Aux 2 Settings

[Mike W]

Hi All,
I'm using Aux 2 in Waste Not Hi mode to divert excess solar power to heat hot water on our boat. I wanted to clarify my understanding of how the Aux 2 settings work.
We're running a 12V system and I have the Aux 2 delta set for -.1 with a width of 1 volt (lowest setting?). My understanding is that the classic should start to PWM the Aux 2 output when the absorb and float voltages are .1V below the charge setpoints (14.3 and 13.3 respectively). If the width is 1V, am I correct in thinking the AUX 2 output will never fully turn on (PWM at 100% on) as the voltage will not rise to 15.2 and 14.2 under absorb and float?

Would I be better off using Aux 1?

Thanks for any advice.

Mike

[toothy]

Hello Mike

I can't tell you anything about how AUX 2 works. I can tell you I can't seem to get as much harvest using AUX 2 waste not hi as I do with a load turned on manually. If I turn on a load after it has been in absorb a while, that is small enough to maintain the voltage, my diverted totals are well above what the PWM diversion totals would have been. I have tried several loads and the AUX 2 never uses the full production capacity of the panels and turbine. I'd be happy with turning on the load with AUX 1 or 2 manually through my ssr if I wasn't afraid I'd forget to turn it off. I'd love to see every watt that wasn't needed for charging headed into the heat tanks automatically.

Wade

[boB]

For Aux 2, since it operates over a volt or so width, what I often
suggest is to set your battery Absorb and Float charge voltages slightly
higher than you normally would.  That is your failsafe.  If you have a 48 V
battery system, 1 volt is usually just fine.  then, set your offset to
closer to -1.0 volt and give that a try.
If for some reason your dump load goes offline, the absolute max voltage
won't be too high.  It's kind of a compromise.

Since you are watching things fairly closely right now, you will get a good
idea how it works and can tweak the adjustments.

That's how I do it anyway.
Chris Olson might also have some good input on how he adjusts his system.
He started out at 24 Volts battery and has since moved up to 48V.

It is all a compromise though, so some tweaking will be necessary for your
system and SSR Relay and dump load you are using.

boB

[dgd]

This pwm width has always been something that has both intrigued and confused me. The common sense seems to be that if it activates, say, half volt below absorb with width 1 volt how can you ever get 100% active when you can never reach the next half volt above absorb set point?  Or does the waste not logic 'know' when the voltage reaches a set point and make a half turned on pwm go full on?  So then the real use of the pwm is how far below the set point you start introducing the load
So balancing the load size to probable excess power availability becomes the real task as you don't want a pwm'd load to stop the battery getting to absorb.
I played with increasing the absorb set point by .7v and starting pwm .3 below proper set point but found that when the water was hot I was over charging batteries.
I have spent many hours playing with this and it was the lack of knowing where the pwm was at, in terms of % active, that makes the settings business somewhat hit and miss.
I often thought that such a % reading for aux2 on the MNGP and local app would have been very useful.

In the end I put it all in the weird basket and had to find some other way to control water heating. Made worse in that I lost the aux2 as my clipper needed it.
Now an arduino UNO gets input when absorb reached and pwms DC ssr as it monitors the PV panel voltage rise over 5v from when it gets absorb started input. It will reach  hard on only if there is sufficient oomph coming from the 1kw, out of 3kw, PVs it diverts to the element. The UNO displays on 2by16 lcd where it's at on pwm linear curve.

Now I'm thinking even this is just too complex, just buy a few more cheap cheap PVs and keep them on the water heating element, a Kw PV here costs under us750, god bless the Chinese 

Dgd

[boB]

I often thought that such a % reading for aux2 on the MNGP and local app would have been very useful.

I like that idea.    One thing I keep forgetting to mention is to set the LED MODE to Mode 1.
Mode 1 makes the  right LED inside the Classic, behind the vent, turn on bright
when Aux 2 is on harder and dimmer since it follows the PWM output.

Aux 1 is the left LED (red ?) and Aux 2 is the right side LED (yellow ?)

Also, the width does not know what the actual setpoint is so it doesn't do anything.
Yes, it could probably work better and it may some day.  Keep these ideas coming.

boB

[dgd]

I really like the concept of 'Waste Not' and choosing this name for an AUX mode was inspired. It added greatly to my feelings of self satisfied smugness knowing that nothing was being wasted and batteries were full and water gets hot.
But the more I think about how AUX2 implemented this using pwm to slide from off to on state the more I wanted something better and something I could control and see what was happening.
So what would be the ideal external load (eg water heating element) control method?
Ideally this should not affect the battery charging, so fenegaling about with upwards adjusting set points to make width for a pwm process to achieve full on state was not exactly ideal. And I like set points because they are just that and probably shouldn't be messed about with too much. I trust the battery manufacturers know what they are talking about when they specify these.
So I would like to have waste-not active  AFTER the set point is reached.
Using AUX1 simply switch in the load may be ok if the load is small but the problem here is it may case the controller to drop out of absorb (or float)
So PWM to gradually slide in the load would be good but since the setpoint is holding the output voltage at the set point then there appears no way of using output voltage to 'range' the pwm width.
The only aternative  I can see is the input voltage. The nature of PV input is that when in Absorb some input power may not be being used so the input voltage will start to rise. Seeing this happen means excess PV input not being used. Ncw Waste-Not can be activated and the PWMing width done over the rising PV voltage.
This should not be too difficult to tune so as to keep Absorb and send excess PV power to the load. Being able to specify the pwm start point at a voltage offset above the input voltage when absorb started and the width voltage range above that along with an input voltage and current display  would be an excellent monitoring tool and probably made tuning waste not a cake walk.

Just an idea, I'm not sure if  wind input could use this concept.

dgd

[ChrisOlson]

This pwm width has always been something that has both intrigued and confused me. The common sense seems to be that if it activates, say, half volt below absorb with width 1 volt how can you ever get 100% active when you can never reach the next half volt above absorb set point?

Pulse Width Modulation has nothing to do with the voltage set points or offset.  The wiki has an excellent tutorial on PWM that may help you understand it better:
http://en.wikipedia.org/wiki/Pulse-width_modulation

If you are not getting the full power to your load then you do not have the offset low enough.  I am using -3.0 offset on our 48V system, and it has no problem ramping the load up to full rated power.  But the caveat is that you need the RE generating capacity to both power the load and meet the battery's charging requirement to get the battery to the correct absorb voltage.

For us, the water heating is priority over battery charging.  If we have enough incoming RE to get the bank voltage up to where it fires the aux load up, that's "close enough" even if the batteries don't get fully absorbed that day.  About one day out of every week, invariably there's an excellent power day where everything gets caught up, and our batteries are fine with that.

Batteries don't live their lives by when the sun comes up and sets, like humans do.  I deliberately cycle our bank from 50-80% SOC for days at a time - and have gone to great extremes to invent and tune aux loading systems that will do that.  Then every 7-10 days we get a "good" day when the bank goes thru the full regime.  The Waste Not operates the load at full power until the stats kick the water heaters out, then the bank snaps up to absorb voltage and the job gets done.

I've been playing with some "boost charging" methods recently to try to shorten that absorb time by using higher voltages.  But that's a different story.

Why do I do this?  Because batteries are more efficient when you cycle them below 80%.  You get more of the kWh that you stick in them back out.  Getting batteries from 80% up to 100% SOC is pretty low efficiency - probably no better than 65% efficiency for that last 20% SOC.  And a full charge from 50% SOC to 100% SOC is probably no better than about 85% efficient.

Most folks think you have to absorb batteries every day or it's not a "good day" or the batteries are going to sulfate and die.  You don't, and the batteries won't as long as you can fully charge them about once a week.  So the Waste Not thing is the best thing since sliced bread as far as I'm concerned because it does exactly what I have tried to do with various other methods for several years, and failed to achieve what it does.

So basically, if you set your offset to whatever - and can't seem to get the bank to full charge stage voltage on the "good" day, then your load is too big or you don't have enough generating capacity.  It's that simple.  So use a smaller load that more closely matches your RE capacity and battery requirements - or strap some more solar panels to the roof if you want to run that big of a load.
--
Chris

[dgd]

]
If you are not getting the full power to your load then you do not have the offset low enough.  I am using -3.0 offset on our 48V system, and it has no problem ramping the load up to full rated power.  But the caveat is that you need the RE generating capacity to both power the load and meet the battery's charging requirement to get the battery to the correct absorb voltage.

For us, the water heating is priority over battery charging.

Your priorities are different from mine. I see hot water heating as secondary to getting the batteries charged. Lowering the pwm offset sure will get the heating element going sooner but, as you say, the caveat is the generation capacity to heat water and get to absorb.
I am also not using AC for heating but simply divert some DC from a panel string to a DC element.
This works good and does not involve pushing the Classics to high output current or loading up an inverter.
This is why I think this pwm controlling of a diversion relay to power an element is interesting but will  be become pointless as PV pricing drops lower. There may even be a time when the cost of adding more panels takes away any advantage in having an mppt controller, much like PV trackers are now of minimal advantage.

If you are into the AC water heating then more PV capacity means more CCs and more inverters, for me it means, eventually, just hooking up a couple of Kw PVs to water heating elements and let a thermostat switched relay disconnect them as needed.

Dgd

[ChrisOlson]

This shows what the offset adjustment does, and what it takes to get full power to your load

http://youtu.be/8JWvDNWGLRY

Water heating is priority for us because we heat all our water with off-grid electricity.  We have two 55 gallon heaters and I have 2000 watt elements in the primary.  So to ensure we always have 55 gallons of hot water stored up, that heater comes on just like any normal water heater does when we draw hot water out of it.

I have continued to tweak our setup over time and the way I have it set up at present is that the top element in the preheater is on all the time, controlled by its stat.  I have the Classic turn on the bottom element at dawn and turn it off at dusk so that element doesn't fire up at night and put a big load on the system when we might not have the available power to run it.  But that bottom element comes on right away in the morning and it usually takes everything the system can put out on a normal day in the morning for 1-2 hours.

In the preheater I've got 4,500 watt 240 elements on 120V power so they draw about 1,080 watts.  Those are the "opportunity" loads on Waste Not, and even with only 1,080 watts going to that preheater we can heat it up to 140 degrees in a day with no problem at all.  If we get that preheater hot, the bottom element in the primary doesn't even kick on in the morning, and that's good.

The top element in the primary is emergency only in the event we run out of hot water.  It kicks on and heats the top ~25 gallons to 140.  And usually, when that element has to be used, it's because of a long stretch of bad conditions with no sun or wind.  So the bank is low anyway, and invariably the generator will start and provide the power for that element.

I used to run 2,000 watt elements in the preheater too.  But we installed a new Trane XR13 2-ton Central AC unit for this summer, so I reduced the aux load by swapping the elements out so there's "extra" to run our new AC unit.

So this system has worked out very well for us using Waste Not.  It's easily "tuned" by simply swapping out heating elements - and I wanted to show that you CAN get full power to your load if you're not afraid to set that offset voltage low enough.
--
Chris

[toothy]

Thanks Again Chris

I very much enjoyed the video, and yes I'm afraid of the -2.7v, -1.2v is as far as I have been yet. For me battery charging is the first priority and water heating is #2. If I go to -2.7, yes the water gets hot but priority 1 goes out the window?

On your system the way you have it set up now, leaving the wind power out of the equation and ignoring inefficiencies, you have a 1.1Kw diversion load and 6kw solar. If your floating with 1.1kw diversion and 1.1kw +/- to hold float and power house stuff, that leaves 3.8kw wasting?

My third Classic arrived but without the required MNPV6-250 combiner box which is being drop shipped by our host, so I'm stuck building timber frame stairs, doing tile and putting up siding, instead of the fun stuff. 

Wade

[ChrisOlson]

On your system the way you have it set up now, leaving the wind power out of the equation and ignoring inefficiencies, you have a 1.1Kw diversion load and 6kw solar. If your floating with 1.1kw diversion and 1.1kw +/- to hold float and power house stuff, that leaves 3.8kw wasting?

No, 6 kW of installed solar capacity will put out 6 kW once in a blue moon when all the planets align with the Galactic Equator and the temperature drops below freezing in hell.  6 kW of installed solar capacity is really about 5 kW installed capacity on the normal "good" day.

Our house loads run anywhere from parasitic draw to 9+ kW if my wife fires up her appliance Load Banks in the house.  I reduced the load on the pre-heater specifically to have more power available for our AC unit.  In that video we weren't running anywhere close to full capacity.  Just showing what it takes to keep your aux load at full capacity because somebody said they can't get full power to the element using Waste Not.

I used to consider battery charging first priority too - until I talked to a guy at Surrette that has been working with traction and deep cycle batteries for 25 years that clued me in on how to properly cycle a deep cycle storage battery without shedding lead off the plates prematurely.
--
Chris