20140612. For some reason, most of the pictures are gone. I’m repairing the site as I can find the pics.

 

 

Currently, ( April 2013), I am working on a WIFI connected RS485 monitor for my 2 x 5KW Solar inverters.

 

The system has 40 panels on the house roof.

RoofView

In this picture, one panel is missing from the top right end. It was damaged in transit, but is now replaced.

The interface I am hoping to use is a TL-WR703n WiFi Router from ebay. They can be found for less than $30.

They are tiny! The firmware has been flashed with OpenWRT and the serial mod done (see below).

http://wiki.openwrt.org/toh/tp-link/tl-wr703n

rs485

The inverters communicate via a full duplex RS485 network. As the router does not have the serial port drivers implemented, I decided to add 2 x MAX3071  3.3V RS485 chips, gluing them to the processor, and connecting them using fine enameled wire.

TL-WR703n-RS485-02

The test setup is shown here. Powered via the micro USB power connector. An extra USB flash drive is added to increase the program storage.

 

Eventually, the power will come directly from the inverters as they supply 7.5V when running. A car 12V to USB power supply (from ebay of course) has been gutted to supply the regulator. These can be purchased for $1.00 so are not really worth trying to build.

 

An MC34063 switcher is on the bottom of the board.

 

The tinned copper link is to replace the connection normally supplied by the ground connection clip.

It looks like the parts will have to be laid down as there is not much room in the router case.

The code is still in development, thanks to my son, who is very good at putting “the magic” in 🙂

The same router will be able to be used, without mods, using a USB hub to connect the flash drive, and a USB to RS485 converter. The first try actually had a USB to TTL converter driving a couple of MAX485 chips mounted on some proto board , but I want to fit it all in the router if I can.

 

20130517….

 

Here is the “finished” article. Parts on the power supply were mounted on the track side and laid flat so the lid would go on.

An 8 way 254 series header is super glued to the USB connector. Connections are numbered the same as the cable from the inverters.

 

Part of the lid had to be cut to fit. A different type of connector may work better but I had these on hand.

 

Here it is sitting on the wall and running on the power from the inverters. Note the blue LED 🙂

Now we just have to get the code going.

My son is sure he can do away with the USB dongle after the development so a USB to RS485 cable can be used.

Then an external 5V reg would complete the setup so not requiring the internal mod.

 

 

 

It Is Alive 🙂
Well, sort off.
Today at last we got to talk to the inverters. It now uses ser2net running on the WR703N and linking via the wifi to another box (Asus EEEBox) on my network that is my VPN and telephone exchange.

One inverter comes and goes a bit for some unknown reason.

Power right now 2516W
Total energy today 15.07kWh
Total Energy 6553.5kwh

I based my setup on…

http://forums.openpilot.org/blog/52/entry-89-tp-link-wr703n-as-telemetry-link-and-camera-broadcasting-part2-software/
But note I have used the on board serial, ttyATH0 in place of the USB serial mentioned in the article.

A problem was I has swapped the TXdata wires over. This was found while running the TL-WR703N on the table and looking with a CRO for serial data out. There was serial data when running…  telnet box.local.IP.address 8081but looping the TXDs to RXDs produced garbage. Then I buzzed the connections and discovered the swapper wires. After that was fixed, what I typed was echoed back.
So, reinstalling it onto the inverter network, and refreshing the Eversolar Inverter Monitor Perl script on another box on my network produced the results!

The files on the WR703N that were modified to run ser2net are…

/etc/ser2net.conf …….
8081:raw:600:/dev/ttyATH0:9600 NONE 8DATABITS 1STOPBIT

/etc/inittab ……
::sysinit:/etc/init.d/rcS S boot
::shutdown:/etc/init.d/rcS K shutdown
## ttyATH0::askfirst:/bin/ash –login          <<<<<<<<commented out

/etc/rc.common
#!/bin/sh
# Copyright (C) 2006-2012 OpenWrt.org
. $IPKG_INSTROOT/lib/functions.sh
. $IPKG_INSTROOT/lib/functions/service.sh
initscript=$1
action=${2:-help}
shift 2
start() {
##      return 0              <<<<<<<<commented out
ser2net                <<<<<<<<added
}
stop() {
##      return 0             <<<<<<<<commented out
killall ser2net        <<<<<<<<added
}

So we are getting closer to the goal of having it all up and viewable on the net.

************************************

Well, it is now the end of September (30th) 2013.

My Very Clever Son has started writing a monitoring system from scratch as we were having trouble getting the other one running reliably. Here is the first day’s info. There still is some adjustment to come and the graphs will be uploaded to this web site each hour eventually, or that is the plan at this time.

 

All the code will eventually become available I believe, for others to use.

Then we need to attack the smart meter readings………

 

One of the inverters fails comms more than the other for some reason. It is the end one on the network so I may need to put terminations on the lines. Also, when the inverters were installed, the main isolator breaker was underrated so in strong sunlight it drops out. As the inverter power approaches full power I will turn one off to keep the power running. The isolator will be changed to a correctly rated one.

I remembered that each inverter has 2 x sets pf panels so I turned one set off on the “Blue” inverter to limit the current to closer to the existing isolator rating.

 

You can see the gap in the left hand part of the graph just above the word “Thu”. That is where the breaker popped.

On the right hand part, the sudden drop in the blue away from the green is where I turned one set of panels off.

 

21st Oct. 2013.

Yesterday I found the “Blue” inverter was jumping in output power. There looks to be an intermittent connection in part of the solar array. As I had previously switched half off to that inverter, I just changed to the other half of the array and the power level stabilized. Just another thing to fix with this installation. Did I mention since the array has been installed, our house roof leaks when it rains?…..

 

Have a look at the code that Daniel has written…

https://github.com/dparnell/evasolar

 

It is now 21st December and since this system was installed in April, it has feed about 6200KWH into the grid.

I am very pleased in that even with our air conditioner running, we are still feeding into the grid on a hot sunny day. Having largely free aircon is pretty good.

Usage20131216

We do use a lot of power as you can see from the graph, running the house air conditioning during the night as it was hot.

Now if only I could store that extra energy and use it later….

Also,it would be nice if the installers would fix the problems with our system, but so far I have not heard from them re a date to do that. Our kitchen ceiling is getting damaged when it rains and although I can get it fixed under our house insurance, that will only fix the damage, not the cause. So I wait……….

Usage-20121002-20140106

 

This shows we are using less power than previously. (click on it for a better res image). Also it gives an indication of power costs here in Oz. A couple of days ago I got a letter explaining how our power and gas prices are about to go up yet again!  Wot Fun (Not).

 

2014….

It is now 18th Jan. 2014 and we have been having a really hot spell, 44C  (111F).

So far, the system has fed in 6.725MW into the grid.  All the sunny time, it has been running with on 3/4 of the panels so it should have been quite a bit more than that.

The installer company still has not come to the party to fix the problems with this installation. I am not to happy with them!!!

Our last power bill was half the previous one, so the extra sunshine is really helping.

 

25th March 2014.

Still nothing from  neeo “the energy experts” ( neeo.com.au )

They may be energy experts but their service leaves a  lot to be desired.

Almost a year has passed since they installed the system and even after a number of emails reporting the problems…

The circuit breaker has still not been replaced.

One bank of panels is faulty so the system is only running at 3/4 capacity.

The roof still leaks in heavy rain.

I don’t think the paper work from the installation has been submitted as there is no mention of solar power on our bill.

Their web site looks pretty good but I would not recommend neeo as an installer. Problems do come up and I have no worries with that, just as long as they do their best to fix them. So far that has not happened. There was one return visit to “fix” the roof leak earlier in 2013 and we hoped that was it but since then the leak has not been looked at. The faulty bank had been ignored, the breaker not replaced (I have purchased one of correct rating and brand, but I will not install it myself) and the paperwork to our power company ??????

Still, to date the system has fed in 8.24 MW so if it was working on all 4 banks, I reckon it would get over 10MW feed in per year.

neeo? No No!

 

7th May 2014.

Well, a couple of days ago (on 5th May 2014) I received a call from P—- at neeo to say they are going to fix the problems with the solar power on our house.  This was after last week I sent an email saying “it was still raining in our kitchen”.

Stay tuned for the next exciting episode 😉

 

 

18th May 2014. (11:40PM)

It has been 11 days since I received the phone call from neeo and I have not had any other action.

Understandably, my Dear Wife is getting really impatient and so in the morning no doubt I will have to ring them for an update….

 

5th June 2014.

Well, my wife rang them and was assured it will be looked into.

We have had a couple of contacts with neeo indicating it will be fixed, but it is STILL raining in our kitchen.

Neeo look to be having a fight with the installers (threatening their license..)  to get the work done. I am not interested in the politics of the situation, I just want this all resolved! My want is to get some local tradespeople to fix the problem and bill neeo, then let neeo chase their installers for recompense on their time, not ours. This is really getting to be very annoying, and we are heading into winter here.  That will make it hard to fix as it requires fine days to allow lifting solar panels and roof tiles, not to mention the danger of working on a wet roof. The way it is going I would not be surprised (but very disappointed) to see this put off to later in the year, if at all.

There has been a lot of talk over many months but no action. I’m not interested in talk, I want work!

 

11th June, 2014.

I have had enough of neeo.

This is a copy of the email (complete with typos) I have just sent to Peter Charitopoulos and Peter Morrison of NEEO Pty Ltd.

“Hello Peter.

It has taken too long to resolve our problems with the solar system you have installed here.
I will engage a local company to remedy your faulty installation here and send you the account for the repairs, along with the bill for our insurance excess payment. Also, it the power company will not back pay our feed in power, that will be added to the bill.

I am very disappointed  with neeo’s handling of this whole situation.  Problems do come up, and I have no issue with that, but your company’s handling of this all has been less than ideal, to say the least.

My contract is with neeo, and I do not need to wait for you to come to some agreement with your installers, that is your problem, not mine.

neeo has earned a good slab of negative feedback, and that is not very wise nowadays, with internet coverage.This whole situation has been ongoing for an unforgivably long time, and should have been resolved many months ago.

Regards,

Denys Parnell.”

 

I am very disappointed with neeo, as I said in the email. This is not a very good indication of their warranty, so maybe it would be worth while rethinking if you were going to engage them for some work.