MPPT comparison

stephendv

Hi all,

I was reading about an issue one of the outback chargers had with it's MPPT implementation and was wondering how much do the MPPT functions of charge controllers differ from grid tie inverters?
Do you think the GTI's would be substantially more efficient in their MPPT functions?

BB.

Re: MPPT comparison

Others can answer the MPPT question in detail better than I...

Basically, MPPT started out pretty simple... Assume that Vmp was some percentage of Voc--and just use that voltage. Newer units would sweep every 5-15 minutes and find the "current" MPPT Vmp.

And the newer ones yet, would optimize some 60-120 per second (like GT inverters)... How that is done--I don't know. And the details may be a trade secret.

New MPPT algorithms capture more energy--but I would be surprised if it was a lot more (probably something like up to 10% better--not 50% better over earlier designs--of course, there are poor MPPT designs that don't work as well as a simple PWM controller too).

The other issue is how to compare efficiency with two very different types of hardware.

One is a single conversion from Solar Panel DC (typically 200-600 volts or so) to ~240 VAC and around 95% maximum converter efficiency.

Vs a MPPT charge controller "RE System" which must down convert from ~17-140 VDC to 12 VDC (or 24-60 VDC), store the energy in a battery (~80-90% efficient) then converter back to 120/240 VAC using an inverter of some sort (around 85-90+% efficient)...

So, just the multiple types of energy storage and conversion probably swamp most of the differences between MPPT algorithms for well designed and functioning controllers.

There are other ways of capturing the energy -- One is to use a GT inverter to "back drive" an Off-Grid inverter to charge a battery bank. It turns out that AC true sine wave inverters are "bi-directional"--they can pump power from a DC battery bank to an AC load... But, they aslo can take energy from a synchronous AC source (Like a GT Inverter) and pump it back into the battery bank.

And interesting function, and a vendor or so has actually used this to charge a battery bank from a Wind Turbine GT inverter (SWWP and Skystream). The back driven Inverter is not designed to be a battery charger--so one has to be added to measure battery charge and (in one case) interrupt the AC voltage so that it stops charging the battery bank.

Is your question curiosity--or trying to optimize your system design?

-Bill

stephendv

Re: MPPT comparison

Hi Bill,

It's a practical question as I'm trying to decide between the backfed inverter option vs. charge controller. We've briefly discussed this here before and I wrote a blog entry where I compared the 2 systems: http://www.casanogaldelasbrujas.com/blog/2009/05/13/sma-sunny-island-ac-coupling/

But I didn't take the differences in the MPPT implementation into account. When I visted the SMA test lab of course the SMA engineers claimed that all the other charge controllers weren't doing "real" MPPT. (But then they would say that). Besides the question of efficiency there are also some other advantages to going with a GTI and backfed off-grid inverter. If you don't fancy reading the whole blog entry to summarise:

Advantages of GTI (SMA Sunny Boy + Sunny Island):

• Can use high string voltages and thinner cabling
• Has higher efficiencies when you use the generated power directly (there's an xls spreadsheet which compares the efficiencies of the two systems).
• Don't need a PV combiner as the GTI has connection for 3 strings
• Don't need a DC breaker as these are integrated into the GTI
• Much simpler wiring, as almost everything is normal AC.
• Only 1 charger and 1 charging algorithm in the system. The off-grid inverter/charger is the only charger for the whole system, so it will control all aspects of charging without having a parallel charger interfere with it.
• When only PV power is being used during the day, the off-grid inverter would be in standby mode as the GTI would be supplying all the loads.

Advantages of Charge controller:

• Better efficiency when more than 40% of the power is used from the battery (this will be my case in winter, where I'll be using about 40% of the power from the array and 60% from the battery)
• Lower costs (although this is debatable as I'd still need a PV combiner, thicker cables, DC breakers and more complicated DC wiring), the GTI solution will still be more expensive, but not by much.
• If the off-grid inverter should fail, then I could still keep my batteries topped up via the solar charge controller. With the GTI solution, the off-grid inverter's charger is a single point of failure.

My main concern with the GTI is the efficiency in winter (low usage of direct PV power and higher usage of batteries), and little sun. My calculations are based on the figures in the blog entry, and published by the manufacturers.
BUT if the MPPT of the GTI can give me only 4% more power than that of a charge controller, then the efficiencies will be inline with a charge controller based system.

mike95490

Re: MPPT comparison

stephendv wrote: »

Hi Bill,

It's a practical question as I'm trying to decide between the backfed inverter option vs. charge controller. We've briefly discussed this here before and I wrote a blog entry where I compared the 2 systems: http://www.casanogaldelasbrujas.com/blog/2009/05/13/sma-sunny-island-ac-coupling/

*snip*

My main concern with the GTI is the efficiency in winter (low usage of direct PV power and higher usage of batteries), and little sun. My calculations are based on the figures in the blog entry, and published by the manufacturers.
BUT if the MPPT of the GTI can give me only 4% more power than that of a charge controller, then the efficiencies will be inline with a charge controller based system.

Very very interesting blog and logic here. Do you have any actual numbers on this yet, or are you still in construction stage ? I'm looking off grid, and cloudy days, and
may have a lot of generator run time ahead of me, unless I move my fridge contents into an outdoor ice box!

I wonder how the Sunny system of frequency signalling compares to the
XW system feedback. Anyone know how the XW does this, via voltage, or the Control Data Buss?

I sure like the idea of the high voltage PV array vs the high amp array.

Sorry about rambling on, but I'm still trying to get this to gel in my mind. I just don't want to loose the thoughts.

Solar Guppy

Re: MPPT comparison

XW is not UL approved for backfeed via another high voltage GT inverter. It works but is not a regulated solution at this time.

There is one major advantage of the Charge Controller vs GT inverter back feed and that is surge handling. I have run the Back feed way and any type of large inductive surge immediately knocks off the GT inverter offline and goes thru the 5 minute reconnect delay, battery based inverters are much more robust in this regard

To the OP question, The XW charge controller is a true dynamic mppt, similar to what Gridtie inverters do. As far as harvest, an XW-6048 with charge controllers is as efficient as any Hybird SMA system would be,

stephendv

Re: MPPT comparison

Solar Guppy wrote: »

I have run the Back feed way and any type of large inductive surge immediately knocks off the GT inverter offline and goes thru the 5 minute reconnect delay, battery based inverters are much more robust in this regard

That's interesting. I wonder how the SMA system would react, since it was designed to work this way from the start. One of their selling points is that you can get away with a smaller battery based inverter since you can parallel the output of the GT with the battery inverter.

Solar Guppy wrote: »

To the OP question, The XW charge controller is a true dynamic mppt, similar to what Gridtie inverters do.

Pitty it's only 60A. Do you know how the Outback and Apollo compare?

mike90045, my system still only exists on paper am weighing up the various options at the moment. I'm sold on the Sunny Island inverter, now it's just a question of whether to get a charge controller (and which one?) or GT inverter.

BB.

Re: MPPT comparison

Just an FYI--from December of last year (Windsun is our Admin from NAWS):

Windsun wrote: »

Just a heads up - we no longer carry the Apollo stuff for the simple reason that far too much was coming back on warranty.

Regarding your setup--do you need a large/wide distributed setup for some reason (solar panels on a hill, loads spread around a valley, etc.)?

Of course, you could put the "solar shed" on the hill with the panels, and just run your 120/230/240 VAC (perhaps transformer enhanced at each end of the span)...

Their system certainly looks interesting. The "closed" nature of their configuration may keep costs high (and make repairs/upgrades more difficult down the road if you fall outside of their, future, offerings).

I am not sure that I buy the GT+Off-Grid add together for peak power usage... If you loads are during the day and can be setup only to run when the sun is up mid-day... Maybe. A cloud goes overhead or birds fly over an array, the GT's collapses--so the loads would need to be shed and restarted.

In any case, do a paper design of the several systems and see which $$$/performance works out best for you.

-Bill

Solar Guppy

Re: MPPT comparison

stephendv wrote: »

Pitty it's only 60A. Do you know how the Outback and Apollo compare?

You didn't indicate what size of a system your considering, 3kW is not small in the RE world for a controller, even the larger GridTie inverters are ~5kw. A Street price of a 60A controller is about 500US, compared to a Xantrex GT3.3 for about 1800US. Need more charge controllers, just add as many as you need, the solar will be the same, its all multi-string anyways, just configured for lower voltage ( typically 1/4 of Gridtie ) The only difference is the wiring.


Unless you can have some automated load shedding, your going to need a battery and inverter setup to power when there isn't solar. The largest inverter is the XW6048, three can be ganged for 18kW ... so I suspect your idea of scale here doesn't match what the available equipment is regardless of which vendor you purchase from

Efficiency wise, the Xantrex XW system is the best on the market and it is very close to battery less grditie system, so that's really not the issue.

stephendv

Re: MPPT comparison

BB. wrote: »

I am not sure that I buy the GT+Off-Grid add together for peak power usage... If you loads are during the day and can be setup only to run when the sun is up mid-day... Maybe. A cloud goes overhead or birds fly over an array, the GT's collapses--so the loads would need to be shed and restarted.

I'd agree if you use off-grid inverters and GTI's that were never designed to work this way, but happen to do so.
But the SMA off-grid system was designed to work this way by design, I've seen it working in their lab, and it's quite impressive how it adjusts the GTI output according to the required loads. The only charge controller SMA sells is a 40A one designed for smaller off-grid systems. ALL of their larger off-grid installation are done with GTI + Sunny Island, that includes village electrification projects and many off-grid homes/lodges/guest houses.

Solar-guppy, my system is sized correctly I believe, the problem is the array will be in the awkward 2.6kW - 3kW range (Need 7kWh/day in Northern Spain). I plan to install 2.6kW to start off with, and see how I manage with it, with the option of expanding to 3kW.
To be honest I haven't investigated the XW system much because Xantrex doesn't seem to have that much of a presence in this part of the world. The SMA guys will ship me a new standby inverter within 48 hours if something goes wrong with my existing one - and they have offices just 2 hours from my place. But I should be fair to Xantrex and investigate their terms and support as they seem to be quite well respected on this board. If not for the complete XW system, then for at least the charge controller.

Solar Guppy

Re: MPPT comparison

Xantrex ( now a subsidiary of Schneider Electric ) is 20X the size of SMA, and has world wide sales and support. I know they have offices in Spain so take a look around. the XW-6048 comes in a EU flavor as well.

Xantrex Technology S.L
Bac de Roda, 52, edificio A
08019 Barcelona, Spain
Tel. +34 93.433.8350
Fax. +34 93.433.8351

Email: europesales@xantrex.com

nigtomdaw

Re: MPPT comparison

Stephen,

Xantrex European HQ is in Barcelona as is Outback. 220 km about 3 hour car trip for you.

Im not in a position to comment on SMA equipment as I have never installed it. Im currently using a 2.4 kw array half on a XW60 mppt the other half on a Xantrex C40 pwm CC.

The best recorded daily input from the XW60 was 5.6 kwhrs on 1200 watts of PV .However I normally see harvest of between 2.5 khrs -4 kwhrs on the XW daily log when the batteries are then fully charged, so Im using around 6/7 kwhrs a day on average. With my current 2.4 kw array I need 3 to 4 hours of full sun to replace the previous day night energy use. I rarely cycle below 90 % from one sunny day to the next. Genny run time for this year to date has been less than 20 hours. I expect that to increase a little when my son moves back home, in September, extra TV Play Station 3 and Laptop running

As Im only a couple of hours south of you that should be of help and not far off your 2.6 kw PV array calculation. I have mains water so I dont need to pump water which can be a energy hog, my biggest indulgence in a American Fridge Freezer which according to its spec sheet uses 1.5 kw hrs a day.

In this region of Spain Outback and Xantrex Equipment is popular along with Studer, the latter being off my fit list because of failure rates I ve seen with the larger units, which to be fair to Studer are being phased out for a new design.
Xantrex also lost there way in market share with the Transition from the SW range of Inverter Chargers to the new XW model, of which I have only seen one in the flesh at the Valencia RE Expo last year. Ive fitted 7 VFX Full Outback systems and upgraded 3 others all are very happy with there Outback Systems and apart from one customer who went back to the UK for a couple of weeks in Febuary leaving his old fridge freezer connected and drained/damaged his batteries I have had no problems or call outs.

The Euro market sadly lacks a small affordable 1500w True Sinewave Inverter Charger with PFC charging , similar to Magnum Energy US 110v 60 hz offering, sos a little off topic.

Do SMA provide any Lightning Protection for there equipment, it is a real problem out here and you should factor in earthing and lightning protection in your research if you havent already done so. Havent come across any of SMA eqipped installs.

I got a free booklet from this company Which was very informative.



Good Luck with your research Nigel

boB

Re: MPPT comparison

nigtomdaw wrote: »

Stephen,
The Euro market sadly lacks a small affordable 1500w True Sinewave Inverter Charger with PFC charging , similar to Magnum Energy US 110v 60 hz offering, sos a little off topic.

There is the Magnum 1300 Watt, 24V 50Hz unit with PF charger. It surges pretty good.

http://www.magnumenergy.com/Literature/Data%20Sheets/Inverters/64-0485%20Rev%20C%20(MM-E%20Series%20Data%20Sheet).pdf

You may pay only slightly less for a 1500 Watt inverter as opposed to say, a 2500 Watt inverter of the same genre, so it may not be that big of a deal to just go for the slightly bigger inverter ??? Maybe ? Those do exist, right ?

boB

stephendv

Re: MPPT comparison

Thanks Nigel, I'll give Xantrex a call on Monday to see what the terms of their support and guarantee are. With regards to a small inverter/charger, Victron do quite a range of smaller units, popular with the marine crowd. There's a 1200W and 1600W model, for 1000 euro's and 1200 euro's respectively.

nigtomdaw

Re: MPPT comparison

boB

I am aware of that unit alas its only modified sine wave. Something like the MMS 1000 watt unit, as its light and very shippable cost wise, was on my wish list

Nigel

boB

Re: MPPT comparison

nigtomdaw wrote: »

boB

I am aware of that unit alas its only modified sine wave. Something like the MMS 1000 watt unit, as its light and very shippable cost wise, was on my wish list
Nigel

Ooops ! Darn ! OK, I got mixed up there.

Doesn't someone make, like, a 2500 Watt 230V, 50Hz inverter with PFC charger in Europe ? A 1500 Watt version certainly couldn't cost much less.

boB

nigtomdaw

Re: MPPT comparison

The MMS 1012 would be ideal for smaller power needs over here in 50hz 230ac format, I see it has a 5min 1200 real watts rating and a 30 secs 1600 real watts rating, with a PFC included and a list retail of $1199 only $300 more that the similar MSW version, I note NAWS stocks Magnum Energy gear but not these smaller inverters.

stephendv

Re: MPPT comparison

Was just reading this thread: http://forum.solar-electric.com/showthread.php?t=5663&page=2 about how the FM80 does it's MPPT sweep. If I understand the comments correctly, it's programmed to do a sweep every 1.5 hours! So if it happens to be cloudy at that time, it will lock on to the wrong voltage and stay that way for the next 1.5 hours. Does the Xantrex do the same?

Solar Guppy

Re: MPPT comparison

stephendv wrote: »

Was just reading this thread: http://forum.solar-electric.com/showthread.php?t=5663&page=2 about how the FM80 does it's MPPT sweep. If I understand the comments correctly, it's programmed to do a sweep every 1.5 hours! So if it happens to be cloudy at that time, it will lock on to the wrong voltage and stay that way for the next 1.5 hours. Does the Xantrex do the same?

The Xantrex is true continous dynamic tracking, the only one on the market that from my testing works, but this is from march/april testing of all available controllers on the market and manufactures are always working to improve their products

The FM60/80 have some type of tracking, doesn't look like true dynamic tracking to my eye. The units I tested this spring had issues ( I saw many days with 20% harvest losses due to mistracking ). It was recently posted Outback has new firmware which I haven't sent our units in to be upgraded ( can't field upgrade ). I won't be able to retest for a few months as I'm swamped with other work.

The MX60 was sweep and sleep and could get stuck on clouds and other environment changes but could be programmed to sweep at a user selected interval. Personally, I'd take a MX60 over the FM series, for starters you could program a MX60 for 70 amps, not so for a FM60, also with the reported amps reporting not quite on, the two FM units I tested couldn't do rated actual amps ( FM60 limited to about 59 actual as it was reporting 60 ) , part of apparent optimistic battery watts reporting that every single Outback charger I have ever tested does, that's 6 or 7 now that I've tested.

Dave Angelini

Re: MPPT comparison

SG, I see my XWCC updating MPPV in bulk about every 3 seconds. Is that whet your seeing?

Solar Guppy

Re: MPPT comparison

Normal operation for the perturb on the XW

mshen11

Re: MPPT comparison

in one of the earlier posts, someone said MPPT is an algorithm... so does that mean on the more expensive controllers (specifically outback 80amp MPPT) the algorithm can be changed from time to time via firmware (as they advance)?

is that something only outback can do or can someone w/ a computer do it (like a computer firmware upgrade of bios)?

BB.

Re: MPPT comparison

Yes, they can be changed--it is just computer code. But, while some can be downloaded by the customers--others require a trip back to the factory for the whole control (from what I have read).

The changes in MPPT firmware is probably limited to bug fixes--you probably are not going to see significant changes in tracking efficiency (again, from what I have read hear) over time.

-Bill

Solar Guppy

Re: MPPT comparison

Its actual much more than code, the underlying hardware plays a large part in the effectiveness in tracking. Current and voltage sensing is the limiting factor , typically resolution and filtering, both analog and digital and the loop control software is critical in giving the Mppt software the correct information to do its thing.

From my experience you need 12 bit ADC resolution for voltage & current and also need input/solar side measurement as battery side can be influenced by things like 120hz inverter ripple.

The XW is the only unit on the market that has this hardware and enables the software to work as well as it does. The up coming MorningStar unit has similar hardware and our testing has re-affirmed this is what it takes to have high efficiency and accuracy tracking

boB

Re: MPPT comparison

Solar Guppy wrote: »

Its actual much more than code, the underlying hardware plays a large part in the effectiveness in tracking. Current and voltage sensing is the limiting factor , typically resolution and filtering, both analog and digital and the loop control software is critical in giving the Mppt software the correct information to do its thing.

From my experience you need 12 bit ADC resolution for voltage & current and also need input/solar side measurement as battery side can be influenced by things like 120hz inverter ripple.

The XW is the only unit on the market that has this hardware and enables the software to work as well as it does. The up coming MorningStar unit has similar hardware and our testing has re-affirmed this is what it takes to have high efficiency and accuracy tracking

Yep. The hardware has to be there. You can't just take a Trace/Xantrex C-40 and apply an MPPT algorithm.

But f the hardware doesn't have efficient power conversion hardware, all that fancy high resolution A/D and current sensing can just get knocked back to a not as good product.

But even great hardware doesn't mean squat if the MPPT algorithm is crap.

It all boils down to how much more energy, or Amp-Hours you get with MPPT as opposed to without MPPT, or one MPPT controller compared to another MPPT controller product.

In general, just having an MPPT controller, as long as the tracking is not crap, is WAY better than a C-40 type PWM controller.

Of course, the word "crap" as used here is entirely subjective. What I tend to call "crap" tracking is where the controller gets "stuck" all the time at the wrong Max Power Point voltage. In that case, you're most likely better off using a regular old PWM C-40 type controller.

boB

mshen11

Re: MPPT comparison

does anyone know the quality of outback's 80amp MPPT?

boB

Re: MPPT comparison

mshen11 wrote: »

does anyone know the quality of outback's 80amp MPPT?

I think 1/2 crazy can probably give you some insight. At this point, if you need an MPPT right now... today... You might want the XW60.

boB

Solar Guppy

Re: MPPT comparison

boB wrote: »

I think 1/2 crazy can probably give you some insight. At this point, if you need an MPPT right now... today... You might want the XW60.

boB

Wow, never thought I would see the day the creator of the MX60 make this recommendation . The XW is not a perfect controller either, but for what is currently on the market, its the best one can buy today and its dynamic Mppt tracking works.

The Flexmax 60/80, while based on boB's hard work, is in no way an improvement. I personally think its a shame what Outback did. With the FM series, Outback attempted to tack on dynamic Mppt software to hardware that really can't support this.

The result of this work is a controller that is the lowest of all the Mppt units harvest wise on the market from my testing, except one, the BZ500. This is no joke, the tracking on the FM series is that bad, or at least on the three units I tested, with the latest firmware as of April 2009. New firmware has been released ( reported by crewzer ) but I am busy with other work and frankly its not my job to be Outback's Quality Control.

Dave Angelini

Re: MPPT comparison

boB wrote: »

In general, just having an MPPT controller, as long as the tracking is not crap, is WAY better than a C-40 type PWM controller.

Of course, the word "crap" as used here is entirely subjective. What I tend to call "crap" tracking is where the controller gets "stuck" all the time at the wrong Max Power Point voltage. In that case, you're most likely better off using a regular old PWM C-40 type controller.

boB

Yes it is subjective because one of the great things about that old C-40 was that it was bullet proof on reliability and harvest means little when you have to drive up that mountain to fix it.

mshen11

Re: MPPT comparison

theres a difference in the outback 80amp and the xw 60amps... what is the difference between the two using a non MPPT as a baseline

BB.

Re: MPPT comparison

Realistically, the difference between an MPPT and a non-MPPT controller is probably an average of 10-15% increase in power collection for most people (and in warm weather, the differences may even be less).

The difference between a well designed and functioning non-MPPT and a poorly designed/implemented MPPT like the BZ -- not even worth addressing.

I very much like MPPT because they support higher voltage strings for the solar panel arrays... Makes it much easier and cheaper to run longer distances. It also better supports the extremes of temperature and voltage (hot panels, low output voltage; cold batteries, need higher voltage).

But if you have a smaller set of arrays (less than roughly 400 watts) and operate in a moderate climate with approprate Vmp rated panels (probably over 16 volts Vmp)--it may be difficult to justify the costs of a good MPPT (right now, the only good small MPPT is the MorningStar 15 amp 12/24 volt charge controller) vs any of the other PWM with long histories of solid operation.

-Bill

mshen11

Re: MPPT comparison

this might be better placed on the outback forum (because i may be [wrongly] getting a sense members here dont have too many good things to say): about the outback FM MPPT controllers... are they worth getting?