I have been using AUX 2 to drive a AC SSR with Waste Not for water heating, with the power coming from our inverter. I wanted to experiment with using AUX 2/Waste Not with a DC SSR diverting high voltage DC direct from the panels to the water heating load. Someone told me this could potentially be "bad" for the Classic, or damage it by attempting to pull battery power backwards thru it to the load.
So when unsure, it's best to ask the experts that actually know if this can be done :D
--
Chris
Hi Chris,
This has come up here as a topic, and believe that Ryan counciled to isolate the PV into the CC with a diode. This isolates the CC from the low Z of the heating element. Shorting the input of most MPPT CCs can fry their guts. The very low resistance of the heating element COULD be enough for this frying.
Believe that this fact is what gave rise to Ryan's signature line regarding smoke emission. The above is, if memory serves. Happy TG, Vic
Quote from: ChrisOlson on November 21, 2012, 12:12:42 PM
I have been using AUX 2 to drive a AC SSR with Waste Not for water heating, with the power coming from our inverter. I wanted to experiment with using AUX 2/Waste Not with a DC SSR diverting high voltage DC direct from the panels to the water heating load. Someone told me this could potentially be "bad" for the Classic, or damage it by attempting to pull battery power backwards thru it to the load.
So when unsure, it's best to ask the experts that actually know if this can be done :D
--
Chris
Sorry no expert but...
Works good, at least with a fairly low wattage dc heating element. 2.4kw array,
Crydom d2d40 Using aux2 waste not hi. And a 48v 600 watt heating element, panel voltage at mppt is about 68v. With panels now so cheap the next idea is to just connect 500watts worth direct to heating element and up load on inverter with more beer cooling.
Also I can't see the pulling battery power thru 150 to element, surely this is no different to unlit solar panels?
Dgd
Quote from: Vic on November 21, 2012, 12:53:58 PM
Shorting the input of most MPPT CCs can fry their guts.
Somehow I doubt this would be the case, you could be correct but to me it seems unlikely.
At the very least I would expect the controller would have minimum protection for reverse polarity input but shorted input is a no input situation and its again unlikely that current can flow out of the input terminals in reverse hence zapping the 'guts' :)
A tree once went airborne and flattened my solar array junction box shorting the inputs. The mppt controller did not die
Dgd
From an earlier Thread:
>> " Re: Classic with Separate Diversion Controller
« Reply #22 on: January 21, 2011, 06:11:50 PM »QuoteQuote from: cdog on January 13, 2011, 10:52:45 AM
Would it be better to divert after the rectifiers or get a 3 phase str and divert the ac?
Would this change what the classic ''sees'' to do, the load being a resistor of some sort, instead of the battery bank?
It is best if you don't short out the input of the Classic or almost any other MPPT Controller.
If you are going to short things, even momentarily, you will want to do that with the AC side.
Loading the DC side just to keep the voltage from exceeding a pre-set voltage (a "clipper") is one thing.... Directly shorting the DC side can break things.
Also, the shorting mechanism (triacs, relays, FETs, etc.) can break because you would be shorting out fully charged up high voltage capacitors which store a lot of energy. That is usually reason enough not to short the DC side.
boB
« Last Edit: January 21, 2011, 06:14:48 PM by boB » << "
Have seen this comment also directed specifically toward a WH heating element connected to the PV input terminals of a MN Classic.
FWIW, many of these things are relative. But even just during experimentation, an accident could happen without some isolation. I have not yet skimmed excess PV input power. Will do so in a year of so. The thought of skimming PV is one of the reasons that one system here runs string Vmp of 106 V -- easier to make more power from available WH elements. YMMV, etc, Vic
I think boB's comments there are probably a good reason not to try it without at least putting a diode inline to insure no backfeed could ever happen.
The only reason I wanted to try it is because with a AC SSR there are some oddities in how zero cross SSR's turn and on off with PWM drive. The PWM freq does not match AC current freq so the SSR turns on and off at half cycles and causes some "flickering" of sensitive loads on the inverter. My thoughts were that using high voltage DC off the array would be more infinitely "smooth" like PWM is capable of.
--
Chris
Quote from: ChrisOlson on November 21, 2012, 05:49:33 PM
The only reason I wanted to try it is because with a AC SSR there are some oddities in how zero cross SSR's turn and on off with PWM drive. The PWM freq does not match AC current freq so the SSR turns on and off at half cycles and causes some "flickering" of sensitive loads on the inverter. My thoughts were that using high voltage DC off the array would be more infinitely "smooth" like PWM is capable of.
--
Chris
I followed with interest the extended discussion on OP about your ac water heating, I also tried the light bulb idea and watch it dimm with a few bright flashes and wondered if this would disturb some of the more sensitive ac loads on the same inverter.
I decided not to go this ac way as my inverter capacity is 3.3kw but the idea I liked was using a cheap ac inverter just to power the heating element. I haven't wired it up yet, I have a spare DR2424 and a Crydom ssr seems happy enough to switch its few-steps squarish wave ac output. I did notice the relay buzzed a bit with this square wave compared to the sine from my 3024e.
I still think the long term solution is driving a heating element direct from solar panels :o
I found reading your experiences about water heating and using excess power very informative and it's given me some good pointers. Thanks for sharing this :D
Dgd
I have given this some more thought as well, and am determined to get to the bottom of using waste not to drive a DC water heating element. As noone seems to know how it works, i will borrow a scope when the classic arrives and try to figure it out.
We can get 0.75 ohm DC elements here for about $60 us.
I want to also say that the classic would benefit from a battery shunt input so that it can distinguish loads from charge, and also that would save us all buying a battery monitor, another box, another incompatible bit, a second whole data logging opreation. But i digress.
Heres what im thinking for my trial board:
-aux2 feeds an optoisolator
-this feeds two AND gates, that allows both a debounced HWC thermostat input and a user adjustable second PWM waveform to either modulate or stop the output to the next stages
-this feeds a high side mosfet driver
-this feeds a bank of n channel power MOSFETS mounted on a block of al, capable of pushing 80A cont.
-various protection-ery
-parts around $50
The thing is :
- elements are only available in certain sizes.
- in order to use every last electron the element must be /bigger/ than the max pv output
- in my case thats two 0.75 ohm elements, total draw around 2 point something kW.
- so it needs some way to stop the duty cycle fully loading the element
- when tuned right hopefully the design will neither over load or under load the diversion load.
- but this all hinges on how well aux2 reflects available power.
It will be christmas before ill get near this, but im happy to publish the design and results of my testing when i get to it. But also happy to collaborate with others on it.
The most $$$$ part will be the heavy cable between your battery and hot water heater if using low Ohm DC heating element..im thinking to use my 3kW inverter with a relay switching (via aux2 control) on the 2.4kW 230 AC hot water element ..2.5mm cable between AC board and hot water heater is dirt cheap too.
Which is cheaper a hundred bucks of cable or $2000 of inverter capacity?
I'm simplifying of course. With load diversity you dont necessarily need 2.8KW of inverter capacity. But once its used its used.
My view on this concept is that its about using surplus capacity, Charging an electric vehicle that needs to be charged everyday is not surplus. Its a base load. How else are you going to charge it? With hot water you make up the deficit with either a wet back, or propane by using the cyllinder as a preheat to an instant unit (some of which can do that).
What type of diode would be suitable to prevent the Classic 150 from backfeeding (if it can) the load when diverting high voltage DC from the panels?
--
Chris
Hi Chris,
If it were me, I'd try to find a Stud-Mount 150 Amp or so 400 - 600 V Silicon Diode. These ratings are overkill, but ... You would want a insulated mount kit, probably as well, and a small-medium heat-sink.
The MN tech/Hardware folks should have a meaningful recommendation, as they have been experimenting with this stuff for some years. Recall that you are a HB WindTurbine guy IIRC, so you may well know just where to find large stud-mount diodes.
I have had trouble finding this type item in main-stream distribution (like Mouser/Digikey) in recent years.
Quite a lot of this sutff is kicking around in the Surplus market. Will look for something.
The above are just opinions from one who has NOT done any of this WasteNot stuff, yet. YMMV,, Vic
Thanks Vic!
--
Chris
My water heating setup was working nice, the classic tickled the relay and pulsed the heating element to take some current away from the panels, a sort of diversion. I also had a second Classic with small turbine input, this never produced more than few amps. Upgraded to 2kw turbine and can see 50 amps into my 24v batteries.
problem is now Classic with turbine can get to absorb then float and this causes the solar Classic to pulse the heater relay. This is good when lots of panel current but not good when panels only making a few hundred watts. Instead of using turbine energy to heat water it gets wasted in the clipper dump resistors.
Seems the best way is to use AC water heater to use PV and turbine excess energy.
Dgd
Try using PV V Hi to drive the clipper and set the voltage up to a point where it's a safe rpm for the turbine running unloaded. Then set the charge stage set points .1 or .2 volts over the solar. That will cause your solar Classic to use Waste Not when excess turbine power comes in and send it to the water heater (assuming you're using AC power from the inverter), and won't unload the turbine unless the wind is really strong so the Waste Not Diversion can't use it all.
I use that setup here with three 2.5 kW turbines and it works great. When the wind is blowing strong, even at night, the solar Classic will have the water heater going full bore, the turbines carry the house loads, plus charge the bank.
This is one thing that has kept me from trying direct panel diversion (yet). I would have to hook up a separate diversion from the high voltage DC input from the turbines because at night there would no power available from the solar to divert. And the turbines, in the long run, heat WAAY more water than the solar does because they put out 5-6 kW in decent wind.
I only use the clippers for overspeed control in the event the controller decides to unload one because of too much incoming power. I smoked a clipper to a crisp here about two weeks ago and the input voltage to the Classic hit well over 200 volts before I got that turbine shut down >:(
--
Chris
Chris,
Yes this looks the way I should be going, it makes more sense to have the excess energy used usefully rather than dumped over air heating resistors.
What ssr are you using for switching the heating element? I have a Crydom d2450-10 , the docs say its a random on ac relay as opposed to a zero crossover type, I hope this will work ok withe the aux2 pwm tickling it :)
Dgd
I took a picture of it for you. Hopefully you can read what it says on it. It is on right now because the wind turbines are going quite strong and low loads on in the house so it is heating water. It is a zero cross type.
I bought it at Amazon.com
(https://lh3.googleusercontent.com/-p4YqyUilzeE/ULfk9QrGlLI/AAAAAAAAGl8/5HKqBznkocQ/s640/100_2275.JPG)
--
Chris