I'm curious to know if the Xantrex SW series
(or any other brand of inverter) support remote discharge/charge capabilities. In particular, we are looking to provide residential/commericial backup power systems for our clients with more precise discharge/charge control and management capabilities via a PC/Server-based software application that we have developed.

Beyond mimicking simply the front panel via a software interface a la Winverter, I'm not aware of this capability in the SW series (or any other brand), or whether or not there is a set of public (or private) APIs that are available and can be written to, but may have missed something in the operational specs.

The ideal scenario would be to dial in the power supplied or retrieved from the grid at any time, or at least to have bidirectional on/off control over the power in and out of the unit. Worst case scenario would be to have on/off delivery to the grid and on/off charge control in a separate unit, but of course this would not leverage the built-in charging capabilities of the inverter.

Eager to hear from either Inverter manufacturers themselves or anyone out there that's come across this capability.

All the best,

Hi Lopez,

Thanks for the excellent questions. Could describe in greater detail your particular application? I agree the ability to remotely control power in/out of the inverter is an awesome capability and it is the wave of future power consumption. However, Xantrex products (SW Series) do not have remote control capabilty at this time. There are a number of groups in our industry who are creating remote control devices that work over the internet, local area networks and the like, however that technology has not made great strides in the RE industry. Any help you can provide on describing your application would be useful for us to determine if this is service we would like to address.

Best regards, Chef PV @ Xantrex Tech

Absolutely, can you send me your contact info.

Lopez,

would it be to much to type up as a response your idea in detail ?? this would be an excellent start to what I have hopes for this site.

Thanks !!

Certainly,

As mentioned by PV Chef, there really are some awesome capabilities enabled by built in remote control of input/output. Forget dial-up line in modem ports and closed-system connectors, any inverter that is IP enabled (think of all the hype in the consumer appliance market with XBoxes, PlayStations, Internet Fridges...OK... so that's out there) immediately appears on a home network - a huge growth area - and can be monitored from a web browser on a PC. Beyond simple monitoring and setting of onboard control parameters by the owner or to allow remote trouble shooting by customer support personnel (a great feature), the ability to control input/output either remotely across the Internet or over a local network opens up huge distributed generation opportunities. For example: peak-shaving in commercial applications using either RE sources, stored energy, or a combination of both; peak demand power 'generation' of energy stored during off-peak hours might be able to reduce commercial/residential utility bills by 5% or perhaps even more. Add the specific abilities to dial-in the exact amount of energy output (or input) by the inverter and you can create some pretty incredible applications.

The truly ideal configuration would be an inverter with a built-in ethernet port (true plug'n play), an onboard micro web server (for monitoring, setting/reviewing control parameters, trouble shooting, etc.) and a set of XML-based (eXtensible Markup Language, similar to HTML) Application Programming Interfaces (APIs) for more discrete control over the input/output flow of power though the inverter

With these features (or at a minimum, the control APIs) built into the inverter, the key of course becomes providing the management and control software to enable these features. That's where PeerP*** Technologies comes in.

For more information on the work we're doing in this space please feel free to contact us at info@peerp***.com (or if the filter on this message board blocks some of the apparently offending letters in our domain name, an e-mail to cmaglaque@hotmail.com would also work in a pinch)

all the best,

Yep the filter did it's thing - it's P-e-er-P-a-s-s Technologies

again, all the best

Some time ago, I queried Trace about publishing their SWxxxx API's for use with other vendor-supplied products. I was informed that Trace/Xantrex does not and never did supply API definitions to other parties.

The Winverter controls people were reported to have reverse engineered the SWCA interface in order to make their program work with the SW. I don't know that this is true, but it is the party line Xantrex's Tech Support uses. They cite it as a concern over liability should an inverter cause damage or malfunction due to easy access to the API by a non-Xantrex device.

Matt Zilmer

Speaking of reverse engineering, my ST2500 has a DB9 connector on the cpu board with obvious connections to pins 2 and 3 (RS232 data-in & data-out). Connecting them to my PC serial port with an ordinary 9 pin modem cable delivers data at 19.2 Kb. There is a repeating sequence of about 50 bytes, with about 12 parameters obvious. The data is strictly binary, so it will take more than a terminal emulator program to decipher.

The article in Homepower states that the STXR DB9 port does not have a standard pin out, and will damage the serial port of a computer. I can't speak for the XR, but my ST's DB9 works just fine with my computer. There aren't any control signals (RTS/CTS/DSR, etc.), so you have to configure the PC port appropriately to disable any handshaking code. The claim of potential damage appears to be FUD (Fear, Uncertainty, & Doubt). I suspect that nobody really knows what the protocol is (they left for Outback?).

It's runs at 9600 baud , not 19.2k and the reason it dangerous is that the ground connections are PV minus , not earth ground. so as the internal switchers transistion on and off there are large voltage spikes injected into the PV minus line. The PV +/- has a large common-mode choke so the spike don't leave the box but its on the ground pin of the rs232 db9 port.

You can damage your computer big time and you will also effect the performace of the CPU since connecting earth ground will in effect provide a path to earth ground thru the CPU card.

I know what the data is an have written a program I have dubbed "Fish Bowl" to mimic the LCD display.

I will have more information on this in about a months time , but I do plan releasing Fish Bowl.

I'm not sure wether to give it away or charge a small fee to help support this site. In any event , it requires an rs-232 opto-isolator which runs about 100 bucks


I'll get some screen shots up in a couple of days so people can buzz about it ....


[img]images/smiles/icon_biggrin.gif[/img] [img]images/smiles/icon_biggrin.gif[/img] [img]images/smiles/icon_biggrin.gif[/img]

That looks better! The data is stable at 32 bytes per block. The field with 0x258/0x259 is obviously the power line frequency times 10 (decimal 60.0 & 60.1). The 0x00F4 field is obviously line voltage (244 volts). The rest will take a little more effort.

I used a portable computer running on batteries, so I guess that's what protected me from the ground problems. An opto-isolator is clearly a safer way to go, but a battery powered PC would seem to be adequate.

So Data-Out is, at least, understood. Data-In is also wired to the DB9. Does it squeal when you send it commands? The ultimate would be to take over the MPPT algorithm from an external PC.

My take on the efficiency of the ST/STXR is that it is just a matter of the proper firmware. It's never going to reach the 95% efficiency of the SMA, but solving the MPPT hiccups seems entirely reasonable.