Sun Power SPR-E19-320 to power element directly series/parallels issue

jimmyaz

Hey guys,

I have a odd situation and I have no idea what's wrong and had called SunPower but they offer no help (rediculous).  They wouldn't even speak to me because I bought the panels used locally and because I don't have an account with them.

Basically I got a few of these panels.
https://us.sunpower.com/sites/sunpower/files/media-library/data-sheets/e-series-home-solar-panels-comparison.pdf

Specs: 
Nom: 320watts
VOC: 64.8 V
VMPP: 54.7
ISC: 6.24 A

I am planing to connect a few of these panels to my water heater directly, currently I have not change the element, it's still on the 220v/4500watts element.  But this is only for testing purposes.  I was hoping to series all the panels together and run high voltage, but upon my testing here's the issue.

If I only connect panel 1 to the element (UNDER LOAD) I am getting 52volts X 4amps = 208watts
If I only connect panel 2 to the element (UNDER LOAD) I am getting 53volts X 4amps = 212watts

So on theory, if I series two panels together, it should give me 104v X 4amps = 416watts right?  Well, that's not the case.  When i series them I get 70volts X 4amps = 280watts.

What the hell?

So just for the hell of it, i tried parallels.

When I parallels 2 panels together I get  52volts X 4.6amps = 239watts (each panel give 2.3amps or so)....  I verify this by leaving one panel unplug and have a DC meter (coil) around one panel and it say 4.5 amps but as soon as I plug in the other panel, it drop to 2.3A or so....


SO CONFUSED? PLEASE HELP.

Estragon

It sounds to me like the heater element resistance is such that it only presents a load of 200-odd watts to the panels. The element is designed for around 20aac@220v so the resistance is way off what you need.

Our host sells such elements but you would need a charge contoller to get the desired voltage.

jimmyaz

Estragon said:

It sounds to me like the heater element resistance is such that it only presents a load of 200-odd watts to the panels. The element is designed for around 20aac@220v so the resistance is way off what you need.

Our host sells such elements but you would need a charge contoller to get the desired voltage.

Ok, I have thought of this as well that it could caused by the element restricted the wattage output to 200-300watts.

But how come using the same 220v 4500watts element on 120V AC give me 1100watts of energy?  Could it be that much of a difference between 120V AC vs 100V DC?  1100watts  vs  200-300watts?

This hurts the brain... lol

Estragon

Just curious - did you actually see 104 v when you wired the panels in series before hooking up the load?

mcgivor

It would appear that the resistance of the element is too low, dragging the voltage output of the panels down. It must be remembered that current is inversely proportional to voltage, the resistance in you equation is the only constant, doubling the voltage will halve the current,. If the panels had the capacity to maintain the voltage "the theoretical 104V" then the results would be as expected, but the panels obviously cannot, so the results will be as you see. More panels in a series/parallel configuration would probably improve the situation as the volt drop would be reduced, or matching the load to the voltage would be another solution. 

animatt

It makes sense that sun power will not deal with you.  They do NOT sell to the public. If you want sunpower panels you have to get them thru an authorized dealer/installer. Or buy them used.


While I would personally like to purchase panels directly from them.  Seems like a very legitimate busy model.  They are not making any money off of someone buying used panels.

Getting resistence values correct like mcgivor says is key.

Getting a different element.  Maybe be an ok idea.

To maximize power off current setup it is exactly as mcgivor says. You will need to 
Parallel and series connect.

It is fairly easily calculated. Ohms laws I believe. 

mike95490

And beware the AC thermostat contacts that will weld themselves shut on DC

jimmyaz

Estragon said:

Just curious - did you actually see 104 v when you wired the panels in series before hooking up the load?

Yes, OPEN voltage for 2 panels in series is actually 128volts (shorted voltage I am not sure as I did not want to short that much of a voltage together.

animatt said:

It makes sense that sun power will not deal with you.  They do NOT sell to the public. If you want sunpower panels you have to get them thru an authorized dealer/installer. Or buy them used.


While I would personally like to purchase panels directly from them.  Seems like a very legitimate busy model.  They are not making any money off of someone buying used panels.

Getting resistence values correct like mcgivor says is key.

Getting a different element.  Maybe be an ok idea.

To maximize power off current setup it is exactly as mcgivor says. You will need to 
Parallel and series connect.

It is fairly easily calculated. Ohms laws I believe. 

Humm.... I can understand if they rejected if I was after warranty or replacement... but this is simply technical question regarding their products.  I feel like a company is obligated to answer any question regarding on their products regardless of who is asking.  It's not like I am asking how they make the panels...  Don't worry, you won't see me buying any of their panels "USED" or "NEW"

I am convince it's most likely the element is the issue here, because there's only two variable in this equation Solar Panel vs Element.  Would be hard to believe if the panels is the issue.

Am I correct to say that I do not need any diodes between the panels when wired in Series or Parralles if both panels is fully in the SUN?  Diodes only would benifit if one panels is being shades?  (diodes built in the panels themselves is a different thing here)?

mike95490 said:

And beware the AC thermostat contacts that will weld themselves shut on DC

Is this because of the amperage rating or because DC arc more than AC?  Weld themselves shut on DC mean it will closed circuit?  I would think even in this case 900watts, it can run all day and still wouldn't be hot enough to trigger the relief valves.

Well, I already ordered a 48volts 1000watts element. 

thanks for all the advise guys.

mvas

Forget the heating element for now.

Do you actually read 64.8 Voc ?
Do you actually read 6.24 Amps Isc ?

Let's establish that the panels are operating at baseline spec before connecting a load.

RE: Welding Contacts - it is more difficult to stop DC current (constant arc) vs AC Current (self extinguishing 120 times/second)

mcgivor

Your panels are probably fine, leave the diodes in place as they have negligible influence. DC will arc when the contact makes or breaks and if the potential is high enough the temperature of the arc can literally weld the contact faces together, AC on the other hand has an, on off on waveform, the frequency and this quenches the arc unlike DC. You Tube should has videos on this subject.

jimmyaz

mvas said:

Forget the heating element for now.

Do you actually read 64.8 Voc ?
Do you actually read 6.24 Amps Isc ?

Let's establish that the panels are operating at spec before connecting a load.

RE: Welding Contacts - it is more difficult to stop DC current (constant arc) vs AC Current (self extinguishing 120 times/second)

I remembered VOC is around that range when checked this morning at noon when I believe I would get most power.  It's around 60-62 for VOC.

The ISC I am not sure.... I did not short the panel to themselves and test it.  I only hook each single panel to the AC(4500w) element and getting voltage under load around 55v and amps around 4.  So the correct way to test specs is to cover the panel, connect negative and positive together and use a DC clamp meter to measure the ISC right?  But how to measure the loaded voltage?

mcgivor said:

Your panels are probably fine, leave the diodes in place as they have negligible influence. DC will arc when the contact makes or breaks and if the potential is high enough the temperature of the arc can literally weld the contact faces together, AC on the other hand has an, on off on waveform, the frequency and this quenches the arc unlike DC. You Tube should has videos on this subject.

So best to replace with these thermostat?  They are advertised for DC usages and can handle (Can handle all voltages from 12 to 220 volts, AC or DC).  I am using one right now in my 600watts 22gal off grid system, but not sure if it benefit anything, as I have used the stock 120v thermostat original from the tank and it seem to work just as well (but I could be wrong in the long run where it turn off and on a lot).  But with so littler wattage, it seem it always remain ON all the time.

http://mwands.com/store/dc-water-heating-elements/water-heater-thermostat

mcgivor

The AC thermostat will "work" for a while, but it the not working that is the problem, personally I would recommend the DC one.

jimmyaz

So according to this calculator.
http://www.sengpielaudio.com/calculator-ohm.htm

If my 4500watts element specs is below.
4500W = I x 240V
I = 4500W / 240V = 18.75A

240V = 18.75A x R
R = 240V / 18.75A = 12.8 Ohms

If I wired 2 panels in parallels, type in 55volts and 12.8ohm I get 4.29amps and 236watts (this match up with my actual result).

If I wired in series, type in 12.8ohm and 4amps and I get voltage result 51v and 208watts.

mvas

jimmyaz said:

So according to this calculator.
http://www.sengpielaudio.com/calculator-ohm.htm

If my 4500watts element specs is below.
4500W = I x 240V
I = 4500W / 240V = 18.75A

240V = 18.75A x R
R = 240V / 18.75A = 12.8 Ohms

If I wired 2 panels in parallels, type in 55volts and 12.8ohm I get 4.29amps and 236watts (this match up with my actual result).

If I wired in series, type in 12.8ohm and 4amps and I get voltage result 51v and 208watts.

Per your measurements the resistance of the heating element is increasing at the higher 70 Volts ...
53 volts / 4 amps ~= 13 ohms 
70 volts / 4 amps ~= 17.5 ohms <<< ???

mvas

You need to draw ~5 amps at 64 Volts to obtain 320 watts from your panel.
Any more or less than 5 amps and you will not be at MPP - not 320 Watts.

What is the Isc measurement ?

jimmyaz

mvas said:

You need to draw ~5 amps at 64 Volts to obtain 320 watts from your panel.
Any more or less than 5 amps and you will not be at MPP - not 320 Watts.

What is the Isc measurement ?

I will try to get more data today.

So the lower the resistance in the element, the higher the voltage would drop causing less wattage?

Also another question, is there any efficiency disadvantage if running 120VAC on a 220VAC element, other than it take longer to heat the water?  Running on high voltage give more efficiency (power consumption wise) only apply to motor right?  For resistance application, wouldn't matter?

jimmyaz

Confirmed it's the element issue.

I hook it to my charge controller and battery and with 1 panel, it put 200watts into my battery, connect a second panel in parallels it put 400watts into my battery.  However I could not test the series connection because my charge controller max out at 100VDC.

I have also test with a 120VAC (900watts or so element) and the result is about 80-100watts increase over the 240VAC 4500watts element.

So I guess the right thing to do here is if your total panels is rated for 1200watts, get a 900-1000watts element would be wise.  Getting the same wattage element or higher and you loose so much power. 

mvas

jimmyaz said:

Confirmed it's the element issue.

I hook it to my charge controller and battery and with 1 panel, it put 200watts into my battery, connect a second panel in parallels it put 400watts into my battery.  However I could not test the series connection because my charge controller max out at 100VDC.

I have also test with a 120VAC (900watts or so element) and the result is about 80-100watts increase over the 240VAC 4500watts element.

So I guess the right thing to do here is if your total panels is rated for 1200watts, get a 900-1000watts element would be wise.  Getting the same wattage element or higher and you loose so much power. 

The Charge Controller is maintaining the Max Power Point.
What forces the Solar Cells to operate at the MPP when they are connected directly to a Heater Element?

jimmyaz

mvas said:

The Charge Controller is maintaining the Max Power Point.
What forces the Solar Cells to operate at the MPP when they are connected directly to a Heater Element?

Not sure what your point?  Are you saying that the solar panels can not output their max power if connect directly to a Element or a load?  I thought MPPT is only useful for if a battery was in the equation. 

mcgivor

When the panels are connected to the controller charging the battery you will get the full available capacity of the panels assuming the battery is not fully charged, the voltages are matched. When you connect to the heating element directly there is no control and the element will only take what it can.

Say for example you had a 1000W  element rated at 120V and you were to apply 60V to it, this is the calculation you would use to determine the wattage.

1000×(60/120)squared = 250 watts

Now you are going to apply 60V to a 4500w element rated at 240V  

4500×(60/240)squared=281.25 watts   60÷240=0.25,   0.25×0.25=0.0625,  4500×0.0625=281.25

This demonstrates why it is best to match voltage ratings, now 1000W element 24V rating using 24v
1000×(24/24)squared=1000W

Go back to your original post and use the formula with the figures you gave, the 52V single panel ones you will see the results are correct, Note, the element is probably rated fo 240V not 220V. 

jimmyaz

mcgivor said:

When the panels are connected to the controller charging the battery you will get the full available capacity of the panels assuming the battery is not fully charged, the voltages are matched. When you connect to the heating element directly there is no control and the element will only take what it can.

Say for example you had a 1000W  element rated at 120V and you were to apply 60V to it, this is the calculation you would use to determine the wattage.

1000×(60/120)squared = 250 watts

Now you are going to apply 60V to a 4500w element rated at 240V  

4500×(60/240)squared=281.25 watts   60÷240=0.25,   0.25×0.25=0.0625,  4500×0.0625=281.25

This demonstrates why it is best to match voltage ratings, now 1000W element 24V rating using 24v
1000×(24/24)squared=1000W

Go back to your original post and use the formula with the figures you gave, the 52V single panel ones you will see the results are correct, Note, the element is probably rated fo 240V not 220V. 

Yes, I have a better idea of how it work now.

I already ordered a rated 48VDC 1000watts element.  But I think even this one might be too much, since my 3 panels all series together isn't going to be able to put out enough power to fully power this element.  Maybe I should had got the 600-700watts element?

What are your thoughts on this?  this claim to be a MPPT controller for PV and element.  Haven't had time to read over all the details, but it said no need to change the AC element, so maybe it's just inverting to higher voltage?  But at 250.00 for just the controller... seem crazy, cost almost as much as my new water heater tank. 

http://techluck.com/

BB.

Sorry Jimmy for the late reply--Life (and flu) sometimes interfere with digital life.

Anyway, as the folks above have discussed, the problem is that you have to "match" the element with the solar array. If you have a fixed resistance (heating element), then you double the voltage, the current will also double (I=V/R). So, if one panel works at ~Imp, then you would need 2x panels in series * 2x parallel strings for 4 panels total.

I do not know anything about the MPPT controller for the Heater... But it does not appear to be a classic buck/boost type controller (which is usually used in an MPPT system).

So, I do not think this is a true MPPT controller--Leaving open to question exactly what it is doing.

If you want to see what a true MPPT controller for a similar application (Linear Current Booster for DC well pumps), take a look at this:

https://www.solar-electric.com/solar-converters-ppt-16-linear-current-booster.html


See the capacitors + relatively large coil (torriod type here)... You need these energy storage units to perform Buck and Boost type DC to DC conversions.

This type of device could work well (I think) for water heater element driving--But it all depends on your heating element/solar array/desired power... I am not sure there is a ready supply for higher voltage off the self MPPT solutions for water heating.

Regarding AC on a DC switch--This Youtube video does a nice job of showing what the difference is:

https://www.youtube.com/watch?v=Zez2r1RPpWY

Lastly--There is also the question of what you eventually want to do... A true Hybrid Water heater (uses a refrigeration compressor to heat the water", electrical heater as backup)--They are about 2-3x as efficient as a resistance type heater and an work well with an off grid power system--Or even with an on grid home.

The plus and minus of this--They output cold/dry air. If you live in a warm humid environment--That is not a bad thing. But if you have a "cold basement" (less than ~55F), more cold air may not be a good thing (folks have ducted outside air to these hybrid water heaters). And if you have cold winters, they may not work well during cold weather (fall back to resistance heating).

http://www.geappliances.com/ge/heat-pump-hot-water-heater.htm

-Bill

Estragon

Bill - great video!

jimmyaz

BB. said:

Lastly--There is also the question of what you eventually want to do... A true Hybrid Water heater (uses a refrigeration compressor to heat the water", electrical heater as backup)--They are about 2-3x as efficient as a resistance type heater and an work well with an off grid power system--Or even with an on grid home.

The plus and minus of this--They output cold/dry air. If you live in a warm humid environment--That is not a bad thing. But if you have a "cold basement" (less than ~55F), more cold air may not be a good thing (folks have ducted outside air to these hybrid water heaters). And if you have cold winters, they may not work well during cold weather (fall back to resistance heating).

http://www.geappliances.com/ge/heat-pump-hot-water-heater.htm

-Bill

Hey Bill,

Thank you for taking your time to reply.  It seem using PV for water heater is getting very popular and now make sense.  On my house, I have been using the extra power coming off my (2.2kw) off grid system to preheat a 22gal tank with just 24v/ 600w and it work amazing.  During summer months my gas heater is completely off, during winter it come on maybe once a day and this has saved a lot.  But this is using charge controller/relay and a bunch of other stuff that just cost too much.

I am doing this new project for my parent house.  I want to simplify thing this time and decided to do PV directly to thermostat/element.  Eliminating everything else to be cost effective and simple.  Therefor with those controller at that price make no sense.   

I guess the right way to do it here is to get the right rating element and more solar panels.. after all, I got them 320watts panel at 150.00ea locally... and can get as many as I needed.  300dollars and I can get a extra 600watts, there's no way those controller can increase 600watts. 

As for the DC ARC using AC thermostat... the video clearly demonstrated this.  But my question is... that's at 220VDC and unlimited supply of amps... or a lot of amps is available.

How different would the result be on 120-150VDC at 5-10amps + voltage drop under load like what we have seen when the element is hook to it...

BB.

Once you are over the minimum voltage (typically 12 VDC) and over a minimum amount of current (not sure, few amps?), the DC arcing can be a problem. People have used electronic power switches (solid state relays) to "protect" the thermostat contacts...

http://www.mouser.com/Electromechanical/Relays/Solid-State-Relays/Solid-State-Relays-Industrial-Mount/_/N-ay0tw/

Not cheap--But can be a help.

I have a long (not pretty) thread from somebody that wanted to do this--The math, and even simulations (focused on winter--because that was his question). Non-winter/sunnier climates, yes this can make good sense. However, for the deep winter up north, did not seem to make a lot of sense.

https://forum.solar-electric.com/discussion/comment/307090/

Sorry for the lost of formatting (we went through a software vendor change--I guess I can fixing if it would be of help the detailed data--if anyone cares).

-Bill

jimmyaz

BB. said:

Once you are over the minimum voltage (typically 12 VDC) and over a minimum amount of current (not sure, few amps?), the DC arcing can be a problem. People have used electronic power switches (solid state relays) to "protect" the thermostat contacts...


-Bill

So what about these thermostat that say it can handle up to 250VDC?  This is what I have been using on my house for a few months... but since it's had never ran the tank above the rated temp, don't think it was ever trip... so not sure if it actually work any better than the stock thermostat.

http://mwands.com/store/dc-water-heating-elements/water-heater-thermostat

Sun in winter isn't a issue here in Phoenix Arizona....  It can be 30F outside and the sun is shining and clear blue sky almost everyday.  Although the day is much shorter, but you can still get 4-5 hours of usable sun light.  This is why it's a no brainer to not use PV in Arizona.    Especially in winter time... sun shinning on the panel and cold temp. to cool the panels. 

BB.

If the thermostat has DC voltage AND current ratings that support your system... Should not be an issue.

Note many switches have very different AC and DC ratings--Read the label information closely.

-Bill

jimmyaz

Hey guys,

Hate to keep posting about this issue... but got 3 panels mounted today... hook it to the 240v/4500 element and getting this.

Noon time, so should be getting 70-80% of total solar power (3x of 320watts/ea)

240v/4500watts element, panels= VOC = 180volts, under load is 75volts and 6amps = 450watts.  As expected. 

Also received my.... 48volts DC 1000watts, put that on and give it a try... and I'm now puzzled.

240v/4500watts element, panels= VOC = 180volts, under load is 15volts and 6amps = 90watts.  ???

 I thought getting the lowered voltage element and the closed rated watts would give me more than the AC 4500watts element... but that's NOT the case.  I haven't got the time to parallels all 3 panels together and run it closer to the rated 48v at 15amps and see what happen... 

jimmyaz

I guess I should have asked.

If I have 4x of these panels. Which configured should I go with and what element would give me the best result?

Specs: 
Nom: 320watts
VOC: 64.8 V
VMPP: 54.7
ISC: 6.24 A

Estragon

If you have panels in series, voltage combines while current in amps stays the same, so 6a would be doing pretty well. In parallel current combines. Put 4 in parallel you should get around 20a to heat the 48v element.

jimmyaz

Estragon said:

If you have panels in series, voltage combines while current in amps stays the same, so 6a would be doing pretty well. In parallel current combines. Put 4 in parallel you should get around 20a to heat the 48v element.

The issue is.... 180VOC or 165VMMP at 6amps going through that 48v1000watts element... I am only getting 90watts because it somehow dragged the voltages down to 15volts... Haha, I didn't even think that was possible.

I guess after all, proper voltage rating on ELEMENT is a must.... Feeding higher voltage to a lower voltage rated element DIRECTLY does not work, it dragged the voltage down too much...