Classic 150 incorrect battery voltage!

[Aftrburnr]

Have had this new Classic 150 installed for about a month & a 1/2 or so, latest firmware. But, from day one, the battery voltage is reading wayyy high. I'd think it should do better/closer to what the actual battery voltage is. As an example, the Classic says 14.1 volts but measured with a couple different dmm, dmm says 12.5-13.0 volts. I test with unloaded & with various loads.

I do know there is an offset and I have been using anywhere from -1.5 to -2.5 volts. But then my stats aren't accurate either. It will be reading low but then once the day goes on then the battery voltage sometimes comes up to reading properly. This being a new setup, I want to keep track of things as I make changes. Any ideas why the battery voltages would be so incorrect?

Thanks!

[grgdgreek]

Can u give us a brief description of your system and pics?:-)
George

[TomW]

As you discovered there is an option in the tweaks menu that adjusts the voltage offset.

There is also the affect of voltage drop from higher current levels. This bit is a moving target but 2 volts is a LOT for that unless you have a bad connection somewhere.

First thing I would do is check every connection in the DC circuit, especially between The Classic and the battery. Pictures help like George mentioned.

Of course, it could be a failure of some sort in The Classic.  It happens even to the best gear sometimes.

Try checking the connections and then let us know what you find and if that doesn't help I will tag Ryan on it and see if he can get you sorted out.

Good luck.

Tom

[Aftrburnr]

Thanks for your expedient replies! Will take me a bit for pics but I can check all my connections. As an aside, I also have a harbor freight CC connected as well and it's near dead on as far as voltages go. It's sometimes off maybe by .1 or .2 volts +/=.

Something else I noticed with the voltage reading high, it thinks my batteries are near charged & I have to manually lie to the Classic CC using the offset and then I'll get some good output...

[grgdgreek]

P.s. what size cables are u using?
Lenth ect.

George

[ClassicCrazy]

Temperature compensation probe can change charge setpoints  depending on what is setup for temp compensation parameter.

Larry

[Westbranch]

Have you tried a factory reset (or VMM  or Vulcan Mind Meld)  as it is also called)?

Power down the PV and battery,
then hold down the left and right arrows on the MGNP and APPLY BATTERY power first and then   pv power...  YOU MAY HAVE TO RESET all your specific values etc..


hth

[Aftrburnr]

Yes, I do agree this situation is a moving target lol

grgdgreek: 10 gauge thhn 10 ft. long but I don't think wire size would affect what voltage is being read from the battery. I do plan on upgrading though within a week or so to 6 gauge thhn

ClassicCrazy: atm, I'm not using the temp probe and have not set that up until I get things dialed in.

Westbranch: Ok, might have to do this as there's some other things I have questions about, like the logs having strange info in them but still want to see if there was some obvious cause for the batt voltages to be very incorrect...

[CDN-VT]

10 gauge thhn 10 ft. long but I don't think wire size would affect what voltage is being read from the battery. I do plan on upgrading though within a week or so to 6 gauge thhn

10gAUGE FROM A CLASSIC  to Battery's is very undersized IF your pushing in a good amount of PV Vdc MPPT.
Mine is 5 feet of 2 gauge with a fancy terminals (I made ) I have on rare occasion seen 96amps & a yellow led showing max . Im over panelled for the winter gray days of drizzle..
VT   

[boB]

Aftrburnr,

how old is this Classic ?  Each and every Classic is supposed to be calibrated in production at the factory.
They should never be THAT far off.   Serial number ?    Maybe this one was missed ?  We will get
to the  bottom of this.

Thanks for posting !
boB

[Aftrburnr]

Hi boB,

This classic is/was brand new and the label marked Q2 2016 Classic 150V (rev 4)   Firmware: - Classic Rev: 2096  - Network Rev: 2097. This may be the problem that calibration wasn't done, done improperly, or possibly calibration has been lost somehow. Any possibility to re-calibrate or check the calibration of this unit myself? I'll try and PM you my serial #.

Thanks!

[Eleceng1979]

Aftrburnr,

As a sanity check,  please measure the voltage at the classic while not charging and again at the battery while not charging.  It is possible to see a slightly higher voltage while charging due to your small wire.  Think of back pressure of water on a small hose...while flowing higher on one end.  While stopped its equalized.  Please verify voltage on each end of the wiring with a meter while idle charging to rule out the cable.

If voltage is equal, then charge at high amps and repeat measurement with meter.  If equal then it's not the wire.  After this it is obviously a calibration/offset issue.

I do agree that the wire is small.  10 awg is good for 40, but fused no higher than 30a per the nec...  the strand size is larger than normal dc battery cabling and doesn't help either.  The nec says you need 4awg copper for 80A at a minimum.  I suspect you will see your voltage drop while charging due to this wire.

See here...
http://www.trojanbattery.com/pdf/WP_BatteryCableGuide_0512.pdf

My math shows a 3.2v drop in a 20' length of 10 awg copper at 80 amps and 48V.  If you are 24v then it's half...or exactly your difference.  If true you are wasting a bunch of power as heat due to voltage drop.
Go here and add your own variables... change to dc.
http://www.calculator.net/voltage-drop-calculator.html

Remember voltage drop is at the battery and higher at the classic due to it being the source of current.

Not sure about classic manual instructions for proper offset.  I would stop all charging, measure with a quality meter and then make offset.  If you offset while charging hard you will be real high on bat volts at the time of finishing a charge.  More charge current = more volt drop.  So offsetting with a bunch of drop will lie to the classic and eventually overcharge the bat.  Imagine playing the wind in golf using a fixed offset with varying wind...eventually you will be in the rough.

[boB]

I think that he checked the Classic without any load so voltage drop wouldn't be it I don't think.

Even if it were loaded, there wouldn't be THAT much voltage drop... I would certainly hope !

Yes, you can calibrate it yourself.  It's pretty easy.

If you could check  modbus register address  4299  (register 4300, FactoryVbattOffset)

It should be stuffed with a value somewhere around    0004  (decimal or hex)

This is the internal factory offset.  We can talk with PM if you don't have means of
checking modbus registers.

Then, it is a simple matter to adjust the gain by that trimpot on the control PCB
just below the bottom side inductor sticking through the board.

More coming about this here I'm sure....

boB

[ClassicCrazy]

Hi boB,

This classic is/was brand new and the label marked Q2 2016 Classic 150V (rev 4)   Firmware: - Classic Rev: 2096  - Network Rev: 2097. This may be the problem that calibration wasn't done, done improperly, or possibly calibration has been lost somehow. Any possibility to re-calibrate or check the calibration of this unit myself? I'll try and PM you my serial #.

Thanks!

Bob told in detail how to calibrate the Classic using the pot on the board  in this post
http://kb1uas.com/mnsforum/index.php?topic=1067.msg7955#msg7955
- though you should check the register first as he says above.
Larry

[Eleceng1979]

I think that he checked the Classic without any load so voltage drop wouldn't be it I don't think.

Even if it were loaded, there wouldn't be THAT much voltage drop... I would certainly hope !

Voltage drop is not present without current flow/no load... I*R = V Drop.  Measuring without load confirms a connection, Measuring with a load confirms a drop.  Only one of these is beneficial to his troubleshooting of his Vdrop problem.  My issue is the amount of drop is proportional to the current flow, but the classic offset is a fixed value.  Big Vdrop + incorrect offset = cooked batteries...

That much drop will happen per the math/scenario I showed.  Remember Current times wire resistance = Vdrop.  Stuffing 2X the current in undersized wiring causes significant rise in Vdrop, wire resistance and heat.  Factor in THHN wire with large strands, not sharing current and being used in a DC service the above math is even worse.  I intend to be constructive, not some NEC code book junkie beating people over the heads with my comments.

I engineer systems with copper wiring transmitting 30+MW of 13.8-69kV and even at these levels we do not tolerate more than 1-1.5% voltage drop in the design.  His design has more than 13% in the example I gave.  Remember I2R = power loss...  In my example there is 256 Watts of heat in the wire while trying to transmit 1920 watts, the net watts to battery = 1664 watts or 13% less... Vdrop = power loss in wire...period.  At some point it is a hazard, and 256 watts of heat in a 10awg wire wont take long to get to 194 deg F to melt insulation and start a fire...