Classic monitoring

zoneblue
zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
Hi Guys,

Thanks to some help from rossw, ive got some daily logging in place on the classic.
Attachment not found.

If anyone's interested I wrote a brief primer on how to do this here:
http://www.zoneblue.org/cms/page.php?view=getting-data-off-midnite-classic

A couple of comments/questions.

1. I've noticed that when the classic goes into float, theres often still 10-20 odd amps going into the battery. Ive checked this with a clampmeter. Could this be a sign that its ending absorb too soon? or that my voltage settings are too low? In the chart above, there was a fairly constant background load of 200W all day and an addtional 500W until 2pm. (i had a small heater on in order to get the controller to not float so early in the day, so i could try to test its full output). Post float you can see the fridge and water pump cycling, and whatnot. But if you take off the 700W after it goes into float you can see theres still a few hundred watts going in the battery even a couple hours after it floated. On a related subject a comment was recently on the midnite forum from someone puzzled over why the morning rest voltage was lower if the bank had floated later in the day. I have also noticed this, and the charts above show why. The bank doesnt reach a steady float current for at least 3hours of float. I figure the long floats this system sees is about the only kind of eq the agms will really get.

2. Theres a 5W discrepency between Vout*Iout <> Pout. Sometimes the calculated controller efficiency from these numbers is over 100%. Reason a, is i dont want to collect redundant data, and b, which numbers are the most accurate of all the classic variables?

3. Does anyone know anything about theoretical pv production? Taking a gander at this graph, ive plotted a model of theoretical pv output, based on algorithms found at www.pveducation.org.

Our array seems to be off the model mid morning. The air mass/refraction/haze side of it is a bit tricky. If only i can think of a way to load the system up all day, one real sunny day. Pet project there.
Attachment not found.

4. What was Midnite thinking that the battery sensor addon, would be an optional extra?!? The demand side of the data is noticeably and problematically absent.
1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
http://zoneblue.org/cms/page.php?view=off-grid-solar


Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Classic monitoring

    20 amps in "float"--Do you have some DC loads (AC inverter idling, etc.) that could be drawing during float?

    If purely battery charging--Absorb current should drop down to ~1% to even 0.1% of battery bank AH capacity (200 AH battery, 2 amps to 0.2 amps termination charge).

    If you see >~2% termination current--You might have a problem with the battery bank such as a shorted cell (or too high of absorb voltage).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: Classic monitoring

    A shorted cell would mean the current doesnt stablise, which this does after a few hours.

    I might be confusing the controller with these unusal loadings, but i dont understand the internal workings of the classic yet. Normally we have about 100W idle consumption, but for these tests i had increased that to 700W. Hence the controller may be being mislead.

    I have the classic in its default absorb settings, and they probably need tweaking. Absorb does seem rather short as a rule, although we are not deep cycling the bank. I got the impression that absorb ought to be around the two hour mark. From the graphs it seems only to be 40mins or so. (red curve steps up, sleep, bulk, absorb, float)

    As per a thread yesterday on the subject, there isnt really any bullet proof way to set the classic. Use ending amps with varying loads and youll undercharge, use time and youll overcharge. It seems the default is time as i understand it, based on the amount of time in bulk. Ill take another look at the settings and manual i guess.
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • inetdog
    inetdog Solar Expert Posts: 3,123 ✭✭✭✭
    Re: Classic monitoring
    zoneblue wrote: »
    As per a thread yesterday on the subject, there isnt really any bullet proof way to set the classic. Use ending amps with varying loads and youll undercharge, use time and youll overcharge. It seems the default is time as i understand it, based on the amount of time in bulk. Ill take another look at the settings and manual i guess.
    Or wait for a model that can measure just the battery current using an external shunt.
    SMA SB 3000, old BP panels.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Classic monitoring

    Which Midnite is on the verge of releasing (integrated battery monitor+shut)(we hope).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • SolaRevolution
    SolaRevolution Solar Expert Posts: 410 ✭✭
    Re: Classic monitoring
    BB. wrote: »
    Which Midnite is on the verge of releasing (integrated battery monitor+shut)(we hope).

    -Bill
    Yep. I just saw it at intersolar in San Francisco. They have the shunt board made. They are now working on the software. Once it is ready the classic may well be the best battery system monitor available for small off-grid systems. It will be able to be connected to a pc, to the web, or you can download stored data from the previous 300+ days.
  • mtdoc
    mtdoc Solar Expert Posts: 600 ✭✭
    Re: Classic monitoring
    Yep. I just saw it at intersolar in San Francisco. They have the shunt board made. They are now working on the software. Once it is ready the classic may well be the best battery system monitor available for small off-grid systems. It will be able to be connected to a pc, to the web, or you can download stored data from the previous 300+ days.

    Can it connect directly to a pc or does it go through the Classic?
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Classic monitoring
    mtdoc wrote: »
    Can it connect directly to a pc or does it go through the Classic?

    You'd have to ask boB to be sure, but I believe it is designed to operate with the Classic specifically, not a stand-alone battery monitor. Of course they may have changed it. They often change things, and always for the better. :D
  • SolaRevolution
    SolaRevolution Solar Expert Posts: 410 ✭✭
    Re: Classic monitoring
    mtdoc wrote: »
    Can it connect directly to a pc or does it go through the Classic?

    It is designed to work exclusively with the classic. There is a single wire running from the shunt board to the classic. The classic will read data from the shunt board and will use the existing negative battery connection as a "common" for the data signal.
  • mtdoc
    mtdoc Solar Expert Posts: 600 ✭✭
    Re: Classic monitoring
    You'd have to ask boB to be sure, but I believe it is designed to operate with the Classic specifically, not a stand-alone battery monitor. Of course they may have changed it. They often change things, and always for the better. :D

    I posted that question on the Midnite forum but never got an answer - though it was during a time when boB wasn't posting much -probably on the road?
  • mtdoc
    mtdoc Solar Expert Posts: 600 ✭✭
    Re: Classic monitoring
    It is designed to work exclusively with the classic. There is a single wire running from the shunt board to the classic. The classic will read data from the shunt board and will use the existing negative battery connection as a "common" for the data signal.

    OK thanks. So to get the info to the PC or monitor on the Web I suppose it will have to work through the Local App or MyMidnite?
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: Classic monitoring
    mtdoc wrote: »
    I posted that question on the Midnite forum but never got an answer - though it was during a time when boB wasn't posting much -probably on the road?

    Yes, sorry, I have been on the road and busy and haven't been posting a whole lot... In fact, I got to go to Intersolar in Munich, Germany and
    spend a few extra days looking around town. BTW, if anybody here gets a chance to go there, make sure you go to the Deutches Museum
    (science museum) and give yourself at least one completely full day there, or two days might be better. Fantastic stuff there !

    Yeah, this monitor is the single wire one that connects to the Classic. It's also a bit delayed but shouldn't be a whole lot longer I hope.

    Also, on the metering issue. the input current is not as calibrated as the output (battery) current on its offset so trust the battery
    current more than the input. I believe the input may show a bit low so that may be why an efficiency of 100+ % was measured
    using that. Efficiency is hard enough with good metering !

    boB
  • stephendv
    stephendv Solar Expert Posts: 1,571 ✭✭
    Re: Classic monitoring

    zoneblue, nice bit of work with the monitoring software! And apologies for not getting back to your PM, I dismissed it while busy then it completely slipped my mind to reply. Anyway, looks like you got on just fine :)
  • Vic
    Vic Solar Expert Posts: 3,208 ✭✭✭✭
    Re: Classic monitoring
    zoneblue wrote: »

    I might be confusing the controller with these unusal loadings, but i dont understand the internal workings of the classic yet. Normally we have about 100W idle consumption, but for these tests i had increased that to 700W. Hence the controller may be being mislead.

    I have the classic in its default absorb settings, and they probably need tweaking. Absorb does seem rather short as a rule, although we are not deep cycling the bank. I got the impression that absorb ought to be around the two hour mark. From the graphs it seems only to be 40mins or so. (red curve steps up, sleep, bulk, absorb, float)

    Hi zone, As long as there is enough RE power available into the CC for it to maintain the temp compensated (if you use it) voltage in Absorb, Float (and EQ), loads should not make a difference to it. A large surge, like a pump starting, etc, may cause a slight blip in the CC's delivered voltage to the batts, but other than that, the CC should not care about your loads on the battery/inverter.

    You DO want to set the correct Absorption Float and any EQ voltages appropriate for YOUR battery bank.

    The required Absorption time is directly dependent on the Depth Of Discharge that the battery has experienced. And, as you note, variable DOD from one day to the next often means that the battery is not correctly recharged using Absorb time settings. EA can work well, it the loads on inverter/s or the battery are consistent around the time that the Absorb stage should end, or if variable loads DO cycle off around the appropriate Absorb end time, such that the CC can see a more accurate battery charge current.

    Having sealed batteries can make knowing the correct time or EA values a bit more difficult to determine, but still quite possible.

    With the Flooded banks here, one can see that if the Abs stage is shorter than the battery needed, the Float current into the battery is increased over what would normally be seen with a correct Absorb duration. This higher current does diminish over time. There is also some battery temperature effect on this current value.

    We do use only EA settings on the banks here to end Absorption, and it works well. In general, batteries are good integrators, and if a refrigerator or A/C happens to be running at the exact time that Abs should end, and therefore extends the Abs stage a bit, this all averages out. And if this effect tends to cause a bit of overcharge of the battery, measurements of SOC will show this, and the EA value can be changed.

    As noted, MidNite is feverishly working on the Shunt-based Absorb terminator, which measures actual current delivered to the batteries. This will simplify setting EA to the correct value and will be independent of system loads. Very Cool!!
    Many, many of us are waiting for this product, with varying degrees of patience. I cannot wait, personally!

    zone, nice job on extracting the Classic data and plotting it.

    And, Hi SolarRevolution - Alex. Nice to see you back here, have missed your useful posts based on your in-depth experience.
    Have Fun all, Vic
    Off Grid - Two systems -- 4 SW+ 5548 Inverters, Surrette 4KS25 1280 AH X2@48V, 11.1 KW STC PV, 4X MidNite Classic 150 w/ WBjrs, Beta KID on S-530s, MX-60s, MN Bkrs/Boxes.  25 KVA Polyphase Kubota diesel,  Honda Eu6500isa,  Eu3000is-es, Eu2000,  Eu1000 gensets.  Thanks Wind-Sun for this great Forum.
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: Classic monitoring

    Looking again at the numbers i get this:

    Attachment not found.

    Which looks ok right? The only thing that pops out is to reduce the temp compensation to -0.004V/C/Cell.
    The slightly higher float voltage to my way of thinking makes up for the fact that its only floating a few hours a day as compared to full 'standby' charge.

    Now for absorb termination, its set to 2.0 hrs, and ending amps 0.1. Those are the defaults, but what does it mean? The latest classic manual (10-0 0 1-1 R E V :C) says:

    p31
    This [absorb] stage is terminated at the end of the Absorb time or the End Amps set point whichever is reached
    first. ... The absorb time is proportional to the bulk time.

    This isnt clear, given that its the manuals only reference to ending amps.
    It seems there are three factors that end absorb. Which ever occurs first.
    - ending amps, 0.1A, really?
    - 2 hours, doesnt seem to get a shot at this either
    - proportional to bulk time. As I recall seeing an absorb up down counter in the modbus registers, this appears to be king maker.

    When you guys say use ending amps, how exactly do you set that?


    The only other vaguely relevant paras are:

    p32
    Absorb, Equalize and Float voltages are fully adjustable. You will need to get the actual voltages from the
    battery manufacturer.

    p33
    To calibrate the Classic battery and PV voltage reading, you will need a Volt meter to check the actual
    battery bank voltage or input PV voltage. ... compare this reading to the reading on the Status screen
    the Classic,(press Status if not in this screen) to adjust the reading of the Classic to the one in the volt
    meter follow these steps...

    Ok i should do this.

    p61
    You can use the Local Application software that was included with the Lite to program it as well.... Please see the
    Local App instructions for more details.

    The local app manual gives no details about the individual config screens.

    The only thing i can see is to increase the absorb voltage. Given these are Agms id do that hesitantly.

    Any thoughts?
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Classic monitoring

    do check from the battery manufacturer what the temp compensation should be at. my sunxtender agms are 4mv and fla types are usually 5mv.
  • Vic
    Vic Solar Expert Posts: 3,208 ✭✭✭✭
    Re: Classic monitoring

    zone,

    YES, looks like the battery mfg is saying 4 Mv/C is correct for Cyclic use.

    And, 0.1 A for EA will almost certainly NEVER HAPPEN, as there is always current jitter, especially at very low values., such that while this EA value may be seen by the CC, the EA value will probably not remain at this low current long enough for the CC to switch to Float, some guessing here.

    Here, EA works on Classics by setting a reasonable Min Abs time, a reasonable Max Abs time, and a REASONABLE EA value - perhaps 0.5 A or so for you may be in the correct range. You can experiment to try to find the lowest EA value that appears to allow EA to end Absorb. Some AGMs have very, very, very low EA recommendations.from the manufacturers, which are unreasonable for almost any CC to satisfy, in my opinion.

    Have never run AGMs, and the banks here are large enough that the EA values are about 15 A. We do see a small bit of jitter. Perhaps on your AGMs the jitter at very low currents is lower than what is observed here ... dunno.
    More later, Good Luck, Vic
    Off Grid - Two systems -- 4 SW+ 5548 Inverters, Surrette 4KS25 1280 AH X2@48V, 11.1 KW STC PV, 4X MidNite Classic 150 w/ WBjrs, Beta KID on S-530s, MX-60s, MN Bkrs/Boxes.  25 KVA Polyphase Kubota diesel,  Honda Eu6500isa,  Eu3000is-es, Eu2000,  Eu1000 gensets.  Thanks Wind-Sun for this great Forum.
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: Classic monitoring

    Thanks alot vic and others for your thoughts.
    Vic wrote: »
    And, 0.1 A for EA will almost certainly NEVER HAPPEN, as there is always current jitter, especially at very low values., such that while this EA value may be seen by the CC, the EA value will probably not remain at this low current long enough for the CC to switch to Float, some guessing here.

    Indeed. Possibly 0.1 is midnites way of disabling end amps.
    Vic wrote: »
    Here, EA works on Classics by setting a reasonable Min Abs time, a reasonable Max Abs time, and a REASONABLE EA value - perhaps 0.5 A or so for you may be in the correct range.

    Thing is the latest version of the local app doenst have a min time and max time, only 'time':
    Attachment not found.

    I have a feeling ive seen min time in a previous version couldnt be sure. I also got the sense that this all was a product of a firmware change last year when the 'proportional to bulk time' was added. Correct me if wrong.
    app.jpg 66.4K
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • Vic
    Vic Solar Expert Posts: 3,208 ✭✭✭✭
    Re: Classic monitoring

    Hi zoneblue,

    Thanks for adding a signature ... Classic Lite, OH !! OK. Have not been using the Local App. So, did not know that there was NO Max and Min time settings.

    And about 18 or so months ago, boB did mention something about EA of zero might not be the way to disable the EA function, he may have said that 0.1 A was the correct value for disabling EA. Ever since the EA code was placed into the Classic code, have used EA, so have never wanted to disable it ... Some of the basic logic of the Classic appears to be based somewhat on the OB MX-60 CC, which, as you know was (basically) designed by some of the same folks that Founded MidNite. The MX-60 needs EA to be set to 00 Amps to disable it. It may have been a leap on my part to believe that this would be the case for the Classic.

    And, all of the previous statements about watching the Battery current jitter may not apply to the Classic Lite, as if you do not have a MNGP to use with the Lite, then you need to use the App, which might not have a fast enough update of things like output current to really be able to see the jitter ... Do not know.

    Absorb time equals Bulk time might not be a horrible way to end absorb in your situation. BUT, here, it would work poorly, as Bulk begins just after Sunrise, but at a very S L O W rate due to the sun rising behind the PVs, which produce little current for a few hours, and if the battery was not cycled moderately deeply, the required Abs is not nearly as long as the Bulk duration, and so on.

    Have little more to offer. Sorry that I did not realize that you were running a Classic Lite. Good Luck, Vic
    Off Grid - Two systems -- 4 SW+ 5548 Inverters, Surrette 4KS25 1280 AH X2@48V, 11.1 KW STC PV, 4X MidNite Classic 150 w/ WBjrs, Beta KID on S-530s, MX-60s, MN Bkrs/Boxes.  25 KVA Polyphase Kubota diesel,  Honda Eu6500isa,  Eu3000is-es, Eu2000,  Eu1000 gensets.  Thanks Wind-Sun for this great Forum.
  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: Classic monitoring
    Vic wrote: »
    Hi zoneblue,

    Thanks for adding a signature ... Classic Lite, OH !! OK. Have not been using the Local App. So, did not know that there was NO Max and Min time settings.

    And about 18 or so months ago, boB did mention something about EA of zero might not be the way to disable the EA function, he may have said that 0.1 A was the correct value for disabling EA.


    In the most recent firmware, 0.0 amps is the proper way to disable EA, not 0.1 which is what it used to be...

    The reason it was changed to the proper 0.0 amps to disable it is that if there are more than one charger in the system
    keeping the battery at absorb voltage, one of them may very well become 0.0 amps (less than 0.1) and would
    prematurely end the Absorb cycle. That was the wrong thing to do, so, 0.0 amps it is for disabled.

    I had forgotten that the MX60 EA was 0.0 amps to disable it.

    Thanks !
    boB
  • Vic
    Vic Solar Expert Posts: 3,208 ✭✭✭✭
    Re: Classic monitoring

    Hi boB,

    Thanks for the added info on the Classic. Things really seem to get complex with all of the interactions of the other charge sources and new/emerging products as well.


    Thanks!, Vic K6IC
    Off Grid - Two systems -- 4 SW+ 5548 Inverters, Surrette 4KS25 1280 AH X2@48V, 11.1 KW STC PV, 4X MidNite Classic 150 w/ WBjrs, Beta KID on S-530s, MX-60s, MN Bkrs/Boxes.  25 KVA Polyphase Kubota diesel,  Honda Eu6500isa,  Eu3000is-es, Eu2000,  Eu1000 gensets.  Thanks Wind-Sun for this great Forum.
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: Classic monitoring

    So the upshot is with the lite you cant use ending amps?
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • boB
    boB Solar Expert Posts: 1,030 ✭✭✭✭
    Re: Classic monitoring
    zoneblue wrote: »
    So the upshot is with the lite you cant use ending amps?

    Sure. With the Local App you can control the EA setting.

    boB
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: Classic monitoring

    Bob,

    Sure there is a end amp setting, but how to make the classic use it? Without a min absorb time field, the bulk counter rule seems to be the determining factor that ends absorb. Im sure that im probably missing something here.

    How do the the three rules interact?

    And a related question is: does the latest firmware for the graphics display panel still have a setting for min absorb time?
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Classic monitoring

    The Absorb time is a maximum. If End Amps occurs before the time runs out, that stops Absorb.

    Usual process:
    Absorb Minimum Time (if equipped) ensures Absorb Voltage is held for at least that long.
    Absorb Maximum Time stops Absorb if End Amps has not been reach or Absorb Time Clock has not been "run out".
    End Amps stops Absorb if reach (after minimum and before maximum or Time Out).
    Absorb Time Clock ends Absorb if time = zero and maximum time or End Amps has not been reached.

    This can be written as an IF/THEN/ELSE statement, but I'm too lazy. :p
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Re: Classic monitoring

    Hi Guys,

    Pondering this, I think my issue is that my local app is newer than my firmware. I just googled varimax and got this informative video:
    http://www.youtube.com/watch?v=XldKMPIRLlE

    It says varimax is present in firmware 1170 and onwards. Im gona go upgrade my firmware first. Should make more sense after that.
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar