diy solar

diy solar

Direct to water heating.

Appreciate the feedback. I've been playing with this off and on for the last year. Currently I'm up to 1200 watt (4x72 cell panels) in series/parallel. I was going to run them all in series and push about 150v into the element but was trying to use DC SSR to do the switching. The issue is they keep blowing and shorting out. I'm only powering the bottom element and I pulled out the 220v element and run a 110 element to better match the watts. Aside from not being able to switch the power I can heat the 50 gallon tank to >180 in the summer and >140 in the winter. So we actually have to use hot water to keep the temp down. hee hee hee. If I could only switch it I'd be happy. I'm using the water heater thermostat to turn off/on a 12v dc power supply and that triggers the SSR. The SSR is rated 5-60 vdc at 40a. I'm pushing the voltage but well under the current rating by more than 50%. So I purchased a DC motor speed controller to try and get a zero crossing pulse so i could use the water heater thermostat but have not tried it yet. I just purchased a mini hand held scope to try and see the wave forms you supplied.
If you are getting the cheap SSRs through Amazon or eBay or the likes, they pretty much all lie on the rating.
 
was trying to use DC SSR to do the switching. The issue is they keep blowing and shorting out.

I'm pushing the voltage

So put panels all in parallel and get it working that way first.

was trying to use DC SSR to do the switching. The issue is they keep blowing and shorting out. I'm only powering the bottom element and I pulled out the 220v element and run a 110 element to better match the watts.

What does gate waveform and voltage look like?
Assuming saturated when on, all the power dissipation is during transitions.

Disconnect heating element and just drive a lamp first, make sure switching is fast.
Calculate energy deposited during transition, make sure thermal mass and heatsinking is going to be sufficient.

Connect top and bottom element in series for lower current as the next step.


Since it fails on, hope you have working temperature/pressure relief on tank.
An over-temperature thermostat probably can't interrupt DC. But a suitable amperage rated NO contact should be able to short out PV panel and drop voltage to zero. Not perfect, because if wire comes loose it stops protecting, but better than nothing.
 
Appreciate the feedback. I've been playing with this off and on for the last year. Currently I'm up to 1200 watt (4x72 cell panels) in series/parallel. I was going to run them all in series and push about 150v into the element but was trying to use DC SSR to do the switching. The issue is they keep blowing and shorting out. I'm only powering the bottom element and I pulled out the 220v element and run a 110 element to better match the watts. Aside from not being able to switch the power I can heat the 50 gallon tank to >180 in the summer and >140 in the winter. So we actually have to use hot water to keep the temp down. hee hee hee. If I could only switch it I'd be happy. I'm using the water heater thermostat to turn off/on a 12v dc power supply and that triggers the SSR. The SSR is rated 5-60 vdc at 40a. I'm pushing the voltage but well under the current rating by more than 50%. So I purchased a DC motor speed controller to try and get a zero crossing pulse so i could use the water heater thermostat but have not tried it yet. I just purchased a mini hand held scope to try and see the wave forms you supplied.
Even if your DC speed controller does work, how would you control the duty cycle? That still makes it pretty useless. You still need a capacitor bank to store the energy in the off cycle to use later. It never ceases to amaze me how people will spend more money and effort to do things the wrong way. Power point water heating is known and working technology. If you have any money after this and are serious, contact me.
 
So where are we with this . Has anyone installed a direct system and seen how it works and been happy with the result . Looking back to the start Offgrid steve said its a waste of time but his current thinking has changed . However he's gone dark now and removed his channel I think.
So I have 4 panels of 250w each and want to get started . Any working Recommendations please?
 
If you are getting the cheap SSRs through Amazon or eBay or the likes, they pretty much all lie on the rating.
I'd gladly purchase more expensive American made but nothing turned up in my search? The highest working voltage I found was 60Vdc. I'd like 200Vdc with 10a to run the 4 panels in series. I can run series/parallel so 73.36vdc @ 16.36a but don't want to run voltage any lower.

So put panels all in parallel and get it working that way first.
Trying to keep high voltage low current. Panels are 36.68 Vmpp & 8.18 Impp. I'd rather not parallel and deal with 32.74 amps and the associated wire gauge and line losses. I replaced the 220vac element with a 110v
What does gate waveform and voltage look like?
Assuming saturated when on, all the power dissipation is during transitions.

Disconnect heating element and just drive a lamp first, make sure switching is fast.
Calculate energy deposited during transition, make sure thermal mass and heatsinking is going to be sufficient.

Connect top and bottom element in series for lower current as the next step.
Bottom is for solar (free), top is for 220vac to catch bad days if solar can't keep up.

Even if your DC speed controller does work, how would you control the duty cycle? That still makes it pretty useless. You still need a capacitor bank to store the energy in the off cycle to use later. It never ceases to amaze me how people will spend more money and effort to do things the wrong way. Power point water heating is known and working technology. If you have any money after this and are serious, contact me.

The point of the motor speed controller was not to regulate power but to provide pulsed DC so I had a "zero" point so the AC thermostat would work without arching. So I'm not trying to store power but run at max frequency, I'd set the PWM as fast as possible for the most power as long as it reached 0. The caps prevent the wave from reaching 0 so want to remove them.

Since it fails on, hope you have working temperature/pressure relief on tank.
An over-temperature thermostat probably can't interrupt DC. But a suitable amperage rated NO contact should be able to short out PV panel and drop voltage to zero. Not perfect, because if wire comes loose it stops protecting, but better than nothing.

Yes, I've popped the over temp serveral times. This system works too well, need to throttle it back and control it. DC power from the panels are switched on and off by the DC SSW. The SSW is switching on/off the AC side of a 220/12v power supply so I don't get arching and the secondary controls the SSW.

I removed the stock 240v 5500w element and found a tankless 120v 1200w element that's a pretty good power match for the panels. Just doesn't make sense to purchase a $2,000-$4,000 solar water heater or $125 "DC" elements. :-0 You'll never recoup your money....
 
Trying to keep high voltage low current. Panels are 36.68 Vmpp & 8.18 Impp. I'd rather not parallel and deal with 32.74 amps and the associated wire gauge and line losses. I replaced the 220vac element with a 110v


I removed the stock 240v 5500w element and found a tankless 120v 1200w element that's a pretty good power match for the panels.

For a given resistance element, lower voltage is lower current. So all panels in parallel or a single panel is going to be lower current than same panels in series. (No reason to put in parallel, actually; just use single panel.) If four in series or 2s2p works, single panel will be lower or at most same current. Just suggesting that for less stress on SSR until everything is working reliably.

A stock 240V 5500w element is a 120v 1500w element. Two of those in series (top/bottom) would have been 120v 750w element. Your 120V 1200W element isn't that much different in resistance, so wiring the two in series would reduce power (perhaps as a lower current test.)

I'd set the PWM as fast as possible for the most power as long as it reached 0. The caps prevent the wave from reaching 0 so want to remove them.

The higher the frequency of PWM, the more power dissipated in SSR.
Put a thermocouple on SSR or use IR gun and see if it gets hot.
 
For a given resistance element, lower voltage is lower current. So all panels in parallel or a single panel is going to be lower current than same panels in series. (No reason to put in parallel, actually; just use single panel.) If four in series or 2s2p works, single panel will be lower or at most same current. Just suggesting that for less stress on SSR until everything is working reliably.

A stock 240V 5500w element is a 120v 1500w element. Two of those in series (top/bottom) would have been 120v 750w element. Your 120V 1200W element isn't that much different in resistance, so wiring the two in series would reduce power (perhaps as a lower current test.)



The higher the frequency of PWM, the more power dissipated in SSR.
Put a thermocouple on SSR or use IR gun and see if it gets hot.
Yea I went back to the drawing board on the element selection. I want to keep the top 220 AC stock on its thermostat set to 120. The bottom I need to change to utilize the full power and set it to 150.

NOTE: it's been my observation that the top element should be set to work off the bottom thermostat as hot water rises. So set the top element to 120 (at the bottom) and the bottom solar element to 150 at the top thermostat)

From my calculations I need to use one of the following;
240v 3000w (19.2 ohms) element - 4S panels - 146.7v @ 7.6a (max Impp=8.18, Ics=8.72a) 1115 watts
240v 3800w (15.2 ohms) element - 3S panels - 110.1v @ 7.3a (max Impp=8.18, Ics=8.72a) 803 watts

NOTE: These are off the shelf Home Depot elements under $15 each.

So I'll still be back to my original problem of switching high voltage DC. :-( The highest voltage SSR I found was a TWTADE SSR-40DD. It's 3-32vdc input and 5-60vdc out@ 40 amps. If there's an American made switch that could handle 110vdc I'd use that and put that extra panel back on my RV. :) I'm just playing around with stock I have on hand and stole 1 of the 3 panels on my RV to experiment with.

I also have a 3500w MSW inverter and x3 12v 4D AGMs sitting around too but didn't want to tie up those with this project, I have other plans for those.

Solar Power36.6836.68
73.36​
110.04​
146.72Vmpp
8.1816.3616.368.188.18Impp
30060012009001200watts
1S2P2S2P3S4SConfig
VoltsWattsampsohms
120144012.010.03.73.77.311.014.7amps
120150012.59.63.83.87.6*11.515.3*this is where I'm set now
120165013.88.74.24.28.412.616.8
120200016.77.25.15.110.215.320.4
24015006.338.41.01.01.92.93.8
240250010.423.01.61.63.24.86.4
240300012.519.21.91.93.85.77.6
240350014.616.52.22.24.56.78.9
240380015.815.22.42.44.87.39.7
240450018.812.82.92.95.78.611.5
240500020.811.53.23.26.49.612.7
240550022.910.53.53.57.010.514.0
240600025.09.63.83.87.611.515.3
Watts
803.3​
1115.1​
DC SSR (1280x590).jpg
 
Last edited:
Some research has indicated that the ideal resistance in direct connect is not ideal at all in day to day total performance. The ideal resistance can almost be doubled without a monthly power loss. This is most true for locations that have overcast and clouds. In most systems there is unneeded power on sunny days and on cloudy days the performance more than doubles with higher resistance.

Pulsing power just to use mechanical thermostats is pointless.

I couldn't be more pleased with how I heat water.
 
Some research has indicated that the ideal resistance in direct connect is not ideal at all in day to day total performance. The ideal resistance can almost be doubled without a monthly power loss. This is most true for locations that have overcast and clouds. In most systems there is unneeded power on sunny days and on cloudy days the performance more than doubles with higher resistance.

Pulsing power just to use mechanical thermostats is pointless.

I couldn't be more pleased with how I heat water.
Thread readers note .
This guy also goes under the name of "opera" and makes a lot of videos on the subject of heating hot water .
However if you try to make some of his circuits work , they don't.
If you look closely at the circuits they cant work and he seems to be trolling the electrical forums posting this rubbish to cause confusion .
He must find this approach amusing but you will waste your time if you try to follow him.
 
Last edited:
Thread readers note .
This guy also goes under the name of "opera" and makes a lot of videos on the subject of heating hot water .
However if you try to make some of his circuits work , they don't.
If you look closely at the circuits they cant work and he seems to be trolling the electrical forums posting this rubbish to cause confusion .
He must find this approach amusing but you will waste your time if you try to follow him.
Members take note.

if someone makes a troll post... do not reply to it...

report it, so it doesn’t cause additional drama.

drama is the desire of the troll... don’t give it.

just Report.
 
If your water heater is connected to a city water system and there is not a backflow device on your water main your water pressure will not get high enough to blow up your water heater, period. IF your thermostat did not open and your high limit did not open then the run away energy would push back into the cities water main to absorb the excess pressure. Now if you shut the water inlet valve off to your water heater that all changes because the pressure will then build up in the house/water heater. This can all be avoided with a simple temperature / pressure relief valve installed on your water heater tank. ALL domestic water heaters come with a temp/pressure relief valve.
 
All water tank installations should have an expansion tank. Tank pressure relief valves will start leaking over time as the result of lime scale buildup from minor discharges due to expansion. With solar heating these wide temperature changes will happen far more often.
 
All water tank installations should have an expansion tank. Tank pressure relief valves will start leaking over time as the result of lime scale buildup from minor discharges due to expansion. With solar heating these wide temperature changes will happen far more often.
You only need an expansion tank IF you have a backflow device between your water heater and the city water system and/or the city pressure is near or exceeds your pressure relief valve setting with a pressure reducing valve installed. When your water heater heats up and the water expands it will push the expansion into the city system with no increase of pressure in your tank under normal operating conditions.
 
I have a 40 gallon tank in a garage I built a dozen years ago with a 40 gallon tank used only for clothes washing. It is against code to run a pipe from house to garage so I run a 100F garden hose. Of course you don't want to keep that hose turned on when you leave. The tank has a 4' 3/4 inch standpipe for expansion. That is what I had around and the tallest I could get without hitting an open garage door. That pipe is topped off with 6 inch pressure gauge. It was sitting around and couldn't find anything else to cap the tube. Normally the pressure is 40psi and I hang around the garage a lot. With the house valve turned off and allowing for expansion in the hose and tube the pressures go well over 130psi before I catch it and drain a half quart off at the sink. I see changes from 80-130F and that is only for the stratified top 15 gallons. A quick calculation for expansion of this tank .1 gallon for every 10C. This year I hope to add more expansion so I don't have to be watching it all the time. Garden hoses and washer valves are expensive.
 
I have a 40 gallon tank in a garage I built a dozen years ago with a 40 gallon tank used only for clothes washing. It is against code to run a pipe from house to garage so I run a 100F garden hose. Of course you don't want to keep that hose turned on when you leave. The tank has a 4' 3/4 inch standpipe for expansion. That is what I had around and the tallest I could get without hitting an open garage door. That pipe is topped off with 6 inch pressure gauge. It was sitting around and couldn't find anything else to cap the tube. Normally the pressure is 40psi and I hang around the garage a lot. With the house valve turned off and allowing for expansion in the hose and tube the pressures go well over 130psi before I catch it and drain a half quart off at the sink. I see changes from 80-130F and that is only for the stratified top 15 gallons. A quick calculation for expansion of this tank .1 gallon for every 10C. This year I hope to add more expansion so I don't have to be watching it all the time. Garden hoses and washer valves are expensive.
What country do you live in that has a code against running water to a garage? I have seen more garages with water run to them for Water Heaters and Washing Machines than I can remember. In your unusual situation I would recommend a small expansion tank on the cold side of the water heater.
 
Last edited:
Check out opera on the youtube. He's been working on MPPT solar water heater controllers for a while. It's DIY and you have to snoop around a bit to figure out exactly what he's talking about, but I built my version, and it works. I'm not happy with my MOSFET driver yet, but I'll get that figured out come Fall when my workload declines a little. My TL494 PWM circuit runs at around 22KHz and dives a 1500W heating element at ~120 VDC.

Check him out:
 
Thread readers note .
This guy also goes under the name of "opera" and makes a lot of videos on the subject of heating hot water .
However if you try to make some of his circuits work , they don't.
If you look closely at the circuits they cant work and he seems to be trolling the electrical forums posting this rubbish to cause confusion .
He must find this approach amusing but you will waste your time if you try to follow him.
I have built a couple of his circuits that did work. His concept is absolutely sound, though he's not very good when it comes to explaining and assumes a lot of prior knowledge, as well as a good understanding of analog electronics. Based on his videos, I built a ~22KHz TL494 based controller connected to 1500W, 120v, IR heating elements. I'm just getting some ringing at the source pins of my MOSFET that I want to calm down. It also helps to put a back-biased diode across the heating element if it has an inductive component.
 
I imaging you are getting a lot of FET heating at 22KHZ. Are you using an external FET driver as the 494 is poor with FET at that frequency? I'd slow it down to 500Hz. A small voltage divider to increase deadtime slightly will give arc interrupt capability.

Here is a recent snapshot using E-test around noon. Not a great program, but cheap and it provides a visual to how your system operates that you might not be aware of. You can see the voltage Green) dropping slightly indicating tracking. The voltage stays near power point even at low light intensity.
 

Attachments

  • 20210407_120620.JPG
    20210407_120620.JPG
    91.3 KB · Views: 0
I imaging you are getting a lot of FET heating at 22KHZ. Are you using an external FET driver as the 494 is poor with FET at that frequency? I'd slow it down to 500Hz. A small voltage divider to increase deadtime slightly will give arc interrupt capability.

Here is a recent snapshot using E-test around noon. Not a great program, but cheap and it provides a visual to how your system operates that you might not be aware of. You can see the voltage Green) dropping slightly indicating tracking. The voltage stays near power point even at low light intensity.
Thanks fo your comments! I'm driving two IRFP250M MOSFETS with a MCP1407 gate driver mounted on a custom board with the FETs to optimize its performance. I had no gate resistors and was getting ringing on the source that exceeded the voltage rating of the FET. I will try putting a 10Ω in series with each gate and see if the knocks down the ringing. I never blew a FET, and for the most part they ran cool. I chose the higher frequency because I was load dumping into basic resistive heaters with coiled heating elements. They would make buzzing / singing sounds at audio frequencies, and bizarre hissing sounds over ~5KHz, and I'd get FET heating at times. The pulse width also played a role in FET heating and audible noise. 20KHz and above seemed to work the best in my application. I'm busy with other stuff over the summer, but plan to get back to this come Fall. Interesting to play with, and it shouldn't take much to build a simple, reliable circuit.

Opera's circuits take the DC power from the solar panels, takes it through a diode, so his circuit doesn't mess up the charge controller, if you have one, and charges up a big capacitor bank. Some comparitors look at the voltage across the caps, and if it get above a certain voltage, it turns on a FET and dumps the energy in the caps into a water tank heating element. When the voltage across the caps get below a certain point, the circuit turns off the FET, and the cycle repeats. Variable resistors are used to set the turn on / turn off voltages. Easy-peasy. I built one, it works. The more "advanced" circuits look at the voltage across the the caps, and use it to drive a PWM circuit. When the voltage goes above the setpoint, the duty cycle increases, and when it goes below the setpoint, it decreases. The duty cycle doesn't go above 90%, so the FET is off for 10% of the time or more. Why? To provide an opportunity for an arc to quench, in case the thermal breaker, for instance, opens. Clever. DC currents are much more likely to arc when a contact is broken, then AC.
 
I was going to say add a series inductor and a diode up from ground so current recirculates and call it a buck switching regulator.
But then you won't have the square wave to help with contact arcs.
Run PWM through the contact, and put the inductor/diode after?
 
Back
Top