New Product Evaluation

[Ric Murphy]

Good Day All,
Not sure how many of you folks are using Outback components in your RE systems but I thought I’d share some personal feedback with you on one of their new products. Last fall they released their FLEXnet DC battery monitor. I was able to get my hands on one a few weeks ago and so far am very happy with it. Outback gear has always had a significant accuracy problem in its data reporting. For the life of me I can’t understand how this problem can exist on such a fine product line but it does. The other problem I have with my system was the inability to log wind data from my turbine. The FLEXnet has solved both of these issues. The FLEXnet allows connection and monitoring of up to 3 shunts anywhere within the system. In my case I have one between the battery and the inverter, a second between the charge controller and the battery and the 3rd between the turbine rectifier and the battery. The unit is connected thru a hub network to the Outback MATE controller the same way all the other Outback components are. By scrolling thru the MATE screens the user can view the shunt data in real time, accumulated data for the day or historical data going back as far as the last 128 days. The MATE also has a rolling screen option. When activated the MATE will scroll thru a number of pertinent screens displaying each for about 10 seconds. I have been collecting system data for the last year using a program called Wattplot (not that the data was accurate). This software was developed specifically for the Outback gear by a guy here in Ontario. Although the software is awesome, it is only as good as the data being rec’d from the MATE controller. Now with the improved accuracy provided by the FLEXnet and shunts it really is a sweet combo. Below are some samples of data collected over the last 2 weeks generated by the FLEXnet as displayed by Wattplot.
The first picture is of the main screen that I leave up all the time. Across the top is the Snapshot Bar showing current status of the inverter (FX-1), the Flexnet shunts 1-3 (DC-3) and the MX60 solar charge controller (MX-4). Shown below that is a penplot for the 3 shunts. Data is rec’d once per second from the Mate and plotted in real time. The data can be compressed 60:1 which is how I usually set it. Wattplot generates a penplot for each device connected to the MATE. Hovering the mouse over any point along the penplot displays the actual data and time. Now that I have the FLEXnet I usually only have to view its penplot. Each night at midnight Wattplot takes that day’s data for each device and dumps it into a spreadsheet (one spreadsheet per device). At the end of each month Wattplot will close out that month’s spreadsheets and start new ones. The spreadsheets make it very easy to chart or graph production data. The spreadsheet below was generated from the FLEXnet data with some revisions. The highlited columns were imported from the Inverter and MX60 spreadsheets. A few other columns were deleted. Although both these products are designed for Outback gear, I think it would be possible to use them with many other products. In addition to Wattplot and the FLEXnet, you would also need an Outback MATE ($295) and hub ($195). Not necessarily the most cost effective option but it will do a good job of data monitoring and logging. So after 3 years I’m finally able to get an accurate picture of my systems performance. I’m looking forward to collecting data over the next year in order to evaluate my specific site for wind and solar.
Ric
448

449

450

[Rob Beckers]

Thank you for the review Ric! I was interested in hearing how the FLEXnet DC worked, and am happy to hear it's working so well.

I was talking to an Outback rep last weekend (who is also one of their tech support people for the "difficult cases"); they are currently in a major overhaul cycle for all their gear. Fixing the low resolution issue is one of the items on the list. Seems the reason for the low resolution is not so much because it is measured so poorly, but to save bytes in storing it. Another interesting tidbit was that the FLEXnet DC was in part made to address the issue of automatic generator start. Most people use the 'aux' port on the inverter for this purpose, but the inverter's internal fan also runs of off this same port. Switching it for generator use means that cooling becomes an issue, and the inverter may auto-derate because of temperature. The FLEXnet adds another 'aux' port to the system, that can be used for generator control (through the Mate).

Besides WattPlot (http://www.intellact.ca/energy/wattplot.htm)there is another source for good third-party software for Outback products, WinVerter by Righthand Engineering (http://www.righthandeng.com/hm_wvmfx.htm). The Outback rep spoke highly of both products.

-RoB-

[Ric Murphy]

Hi Rob,
I looked at Winverter when I first installled my system about 3 years ago and wasn't impressed. They have been around much longer than Wattplot so not surprised Outback speaks well of them. Not sure but I think Winverter has been updated and may be more functional now. The spreadsheets are what sold me on Wattplot. Glad to hear Outback is finally addressing the accuracy issue. However it doesn't help anyone with existing product.
Ric

[Julie Haugh]

Hi Rob,
I looked at Winverter when I first installled my system about 3 years ago and wasn't impressed. They have been around much longer than Wattplot so not surprised Outback speaks well of them. Not sure but I think Winverter has been updated and may be more functional now. The spreadsheets are what sold me on Wattplot. Glad to hear Outback is finally addressing the accuracy issue. However it doesn't help anyone with existing product.
Ric

OutBack doesn't speak well of any of the products -- Intallact has taken out-of-context statements by employees of OutBack and turned them into claimed product endorsements.

OutBack is aware that Intallact is making unapproved claims of product endorsement, but has been unable to make them stop.

It does appear that Intallact has quit the business as there hasn't been a new WattPlot update in =months=. Some of my customers have told me that Intallact has been unable to solve problems with Windows 7, as well as 64-bit versions of Windows. I do know that the most recent (Feb '10) version of WattPlot is still for Windows XP -- and the installation instructions for Vista and 7 both suggest using a separate hard disk (as if!) as well as XP emulation mode.

[Dan Lenox]

Ric,

I've recently ported my PowerDashboard software to work with the OutBack product line. What I found out is that for some data values OutBack use whole integer values.

Specifically rounding is done with all the current readings from the inverter and charge controllers, while the FlexNet device does report current at 1/10 amp intervals. Voltage data values are given in 1/10 volt increment.

It appears that everyone that makes software for Outback provide a 'fudging' factor to help alleviate accuracy, and it is not a fix for the issue but does help bring things back in line - to some extent.

Dan Lenox

[Andrew Welch]

Besides WattPlot (http://www.wattplot.com)there is another source for good third-party software for Outback products, WinVerter by Righthand Engineering (http://www.righthandeng.com/hm_wvmfx.htm). The Outback rep spoke highly of both products.

-RoB-

Hi Rob:

I'm not surprised that your OutBack rep spoke highly of WinVerter as well. Randy is a great guy who writes solid software. He was a trailblazer for both OutBack and Xantrex products. I hear he is now working on new tools for Bergey equipment too. Well done!

As for WattPlot, we are as busy as ever, and continue to focus on tools for monitoring OutBack devices on any Windows platform. You should note that our URL has changed since your original post though - it is now simply WattPlot.com (http://www.wattplot.com).

By the way, Dan, the OutBack FLEXmax 60/80 line of charge controllers now reports charging amps at 1/10 intervals as well, further improving the precision of the available data.

[Julie Haugh]

It appears that everyone that makes software for Outback provide a 'fudging' factor to help alleviate accuracy, and it is not a fix for the issue but does help bring things back in line - to some extent.

Dan Lenox

The truncation problem with OutBack data is fairly well understood. Back in 2007, I introduced the first of the "truncation correction" algorithms to the software that became greenMonitor (http://www.greenhousepc.com/greenMonitor.html). The basic idea is to calculate the statistically most likely correct value for the truncated amount, then insert that value into various calculations. The result is that I can do far better than the 50/50 solution that certain competitors of mine "borrowed" without giving credit. How I do that is a secret -- certain competitors already "borrowed" the basic idea, not going to give them more secret info ;)

The problem isn't rounding, in the 4/5 sense of rounding, it's straight up truncation, with a loss of all values less than a whole number -- floor() function, not ceil(). For example, if the inverter is producing 1.999 amps, it reports 1, the same as if it's producing 1.000 amps. We've been looking at adding split core transducer support to our product, but they are all very expensive. The TED 5000, which is readily available, has the ability to support CT's (you buy them when you buy the TED 5000) and we can read those values from the TED 5000 gateway. That's what my clients who need more accurate values for power production and consumption use. That's where I'd go -- add either TED 5000 or some other AC power monitor support to your product.

[Julie Haugh]

It's been a while since I posted to this thread, and a lot has changed, and a lot has been learned.

First, it turns out that the truncation problem is worse than expected. The accuracy of the amp-meters turns out to be "really bad", to the point that they are fairly worthless. The solution for the greenMonitor software is to enable high accuracy DC amp-meters to be used, and the inverter output to be calculated from that.

Second, we have a new name! greenHouse Computers was founded to provide low power computing solutions to off-grid users, as well as people who didn't like the fact that their desktop computer used more electricity than the machine cost. I felt I had a better solution. Within 9 months, the renewable energy system software I wrote became the #1 product, and by the first year, the original purpose of greenHouse Computers had faded into history. What was originally sold as a way to induce customers to buy hardware, had become the main product.

After almost a year of planning, re-evaluating products, and consolidating models, we're now ready to start doing business as greenHouse Gas and Electric, focusing solely on what we've come to do best -- delivering the highest quality, highest function products available to small to medium sized systems (1 to 50kW) owners. The website is just now getting to be useful, but there will be more content at http://www.ghgande.com/ in the weeks and months to come.

[Andrew Welch]

Yep, quite the history going on in this thread. And in case anyone stumbles on this post and is perhaps thinking that WattPlot quit the business - well, it was never the case.

Thanks to the good people at OutBack who continue to recommend the programs to their clients and who send us new OutBack communication hardware to write software for, we're still as busy as ever. Over more than a decade, we have installed software with hundreds of clients on every continent.

But, to be more specific to the original content of this thread, WattPlot 5.1 can now replace poor-resolution device-reported values (especially inverter AC Amps) with FLEXnet DC Monitor shunt values automatically - thus greatly improving wattage accuracy. See this post on their user forum (http://www.outbackpower.com/forum/viewtopic.php?f=25&t=8393) for more details.

Thank you all for your continued support! :smile1:

[Julie Haugh]

Andrew -

If it takes you almost SIX YEARS to get around to responding again, I think that says something.

That said, yes, I'm glad to see it would appear you've finally caught up with the "use shunts to calculate AC power" feature I added, what, 8 years ago? How do you calculate the efficiency curves of the inverter so that you provide the correct AC power output from 0 to 100% rated inverter output? Did you know that most of your customers are operating so low, as a percentage of maximum inverter output, that any AC values you compute without the efficiency curves are meaningless? Can you provide certified (ANSI Standard) revenue grade power and energy accuracy? Please let us all know when you integrate revenue grade meters, weather stations, contactors, Modbus master and slave interfaces, SunSpec device support, and everything else I supported many, many years ago.

The facts are as I stated - OutBack doesn't endorse products from any 3rd party vendor. Sub-forums are provided as a service to their customers. Andrew sent OutBack his software unsolicited, and as of the last time I spoke to them in person (I did a presentation at their office and spoke to their entire customer support team, as well as sales and engineering) they ignore all the various logs and whatnot Andrew's customers send in.

If you want software to monitor your OutBack system, the new Mate 3 does a nice job and it's supported by a company that does millions of dollars a year in sales. They have a fully-staffed customer support team waiting to answer your questions. If you want software to monitor, manage and control your OutBack system, you're not going to get anything from Andrew that supports SCADA. AlsoEnergy has SCADA support and they are a great bunch of people. They are also another multi-million dollar company with a fully-staffed customer support department.

[Andrew Welch]

I knew that updating this thread would have the expected repercussions. You're so wonderfully predictable! :amuse:

The facts are as I stated - OutBack doesn't endorse products from any 3rd party vendor.
I guess you don't have a copy of their current MATE System Controller and Display User Manual (900-0012-01-00 REV C). WattPlot is listed on page 8. Of course, they're not going to formally endorse products from a 3rd party vendor - and they certainly wouldn't say otherwise to a former 3rd party vendor who was not listed in the manual.

But enough of the bickering. It's a shame really. You have so much expertise to contribute.

Oh well. I'll let it drop (and you'll reply one more time).

[Julie Haugh]

I knew that updating this thread would have the expected repercussions. You're so wonderfully predictable! :amuse:

You replied to a thread after an almost six year absence from it. The only thing that's predictable is you run around the Internet touting your software every time you make some change that is years behind every other product on the market.

I guess you don't have a copy of their current MATE System Controller and Display User Manual (900-0012-01-00 REV C). WattPlot is listed on page 8. Of course, they're not going to formally endorse products from a 3rd party vendor - and they certainly wouldn't say otherwise to a former 3rd party vendor who was not listed in the manual.

But enough of the bickering. It's a shame really. You have so much expertise to contribute.

Oh well. I'll let it drop (and you'll reply one more time).

Of course I'll reply one more time.

I worked in the industry as a full-time professional. Yes, I know far more about how solar power plants work, how all the various equipment works, relevant communications and operating standards, etc. The entire industry is engaged in a race-to-the-bottom on all aspects of an installation. That includes panels, inverters, monitoring equipment, switch gear, transformers, interconnections, etc. But being cheap doesn't make it good.

What you do, you do as a hobby -- by your own admissions. You target a single product line (OutBack) on a single platform (Windows). You've never been, or worked for, a hardware manufacturer, you don't integrate with other vendor's equipment, or SCADA equipment, or weather equipment, or meters, or switch gear, or anything that doesn't hand you data via an OutBack provided hardware interface. You've never interacted with or contributed to national or international standards bodies. You've never conducted research into electrical grid management or produced technical innovations in the field of renewable energy systems management. That's the difference between us -- all those things you've never done, I've done.

You've repeatedly treated me like I'm a competitor, and I've never been your competitor. Long before I ever sold a product, it had far more capabilities than anything you've ever sold. It supported products from more different vendors, with far more capabilities.

This doesn't make you a bad person, a horrible programmer, or much of anything else negative. It means that you and I have simply NEVER been competitors. The most absurd thing about this imaginary conflict you've got going is that I've offered on numerous occasions to help you with this hobby of yours.

Best of luck with your hobby. I most certainly wish you well, which is more than you've ever done for me.

[Brian McGowan]

While you guys are discussing the finer points of monitoring I find two major drawbacks to my Outback FM80. I have found the same drawback in other charge controllers and equipment also.

First drawback is the lack of Ethernet connectivity. In this day and age, the fact that I can’t put this unit on my home network is a dismal failure. There should be an Ethernet port directly on the unit. Nothing produced today should not have an Ethernet port. Even after calling outback directly the only suggestion they had was to get the RS232 device and then attach that to a computer on the network. There is an Ethernet device but the rep told me that is for industrial use only with multiple units although I can’t imagine it can’t be used for a single unit. Still either option is 1/3 the price of the controller which is unacceptable.

Second is the lack of control for anything except the panels and one other thing controlled by the single available relay in the most minimalistic way possible. As good as they are at converting panel power into charge current, it seems to be their singular focus to the exclusion of all else.

As far as I am concerned the charge controller should be the “brains” of the system. It is the one device that has the ability to know how much power is coming in from the panels and how much power is needed by the batteries and therefore how much power is available to be used in other places. Therefore it is the one device that can keep the panels producing at the highest amount instead of just choking the panels down to save the batteries which seems to be the common method. It is the one device that knows the battery state and therefore also would be able to control low voltage disconnects.

I can buy other devices and have bought a Victron battery monitor but that means another opinion of what the battery state is. While this is somewhat useful it doesn’t really do anything for me but tell me what I basically already know. There are low voltage disconnects and dump load controllers but they are all limited and each device will have its own opinion of what the state of the battery is.

In short, charge “controllers” are a dismal failure at doing anything but providing MPPT capability to make as much charge current as possible from the available panel power. Because of this I have come home to find my system has spent as much as 5 hours a day either absorbing or floating which means the controller has been choking my panels down for this time instead of making as much power as possible from them.

To solve this, and since charge controller companies seem not to be interested, I am making a device that will be a low voltage disconnect and dump load controller. It will have 8 available circuits to be assigned as either. There will be programmable time delays so things don’t get shut off because the fridge kicks in and the voltage drops for a half a second. It will also have the ability to bring on dump loads, determine when those loads have been satisfied and bring on other loads or combine loads to use all available power that is being produced above what the batteries need.

I realize this will be yet another opinion of what the state of the battery is but it will now be my opinion. The charge controller will be set to do what it does best. Create as much power as possible at all times. I will deal with the rest and see to it that the power is used to best advantage at all times. This will make the best use of my generating capability.
Brian

[Julie Haugh]

Brian -

What you've described certainly can be done, but the problem with selling to the OutBack market is downward price pressure has made products with that level of control cost-prohibitive. The last multiple-load SCADA control system I did sold for $7,500. You likely don't need all the gear I sold, but you'll need other equipment, or else you'll just need to guess. They could be made cheaper, if there were more demand, but there's just no enough demand to justify the development costs, electrical certifications, marketing, etc.

ANYWAY, what you need is two things - a plane of array irradiance sensor and something that can tell you if the charge controller is in absorb. If the charge controller is in absorb and the spilled power (theoretical output minus actual output) is greater than the next available load, turn on the next available load. If the charge controller is no longer in absorb, disable the lowest priority load.

That's it - real simple stuff. It just costs time and money to implement.

[Brian McGowan]

Julie,
Not to worry. I have this covered. I was just wishing the charge controller people would do this since it would be easy and logical for them to provide the capability.
Brian

[Julie Haugh]

Julie,
Not to worry. I have this covered. I was just wishing the charge controller people would do this since it would be easy and logical for them to provide the capability.
Brian

Brian -

They provide part of it, but the rest of it - knowing how much power you can get from the array without leaving Absorb or Float - requires additional equipment and software.

[Brian McGowan]

Julie,
Again not to worry. I am creating the hardware and the software.
Brian

[Peter Klaassen]

Hi Brian,
I am off grid with Outback equipment. I also have a Windspot Turbine with it's controller. I would be interested in the power controller you are working on. I am not much of an expert on this computer controller stuff but I have been off grid for a few years. I am interested in using extra energy from the turbine (when available) for water and home heating. We have hydronic floor heating.
Pete

[Brian McGowan]

Peter,
I will keep you posted as development progresses.
How are you heating your water and the hydronic system? Is it all electric? Are you using inverters to do this?

Considering what others may want to do with this will help me make it as flexible as possible.

Brian

[Peter Klaassen]

Hi Brian, there are two 3600 W inverters for electric supply and gas heats water and food now. Mainly we use wood heat from a masonry stove. The solar array is 1350 W and is only enough in the summer. It would take some more equipment for electric water heating.
Pete

[Julie Haugh]

Hi Brian, there are two 3600 W inverters for electric supply and gas heats water and food now. Mainly we use wood heat from a masonry stove. The solar array is 1350 W and is only enough in the summer. It would take some more equipment for electric water heating.
Pete

Pete -

You may be able to use your dump load control (diversion control relay setting in a FlexMAX charge controller) to heat water in either your main hot water tank, or an additional tank that you recirculate. APRS World has thermal controllers pre-made. The product I was working on (... and ran out of money to fund ...) had four relay controllers which could be programmed using 5 analog inputs and 3 digital inputs. You'd need multiple devices from APRS World, but they most definitely have gear you could put to good use.

[Brian McGowan]

Peter,
We are discussing the controller in the
Arduino LVC-DL controller thread.

Brian