DIY Solar Energy Diverter Part II using PWM for Hot Water

[@peterball9771]

Thanks for the great videos, it's exactly what I've been looking for. I recently got 7kw solar system and have been running an emporia energy monitor on each circuit. It's a shame to see so much solar going back to the grid for next to no return. I plan on dumping it into our hotwater system and in the future charging some sort of energy storage. Thanks again for sharing.

[@greenenergygav8053]

Hi Peter, my pleasure.

[@substandard649]

Great video, really interesting. It would cool to see an overview of the hardware, your design choices and build methodology.

[@stevehorton8889]

I did a similar thing using a solid state relay and zero voltage switching to enable 1 to 4 half cycles every 2 mains cycles at 50hz to give 25% steps (750W steps) to match hot water power to surplus power. Zero voltage switching was done to reduce EMI.

[@tamerpehlivan453]

Sounds pragmatic, can you share your design?

[@PaulRBinding]

Great video Gav & very useful information. I am using a more crude arduino based all or nothing system at the moment to try and a) run the pool pump & b) harvest as much of the surplus as possible to for the immersion heater & draw as little from the grid as possible so I think I’ll adapt your code to incorporate this burst technique now. Thanks, I’ve subscribed!

[@ratgreen]

Nice build. I've been brainstorming making something similar for years, after seeing the insane prices of off the shelf solar diverts like iboost and eddi's. But I was planning on using a CT on the solar and the grid side of my incoming so I could compare the two, and PWM a mosfet or something at high frequencies. Which would mean I wouldn't have to have it turn on off for a few seconds for sampling like your design does. But I dont know how to code an Arduino. I struggle to get a LED to blink without sytax errors. I didnt realise car chargers could be turned on and off like that. I assumed they needed a smooth source of power and then reduced duty cycle themselves to get the desired charge current.

[@greenenergygav8053]

Hi Ratgreen, Apologies for the delay in responding to you. Yes, I'm no programmer myself! Lots of copied code and examples are evident in my sketch! In terms of EV charging, these need to be stable. The minimum cycle time I have set is 20 minutes and I don't think you want it any less than this. Both our EVs seem happy with this arrangement. Good luck with your project. Cheers, Gav.

[@chaswinder]

Really interesting, thanks for sharing!

[@DanielG0m3z]

This is great! Could you please share the circuit diagram/schematics as well? Thanks

[@greenenergygav8053]

Hi Daniel, Unfortunately I haven't drawn up a circuit diagram. The best source for the voltage and current sensing is the open energy monitor learning website. This link takes you to the simple voltage dividers for the voltage and current sensing: docs.openenergymonitor.org/electricity-monitoring/ctac/how-to-build-an-arduino-energy-monitor.html The Arduino script identifies the pins of the Arduino which are used for voltage and current sensing (in my case 3 phase, but single phase just requires two inputs rather than 6). The pin that's used for the mode switch is also identified. The digital pins for the output control are also easily identified from the script. Apart from power connection to the Arduino, that's about it. I will be posting another (and I assume final) video at some time covering a further enhancement using phase angle control for the HWS heating. This was a relatively easy enhancement and has some benefits over the solid-state relay on/off switching. I just want to test the phase angle controller for a few months before I post the video. Cheers, Gav.

[@jadsakr]

Good job Gav, I was wondering if you tried to work on something for off-grid inverters

[@btapson]

Hi Gavin, do you have more details on the hardware you’ve used. Clamps, SSR’s ect? I’d love to build my own

[@greenenergygav8053]

Hi Brett, My apologies for the delay in responding to you. In terms of components for the system, there's nothing special here. The current transformers and SSRs were low cost items purchased on eBay. The OpenEnergyMonitor site has good sections covering current transformer selection and setup (you'll need a simple voltage divider circuit to enable the Arduino to read the current transformer output). This is a link to a CT I used: www.ebay.com.au/itm/321424072757?hash=item4ad65e2035:g:lKYAAOxyNa9SJqTf The SSRs I used are all low cost devices. Obviously you need ones that accept DC input and are able to switch AC loads. I used 40A versions and have had no issues switching our 16A HWS element. There's a fair bit online regarding the reliability of Chinese made SSRs, but I've had no issues. I do use a bigger (quite meaty) heatsink on the HWS SSR. The AC voltage sensing can be done using small AC transformers (150mA is more than sufficient)or plugpacks to keep things simple. Again the OpenEnergyMonitor website has details of the voltage dividers you'll need to build. The rest is basic Arduino componentry. Again, here I've used clone devices with no issues. Good luck! Cheers, Gavin.

[@jerryp9647]

Isn’t there solution to limit your 3.6kW water heater electronically.? I mean if there is only 1kW generated the system will allow only 1kW to your 3.6kW heater.

[@greenenergygav8053]

Hi Jerry, Yes, you can utilise phase shift modulation to limit the actual power delivered to the hot water element. This approach is more complex and I didn't want to alter the switching hardware setup I was already using, hence the PWM approach. Cheers, Gav.

[@jerryp9647]

can you please suggest some device to utilise phase shift....thank you

[@greenenergygav8053]

Hi Jerry, I did a quick search and there are some products available for control of single phase resistive loads. I found quite a few in AliExpress, but I can't vouch for their performance or reliability. In general they accept a control signal of 0-10V or 0-5V and use phase shift control to vary the AC output. You'd need to the test the set up to firstly see if the device will accept a raw filtered PCM 1-5V DC (like) output from the Arduino. Assuming this works, you need to test the linearity of the device (ie: if 5V delivers near on 100% output, does 2.5V deliver 50% - I suspect not). The charts I saw look roughly linear from about 1.5V to 4.5V, but you'd need to test and model this and deduce and algorithm that produces the required power control. I'm a bit buried at the moment, but might have a play with one of these devices at some time. I hope this helps. Cheers, Gavin.

[@blake9358]

That's silly considering solar panels are only 23% efficient, and Solar hot water heaters convert just about all the heat from the sun into hot water

[@greenenergygav8053]

Hi Blake, Our evacuated tube water system comprises two banks of 30 tubes each of 1M in length. They are tilted at an angle to catch the winter sun. During winter and the shoulder months, they are not generally able to fully heat our 315L storage HWS. During this time, we divert the energy from our rooftop electric solar array to assist with hot water heating. In this way, the diverter is invaluable. During the warmer months, the evacuated tubes heat the water on their own with no assistance required from the electric solar array. Not silly at all really. Cheers, Gavin.