Fronius MPP Hunting

13

Comments

  • mradtke
    mradtke Solar Expert Posts: 53 ✭✭✭✭
    Re: Fronius MPP Hunting

    Hi Solar Guppy,

    Thanks again for you comments.

    Of course IR measurement would be best, but I'll have a long wait until summer for the test. I think that I should trust the various string calculators but stay away from the boundary conditions. On the other hand, I was well within the Fronius specifications and look where it got me.

    Neither of us is very happy with PVPowered. Do you (or anyone else) have opinions of KACO or Power-One?

    Thanks,
    Mike
  • mradtke
    mradtke Solar Expert Posts: 53 ✭✭✭✭
    Re: Fronius MPP Hunting

    Hello,

    With the holidays, little has happened. However, yesterday all parties came to a tentative agreement to install a Power One PVI-5000-OUTD-US. This is a transformerless inverter so my installer is having to make a number of electrical modifications to my system in order to support it. I chose this inverter because it has dual MPP tracking inputs which matches my system requirements, is sized correctly, and will at least match the expected efficiency of the flawed Fronius.

    If anyone has reservations about using this inverter, or transformerless inverters in general, please reply to this post as soon as possible.

    Thanks,
    Mike
  • Solar Guppy
    Solar Guppy Solar Expert Posts: 1,989 ✭✭✭
    Re: Fronius MPP Hunting

    full power requires a listed a 200VDC minimum input, no better than the Xantrex inverter

    http://www.power-one.com/resources/products/appnote/AlternativeEnergy/safety_certification_pvi-5000-outd.pdf

    Section 8.5 discusses voltage and current derating ... Fig 27 in the link below showes the maximum current is decreased as the volts falls ... from a technical aspect, I would expect this. The UL minimum is listed as 200V, the manual showes 180V ... I would trust the UL listed value

    http://www.power-one.com/resources/products/appnote/AlternativeEnergy/manual_instruction_pvi-6000-outd.pdf

    As far as transformerless, you will be the first install I have heard of, only the most recent NEC2008 allows this , better check that your power company and your local building department both recognize and allow installs to NEC2008

    Your going with a manufacture that at a minimum is brand new to the US market, that's a risk in both support and unproven operation
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: Fronius MPP Hunting
    mradtke wrote: »
    If anyone has reservations about using this inverter, or transformerless inverters in general, please reply to this post as soon as possible.

    Without a transformer, your power output stage is sitting right on the AC feed. Any spikes or glitches will have a straight shot at the transistor junctions. If someone on your block generates "noise" with a backyard welder, or a hairdryer with nasty brushes, you may find how well noise rejection works in the inverter. Sure, the stuff can be filtered, but the transformer simplifies that filtering.

    I'm not even sure how they do it without a transformer, but I guess they have a way.
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • Solar Guppy
    Solar Guppy Solar Expert Posts: 1,989 ✭✭✭
    Re: Fronius MPP Hunting

    To do an inverter without a transformer, one uses inductors, still magnetic's but slightly better efficiency, and probably a SEPIC converter ( can boost or buck )

    Current multi-stage inverters are no different than transformer less units on the AC side, its a PWM H-Bridge driving the AC from a DC rail ( 430 VDC ), the transformer is used to step up the DC input side and the isolation is before the AC inverter stuff. So I don't see and difference in regards to Mikes comments
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: Fronius MPP Hunting

    Ah, SG, thanks for that. I just could not see how to not fry stuff w/out coils somewhere.
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: Fronius MPP Hunting

    Actually, I don't think there was anything specifically in UL1741 that kept a manufacturer from listing a non-transformer isolated inverter.
    Some may have "thought" they weren't supposed to for one reason or another. And there has always been requirements for things such as DC-Injection limits, etc.

    There was at least one utility that said it was fine to have a non-isolated grid tie inverter as long as you used one of their huge transformers between that inverter and their grid.
    There went all your efficiency gains down the toilet.

    boB :D
  • Solar Guppy
    Solar Guppy Solar Expert Posts: 1,989 ✭✭✭
    Re: Fronius MPP Hunting

    Hi boB,

    I was told in 2004 it was not allowed in the US by Xantrex as they were in the middle of the GT program and I specifically asked about the typology. UL1741 isn't free, so I can't scan to give the exact text/section for reference but I seem to remember it was the most recent UL/NEC changes that allowed what has been done in Europe for years

    This is a link to a white paper discussing the benefits of going transformer less and specifically calls out UL1741 and the galvanic isolation requirement. Also, GFD is much harder without the isolation as one can't just bond a PV leg to ground for sensing

    http://74.125.77.132/search?q=cache:OIScv7NQ4wYJ:www.elkraft.ntnu.no/eno/Papers2008/Schimpf-norpie08.pdf+UL1741+galvanic+isolation&hl=en&ct=clnk&cd=12&gl=us&client=firefox-a
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: Fronius MPP Hunting

    Thanks Guppy. I will look again for this.
    I don't remember looking for the word "galvanic" so maybe that word is/was in there. However, just because someone (even someone at Xantrex), said it isn't allowed, doesn't necessarily mean it isn't allowed. I've gone over UL1741 many times, but it IS a large document and I could certainly have missed it somehow.

    And, thanks for that link.

    boB :D
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: Fronius MPP Hunting

    "At the
    moment this is the case in countries where the safety standards
    for PV-systems do not yet cover transformerless topologies.
    One example is the UL-standard UL1741 [11] which is valid in
    the USA."

    This paper tends to agree with my findings... "Transformerless" PV inverters are not covered or mentioned in UL1741.

    I will still look again in the spec for galvanic.
    Thanks again for the link to that paper. I can't get enough of those it seems. (even though I have a bunch that are fairly worthless).

    boB
  • mradtke
    mradtke Solar Expert Posts: 53 ✭✭✭✭
    Re: Fronius MPP Hunting

    Hello,

    Thank you all for the replies. And thanks for the link to the interesting paper.

    I'll try to address the issues.

    Unknown company: My answer to that is that Fronius is well known and well respected and look where it got me.

    UL1741: The inverter is newly compliant although I would expect that as experience grows with this type of inverter, there may be some changes. The code complications mostly deal with having 240 VAC on the array and my installer has addressed them. Because the array is floating, when the disconnects are opened it is much safer to work on.

    200 volts: This CSA statement seems to point out what should be obvious. But since the CSA flagged it, it had me going for a while too. But, here is the math for the PVI-5000

    Maximum Utility Backfeed Current (AC) = 30 A @ 240 V, or 7200 peak AC watts
    Maximum Input Current (DC) = 2 x 18 A = 36 A

    Therefore (7200 peak watts)/( 36 peak A) = 200 minimum DC input volts.

    Using a similar calculation, 138 DC volts is necessary to run at the continuous 5000 watt rating. The MPP tracking is from 90-580 volts.

    Derating: As far as I know, the 200 volt issue was as stated above and applies to derating from the peak power, not the continuous power and is just a re-statement of the peak specifications. I believe that the 180 volts on the graph applies to continuous duty and is well below my array voltage.

    Thanks again for all of your comments.

    Mike
  • Solar Guppy
    Solar Guppy Solar Expert Posts: 1,989 ✭✭✭
    Re: Fronius MPP Hunting

    A 6kW inverter is rated for 6000 watts continuous output, UL doesn't test or list peak values, what the DC voltage deals with is the maximum input current. Like most GT units, they power limited on the DC power, not the AC power

    The maximum input current as the graph shows in figure 28 is the unit is DC current limited from 90 to 200 VDC ( notice the 90 and 200 have been added to the graph to meet the update listing changes )

    Whatever the DC power is ( each channel on the 6Kw unit ) is listed as 18 amps DC

    90VDC * 18 amps = 1620 watts DC
    200VDC * 18 amps = 3200 watts DC

    as a guess for 5KW unit ( 0.83 of the 6K )

    90VDC * 15 amps = 1350 watts DC
    200VDC * 15 amps = 3000 watts DC

    So as your hot panels drop, lets say to 180V , the maximum the inverter can harvest is 180V * 15 = 2700 watts per channel or 5400 watts DC ... If the inverter is efficient as it claims, then even at 180V it will be full power, maybe that's the difference between the 5k and 6K units, but certainly not 135VDC, that would be only 4Kw of DC power
  • mradtke
    mradtke Solar Expert Posts: 53 ✭✭✭✭
    Re: Fronius MPP Hunting

    Solar Guppy,

    First, I owe you an apology, because I noticed this and forgot to correct you. The manual that you are referencing is not correct. The correct one is:

    http://www.power-one.com/resources/products/appnote/AlternativeEnergy/pvi-6000-outd-us.pdf

    Please note that it is 100 pages long and includes the certificates.

    I agree that the calculation that the CSA did is not correct because it apparently mixed AC and DC and ignored the efficiency of the unit. My reference to 138 volts was built on that incorrect calculation, not to support it, but to show that even this incorrect calculation supported my application. I think that we are justified in ignoring the whole thing and using the graph in the manual for derating. You may note that the "correct" manual does not have a reference to the 200 volts outside of the CSA certificate.

    I appreciate your calculation for the 5000 watt maximum input current. It is not listed in the manual as it should be, but the included CSA certificate for the 5000 lists it as 18 A. But then, I have lost confidence in the CSA certificate.

    In the end, I think that you are saying that other than being "bleeding edge" technology, the unit is a good match for my arrays. Please take a look at the referenced new manual and let me know if you spot anything else.

    Thanks for your careful and thoughtful effort.

    Mike
  • Solar Guppy
    Solar Guppy Solar Expert Posts: 1,989 ✭✭✭
    Re: Fronius MPP Hunting

    All sorts of errors in that document

    Search for UL1741 and it gives 50hz and 230VAC for the parameters, which is wrong, the 5Kw unit has the same input current specs, which it shouldn't

    Its also CSA , not UL listed ... this is SUPPOSE to be the same I have been told CSA is no where near UL in checking things out ... the errors in this document and the listing signoffs seem to back this up

    I wouldn't use it ( Aurora unit ), I would ( and do ) use Xantrex GT's, 4+ years of proven performance and similar efficiency, but that's just one guys opinion.

    You might have to start you own testing service ...
  • mradtke
    mradtke Solar Expert Posts: 53 ✭✭✭✭
    Re: Fronius MPP Hunting

    Hi Solar Guppy,

    Thanks again for your thoughts.

    You are certainly correct about the errors, and as I have said before, the Xantrex GT's would be an excellent choice if they would work on my low voltage.

    Besides the P1 documentation and the CSA certificates, I looked at the CEC testing of the unit. I think that everything is OK, but not at all proven when compared to the Xantrex. That doesn't mean that it's a bad choice though. It means that it is not the safest choice.

    I think that I am about 1 week away from installation now. I'll report back as current begins to flow.

    Meanwhile, the well documented and popular Fronius IG 5000 continues to hunt down to 232 volts for the MPP and then resets ... over and over and over.

    Please stay tuned,
    Mike
  • mradtke
    mradtke Solar Expert Posts: 53 ✭✭✭✭
    Re: Fronius MPP Hunting

    Hello,

    No long after my last post, my installer contacted me and said that Fronius had contacted him with a fix for the problem that I have been whining about. I quit talking directly with Fronius some time ago, so this information is second hand. I was told that the fix was coming from Austria and would arrive in a few weeks. I was caught completely off guard by this information. My installer asked me to wait for and evaluate the fix, rather than going ahead with the Power-One inverter installation.

    It took me a few hours before it dawned on me that this might well just be another delay in getting to a final solution. I had no way of knowing whether the Fronius fix was real or not. The earlier Fronius behavior documented in this chain was enough evidence for me not to change my plans. So, I asked my installer to continue with the Power-One inverter installation.

    If I become aware that there really is a demonstrated fix, I will post that information here so that everyone may benefit.

    The Power-One inverter was installed yesterday. Everything went smoothly and I find it interesting, but perhaps not all that useful, to look at the energy being harvested by each of my two arrays separately. It will take a few days to know about how much more energy I am capturing with the inverter not oscillating, and I'll post again when I can better quantify that.

    The inverter looks to have good workmanship, even if its manual is not all that great. It is convection cooled and during our 75 degree afternoon, the heat sync is just barely warm. The Fronius was not noisy with its forced air cooling, but it could be heard most of the day. By afternoon this time of year the exhaust air was a lot warmer than the Power-One heat sync.

    Thanks for listening,
    Mike
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Fronius MPP Hunting

    heat is the enemy of electronic components and your saying it's cooler even without a fan is a very good sign it may outlast the other and be quieter too.
  • mike95490
    mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: Fronius MPP Hunting

    HEAT. That's a new clue.

    Perhaps the inverter was overheating, and throttling back, and as it cools, throttles up. Oscillation !
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Fronius MPP Hunting

    i don't think it was heating and cooling quite far and fast enough to have been the cause, but good thought.
  • mradtke
    mradtke Solar Expert Posts: 53 ✭✭✭✭
    Re: Fronius MPP Hunting

    Hello,

    Thanks for your comments.

    There is no evidence that heat had anything to do with the problem. I was just commenting that the Power-One seems to run a lot cooler and quieter without the fan noise of the Fronius. At some point a Fronius engineer talked about the performance hole being related to an MPP in the area of 230 volts with a high AC line voltage of around 250 volts.

    I haven't heard anything more from Fronius on their recently promised fix.

    The Power-One inverter seems to be working just fine. I took some temperature measurements of the array back side and calculated the MPP for that array. My calculations were within 1 volt of what the inverter had tracked.

    I also looked at my personal method of seeing whether the MPP tracking was working by looking at the changing voltages and currents. I think it was SolarGuppy that predicted that I would fail at this and he was right. The Fronius walked around a lot, and that's what I based my method on. The Power-One stays very stable with essentially constant voltage and current values unless a cloud passes by.

    I figured that the Fronius was oscillating about 1/2 the day this time of year. I also made a rough calculation of the losses while oscillating as about 10%. On the sunny day just before the new inverter was installed the Fronius was collecting 18 kWh. Under the same conditions, the Power-One is harvesting 20 kWh.

    From the above, I conclude that everything is working as it should and I am pleased with the replacement (upgrade?) so far.

    Thanks,
    Mike
  • autoxsteve
    autoxsteve Solar Expert Posts: 114 ✭✭✭✭
    Re: Fronius MPP Hunting

    glad you're getting output from your system. I find it disheartening that Fronius has not bellied up to the bar and altered their specs or fixed this.

    Power One - I believe that they are a Camarillo, CA HQ'd company....
  • mradtke
    mradtke Solar Expert Posts: 53 ✭✭✭✭
    Re: Fronius MPP Hunting

    Hello,

    The Power-One is silently inverting power just as it's supposed to.

    As was discussed in this chain, the Power-One is a transformerless inverter, so the arrays are floating. This makes them much safer to work on when the disconnect is open, but also much more sensitive to a ground fault than the type of inverter that is usually installed. Besides the GFCI protection that is always present, the inverter checks the array resistance to ground and leakage current to ground during startup. This has been 18 megohms an 0 milliamps consistently, which probably means off scale. I live in the desert, but there was a driving rain one day during the startup last week. That presented a good test of the array isolation. The resistance reading that day dropped to 6 megohms, and the current stayed at zero. My feeling is that those are pretty good readings for the conditions, so I can stop worrying about the isolation.

    I am very pleased with my choice of Power-One. I guess it's because of all of my earlier difficulties with the Fronius, but I just can't get over how this inverter just quietly sits there doing what it's supposed to do. I keep stopping to scroll through all of the information on it looking for some sign of trouble, but everything always looks great.

    No further word from Fronius.

    Thanks,
    Mike
  • mradtke
    mradtke Solar Expert Posts: 53 ✭✭✭✭
    Re: Fronius MPP Hunting

    Hello All,

    The Power One inverter is still working just fine. I look for trouble every day and can't find a thing to complain about. I am harvesting more than 25kWh per sunny day now which seems more than satisfactory for a 5kW system this time of year in Phoenix. Since I didn't have the Fronius last February, I can't do a direct comparison. I'll be sure to talk about that in June.

    No word from Fronius about their fix.

    Thanks,
    Mike
  • autoxsteve
    autoxsteve Solar Expert Posts: 114 ✭✭✭✭
    Re: Fronius MPP Hunting

    excellent you're getting good production from the replacement inverters.
  • mradtke
    mradtke Solar Expert Posts: 53 ✭✭✭✭
    Re: Fronius MPP Hunting

    Hello,

    The Power One inverter is still doing just fine and it is now harvesting more than 30 kWh per sunny day. According to my installer, the best harvest should be in May for my installation. I still can't do a direct comparison to the Fronius since it wasn't operational until June of last year. However, the best ever for the Fronius was 29 kWh in a day and it only did that once.

    My installer says that Fronius has still not provided the fix that they unexpectedly offered two months ago. He says that Fronius has reverted to an earlier claim that any fix would require re-certification before it could be used in a customer system. Fronius now says that when a fix is available, it will be only for beta testing. I have no further word about whether Fronius has changed their mind about keeping the fault secret. Please draw your own conclusions.

    Thanks,
    Mike
  • rsl360
    rsl360 Registered Users Posts: 14 ✭✭
    Re: Fronius MPP Hunting

    I also have a system with a Fronius inverter, the IG-5100. It also hunts and fails to find the MPP. This has been going on for a while. I contacted my installer and Fronius and everyone agreed that the system is hunting and Fronius agreed to replace the inverter. 11 days ago the inverter was replaced. It's still hunting. Tomorrow I will contact my installer and Fronius with the bad news, not sure what Fronius will do this time. This would appear to be a problem that Fronius does not want to fix, or is incapable of doing so.

    I held off on posting anything until the new inverter was installed and tested, hoping that I could say Fronius had the problem solved. Obviously they have not. Or, Fronius could issue a bulletin, stating that the inverter was inappropriate for certain installations, (assuming it can be defined), such as ones with high line voltage. (which seems to have been suggested as a possible problem.)

    More info as I get it.
  • solarvic
    solarvic Solar Expert Posts: 1,071 ✭✭✭✭
    Re: Fronius MPP Hunting

    I have 2 Fronius inverters and don,t have the problems you all seem to have. ! is an Ig-2000 and other is Ig 3000 Due to 2 different brand of solar panels. I am running them at a lot higher voltage than indicated here and I am thinking that might be your problem. Your voltages aren,t ideal because of your high temperature. Don,t know how hot it gets in Phoenix but Idid a string chart @ 110 f at highest temp and 32F at lowest temp and my results is :
    strings of 7 = 203 v
    strings of 8 =232 v
    strings of 9 = 261 v
    strings of 10 =290 v. According to string chart you can run 3 strings of 10 on that inverter. operating voltage is 150 to 450 v with a buffer to 500 v. So my theory is you should have either run 3 strings of 9 panels or bought 2 more so you could run 3 strings of 10. Or got 2 smaller inverters since your panels are aimed in 2 different directions. I am running a lot higher voltage at my location where the record low is -26f and the normal high never gets above 97f. With a total of 16 Kyocero KC- 158 g ans 12 Sharp 167 sharp ND 167 ndu3a I have benn getting 25 to 29 KW on sunny days in Western pa. which has plenty of cloudy weather.
    S:Dlarvic
  • Windsun
    Windsun Solar Expert Posts: 1,164 ✭✭
    Re: Fronius MPP Hunting
    rsl360 wrote: »
    I also have a system with a Fronius inverter, the IG-5100. It also hunts and fails to find the MPP. This has been going on for a while. I contacted my installer and Fronius and everyone agreed that the system is hunting and Fronius agreed to replace the inverter. 11 days ago the inverter was replaced. It's still hunting. Tomorrow I will contact my installer and Fronius with the bad news, not sure what Fronius will do this time. This would appear to be a problem that Fronius does not want to fix, or is incapable of doing so.

    I held off on posting anything until the new inverter was installed and tested, hoping that I could say Fronius had the problem solved. Obviously they have not. Or, Fronius could issue a bulletin, stating that the inverter was inappropriate for certain installations, (assuming it can be defined), such as ones with high line voltage. (which seems to have been suggested as a possible problem.)

    More info as I get it.

    Or it could be something wrong with your panel installation. Since this is your 2nd inverter, that seems more likely. If you have mismatched strings it will cause hunting in any inverter.
  • rsl360
    rsl360 Registered Users Posts: 14 ✭✭
    Re: Fronius MPP Hunting

    Hmm, maybe I should have started a new thread. I posted on this thread because I have the same problem as the original poster and wanted to continue the discussion of what seems to be a flaw in at least one model of Fronius inverter. I'd like to be able to say that Fronius fixed it, but that does not seem to be the case.


    To solarvic: I'm not in Phoenix, that's the original poster. And it any case, it's not a hot weather issue, the hunting occurs in very cold or moderate weather. Not sure about hot weather. The problem is not very obvious unless one is logging at a fairly rapid interval. Logging at a one minute interval will show the problem, but what shows up on a graph is actually an alias of the hunting. The hunting takes place on about a 45 second interval. Recording data at a two second interval shows this well.

    My System: Two strings of 12 Evergreen ES-A-205 panels (total 24 panels) and Fronius IG-5100 inverter.

    To Windsun: I suppose until the problem is solved, I won't know for sure what the problem is. As to it being mismatched panels- When I sent the data to Fronius, there response was that we needed to swap the inverter. While I'm pleased that Fronius has been responsive to the problem, I'm not sure why they would swap out the inverter if the problem is in the strings. Having the problem recur just makes Fronius look bad.

    What is your authority for "If you have mismatched strings it will cause hunting in any inverter." This is not obvious to me. Even if strings are mismatched, there still exists an MPP. Neither string will be at it's MPP, but the system overall will be, given the restraint of having to pick the same voltage for both strings.

    rsl
  • Windsun
    Windsun Solar Expert Posts: 1,164 ✭✭
    Re: Fronius MPP Hunting
    rsl360 wrote: »
    What is your authority for "If you have mismatched strings it will cause hunting in any inverter." This is not obvious to me. Even if strings are mismatched, there still exists an MPP. Neither string will be at it's MPP, but the system overall will be, given the restraint of having to pick the same voltage for both strings.

    rsl
    My "authority" is the simple fact that if you have two or more strings, you can have two different MPP's if the strings are not matched. In some cases the inverter will try to pick the best overall point, but some will settle on the highest peak.

    But I suspect that Guppy's analysis is probably more correct - panels can be anywhere from 5 to 40 degrees hotter than air temperature on a roof, so you may be getting too low.