DC water heater element?

[GLC]

Sounds like it should help what voltage are you getting at the element ?

Ok, starting the experiment. Phase 1 just two panels wired in parallel. Each panel 305 watt, 40 volts, 8 amps. 24 volt dc element with thermostat. Panels wired directly to element. Amps at element is 15 amps in max overhead full sun with 22 volts. 5 gallon bucket 75 degree water measured 115 in 1 hour, 130 in 2 hours. Have had accumlative kilowatt meter on hot water heater for a week as of tomorrow. Will replace lower 220 volt element with dc element to preheat the water in tank. Will check kilowatt meter in one week to see the difference in kilowatt usage.

 

[Craig]

Ok, starting the experiment. Phase 1 just two panels wired in parallel. Each panel 305 watt, 40 volts, 8 amps. 24 volt dc element with thermostat. Panels wired directly to element. Amps at element is 15 amps in max overhead full sun with 22 volts. 5 gallon bucket 75 degree water measured 115 in 1 hour, 130 in 2 hours. Have had accumlative kilowatt meter on hot water heater for a week as of tomorrow. Will replace lower 220 volt element with dc element to preheat the water in tank. Will check kilowatt meter in one week to see the difference in kilowatt usage.

SO it appears that the panels can produce almost full amps ut obviously at a lower voltage. You may be wasting energy this way but right now my batteries are fully charged the sun is shining bright and my panels are producing O watts so any use of the available energy is a good use.

Thanks for the update.

 

[GLC]

So i did another little test today. I had the 400 watt 24 volt element but also had a 900 watt element. Using four 305 watt panels hooked in parallel. The 900 watt produced only 16 volts and 26.4 amps at the element contact . Took one hour for it to raise 75 degree water to 125. Went back to the 400 watt element with same setup. Registered 32 volts with 22 amps on the element. In thirty minutes the water temp was already 145 degrees. Since it is a 24 volt element , it may not last long at 32 volts? I probably need to find a 600 watt and see how it reads. Anyways i installed the 400 watt dc element replacing the lower 220 element in the water heater. Will let it run for a week to see the kilowatt difference reading on the hot water heater. With both ac elements, heater consumed 37.2 kw in one week. This is based on only me and my wife no kids.

 

[GLC]

I will post the good and the bad. 400 watt , 24 volt thermostat on the element fried in about 4 hours yesterday. I am ordering a different dc thermostat tomorrow. Good news is that the 400 watt element is working great. Still don't know how long it will last since it is pushing 700 watts for a 400 watt element. Since the 900 watt i had was a bit much and did not produce enough heat for the panels , i am ordering a 600 watt to see how it does later. Experiment is still on till next Saturday to see the difference in the kilowatt meter. Thinking about something else. I have the panels on the sw roof of the house. Thinking about possibly adding same panels on the se roof. I know that that over panels it but with the sun not shining on both sides of the roof at the same time only one side will be producing amps at one time. Just a thought for the future experiment. Nothing to burn up, no batteries or controller hooked up. Just thinking.

 

[Craig]

I'm not sure exactly what your end goal is here not because I'm confused but because I didnt read the whole thread again. What about putting 2 400 watt elements in series which should cut volts down to 18 per unit and not blow it out.

 

[GLC]

Right now just looking at the effects of decreasing kilowatt usage on the hot water heater by using one dc element on the bottom side of the hot water heater to preheat the water while trying different elements.You are correct that 2 , 400 watt elements would probably work pretty good. Remember , right now this just an evaluation. Depending how things work will determine what i do next. I still have to have hot water on demand 24-7 for the wife. Probably will wind up putting a second hot water heater in powered solely by solar to feed the main water as a pre- heater.

 

[efficientPV]

I have heard that experience before of these thermostats going bad. To my knowledge there are no suitable DC thermostats at that current and voltage. Hate to see people just burn thru a lot of parts.

Series elements offer many options. Many DC elements have two parallel elements which are just connected by shorting bars. In theory that gives the option of two parallel elements, one element ot two elements in series. Using just a small independent pilot PV panel, sun intensity can be measured. When enough sun is available, the second element can be paralleled. This gives greater low sun performance increasing daily output.

I know there is a comfort level with relays, but they are a poor solution. High power FET are an easy cheap solution and could be powered with just a 9V battery.

 

[GLC]

I have heard that experience before of these thermostats going bad. To my knowledge there are no suitable DC thermostats at that current and voltage. Hate to see people just burn thru a lot of parts.

Series elements offer many options. Many DC elements have two parallel elements which are just connected by shorting bars. In theory that gives the option of two parallel elements, one element ot two elements in series. Using just a small independent pilot PV panel, sun intensity can be measured. When enough sun is available, the second element can be paralleled. This gives greater low sun performance increasing daily output.

I know there is a comfort level with relays, but they are a poor solution. High power FET are an easy cheap solution and could be powered with just a 9V battery.

Can this parallel switching between the two be accomplished without doing it manually?

 

[Craig]

Can this parallel switching between the two be accomplished without doing it manually?

Seems to.me Drok makes some sort of current sensor that could turn on the second element when current reaches a set level. I'm not sure it would be necessary though. Especially in your case. I think the pre heater idea is great!

 

[tazmann]

I'm not sure exactly what your end goal is here not because I'm confused but because I didnt read the whole thread again. What about putting 2 400 watt elements in series which should cut volts down to 18 per unit and not blow it out.

How would you hook them up in series ? There is no + or - I think they would come out in parallel

 

[GLC]

How would you hook them up in series ? There is no + or - I think they would come out in parallel

In this case series would mean on the same circuit not as terminology of batteries and solar panels.

 

[tazmann]

With the amps your drawing you could use a relay, rather than pushing through thermostat, thermostat just operates relay.

 

[tazmann]

In this case series would mean on the same circuit not as terminology of batteries and solar panels.

Ok Thanks

 

[Craig]

Ok Thanks

No I meant in series like batteries or panels. I have not tried but I think it will work as expected. Positive to one element wire from other side of element to next element then negative to other side of that. The polarity does not really matter with some things.

 

[GLC]

DC Elements do not have a positive or a negative. You can hook them up either way, does not matter.

 

[tazmann]

No I meant in series like batteries or panels. I have not tried but I think it will work as expected. Positive to one element wire from other side of element to next element then negative to other side of that. The polarity does not really matter with some things.

Ok I thought that was what you meant . I have a 600 watt 12 volt element it is actually 2 300 watt elements in one . the wire studs have flat straps connecting them together. As per manufacture you can remove straps and run 1 at 300 watts 12 volt or with straps 600 watts at 12 volt , I'm not going to try hooking 24 volt to them because I dont want to burn them up

 

[GLC]

Ok I thought that was what you meant . I have a 600 watt 12 volt element it is actually 2 300 watt elements in one . the wire studs have flat straps connecting them together. As per manufacture you can remove straps and run 1 at 300 watts 12 volt or with straps 600 watts at 12 volt , I'm not going to try hooking 24 volt to them because I dont want to burn them up

Although hooking up a 12 volt to a 24 volt would work very good for a short time. Haha. I am finding out that elements actually work better being driven harder. Example, i have tried a 24 volt , 900 watt that was sized for the amount of panels i was using but it did basically nothing. I installed a 400 watt on a 900watt system and it worked great. I dont know for how long it will work though.

 

[efficientPV]

You could use a module like this https://www.ebay.com/itm/50A-Switch-Output-DC-Current-Sensor-Detection-Module-with-Base-5-24V/292903866720?_trkparms=aid=1110006&algo=HOMESPLICE.SIM&ao=1&asc=227684&meid=1f7be142709341b2b567aab198e203ad&pid=100005&rk=5&rkt=12&mehot=co&sd=233709018723&itm=292903866720&pmt=1&noa=0&pg=2047675&algv=SimplAMLv5PairwiseWebWithBBEV2bGenderDemotion&brand=Unbranded&_trksid=p2047675.c100005.m1851

Panels are current sources and below power point it doesn't matter if you are a quarter ohm or one ohm. You get the same current. You set the current adjustment to half of what you would expect to be the maximum current. At that time it will switch on the second heating element. Use a couple FET in parallel for lowest voltage loss. FET are cheap and there is no need to use a heatsink bigger than 4 square inches. At that you would barely know it is on. Use three 75N75 or two IRFB4321 Total amp capacity of FET should be 10 times your element current. The relay drives the FET gate and you need to guarantee 10V to insure full turn on. A 9V battery could do it or a zener protecting the gate so it never sees over 18V. So what do you have as a voltage source? I could draw something up.

Shocker...... At the time it does switch, the power instantly drops to half. But overall daily totals will increase. When this system is used it is often with three or four elements so the change isn't as dramatic. That is why people quickly determine full electronic is the way to go.

 

[sunshine]

Re-thinking common issues of direct heating.
From my setup https://diysolarforum.com/threads/dc-water-heater-element.11699/page-2#post-135871

Readings at 10am: PV end of cable 39.7amps 11.57volts
element 9.13v ......362w through element 100w wire loss (thanks ePV)

This thing works. The metal inlet to the tank gets untouchable at full sun at the 362w level.
But why only 9v at the element when all the theoreticals pointed above the wanted 12v and how best to get that 12v??

My panels are old and the aspect is set so instead of the original 190watt panel rating I should be using 150watts in the calculations. 20% output loss is somewhat of an industry standard I read somewhere.

I will need 4 panels at the mppt plus the 1 extra to provide that power leading up to the mppt (5 panels and not the original 4)

This underestimation of panels required would be the common problem in many of these setups!

With high wattage heating elements we are at the far right of the mppt curve. Any slight error pushes us over the ledge.
5 amps through any cell string is the limit so increasing the voltage to get more amps through the fixed resistence element isnt the answer if the panel can not supply those amps- as efficientPV correctly keeps pointing out. Thus the extra panel in parallel is the answer. Or more if we want to keep heating in poor sunlight.

Then we come to the area required!!

 

[sunshine]

Readings for 5 panels full sun
10am at element 42.6 amps 9.94v 423 watts
at pv combined junction 42.6 amps 12.53 v 534 watts

For 4 panels
Readings at 10am: PV end of cable 39.7amps 11.57volts
element 9.13v ......362w through element

I had connected a 50watt 12v halogen globe above inside fish tanks (disconnected for the readings) so the output could be observed. My substitute for a whr meter. For 4 panels it was constantly cycling from bright orange to off. Now with the added 5th panel it cycles from bright orange to nearly the intensity of the globes next to it even in thin cloud cover that extinguished the globe prior.
A proper watt/hr meter should confirm the magnitude difference adding that extra panel makes.