The blackbox project

[Gregy]

My understanding (and I admit im far from knowledgable on this)
Is that the MPPT in the CC will always maximize against the available power given the current
Solar radiation - but only when it "needs" all the avail power for the given charge state
Eg when in bulk it's trying to push max power into battery, but when it goes into absorb and
Particularly float ... It's not in MPPT mode - because it simply doesn't need all the power.
(This is a very simplistic of a much better explanation that I've read from bob)

So unless a load is applied, it's difficult (impossible?) for CC to determine how much available energy
Could be avail  - because float (for example) uses only a very small ammount
Of potential solar capacity
Hence where the PWM opportunity load mode of operation becomes useful ... To slowly ramp up (or down) the load to find the sweet spot  .... But it needs a load applied to do this, and by using a
Variable,load (controlled by PWM) it can (using feedback mechanism) vary to optimise for avail energy

So "our" (Ross and Greg .... No doubt others) dilemma is that we have "no use" for variable load
Heating ... And hence we dont have a  (useful) variable load we can apply via PWM
And we are looking for a way to "forecast" potential nett excess energy (that's not being used for charging) and then switch on an appropriate combo of known, fixed value Discretionary loads
- without impacting battery cycle/charge state

Hopefully my poor explanation may make sense

[TomW]

Westbranch;

I don't seem to have the power to add a section? I will tag Ryan on this see if he can do it.

And, I know you knew what insolation is but insulation crept in during the discussion.

Future searches will work best with proper terms.

No offense intended.

Tom

[Gregy]

Hi tom
Also pls guide as to whether my posts are appropriate for this thread .. Or better elsewhere
BR
G

[TomW]

Hi tom
Also pls guide as to whether my posts are appropriate for this thread .. Or better elsewhere
BR
G

Gregy;

Your previous post would probably have been better over in the Classic charge controller board:

http://midniteforum.com/index.php?board=2.0

It may be a bit off the topic in this thread. But we don't get too excited about that kind of thing.

Tom

[dgd]

So unless a load is applied, it's difficult (impossible?) for CC to determine how much available energy
...

So "our" (Ross and Greg .... No doubt others) dilemma is that we have "no use" for variable load
Heating ... And hence we dont have a  (useful) variable load we can apply via PWM
And we are looking for a way to "forecast" potential nett excess energy (that's not being used for charging) and then switch on an appropriate combo of known, fixed value Discretionary loads
- without impacting battery cycle/charge state

So how do you envisage this 'forecast'  process/calculation gets data to provide meaningful results?
The potential maximum available power should be relatively easily calculated and adjusted using actual historical data. However, the biggie issue here is the highly variable solar insolation and how to get real time data on this - obviously without loading the array and causing that impact on battery charge cycle.
Perhaps a separate tiny pv to use as a loadable indication of available power in main array?
Dgd

[Westbranch]

Greg, that is a good explanation, at least for me..  What non PWM options are you looking at?

Tom, none taken,  my typing skills need a lot of improvement, as well as proofreading..  tks. Have to watch that spell check too

[Gregy]

DGD
Yes - that's one method
Pls see my post #37 above - item 5) explains my proposed options (and yes one option is the "tiny PV array) - for which I have a proposal (Ross and I have separately been exchanging ideas)
Happy to post here also if there is interest

Westbranch
.. I'm glad you asked

Gregy's Diversion opportunity definition/categories

Fixed (known) loads for variable periods
DC load (at battery voltage) or AC load using inverter
- Cat 1 Opportunity : time shifted fixed loads
Can Simulate a variable load (but over a period of time)
Eg water storeage pump, desalination system, Electric vehicle battery charge, ironing
- Cat 2 opportunity : discretionary fixed loads
High pressure washer, boat wash, irrigation, A/C (wife may disagree)

2) variable power opportunity loads
Eg hot water storage, heating, fans
Can be controlled using PWM output with SSR
Works well for loads that can work with "variable power input"

I guess the above starts to give away my intended scenario
Think ... Surrounded by water, with a lot of sun shining ... definitely "way off grid" !

[Gregy]

DGD

"Low cost Solar test array for measuring PV potential"


http://www.adafruit.com/products/390

Above is the link for circuit for connecting to a cheap 6v 2w solar panel
It's got a lot of stuff not needed for a "test array" (but it's all built to go at $18)

If you read the accompanying tutorial and in particular the "design items"
it seems they have done an admirable job to
Emulate a max power type of operation - so it's "MPPT like" for loading the solar panel

I was proposing to simply put a load resistor (in place of a LiPo ) and up the load current
Resistor control - So panel was always max loaded
And then simply measure voltage To calculate power
(May need a resistor voltage divider to keep ADC within 5V or 3V range )

So for around $18 (Adafruit board) + $12 6V 2W panel (eBay)  = $40
Could deliver a good "test array" that operates in a max power style operation
And hence be a low cost predictor of available power once calibrated against the
Production array power measurements
..... $40 beats a pyranometer at $300

Further food for thought (yeah I know I have way too much thinking time)
If above works - I was considering using a second "test array" at different tilt (optimum
Seasonal tilt)
To determine opportunity benefit/loss of fixed tilt ..
. But this could also be Extended to tracking vs non tracking !!
(Projects for when we have idle time! Hmmm )

[dgd]

Gregy,

Thanks for link, excellent info there, looks like I will be busy constructing this sensor

Although my main diversion interest is water heating I have used a simple non pwm method using an external processor to simply divert output from a section of my solar array. Nothing too intelligent about the processing, the 'optimal' settings to ensure non interference with the battery charging process were arrived at by a trial and error process.

However, on good days a lot of my pv power is still wasted.

Dgd

[Vic]

Others have pointed out,  elsewhere,  that a single Photodiode with a heavy load,  measuring its Isc,  will be a good indicator of Insolation.   Ideally,  it (or any other such home-brewed detector) should have a diffuser.   This is probably the tricky part.  FWIW.   Vic

[Gregy]

It's getting late for me for sure ..
18+12=30 (not 40) ... Before anyone points it out!

DGD
At present I'm "workshop challenged" (long story ... Basically living in a hotel, no soldering iron ...
Getting withdrawl symptoms)

So I can't build this to prove it works .... Pls,let us know how you go.

Note however ... The key to making this work is the capability to chart "power" measurements
From test array vs production array (when both fully loaded) .. To calibrate and develop algorithm (as I'm guessing it
Will not be linear relationship ... That would be too easy)

Hence the interest in BPP ... Because that would make it's oh so much easier to capture and calibrate...
And then forecast and control
(Bringing us back to the thread topic !)

[Gregy]

Vic

Yes your correct - diffuser required to address below "cosine" matter
I've seen various diffuser solutions
Ping pong ball being one of the more popular! .. I guess you just replace them as they
Deteriorate from UV, bird attack et al
And mount the sensor collocated at and orientated per the array (tilt and direction)

Quote clipped from
http://www.instesre.org/construction/pyranometer/pyranometer.htm

"important design consideration for even the simplest pyranometer. Ideally, pyranometer detectors should respond to direct sunlight in proportion to the cosine of the zenith angle of the sun. When the sun is directly overhead, define the "normalized" response as 1. As the zenith angle increases, the normalized detector response to direct sunlight should decrease as the cosine of the zenith angle. Real detectors do not have perfect cosine response. To improve their response, some kind of sunlight diffuser can be placed over the detector. Teflon® is often used because it has good spectral transmission properties, is very stable, and is relatively unaffected by long-term exposure to sunlight"

Given that calibrated high quality pyranometer can be $250+
I was searching for a lower cost method that achieved the stated intent ... Which will
Not give W/m2 absolute measure of insolation (pyranometer )
But will enable a forecast that can be confidently related to each systems individual production array

Similarly the photodiode light sensor options could also be measured and calibrated against production
Array ... We are not looking for absolute insolation measure, but rather relative measure of potential
We should try it also! If it works reliably with desired accuracy and it's cheaper solution - I'm all
For it

[Vic]

Hi Gregy,

Thanks for the additional info.  Some day I will try to progress with instrumentation here,  will SIT and wait for all of you folks to do the heavy lifting.

Thanks,  Have Fun,    Vic

[RossW]

So how do you envisage this 'forecast'  process/calculation gets data to provide meaningful results?
The potential maximum available power should be relatively easily calculated and adjusted using actual historical data. However, the biggie issue here is the highly variable solar insolation and how to get real time data on this - obviously without loading the array and causing that impact on battery charge cycle.
Perhaps a separate tiny pv to use as a loadable indication of available power in main array?

OK. A few years ago, I installed the pyranometer at my own site so I could measure the actual power available at any given time. ONE of my purposes for this was to calculate the actual efficiency of my entire system. By measuring the power from the arrays vs what was actually available tells me how well I'm collecting power. By measuring how much my charge controllers delivered into my batteries tells me their efficiency, etc.

HISTORICAL INFORMATION IS USELESS here because it can't predict what atmospheric conditions we have (cloud, smoke, dust etc).

I also calculate every minute, where the sun should be and from that what the radiation should be. By comparing that to the pyro output, I am detecting (quite accurately) small amounts of "lost radiation" due to high cloud, haze, smoke etc.

Additionally, by knowing the radiation present, I am able to calculate a "predicted power" from my arrays, knowing where the sun is (by calculation) and how much sun there really is (from the pyro) and calculating the error angle (absolute off-axis vector elevation and azimuth) from each array to the actual sun position.

Between all these, I've a pretty good idea of just how much power I *COULD* be getting from each array (and in total).

Now, here's the kicker (the bit I haven't done yet).

As explained elsewhere by others, the classic is only drawing as much power from the arrays as it needs to supply the batteries and instantaneous load. If I "could" be making 3KW but the batteries only need 400W (because they're in float) and the load is only taking 800W (because that's all that's running now), I have a "potential" of (3000 - 800 - 400) = 1800 watts of "potential power" that I could be using for something else.

Ideally, I should "switch on" about 1800 watts worth of opportunity loads.

It also needs to be "smart enough" to drop opportunity loads when we don't have enough headroom to maintain them all AND keep the battery and load supplied.

[zoneblue]

Such a rash of interest in the project. Cool.

Work's a bit crazy at present but ill try to catch up reading  in the weekend.

Briefly. Insolation: how much power arent i using, but could be?

The latest trunk code has a solar insolation module built it. It calculates insolation from first principles, using a variety of algorithyms, the most promising of which are from pv-lib here:
http://pvpmc.org .Those are "clear sky" models.  Ross's insolation measuring approach is better.

Secondly, i agree that the CAI is more or less perfect companion for the black box, such that ill be mounting at least one inside the 'box' here.  My interest in co-collecting weather data is shared with you guys it seems. We may as well have one device powered up and storing everything.

I also have been reading about reference panels on the same plane as the array, and ordered a small glass covered mC panel from dx.com last week to test. From what i read about reference panels its the Isc that is of interest, so shunt, feeds adc is probably all you need.

More about pv modelling and monitoring here:

http%3A%2F%2Fenergy.sandia.gov%2Fwp%2Fwp-content%2Fgallery%2Fuploads%2FSAND-2004_PV-Performance-Array-Model.pdf
http%3A%2F%2Fprod.sandia.gov%2Ftechlib%2Faccess-control.cgi%2F2009%2F098258.pdf

the latter of which covers reference panels.

Lastly, what if we had 12+ adc channels. Then we could monitor each battery cell's voltage. Now THAT would be cool. Pick up imbalances before they get out of hand.

Ross. I havent hooked up the webcontrol yet, but i noticed in the manual, it says you can also add additional adcs using 1 wire, of which there are several ports on the board. Will that work? I also note that there are now arduinos with 12 bit adcs.

WB jr. You can bet BB will have support for this soon. I hope Bob will be willing to swap ideas on SOC modelling, and im particularly interested in also modelling the decay of battery capacity over time, similar to how linux laptop battery monitors work. With the extra cpu BB is ideally placed to do these sorts of computations.

Re using BB to control/manage. My focus for now is just monitoring, but once thats mature,  i have little doubt that things like weather forecast lookup, CC and load managment will be candiates for future devleopment.

Now that we have our own sub forum (thanks!), i guess we can break this conversation up into threads. Phew.