of specific implementations, you're creating an Extensible Plug and Play Energy Protocol ? I'm seeing some kind of layer model here, similar in context to the OSI for the internet? Extensible means flexible.

Second guessing, the sector's big players may either embrace an open standard or judge it somewhere between amusing and irritating? It's the developing BRICK economies where open standards are taking hold. They are relatively free from the constraints of North American and European corporate interests, and seek ways to solve their own problems at a local level. Avoid big business interests: Tail > Dog > Wag

Why just limit your ambition to 10,000 customer end points; why not 10 to 100+ million. With such wide scope of ambition, I suggest first you need to establish a working group of people with the necessary energy provision skillsets to define a clear mission statement. Later, evolve this to a skeleton/framework to achieve the mission. Then you can make residential energy management as easy as Lego.

[Bud]

Who is the Customer (consumers, installers, DIY-ers) and what is the plan to provide support to the Customer, if any?

[nctnico]

Floating away from the technicalities of specific implementations, you're creating an Extensible Plug and Play Energy Protocol ? I'm seeing some kind of layer model here, similar in context to the OSI for the internet? Extensible means flexible.

Second guessing, the sector's big players may either embrace an open standard or judge it somewhere between amusing and irritating? It's the developing BRICK economies where open standards are taking hold. They are relatively free from the constraints of North American and European corporate interests, and seek ways to solve their own problems at a local level. Avoid big business interests: Tail > Dog > Wag

Why just limit your ambition to 10,000 customer end points; why not 10 to 100+ million. With such wide scope of ambition, I suggest first you need to establish a working group of people with the necessary energy provision skillsets to define a clear mission statement. Later, evolve this to a skeleton/framework to achieve the mission. Then you can make residential energy management as easy as Lego.

But the first question is: why would people even need energy management at home?

[snick]

This is getting a very interesting discussion, and I'm really enjoying it. Thank you for engaging.

Addressing some of the latest comments.

ebaster's:

but I beg to differ. Getting your design certified may be somewhat akin to getting automotive software approved. But you will also need to deal with supply chain, inventory cost, manufacturing infrastructure, sales and physical distribution of your product, warranty, repair, electronic waste recycling schemes. Given the nature of the intended products, you might also want to think about liability insurance.

You are most likely correct. I have no idea, but that's why I'm looking for a partner, to know what I don't know, and get help with it. That said, bureaucracy and complexity don't usually scare me.

What was your estimate regarding "the cost of what you are proposing"?

I think we are roughly at 2M per design, just to get the manufacturing setup. From there, is Cost per units economics. Ideally that 2M/design get's smaller over time (with iteration or part of the processes that can be shared accross designs). I mean, just an injection moulding of a case at scale is 250K usd.

Would you have a recent example of any industry which has moved in that direction? (in relationship to decentralisation over consolidation)

In Europe telecommunication, transportation, and metropolitan heating systems,  are going towards a more decentralised structure. You can start your own train company, and use the country infrastructure. My intuition is that the energy market is going, slowly admittingly, towards that direction.  You might be question if there is a viable economic at small scale / decentralized scale. My assumption is yes, but I don't have much to back it up.

From AndyBeez

you're creating an Extensible Plug and Play Energy Protocol ?

again your choice of words is spot on, I think extensible is a better term than modular in this context.

Why just limit your ambition to 10,000 customer end points; why not 10 to 100+ million. With such wide scope of ambition, I suggest first you need to establish a working group of people with the necessary energy provision skillsets to define a clear mission statement. Later, evolve this to a skeleton/framework to achieve the mission. Then you can make residential energy management as easy as Lego.

That is indeed what I'd like to do. I think my 10.000 is a mental limit to my investment in case we need to go fully vertical and build all the hardware that is need. But, yes, there are no upper bound here.
The only limit is that it's not a 1 person job, especially if that person is as ignorant in EE as me

Bud's

Who is the Customer (consumers, installers, DIY-ers) and what is the plan to provide support to the Customer, if any?

In go-to market strategy, you differentiate who's the customer from who's the channel.
In this case, the customer is the consumer, but the installer is the channel.
 
To the consumer you sell the cost/effective 'make your home energy positive in 24h, and live worries free forever*'.
To the installer you explain why the solution is easier to install, and makes the end customer happier.

I don't have a strategy flashed out for customer support, at this point, which is a big gap. I would need to understand more where today's pain points are. I assume, it's the fact that when something break you need to replace a lot of stuff ($$) or that is an highly specialised system to touch.

AndyC_772's

Your use case is already quite advanced (as in, you already have everything that you needed to live energy positive). Imagine if this was day 1, and you didn't know where to start.
Maybe we could have sized that systems better and saved you some extra $$, or maybe we could have make the system more efficient or automatic, saving $$ or time.

On monitoring: 100% agree with you, this is trivial, and a pro-sumer should be having those informations
On privacy and where the data lives: also agree. This thing cannot stop working (even monitoring) just because there is no internet.

Nctnico's

But the first question is: why would people even need energy management at home?

People don't need/want an energy management at home.
People need to pay the least amount possible for the energy they need. If there is a positive externality for the environment, that's  a cherry of top. They don't need / want anything else.

How you get to that, is by giving them a system that is easy to install, it's well managed (you can save up to 30% energy), and has a great user experience (so they don't need to become EE to operate it).



 

[nctnico]

Nctnico's

But the first question is: why would people even need energy management at home?

People don't need/want an energy management at home.
People need to pay the least amount possible for the energy they need. If there is a positive externality for the environment, that's  a cherry of top. They don't need / want anything else.

How you get to that, is by giving them a system that is easy to install, it's well managed (you can save up to 30% energy), and has a great user experience (so they don't need to become EE to operate it).

I don't see where a 30% energy saving is coming from. I know that in the UK there are super low nightly tariffs but as far as my knowledge goes these do not exist at a large scale on mainland Europe because there is a lot of interconnection going on (including using pumped hydro storage in Norway) that evens out the day / night tarifs. Now you could argue that people should go for hourly tariffs and cherry pick times that electricity is the cheapest but realistically that is only interesting for doing the laundry and / or the dishes. You will want to watch TV / use the computer when you come back from work, cook dinner at set times, have a warm home, etc. And with too many people doing that, those tariffs will start to even out as well so there is little benefit. It also requires to kind of organise your life around the electricity prices. That gets tiresome quickly; most won't bother and prefer to pay a flat fee. On top of that, the major electricity companies are also working on installing storage to even out peak production and peak demand.

So what is left? Maybe making people more aware of what appliances are actually consuming but for that the energy labels and public awareness campaigns are more appropriate. In the end: if you want to save electricity (and money), buy efficient appliances and switch off what you don't need.

Also keep in mind that a lot of people don't really want to bother even with simple 'technical' things. This is a trend that has been going on for a long time. My first employer was into professional audio and telecom. The professional audio department did a lot of audio installations in churches. In the past they would deliver a box with 5 or 6 volumes controls so the 'audio guy' from the church could adjust the volume just right. Later on more and more churches told us they don't have a person that is interested in controlling anything audio related. Just install the system and make that it is always ready to go. To at least have some control we installed a box with a rotary switch that had 3 positions for the overall volume: soft, normal and loud.

[redkitedesign]

I don't see where a 30% energy saving is coming from. I know that in the UK there are super low nightly tariffs but as far as my knowledge goes these do not exist at a large scale on mainland Europe because there is a lot of interconnection going on (including using pumped hydro storage in Norway) that evens out the day / night tarifs.

This will change. Rooftop solar is getting so cheap, it becomes profitable for just replacing the home electricity usage while the sun shines, even without getting paid for feeding back into the grid.
The addition of large amounts of wind power (which is the cheapest form of energy now commercially available from new-built plants) will add to the variability, making hourly changing tariffs very likely in the near future.

Home batteries will become more common, to use some of that solar in the evening and night. But as a well-managed Li battery easily has 10000 cycles of life in it, it is worth to actually charge and decharge it every day. (That will still give you 30yrs of life)

In summer, you can only charge it if its empty in the morning. So you must sell or use all charge before the PV starts producing.
In winter, you cannot charge it from solar so you'll have to use the cheapest tariff.
Nobody will want to time the laundry or the cooking to the moment the tariff is lowest. Perhaps only the car-charger will adapt automatically (most people drive less then 60km/day, thus only need 15kWh of charge/day. Thats 4 hours at a simple 16A granny cable, thus easily scheduled at the lowest tariff period between 6pm and 8am the next morning.)

Thus I think there is a huge market for devices that schedule home battery charging and discharging, based on (expected/forcasted) energy prices.

[Phoenix]

On SwithDin
I have no idea what SwitchDin is trying to do. It sounds like an optimizer (great!), but it doesn't solve the user experience that I want to target.
'I have a home, I want to make it energy positive, what do I buy to get a modern user experience that just work' ?

They are doing a few things, sorry I didn't spell it out in my last post.

In a residential setting their "Droplet" device interfaces with devices like solar inverters, battery systems, EV chargers, other loads (aircons, hot water, heaters etc.) in the house. The Droplet does data capture and control. You can see their list of partners here:
https://www.switchdin.com/oem

The data and control is cloud based (yes, it needs the internet or cellular data connection to operate). Aggregating the data and control into a single manufacturer agnostic platform.

For residential and commercial customers this platform allows the user to configure their energy management, schedule battery charging/discharging and load profiles. When I spoke to them they were also looking into using weather predictions for better energy optimisation. Part of their buisness is making this part useful yet simple.

For utility companies the back end they offer allows groups like Virtual Power Plant (VPP) operators simple control access to individual residential customers and bigger commercial customers with almost any brand equipment. They are the main VPP "middle man" in Australia.
https://www.switchdin.com/virtual-power-plants

I would say the VPP is one of the most attractive aspects of actively managed home energy! You can't get this operating your own stand-alone energy management.

Not even proper case studies showing equipment installed plus software. I had curiosity to see their hardware solution based on the raspberry pie but nothing...

https://www.sapowernetworks.com.au/industry/flexible-exports/compatible-equipment/switchdin-droplet/
https://static1.squarespace.com/static/592cdfd3e6f2e1b24180305b/t/605348e1119d782953ab1b4d/1616070897521/SwitchDin+Droplet+Installation+and+Commissioning+V5.pdf
https://static1.squarespace.com/static/592cdfd3e6f2e1b24180305b/t/5f56ea31020c314b1f095f72/1599531590809/Droplet+Quick+Installation+Guide.pdf

It's a black and white box. Who cares what it looks like, it's being put behind an opaque panel and forgotten.

Here's a glossy picture inside a control box with their unit being installed.
https://images.squarespace-cdn.com/content/v1/592cdfd3e6f2e1b24180305b/1618371212665-9KGZVNNMHNLOAGNCKL7T/switchdin+installer.png?format=2500w

[I'm not associated with SwitchDIN; I have spoken to them on a few occasions, with my previous employer, about becoming an OEM partner]

[Phoenix]

I suggest first you need to establish a working group of people with the necessary energy provision skillsets to define a clear mission statement. Later, evolve this to a skeleton/framework to achieve the mission. Then you can make residential energy management as easy as Lego.