The blackbox project

[Halfcrazy]

Guys there is a possible firmware issue that we may have fixed. We should have new code today or tomorrow with that fix and the current sense module code as well as Days between Bulk charges. Stay tuned.

Ryan

[Vic]

This sounds great Ryan.  Thanks,  Vic

[zoneblue]

I have added support for multiple views, and an example here:
http://www.zoneblue.org/blackbox/index.php?id_view=2

My artistic ability is not that flash but anyone who can make an html page can make anything they want.  The basic concept is that a view template has a number of 'Panes' and you can put any datapoint in any pane.

Im still waiting for the modbus-fix firmware, once thats in, ill make a zip. Meantime if you use SVN you can check out the code at http://code.google.com/p/theblackboxproject/ It runs stably enough if you limit the sample time to 2 minutes.

The next tasks are:
-  to make it more flash friendly
- add better support for multiple devices of the same type.
- along with that will no doubt come hybrid datapoints that combine data from multiple devices, eg total energy today from all devices.

If anyone needs some help trying it out, im happy to help as it will provide insight into where the  documentation (minimal so far) is weak. Also ideas for future development.

And oh if i could get one of the whiz bang things....

[SolarVet]

So is there a Software for Win 7/// I love this Interface.

[zoneblue]

With a fair bit of messing about it probably could be made to run on windows, but that isnt the point of it. Its designed to run on a network 24/7 so that it can collect complete data, and serve that information to computers and hand helds. If the computer isnt always on, there will be holes in the data.

The actual cost in running blackbox on its own computer is, what, about US$60 for the board, and a couple of hours installing the OS and other software. The latter is probably the dealbreaker for most windows folk, but we will hopefully be able to do some prebuilt micro sd card images and such later on. All you need to do is write the image to the card, stick it in and itll go. Well thats the plan.

[SolarVet]

I have a spare motherboard, I could run win Xp or even win 98 If I had to..

[RossW]

I have a spare motherboard, I could run win Xp or even win 98 If I had to..

None of the participants of the "black box project" are currently inclined to write for windoze.
The whole premise of it is "lean and mean". Very low power draw, low hardware requirements. Not a power and resource hog like windows. Of course, once the "black box" is on the network and monitoring things, you would be able to view the details from your doze box (or indeed, anything with a standard browser).

[Gregy]

Have been reading and following with interest the opportunity and diversion debate in NWAS forum
(where I have also posted similar to below)
and some very creative ideas for how to better harness and manage opportunity.
After reading as many posts as I can find, I would summarise that much of the effort is focussed on using opportunity
for "variable load heating" of one form or another - with various implementations using PV, DC or AC inverter loads.
Which no doubt works great for those in cooler climates ...

However my scenario is somewhat the reverse, and with some other less common characteristics.
I have no reqt for "heating water or similar" and hence can't use this variable load method.
But I have numerous "opportunity" loads that can be used in a combination over time to to simulate "variable"

Anyway to my point .... I've been speculating about a method to Measure the solar potential energy "solar irradiation"
and then use it as an "input" to a decision making process for "opportunity" load control
(For loads that cannot use the PWM methodology .... Hence the need for more informed decision Making)

Which brings me to the "blackboxproject" (BBP) - of which Im very excited (given my passion for home automation)
Here we have a fantastic opportunity to bring various inputs into a BBP that will already have access to much of the
data needed, and use it to make (or assist) informed decisions that can be used for load control.
Note this is not in anyway a plan to modify, interfere with or substitute for the core CC functionality it would simply
"wrap around" the Existing CC/inverter and Other components... And leverage the work and inputs already underway in BBP.

This would enable
- graph/charting etc of additional variables
- help understand better the RE system status and capability
- fault finding/preventative Maintenance (with future expanded sensors)
- better decision making for load management
- better data for planning
-  some limited decision making and control capability

Required/suggested BBP sensor/inputs:

1) battery current/direction via current shunt
Q: can BBP get this data from CC via the modbus interface? (When CC fitted with the whizbang)
Hopefully yes soon (if not, need to add yet another voltage sensor across the shunt
To also measure battery current - direction AND value)

2) PV array
Voltage : BBP  "vpv" - yes; Current : BBP  "ipv" - yes

3) CC output
Voltage and current BBP - yes

4) solar panel temp
One or more temp sensors (1wire) mounted on rear? Surface of solar Panels 
- measure/prove the temp vs power/performance Relationship
- relate to measured array voltage and Vmp (temp compensated from panel Temp vs voltage curve)

5) solar radiation (aka potential PV RE)
method 1 : solar radiation sensor
What type? How representative of potential power from real world panels? How to adjust for MPPT?
How does this allow for panel tilt ? And sun orientation? (Perhaps a shield to simulate panel orientation?)
It strikes me that with a database (collected over time.under actual operating conditions) of actual power
produced by CC when under "full load" ... And comparing this to the radiation sensor, it would be possible
to determine some algorithm that can use radiation sensor to estimate instantaneous production potential.

method 2a : use an actual solar panel (separate panel - not part of production array) with a permanent
dump load and measure instantaneous power produced "solar potential"
- a small cheaper panel could be used
- no battery required
- to be representative it would need MPPT capability?
Even with minimal panel, Charge controller and load, After adding all costs ... Would this be worthwhile... Hmmm

OR

method 2b :  Lets find a very cheap way to do it - accuracy may suffer (could Be negated by the same method above to calibrate)
- Use a very small panel (eg 6V 2W) as "measurement array", permanently Fully load it (resistor load),
use a method For keeping it in a crude  MPPT like region of operation ..(Research on this underway) and Measure actual power produced
Q: How representative could this be?

Summary : using all available data (much already avail from CC via BBP) and augmented with additional sensors,
Use real world measurements (charted over time) of "production array" vs "measurement array" to calibrate the relationship
and then use it to predict/ measure in real time, available energy - to intelligently switch  Opportunistic (fixed or variable) loads,
And of course have a much greater insight into our solar PV system operation

Apologies that my first post is a big one ... but have been thinking this thru and researching how to get RE system data into
my existing HA system .... and finally was ready to put my (lengthy) thoughts into print.

To achieve my suggestion - requires a number of additional sensor inputs into BBP - but this is easily achievable using
arduino or RPi shields and readily available low cost sensors
(I personally run HA application on Win7 environment - but note and respect the plan for BPPto be Linux based on a Rpi or similar)

Thoughts from the BBP team....

[RossW]

Gregy, I have similar needs to yours. "oportunity heating" isn't of any interest.

Amongst my other meterological monitoring equipment however, I DID install an Apogee pyranometer. These are available as bare units, or amplified with a 0-5V or 0-10V output.

They are cosine corrected and are designed to be mounted horizontally, however for the purposes you and I want, mounting it in the same plane as your array is absolutely perfect. Its output will tell you exactly how much solar radiation is available to you. (My circumstance is complicated by having 4 tracking and 2 fixed arrays).

I use a CAI WebControl board to monitor ambient temperature, the pyranometer and various other inputs. It would be trivial to add a temp sensor (it uses the DS18B20 sensors) to monitor the temperature of one array for direct calculation of available power based on panel temp + solar radiation.

I won't post more here incase it's not of interest, but if it is... just ask specific questions for more info.

[Gregy]

Hi Ross
If I recall from reading various posts - you may be a fellow countryman - albeit I don't live down under at present - indeed where I currently live, we use water from storage tank which is just fine for showering... No heating read

I don't want to hijack this forum unecesarily .... Maybe we should direct communicate or start another thread?

I use Homeseer (HS)  HA pkg and have various sensor input capability,
I've seen the CAI device - however it's analog input (3 only) is way too limited for what I have in mind.
I've been anguishing over how
To get PV system data into HS And after much research found this BBP forum.
(Would be happy to separately share my research to date)
BBP will already have a lot of heavy lifting "done" eg getting access to modbus data and hence it's
One less thing to do, to achieve my desired outcome. Hence my suggestions above to enhance BBP

Edit - have sent you PM

[RossW]

Happy to start another thread, but also happy enough I think, for it in here as it is relevant.
The old version of the CAI device only has 3 0-10V inputs, plus another 0-5V used by a humidity sensor.
The new version (I have a beta one here) increases that to 8 analog, plus 16 digital in and 16 digital out, plus 16 temperature, plus a few other bits. Its the same formfactor, similar power consumption, same interface.
Additionally, you can read values from it by http, or it can push values to something else via http.
It is straightforward for example (as I have), to have it running code in the background to constantly monitor the maximum solar input, and the cumulative watt-seconds over a reporting interval (I use 5 minutes) and either push that to another device, or have something poll for it as required.

The aussie ripoff merchants wanted almost a thousand bucks for the pyro, I bought mine directly from the manufacturer for around $300, so it's not "trivial", but it is a proper scientific, calibrated, instrument.

If you didn't want something so "pro", it would be very straightforward to use a small cell (perhaps from a scrapped garden light) and an opamp. Mixing it with something like http://webcontrol.rossw.net/ makes a fairly neat solution.

[Gregy]

Newer more capable version of CAI - very interesting!
Current CAI already has support for xaP, as does HS - hence I can easily communicate between HS and CAI  (Aka heavy lifting done) and thus read it's analog and digital variables into HS And also send commands to control CAI digital IO
My planned alternate was to use Arduino (for which an Interface capability with HS exists) and
Leverage it's low cost, extensive Io capability (also easy to add low cost Wireless ) and range of low
Cost sensors.

Ouch that pyranometer is not cheap - get what you pay for I guess - as it's been prof calibrated.
I can see your dilemma with multiple fixed and tracked panels ..... You don't want to be buying more than one pyranometer !
I'm searching US sites on line ... But looks like a $300+ device ...

Your suggestion of using a cheap panel and opamp is along the lines of my above suggestion
However my proposed method has the advantage that if we can measure production array output (under full load) AND "test array" (aka low cost cheap solar panel) and chart them over a period
Of time and environmental conditions - we should be able to develop a reliable algorithm to
Use the "test array" to predict solar energy
I guess this is just a different way around using a $$$$ pyranometer - however my proposed method
Holds true either way - run a calibration period using data from production array to "calibrate"
The pyranometer or cheap test array
Either way we need access to all the data from CC ... Hence what led me to BPP - as it will access
CC via modbus and have access to all? The CC data
So using CAI or any other method to read sensors is fine, but we end up with data in multiple devices/applications
... So how to bring it all together is the BIG question

I've looked at using various modbus applications avail for win OS (my. HA environment) .. Which is possible, but there
Is a lot of work to interpret and translate the various modbus registers - as evidenced by the
Great efforts in this forum ... Basically I don't want to re-invent where it's possible to leverage
Existing capability

My alternate - is to work out how to get the data out of BBP (different Device and OS - I have no Linux experience) across a TCP/IP interface .... But this seems like reinventing also !

Would like to share some other thoughts on PV system automation and sensors with you - but don't
Want to further hijack this thread .. Untill we determine desire/capacity for BBP to be expanded to inc these
Additional inputs/sensors

[Westbranch]

RossW, Gregy, I have a question about the need to be measuring solar insolation:   

Doesn't the CC do a similar function by 'determining ' the sweet spot for maximizing the array output?
I can see this if the array was tracking, but is there much benefit if it's fixed?


Tom I just noticed that the BBP is not a 'child' of the Open Source Software Section.

Can it be made one? and if there is a lot of feedback on this 'monitoring insulation' then make this a child of the BBP?

[TomW]

Tom I just noticed that the BBP is not a 'child' of the Open Source Software Section.

Can it be made one? and if there is a lot of feedback on this 'monitoring insulation' then make this a child of the BBP?

Westbranch;

Not sure what you mean?

it is already beneath The Open Source software/hardware corner:
General info>> The blackbox project

And, not to nitpick but it is solar "insolation"

Tom

[Westbranch]

Hi Tom,  duh! ya I know about solar insOlation, had to study that plenty.
Just didn't re-read before hitting SEND... will fix it.

I'll try to explain my thinking:
The Open Source Section has Beagle Board, Arduino and rPi sections and I see the BBP as another Software/Hardware type as it looks like it is/has become a separate entity by evolution.

Ciao