Solar Freaks

Hi Guys,
Just got a Sunezy2800 installed (in Switzerland) and like everybody else was shocked at the crappy software that comes with it.
I want to integrate it into my existing environmental monitoring system, 1-wire + linux based and Matts' script certainly looks promising. Thanks Matt!

If you are still interested in a dump from a sunezy/schneider Electric/Phoenextec system then here is the first few mins I pulled from it this morning before hitting the coal face.
Attached is txt file with serial log and what the crappy w98 sw displays at the same time.

-D

Hello fwmeister

I took a quick look at your output, and the first thing that struck me is that your output is very similar to mine - you have one additional parameter between the Temperature and the Vpv (which I call Vdc in my script), the rest is all the same.

This is apparent both in the "2nd request" where the inverter lists the parameters it will be sending - you have an "0D" between the "00" (temperature) and the "40" (Vdc1) - and also in the main data dump where if we look at your last "Answer" you have 01 6B (363 = 36.3 degrees), then the unknown 00 3F (63) and then 0C 15 (3093 = 309.3 Volt)

This means if you want to be up and running very quickly all you need to do is in my script change this line:

Code: Select all

$Vdc=(hex(&fromdata($d[11]) . &fromdata($d[12]))) / 10;

to

Code: Select all

$Vdc=(hex(&fromdata($d[13]) . &fromdata($d[14]))) / 10;

and the same with all lines below it - increase the array ref by 2.

I can't figure out from your screenshots what the unknown value represents - in your log it starts at 0x39 (57) and slowly increases to 0x3F (63), and all the values that are in your screenshots are already accounted for anyway.

Hi JinbaIttai,

Worked almost from scratch !

Only differences are:

1. my data request string is aaaa0001010001822a
2. the data you couldn't identify is power produced today (Etoday)

btw. What's mod ?

With rrdtool it's a winner.

Now I just have to figure out how to safely get the temperature of the panels and the solar exposure.
I don't think my 1-wire sensors or any of the computers would survive a lightning strike.

Thanks a lot

So for the search engines; Schneider Electric SunEzy 2800 inverter rs232 serial port hack PHOENIXTEC (European version?)

These are relatively new on the Australian market...

Are these also the same internals as the CMS/Orion???


http://kinglongpower.en.alibaba.com/sea ... &d_type=sp

compare the specs with Orion/CMS - even the bottom of the inverter casing
and the rear looks the same as CMS/Orion

http://www.ecvv.com/product/2328152.html

Hi fwmeister

"$mod" = "mode" - either 1 (normal) or 0 (wait) - can give an early warning of impending shutdown of the inverter due to lack of sunlight, but the reality is there are plenty of other indications of that too, so pretty much useless

Inspector: indeed they look very familiar. Looks like Sunteams are yet another Phoenixtec rebadge.. Those guys sure get around

Inspector wrote:These are relatively new on the Australian market...

Are these also the same internals as the CMS/Orion???


http://kinglongpower.en.alibaba.com/sea ... &d_type=sp

compare the specs with Orion/CMS - even the bottom of the inverter casing
and the rear looks the same as CMS/Orion

http://www.ecvv.com/product/2328152.html

They look mechanically the same as the CMS/ORION.
Also look electrically the identical too.

The CMS is made in Taiwan, those units are made in China.
If they aren't a CMS, they must be a very close imitation.
Also, CMS has come out with the CMS4000 and CMS5000, which look identical to those units.

Its all starting to make sense. The SEA Orion is a rebadged Suntems1500, and the CMS2000 is a rebadged Suntems2000. Look at the voltage and power ratings.

Hi Guys,
Yep, these inverters all look the same. Just different colours, badges and slightly different firmware.
The mode just confirms what we already know. 0 watts. I think I'll use it now to schedule a restart of the program at sunrise (at cmd) then exit.

Do you guys notice a big drop in power as the ambient temp rises?

Fwmeister

Question for anyone using ProControl to log their inverter data...

Has your logfile ever become corrupt? Mine "reset" itself this afternoon, after having been logging for approx 10 weeks continuously (with sample rate set to 10s intervals). The weekly/monthly/yearly logfiles are still there. I did notice last week the daily logfile was a very huge file, so I suspect it reached a limit (ProControl or DOS, not sure) and now only has data from 3pm this afternoon. I'm rather disappointed I didn't make a backup of it - I would've liked to have had the data to compare with when I upgrade my system in the next few weeks.


Also I seem to have found a solution to ProControl not recognising the inverter every morning (PC and ProControl left running 24/7 did not work - I had to restart ProControl every morning after the inverter had powered-up for it to be recognised). I borrowed an RS232-USB adaptor from a friend and magically ProControl found the inverter for two mornings in a row. Perhaps it's something to do with the "plug-and-play" feature of USB which the older protocols from the 1980s did not have

TonyB wrote in May:

Hello Inspector,
The RX from the Inverter goes to TX of your PC.
The TX from the PC goes to RX of your Inverter.
That is, Pin 2 of inverter to Pin 2 of PC RS-232.
And, Pin 3 of inverter to Pin 3 of PC RS-232.
Also, Pin 5 to Pin 5, Common Ground.


I guess I am asking (with all the respect I can muster), are you sure? When I read Tx->Rx and Rx-Tx I think that the 2s and 3s are crossed over.

I am currently battling to connect my new CMS2000 to my PC via RS232. I have been trying for quite a few days and reading scores of posts I am also experiencing the same pain as I have seen you guys; Inverter's being powered down by the time you get home from work and then having 30 minutes before work to madly try and get it going.

My (too) modern PC has no COM port exposed, only an ASUS header on the mobo. So I built a module myself using:
http://jaycar.com.au/productView.asp?ID ... rm=KEYWORD and
http://jaycar.com.au/productView.asp?ID ... rm=KEYWORD and
http://jaycar.com.au/productView.asp?ID ... rm=KEYWORD

I had ALL SORTS of trouble lining up the pins between the mobo and the serial cable. I built the serial cable also (20m long) with good quality 5 core data cable and two DB9 (MALE) plugs. I wired pin 2-2 and 3-3 and 5-5 as the inverter manual suggests.

I cannot for the life of me get any data onto a hyperterminal program or into ProControl. Just blank. Tonight I am going to try crossing over 2-3 and 3-2 (null modem style), but I am otherwise out of ideas... help

Oh also, I had to get a http://jaycar.com.au/productView.asp?ID ... rm=KEYWORD because I bought male DB9s for the serial cable. Does this flip/messup/confuse the pinouts? My cable has:
Inverter -> BD9 male plug -> my data cable -> DB9 male plug -> DB9 gender bender -> DB9 plug -> ribbon -> Mobo Header block. Son many chances for silly errors.

Thanks for any insights!

PS: I must add, I am surprised about the affordability. Here in Brisbane I only paid $1990 for a 1.5kw system from Eco-Kinetics, and I got a 2kw inverter and a very tidy system. Pics here (http://glenhendry.blogspot.com/) if anyone is interested.

I finally fixed it!!! Relief. I am not a fan of serial comms. I was right, in the end I had to route pin Inverter2->PC3 vice versa. Although maybe I only needed to do that because the DB9 gender bender plug switched them to the wrong cross over?

Either way ProControl can see the inverter and is capturing the data. Hope my confusion helps someone in the future.

If anyone is interested, I stumbled across the CMS webpage, which I could find on an earlier attempt:
http://www.carbonmanagement.com.au/html ... d_encl.php

No mention of firmwares or better software than ProControl though...

Regards

I have emailed them (they are an Aussie mob) requesting firmware and software updates. Fingers crossed. However, a few threads on other forums have mentioned that they have had no luck contacting them. I kinda wish now that I had spent a lot more on something with all the bells and whistles - but, that's life.

I've also emailed them, but have also been liasing with a member of another forum who has reported that their latest firmware patch still does not resolve the problem of comms dropouts within 20 minutes and continually needing to reset the comm port on their PC. Early stages yet, so I'd recommend waiting until all the bugs have been ironed-out. In the meantime, I'm more than happy with ProControl and hope the RS232-USB adaptors solve the restriction of logging only one inverter at a time, and the problem of me having to restart ProControl every morning (after the inverter starts up) so it recognises the inverter is physically still connected.

Interesting news, Dean Edmonds (dean.edmonds@carbonmanagement.com.au) from CMS was kind enough to return my email about software for the CMS2000 inverter, but when I anxiously cracked open the ProControl he sent me and went to Help|About, it said version 1.0.0.0. Doh. Obviously this is an earlier version of ProControl than we are all using (which is 2.0.0.0). So no luck yet.

I hope I am not being captain obvious here, but maybe my lessons getting my CMS2000 inverter hooked up to my PC over serial cable via RS-232 will help others in the future.

I have used ProControl successfully, but also there is a Perl script to poll the CMS2000 written by jinba @ jinba-ittai.net. (talked about quite a bit in whirlpool forums).

Also there are Python scripts written by Edmund at http://www.edmundtse.com. To run this you need to install Python from (http://www.python.org/download/) and also install the pySerial python module that it uses from (http://pyserial.sourceforge.net/). Then you need to run from the command line "python __main__.py COM1". If that fails, its because your PATH environment var is not set correctly and the cmd.exe window cant find python.exe (normally in c:\python27) or it cant find the __main__.py script. Either way, you can use fully defined paths such as "c:\python27\python.exe c:\temp\pv\__main__.py COM1"

You would use these scripts to capture the data from the inverter if you wanted to analyse and chart the data or upload to RRD database or http://www.pachube.com

HTH
Glen

Hi,

I've been working on a C# .net data logging windows service for my own CMS2000 and I ran into a problem after a power outage here in Sydney.
The power outage happened in the middle of the night and caused the Inverter to reset itself.
The LCD was displaying German and a new value of Etoday.
This new value Etoday has changed the position of the other data values in the return byte array.

It seems we need to cater for 2 flavours of return data from the CMS2000.

Anyway it was easy enough to code for but just thought I would share this info with others who may be having difficulty setting up their logging.

The key was to look for "FFFF" in the return array. If position 23 and 24 are "FFFF" then the rest of the data is now 2 bytes out.
ie. position 11 and 12 now contains new value Etoday and Vac is now shunted to new position 13 and 14, Iac now at 15 and 16 etc..

The RS232-USB adaptors arrived so I connected them up and left everything running for a few days while I was away. I am happy to report that the datalogging worked flawlessly, and I was able to flick back through the graphs and see we had perfectly clear skies on Saturday, with my best output yet (after tallying up the logged data from the daily history files)...

The RS232-USB drivers are the "prolific" ones others have mentioned. It works fine with two inverters running one ProControl task; each adaptor has it's own COM port # (5 & 6 in my case).

I was emailed details of the Ez-Logger and it's not cheap (over $800 I think I saw on a website) but it looks like it would do a similar job, but I can't justify the cost of almost another inverter for somemthing I managed to get working in a similar fashion for under $30!

glenhendry wrote:I have used ProControl successfully, but also there is a Perl script to poll the CMS2000 written by jinba @ jinba-ittai.net. (talked about quite a bit in whirlpool forums).

Yes indeedy, and in fact if you go back to the beginning of this thread, you will see I did most of my development here

Mal wrote:The power outage happened in the middle of the night and caused the Inverter to reset itself.
The LCD was displaying German and a new value of Etoday

How.. bizarre, I haven't seen a CMS2000 do that before. Maybe I should trip the AC power to mine overnight and see if I gain an E-today value tomorrow

- Matt aka Jinba Ittai

JinbaIttai wrote:

Mal wrote:The power outage happened in the middle of the night and caused the Inverter to reset itself.
The LCD was displaying German and a new value of Etoday

How.. bizarre, I haven't seen a CMS2000 do that before. Maybe I should trip the AC power to mine overnight and see if I gain an E-today value tomorrow

I expect the inverter received a power surge while the power dropped out or came back on, which fried it's memory and luckily for you, nothing else...

Hi guys

Maybe you have seen already, but I just noticed version 2.5.0.6 of ProControl available at:
http://www.solaraustralia.com.au/solar_ ... orion.html

Looks similar to v2.0.0.0 at first glance...