I have my Classic 150 on a 48 volt system set to absorb at 58.9 volts and the voltage limit set at 60.4 volts. The ambient temperature where the batteries are located is 45 degrees F and the temp compensation is set at -2.5 mv. Typically the batteries absorb around 59.6 give or take a few tenths. This has worked well for several months but yesterday when monitoring remotely I saw 60.0 volts then 60.4 volts. This was not normal behavior but I didn't pay that much attention. Today it was working normally for most of the day absorbing between 59.6 and 59.9 and then I saw 60.4, 60.8, 61.2 and higher. When I got home I looked at the graph on the Classic and the high reading on the graph was 61.8. It was above the 60.4 volt limit for approx. 30 minutes. Given that I have agm batteries I can not afford to charge the batteries outside of their recommended charging parameters so this has me very concerned. Especially when you see it happening remotely and there is nothing that can be done about it. I don't have the newest firmware that was recently released but I have the previous version and I don't think this behavior has anything to do with firmware because I've been running that firmware for many many months and I've never seen this before.
Other than setting the voltage limit which doesn't seem to be working what can I do to prevent this from happening again?
WXboy, what is the service history of those batteries? Age? Max DoD achieved? How often do you achieve Float?
Westbranch, the batteries are healthy.
It just dawned on me what caused this and I went down to check the Classic graph and my suspicion was correct. A year or two ago I saw the Classic do the same thing on my 12 volt system and I think I have an old post about it.
Every once in a while my Classic's power output gets out of calibration by 80 to 100 watts(I also have a post about this). What happened today, and it's the same thing that happened a while back, is the calibration went off and the Classic was in absorb but it was registering that it was outputting zero watts of power. What that meant is that the batteries were taking less than 100 watts of power but the Classic was registering zero. For whatever reason when the Classic is in absorb and it thinks it's putting out zero watts it causes the voltage to go over it's voltage limit. I do not have another charging source connected to these batteries.
The reason I know the Classic gets out of whack is because my Trimetric and Classic are usually relatively close but when the calibration goes off the Trimetric will show between 80 and 100 watts more going to the batteries than the Classic displays. Normally not a big deal until it causes a an over voltage situation. The calibration gets corrected when the Classic goes into resting and wakes back up but when the voltage is high it doesn't rest and it doesn't get corrected. Luckily we ran out of daylight today otherwise the condition could have gone on for more than 30 minutes.
I think it's time for a call to Midnite.
Hi Wx ...,
Do not run AGMs here, but it has been reported in a number of places, that when AGMs are nearly fully charged, that there appears to be a Surface Charge effect, which causes the battery terminal voltage to rise. This is especially noticeable with MPPT CCs when they perform a Sweep, and the loads from an inverter or other loads are low.
At times, when AGMs are essentially fully charged, it has been noted that AGM batteries display an apparent over-voltage condition during these Sweeps. A battery that is fully charged or nearly so, represents a high impedance. When the MPPT CC sweeps, there needs a place for the current that the CC delivers to the batteries/loads to go during Sweeps. These Sweeps can cause Vbat to rise above Set-points.
Folks smarter than I have noted that this situation should not be detrimental to the AGM batteries. I do not know enough to know all of the ins and outs of this situation.
Of course, this may not be the exact situation with your system.
FWIW, Vic
It may overshoot for a moment but should not see that for 30 minutes.
If you saw it in Absorb for very long at 000 watts or 0.0 amps, that USUALLY means that there is another
charging source...... But you don't have another charging source...
I'm not sure about what Vic was saying about the voltage going up by itself but maybe ?
Maybe some kind of temperature variation thing ?
So you have your LIMITS MAX Voltage set to 60.4 volts ? 61.8 should not occur unless it
is ~maybe~ for a quick moment IF there may have been a load (and Absorb was being held),
and THEN that load turned OFF. But it would not stay at this high voltage.
The fact that you showed 000 watts means, even if the current sense if off slightly, that the
current to keep it at Absorb is VERY low.... I do not see how a Classic that has been
working fine for quite a while can mess up like this... I have not seen a problem with
the MAX voltage limit not working correctly unless the Classic is not quite calibrated
correctly and the reading is coming from another meter. If you ARE reading it on
another meter and not the Classic's meter, it may be that you need to calibrate the
voltage reading. If so, what I would do would be to set the TWEAKS voltage offset to
say, +0.4 volts and then adjust the trim-pot battery voltage adjustment to read
the same as your Fluke meter or Trimetric.... Assuming the Trimetric is correct to
the tenth of a volt, and it most likely is.
But, if you really needed your Classic to be fixed/swapped/looked at, then maybe it's time ?
boB
Hi boB,
I claim to nothing specifically about these reports, but, for a number of years this has been a not infrequent Topic on other Forums. AGM batteries seem to be the most-reported battery type, but there have been some reports of this when Gel batteries are used:
Please plug this Search Term into Bing:
agm battery charge voltage overshoot site=outback power forum
There are many Threads on the Outback Forum on this behavior.
It seems difficult to post a direct Link to (at least for older Threads, using a Search Engine these days).
The Wind-Sun Forum also has had a number of discussions on this this as well.
FWIW, Vic
Quote from: Wxboy on February 17, 2015, 09:18:35 PM
the batteries are healthy.
that is an assumption, based on ?
I did some guessing and I think you have 3 - 255W pv's... for 765W
and 765W / 48v = ~15.94A max so that is about a 15.94A/ 200A =~ 7.9% charge rate or
at 60.4V, (765W/60.4V ) +~ 12.64A=~6.3% charge rate.
Therefore depending on the DoD you appear to have a low charge rate,
so what makes you believe your batteries are healthy?
Do you have a WBjr, or similar meter, to verify the actual amount of charge your bank gets?
Does the BTS (Battery Temperature Sensor) compensation cause this ? I ask, because in really cold weather, the chargers really boost up the voltage a bit (on a 48V bank) and that surprised me. I had to tweak things back a tad to keep the inverter out of OverVolt shutdown.
The meters/settings are not always plain if it's actual voltage or TempComp voltage.
But reading, maybe it's an AGM thing, or something totally different.
WX ..,
You did start a Thread on this behavior about 23 months ago - go to the top of this:
http://midnitesolar.com/Forum/index.php?topic=1120.msg8482#msg8482
In that you Thread, you seemed quite adamant that this behavior was NOT Surface Charge effect. In my opinion, it probably is surface charge. When all loads are removed from the battery by turning the inverter off, and the voltage rises a bit, this is exactly what surface charge looks like. Your configuration was different at the time of this other Thread, I realize.
In VRLA batteries, electrolyte migrates away from the plates slowly. This often means that the effects of surface charge are visible for quite some time, IMO.
Not to try to put too-fine a point on this. But, there are many, many Threads of other Forums about very similar behavior with AGMs, and Gels.
FWIW, Vic
Quote from: Westbranch on February 18, 2015, 11:27:23 PM
Quote from: Wxboy on February 17, 2015, 09:18:35 PM
the batteries are healthy.
that is an assumption, based on ?
I did some guessing and I think you have 3 - 255W pv's... for 765W
and 765W / 48v = ~15.94A max so that is about a 15.94A/ 200A =~ 7.9% charge rate or
at 60.4V, (765W/60.4V ) +~ 12.64A=~6.3% charge rate.
Therefore depending on the DoD you appear to have a low charge rate,
so what makes you believe your batteries are healthy?
Do you have a WBjr, or similar meter, to verify the actual amount of charge your bank gets?
Westbranch, the batteries are about 2 years old, they get cycled an average of 15-25%, and they've never been lower than 55-60% DOD. They never sit in a discharged state for more than 24 hours and they get a full charge at least twice a week. The batteries are healthy. My older smaller bank is 7 years old, they are the same type of batteries, and even those are doing well for their age.
As for the other ideas I value all insight but it isn't surface charge in my case. Other people may experience this but watching my batteries every day for years I can say they don't behave this way. As soon as the charging current stops the voltage drops quickly. If the charge controller was resting it would draw about 5 watts and my inverter draws about 7 watts while in standby. With a combined 12 watt draw my batteries would never be above 60.4 volts for 30 minutes. I have a Fluke meter, a Trimetric, and a Mate 3 all to verify voltage. The Mate 3 is what first tipped me off about the high voltage when I was monitoring the system remotely. The charger on my Radian was off as it is every day. I only use the charger to top off the batteries if there is a storm coming that could knock out power. Also the Radian was in Grid Zero mode which is a mode that requires the charger to be turned off to operate.
Maybe I'll get lucky and it will happen on a weekend when I can take a video and show with a clamp meter what's happening. I will be the first to say that this is a rare event(3 times that I know of in 3 years, although 2 days in a row) so chances are I won't be lucky enough to get it on video. I suspect the Classic was putting out between 1.1 and 1.5 amps when this occurred because I have the whiz bang jr. set to trigger float at 1.0 amp.
I don't claim to be a solar expert but this was not normal for my system.
Bob, I am about ready for a swap out if you guys are willing. I'm working with Roy on this situation and I sent him a few pictures.
Quote from: Wxboy on February 19, 2015, 09:57:04 PM
Bob, I am about ready for a swap out if you guys are willing. I'm working with Roy on this situation and I sent him a few pictures.
That is good. Roy is a pretty sharp guy and we're very glad to have
him and the other great support people here at MidNite !
boB
Another event today. Voltage limit set at 60.0 volts. Classic registered 60.8 volts at one point on the MNGP display and 60.5 at the next interval. I knew to check it today because I saw the high voltage while monitoring Optics RE which shows the Mate 3 voltage. This was a short lived event because shading hit the panels about 10 minutes after the event started.
Same exact conditions as last time. No loads on the system, batteries almost at 100%, Classic showing Zero watts out(real watts out would have been in the 60-80 range), voltage overshoots the set limit by an unreasonable amount. No other charging source.
Wx, Just looking at the code... It looks like it should be limiting OK. Have you seen the battery voltage
stay more than a couple of tenths of a volt above the maximum voltage set point for any length of time ?
When the battery voltage was above the maximum voltage set point, have you gone into the T-Comp--VIEW
menu and looked at the temperature compensated target voltage at that time ??? If you catch the
voltage above the set MAX T-comp'd set point, I am very curious as to what that value is at that time.
That value may give a clue as to what is happening.
boB, I've seen it go more than a few tenths above the set point for about 30 minutes on one occasion. I believe it would have gone on longer but I usually run out of sun this time of year when the batteries are getting topped off.
I haven't looked at the temp. compensated target voltage while this is happening because I'm typically at work but I've checked it during normal conditions and it will be in the 59.8 range. I believe this problem is linked to the fact that the Classic thinks it is outputting zero watts and maybe because of that reason it doesn't try to scale back it's power output. If it was showing that it was putting out even 20 watts I'm confident that it would not overshoot the voltage. Every time I've seen this happen the Classic was registering zero watts out. It sees that it's still at or above the absorb voltage so it doesn't go into resting. If it went into resting and woke back up it would reset it's calibration and when it woke up it would see it was outputting power and I believe it would correct the problem. Of course at the heart of all this is the problem where my Classic's power is sometimes spot on and sometimes it reads low by approx. 75 watts. I would prefer to have that corrected and that would also solve the problem.
Thanks boB.
WX, are you running in SOLAR mode ? If so, there is a new revision coming up that recalibrates the current sense extremely well.
This is done without any extra resting involved which is good.
boB
Yes, I am in Solar mode. Looking forward to the update. Thanks.
The Classic went over voltage again yesterday under the same conditions. No load, Classic showing Zero watts out but clamp meter was showing about 1.4 amps out. Voltage limit 60.0 volts. Temp compensated target 59.8 volts. I saw the voltage go as high as 61.2 on the Classic while I was watching it. I was losing daylight as it was happening but every time there was enough light the voltage was going over the limit. Once I saw it hit 61.2 volts I had to shut down the PV breaker to allow the Classic to reset itself.
Bob, I will email the video to Roy. The video is dark but it should show what you need to see.
I can't afford to risk losing my batteries to this situation so hopefully we can work something out .
Hmmm, reading all this is making me nervous to purchase a classic if it goes over its set point, as I will be running AGM's. Its very simple to make a remote disconnect with a wifi controller box from china ebay for $60-100 and integrate it a contactor to shut your pv array off or classic etc. 8 channels of remote on off could come in handy.
As long as the Classic is reading close to the real battery voltage, this must have something to do with a setting or temperature comp or
something. The software doesn't just over-voltage batteries without something making it do that...
Now, it ~could~ be a defective Classic I suppose. Now that it's about 3 weeks after these posts, maybe the Classic
has been RMA'd and fixed and I haven't heard what was found yet.
Hopefully I hear what happened.
boB
boB, the Classic has not been RMA'd yet. Roy wanted me to try updating to the latest firmware before giving up on the Classic. I tried that yesterday and it bricked the unit. I had to go back to the previous version to bring it back to life. The Classic has demonstrated this charging behavior going back several versions though.
I made a video and sent it to Roy showing that all settings are correct including the temperature compensated target and the voltage limit and that the Classic was outputting current when it was showing zero output. That appears to be the issue. The Classic doesn't stop outputting current because it thinks there is another charging source and it's just along for the ride. It doesn't limit it's output because it shows Zero output.
Apparently I'm the only one that's seen this problem or maybe other people haven't seen it because it takes a very specific set of circumstances for it to happen. The Classic's output power accuracy/reading has to drift off so that it's outputting more than what it's displaying otherwise the problem won't occur. And it has to drift more than the amount of power that the batteries require to absorb. Since agm batteries absorb down to a low end amps that makes the issue more likely with agm's or with a very small battery bank.
Quote from: Wxboy on May 17, 2015, 09:06:17 AM
I made a video and sent it to Roy showing that all settings are correct including the temperature compensated target and the voltage limit and that the Classic was outputting current when it was showing zero output. That appears to be the issue. The Classic doesn't stop outputting current because it thinks there is another charging source and it's just along for the ride. It doesn't limit it's output because it shows Zero output.
Apparently I'm the only one that's seen this problem or maybe other people haven't seen it because it takes a very specific set of circumstances for it to happen.
Or you are the only one to notice it...
When you see zero output current and the voltage is high, have you looked at the "secret" screen from main status ?
i.e. Hold down the left-arrow key and then tap the ENTER key. Look at the two middle (ish) numbers to the right of
the + and the - signs. These are the RAW battery output currents. Near zero amps, they should show somewhere
around 40 or so. I am wondering if the current sense has maybe drifted since last turning on from Resting.
What mode are you in ? SOLAR ? Either way, during this time when you see high voltage and zero amps,
have you tried turning the MODE to OFF and back on again ? Don't need to turn the power off.
New firmware coming up makes the current input and output much better in SOLAR and U-Set modes.
Also networking.
boB
Quote from: boB on May 18, 2015, 01:32:14 AM........
New firmware coming up makes the current input and output much better in SOLAR and U-Set modes.
Also networking. boB
Does that mean it will transform my 20a input and 35a out to 25a in and 42a out ? That would be MUCH better ;)
or does it mean it will report more accurately ?
boB, that is exactly what's happening, the current sense drifts during the day, and on many days by the afternoon the Classic is outputting more than it reports. I haven't changed the mode to off and back again but I've shut the input breaker off to get it to rest and turned it back on and the problem is solved so it is definitely the current sense being off. That being said I would think you could add something to the software to force the Classic into resting when it goes above the voltage limit as a fail safe.
I am in solar mode.
Quote from: Wxboy on May 18, 2015, 01:40:11 PM
boB, that is exactly what's happening, the current sense drifts during the day, and on many days by the afternoon the Classic is outputting more than it reports. I haven't changed the mode to off and back again but I've shut the input breaker off to get it to rest and turned it back on and the problem is solved so it is definitely the current sense being off. That being said I would think you could add something to the software to force the Classic into resting when it goes above the voltage limit as a fail safe.
I am in solar mode.
OK, the current reporting will be more accurate but that shouldn't help high voltage regulation. Current doesn't have anything much to do with voltage regulation (I don't think anyway).
Yes, the update fixes current sense drift which was pretty good to begin with... WAS anyway.
Anyhow, now it's perfect.
boB
I have for years used the current limit function on the Classic configured at 63amps and for what's it worth it never tripped my 63amp breakers in/out with over current. So whatever variance there may in current sensing during the day, it's probably in the tolerance range of the MN 63 amp breakers. Wx is trying to show a condition that is *really* hard to catch and reproduce, and I appreciate the fine empirical observations he made and the discussion that ensued.
Going back to the problem I had where my Classic was showing Zero watts out and was regularly pushing my AGM batteries over the max. voltage set point, was there ever a firmware version that corrected this? I realize I may be the only one ever to experience the specific circumstances where this occurred but it did occur and it fried 2 out of 8 agm batteries forcing me to take my bank down to 4 12 volt batteries in series for 48 volts. I haven't used my classic in over a year because the warranty refurbished model has a loud annoying fan(my original fan sounded much better) and I just don't trust the controller anymore. I don't want to harp on this but I have to mention the fact that the event that pushed my batteries over the edge happened 3 months after I first reported the issue to Midnite. Had an RMA been issued in less than 3 months my batteries never would have died. Had to get that off my chest because I was originally a huge fan of Midnite products and I want to be a fan again.
The reason I ask if there was a fix is because someone I know is looking at the new Morningstar Prostar 40amp controller and I really want to suggest a Kid or Classic but since this person also has agm batteries I don't want them to have the same problem I had.
As a refresher this happened late in the day when the Classic was outputting more power than it was reporting. Eventually the batteries would be in absorb taking only about 70 watts or less with no load. The controller was outputting the 70 watts or less but showing zero on the screen. This condition would send the batteries up and up and even the max. voltage set point would not stop it. 2 of my batteries started bubbling internally and then gassed and now their capacity is much lower than the other 6 batteries and they can't be used in the bank anymore.
Quote from: Wxboy on March 20, 2016, 09:12:27 AM
2 of my batteries started bubbling internally and then gassed and now their capacity is much lower than the other 6 batteries and they can't be used in the bank anymore.
[/quote
IS THIS SOMETHING YOU COULD SEE OR HEARD? I USE AGMS AND LOST A FEW THIS WINTER AND LOOKING INTO THE CAUSE BET FEEL I LET THEM FREEZE BY LETTING THE SOC FALL WAY TO LOW BEFORE CHARGING BACK BUT NEVER HAVE HEARD BUBBLING FROM AN AGM, I DO KNOW I OVER VOLTED THEM. BUT STILL LOOKING INTO THE CAUSE OF DEATH.
WALT
Walt, the gassing comes from over charging, like an EQ on FLA batteries .... the problem is AGMs are sealed, but they have a release valve, hence the term,VRLA, valve regulated lead acid, the valve protects you by releasing excess gas H and O2..
Hth..
Quote from: Wxboy on March 20, 2016, 09:12:27 AM
As a refresher this happened late in the day when the Classic was outputting more power than it was reporting. Eventually the batteries would be in absorb taking only about 70 watts or less with no load. The controller was outputting the 70 watts or less but showing zero on the screen. This condition would send the batteries up and up and even the max. voltage set point would not stop it. 2 of my batteries started bubbling internally and then gassed and now their capacity is much lower than the other 6 batteries and they can't be used in the bank anymore.
So how did you establish there was power , the 70watts, going to the bank?
Was that from the WBjr screen which shows actual battery current or maybe a DC clamp ammeter on the main battery cable?
The zero power just shows there was no power being provided by the Classic BUT if the WBjr shows amps flowing into the battery then something else is likely providing that power OR if the WBjr is connected wrong way around then the 70watts could be the base (tare or minimum) load.
dgd
Guys you may recall that i also have confirmed the same rare classic output reading anomaly previously, if you need me to to find the threads holler. I dont recall seeing it occur for a while but ive probably not been so studious.
I always noticed it as a output current discrepancy of a couple amps, that came then cleared next day, never thought to study the voltage implications of it though.
Edit:
heres the post, graph shows the phenomenon clear as day
http://kb1uas.com/mnsforum/index.php?topic=1959.msg18186#msg18186
Quote from: dgd on March 20, 2016, 11:11:55 PM
Quote from: Wxboy on March 20, 2016, 09:12:27 AM
As a refresher this happened late in the day when the Classic was outputting more power than it was reporting. Eventually the batteries would be in absorb taking only about 70 watts or less with no load. The controller was outputting the 70 watts or less but showing zero on the screen. This condition would send the batteries up and up and even the max. voltage set point would not stop it. 2 of my batteries started bubbling internally and then gassed and now their capacity is much lower than the other 6 batteries and they can't be used in the bank anymore.
So how did you establish there was power , the 70watts, going to the bank?
Was that from the WBjr screen which shows actual battery current or maybe a DC clamp ammeter on the main battery cable?
The zero power just shows there was no power being provided by the Classic BUT if the WBjr shows amps flowing into the battery then something else is likely providing that power OR if the WBjr is connected wrong way around then the 70watts could be the base (tare or minimum) load.
dgd
dgd, the power out was confirmed by a clamp meter and the power was coming from the Classic.
After reading ZB's previous posts on this issue this seems to be an issue, maybe with just some Classics :-\
70 watts output into a fully charged bank after Absorb completed for 48v battery bank, thats about 1.25Amps @ 56v
I'm surprised that would be sufficent to take an AGM bank voltage high enough to cause battery destruction due to gassing/venting. Perhaps so if the bank capacity was quite small hence allowing the bank voltage to rise fairly rapidly
And that would only be for a few hours while PVs were illuminated each day then just the tare load of the Classic overnight (and any other connected loads) would soon reduce the bank voltage.
I checked through 6months+ of 10 minute logs I have for two RE systems with Classic 150s, 24v and 48v
AGM banks and thankfully found no real evidence of this behaviour. Or if there was then its likely insignificant and lost in the Amps reported for Classic output, WBjr battery current and tare/running current used by loads.
dgd
Its actually quite hard to spot. What you are looking for is when the positive WbJr current exceeds the classic output current. In order for that to stand out you need either no loads or fixed loads. Or if it happens during bulk on a nice day. But i only ever saw that once, usually it was in abosrb that it does it. Personally ive only noticed it a half dozen times.
Quote from: zoneblue on March 22, 2016, 02:15:13 AM
Its actually quite hard to spot. What you are looking for is when the positive WbJr current exceeds the classic output current. In order for that to stand out you need either no loads or fixed loads. Or if it happens during bulk on a nice day. But i only ever saw that once, usually it was in abosrb that it does it. Personally ive only noticed it a half dozen times.
Was there possibly a semi-heavy DC load that suddenly turned off during Absorb ?
boB
[/quote]
Was there possibly a semi-heavy DC load that suddenly turned off during Absorb ?
boB
[/quote]
Not in my case.