midnite not following float set point voltages

Started by australsolarier, September 13, 2017, 10:14:14 PM

Previous topic - Next topic

australsolarier

whilst watching the charging and other things on the status panel, i noticed two interesting facts:
the midnite charges with all possible power when in bulk. it then drops into float mode. generally  it keeps to the set float voltage. but...

-    when only light use, it slowly bleeds battery capacity. this happens in particular when i am at work. when i at home and the computer runs, etc,  there is no problem with that. (it's like the midnite thinks, wow, wow, he is not at home, we will just do a pretence of work and do as little as possible)

-    at the other hand, and i am talking about a 400ah lifepo4 battery and 10kwp solar panels with 3 midnite classics in follow mode, when in heavy use after float, say like 4kw or more, that battery capacity starts to bleed too, even though there is plenty of reserve solar power available. it will just slowly bleed down to under 380ah. (it is like the classics are thinking , wtf, we just worked so hard topping up the battery and now he wants us  heat up the shed. bugger that, will teach him a lesson and just passive aggressively  drain the battery. take the load off to say 1kw and the battery is slowly recharged up to 99% or 396ah again. (in float mode)

in the status panel the float voltages are down to 53.1V or so to the set 53.6v. so it is not high current  from the battery, it is not coming from there, well only a few amps.

it is possible the midnite reacts this way because i have a lifepo4 battery.

3 midnites are connected in follow me mode. they all 3 show the same symptoms.

whilst not doing much harm, i would prefer, the midnite would keep to the float set point. so, when the sun goes down the capacity  is at 99% and not at 96%.

it is almost like in California. you go to a shop and try to pay the advertised price, but oha,  there is a sales tax added. so the price tag is somewhat  irrelevant.

ClassicCrazy

That is how it works with lithiums because the voltage stays up so high after it goes to float if there is not much load on it. The Absorb stage is very short so looks like it just goes from Bulk to Float.

Midnite is developing a lithium module to go with the B-17 inverter charger system so maybe they will get a new firmware that works better with lithiums.

Larry

system 1
Classic 150 , 5s3p  Kyocera 135watt , 12s Soneil 2v 540amp lead crystal for 24v pack , Outback 3524 inverter
system 2
 5s 135w Kyocero , 3s3p 270w Kyocera  to Classic 150 ,   8s Kyocera 225w to Hawkes Bay Jakiper 48v 15kwh LiFePO4 , Outback VFX 3648 inverter
system 3
KID / Brat portable

australsolarier

yes i agree it might be good for us lithium battery users if midnite themselves start to work with them. it seems to be that north americans are somewhat reluctant using lithium batteries.

Vic

Hello austr... ,

IIRC,   you hate Solar charge Mode  --  I have noticed that, in Float,   at times,  Legacy P&O  appears to regulate battery voltage less well than does Solar Mode.

To a certain extent  it can probably be attributed to some voltage ripple on Vbat.   This effect seems to be greater when running large Opportunity Loads.

Know  that  generally,   LiPo  batteries are considered to be stiff current sources,   and also that the Classics and almost all inverters have fairly large capacitors on the battery terminal connections.

Also,   IIRC,   the Local App  displays readings about every four seconds,   which can sometimes make things appear worse,   or better than real-time voltage readings.

I would be quite timid in mentioning checking connections,   so will not do that ...

Also   would be timid in saying that voltage drops in the connections between the Classics and the main battery buss,   so will not mention that either  ...

FWIW,   Vic
Off Grid - Sys 1: 2ea SW+ 5548, Surrette 4KS25 1280 AH, 5.25 KW PV, Classic 150,WB, Beta Barcelona, Beta KID
Sys 2: SW+ 5548s, 4KS25s, 5.88 KW PV, 2 ea. Classic 150, WB, HB CC-needs remote Monitoring/Control, site=remote.
 MN Bkrs/Bxs/Combiners. Thanks MN for Great Products/Svc/Support&This Forum!!

australsolarier

vic,
your timidness is justified. in  bulk mode, even drawing 7.5kw the voltage drop is about 0.3V.

aha, yes, ripple voltage might explain that bleeding effect in high current situations,  but not so much on low current.

and yes, solar mode seems to have problems. it is not just that it cycles in a blue sky, but i also noticed whilst in bulk charge, solar mode decides after an hour or so it is only going to charge with something like 70% of the available power. (nowhere near absorption voltage)  now i have not tested this with the latest firmware upgrades. P&0 works for me. mostly.

i suspect the devil is in the software. those problems just might not be a problem with lead acid batteries. their voltage is much more elastic. (so the voltage drop is much bigger, float has much more room to kick in) the voltage of a lifepo4 is very stiff. there is not much give even in high current applications. in the low decimals.

australsolarier

ok, so i might have found what is causing that bleed in float mode under high use.

i want to say here right away, before half of the forum member jump on the bandwagon, all voltages have been measured with a fluke meter and adjusted. pv in and battery out.

midnite A has 2.4kwp connected
midnite B has 5.25kwp connected
(midnite C  has 0.75kwp connected and is irrelevant for this explanation)

all 3 classics are in follow mode and have the same firmware

so during that heavy use i noticed, that midnite A has full output, B however only 2/3 output.

now my float voltage for the lifepo4 battery  is 53.6V. that is when you measure the battery terminals.
midnite A's float voltage is set to 53.7V, but in reality it floats on 53.6V.
midnite B's float voltage was also set to 53.7,  but for some reason or other needs to be set to 53.8v to float on 53.6.

i repeat here, the pv and bat voltages were adjusted.

this discrepancy probably is not noticeable on a led acid battery. but it certainly was with my lifpo4 setup.

so i am happy altogether figuring out it was only such a small thing.

i will keep on watching. in case i am wrong, i will let you know.

dgd

Quote from: australsolarier on September 13, 2017, 10:14:14 PM

3 midnites are connected in follow me mode. they all 3 show the same symptoms.

Try unhooking the Follow-Me serial loop wires and let each Classic run independantly. they should reach Float at near the same time or within a minute or so.
When your load applies then each Classic can max its output to reach the float voltage.
I have found Float MPPT is a bit weird when charging and in Follow-Me it does not change all Classics to Float-MPPT mode and can leave some in Float

dgd
Classic 250, 150,  20 140w, 6 250w PVs, 2Kw turbine, MN ac Clipper, Epanel/MNdc, Trace SW3024E (1997), Century 1050Ah 24V FLA (1999). Arduino power monitoring and web server.  Off grid since 4/2000
West Auckland, New Zealand

australsolarier

dgd,
i will check about that float mppt mode tomorrow, whether it is synchronising or not.

another interesting thing i noticed, it takes a while to ramp up the production whilst applying a 3kw load in float. it takes about 2 minutes to ramp up to full compensation.

raising the float value one decimal in one midnite made the situation much better, though still a bit bleeding off capacity, so  1 to 2 ah. much better than before 15ah or more.
anyway, thanks for mentioning it, i will report back tomorrow.

australsolarier

dgd,
yes you are correct, about the float mppt mode. the classics just do whatever they want, just like the pupils do when the teacher leaves the class room. follow me mode obviously does not include float mppt mode. not that it matters that much.

i have previously described i raised the float voltage set point one decimal in one of the midnites. the midnites still bleed the battery under  a 4kw load ( i want to make this clear, there is plenty of solar there, force bulk and it will instantly charge the battery, also looking at the pv voltage you get a good idea whether they are working at the mppt point ) but not so much anymore.  it will bleed the battery at -4 amperes. reduce the load to 1kw, the charge slowly ramps up again. even though the float set points are set to 53.7 and 53.8 volts, all 3 midnites keep hoovering around 53.2 and 53.3 volts.

i also noticed when applying a large load, the midnites need about 2.5 minutes to ramp up to the required output. this might be partly explained because of the legacy mode. i might test this tomorrow and see how it works under solar mode.

i want to make it clear here, i do not think this is a big problem. the midnites work great and are reliable. it is just one of the things that does not make sense to me. i suppose i have too much time on my hands.

WizBandit

Most Lithium users set ABSORB very low, 3 minutes is as low as the Classic goes.  I suggest setting REBULK to a voltage that kicks the Classic out of FLOAT/FLOAT MPPT to deliver max power until the ABSORB V is reached again then in 3 min go back to float.

FLOAT and FLOAT MPPT will not make max power until about 8-10% of the top charge of the battery is gone, then it will bring it back up but will take hours compared to the BULK/ABSORB mode, which may take minutes.

ClassicCrazy

#10
Quote from: WizBandit on September 28, 2017, 11:37:22 AM
Most Lithium users set ABSORB very low, 3 minutes is as low as the Classic goes.  I suggest setting REBULK to a voltage that kicks the Classic out of FLOAT/FLOAT MPPT to deliver max power until the ABSORB V is reached again then in 3 min go back to float.

FLOAT and FLOAT MPPT will not make max power until about 8-10% of the top charge of the battery is gone, then it will bring it back up but will take hours compared to the BULK/ABSORB mode, which may take minutes.

Interesting - I hadn't thought about setting up my lithiums that way.
Will have to think about what you said and then maybe make some changes.
I bottom balance and use ending amps now but notice the top charge delay you mention.

Larry
system 1
Classic 150 , 5s3p  Kyocera 135watt , 12s Soneil 2v 540amp lead crystal for 24v pack , Outback 3524 inverter
system 2
 5s 135w Kyocero , 3s3p 270w Kyocera  to Classic 150 ,   8s Kyocera 225w to Hawkes Bay Jakiper 48v 15kwh LiFePO4 , Outback VFX 3648 inverter
system 3
KID / Brat portable

australsolarier

whizz bandit,

theoretically yes, practically definitively no. i do not want to keep the lifepo4 up at absorption voltage for longer than absolutely needed. practice has shown it shortens the lifespan.
and yes, in absorption mode the midnite classic keeps to the set absorption voltage, no matter what you throw at it, (if there is plenty of solar available)
and yes, sometimes i use "force bulk" to quickly get it back up, particularly before sunset.

on light use like 1 kw, it pretty much keeps to the set float point. at 3kw it starts to bleed. down to 384kwh (400ah battery). take the load back to  1kw it slowely returns to 396/398ah.

there is also that slow delay with ramping up the power when you turn on a 2kw load like a kettle. it needs about 2,5 minutes to fully compensate the load. and before that it draws a considerable current from the battery. (again in bulk mode it instantly ramps up without blinking an eye)

i think there is something programmed in the software for doing that.(there might be a reason for that, i just don't know)   it might be that lead acid batteries are not affected to that degree. because their voltage is much more elastic, so the classic has more room to keep the battery full. the lifepo4, just 2/10ths of V makes a largish difference in charge levels.
just a minor nuisance, i suppose with one of the updates maybe in the future  .........