Eric Hammond

How did the replacement do?

Land is a tremendous amount of work, I purchased 10 acres and can't keep up with it. I'd suggest 5 or less

Everyone here has offered great suggestions so I wont repeat any of that, but I have one thing for you to think about.

When you buy your inverter, consider one with a integrated battery charger that you can use with a generator to charge the battery bank.

The sun doesn't always shine, and a bank at a low state of charge for several days is hard on the plates in the battery.  It's nice to be able to fire up a generator and charge the bank quite easily

Paul seems to think that when multiple people start making lists of things throughout a post, when someone replies to item number 1, it's hard to tell who they are actually replying to.   By making your numbering completely different, there is no question who you are replying to.....there seems to be logic there for me

In the simplest terms, judging from what I see, the battery monitor is bad.

(using Paul's numbering system)



12s)  The voltage drop of .1 mV seems right for the 2 amps coming in.

145)  The battery monitor being plugged in to the circuit should not increase the voltage in the circuit, making me think there is a voltage leak internally to it.

198r)  The very last picture tells me the monitor thinks amperage is always coming in to the battery all the time, and never leaving the battery at night, still charging through some miracle

23b)  The amperage of 1000 amps at 48v = 48,000 watts from the panels at night seems excessive  (this one was slightly a joke)

1s)  The 1000 amps is positive, meaning that's what it thinks is going in to the battery.  The shunt measures voltage drop across it to calculate amperage.  In order to figure whether that voltage drop into the battery or out of the battery, it does an open circuit voltage test of each side of the shunt basically.  In our example of 50 mV drop at 500 amps, say 48 volts,   If we are in a discharge state, voltage between the positive terminal and the battery negative terminal at the shunt would be 48 v, the opposite side of the shunt to B+ will read 47.950.   If we are in a Battery charging state, open circuit voltage B+ to battery negative side of shunt would be 48 and B+ to the opposite side of shunt will be 48.050   (these numbers aren't real obviously, but gives you how the battery monitor will interpret if we are in a charge state or discharge state)

26t)  That means in order for your battery monitor to read +1000 amps continuously, the side of the shunt away from the negative battery will always have .100 V DC more then the negative side, even at night, and that just cannot be possible

34d)  New parts are just untested parts

Based on this logic in my in-expert opinion I would replace the battery monitor.


I also at this point would like to razz you about the wire nuts, and that you should get rid of them, we aren't building a house ;)

Gerry Parent wrote:

Eric Hammond wrote:
Hey Gerry, sorry it wasn't something simple!  I'm a little confused by some of the readings you said, mostly because without pictures its not clear what your referring to.  Also, not to second guess you in any way, but I would like a picture of the meter screen when your make your measurements.  We are dealing with REALLY small amounts of voltage here and the measurements are critical.

I'll snap some photos today

First, you did install the shunt in the NEGATIVE battery cable correct?  Were the fasteners torqued with a torque wrench?

Yes, Shunt is on the negative cable but not secured with a torque wrench. Couldn't find any info in the setup about what torque to use....any suggestions?

You need to set your voltmeter up to read mV.  I highly recommend this specific meter for ease of operation and accurate results/ with cost in mind.

For now I'm using a cheap Home Depot special volt meter but will order one of those your using as I need a better one anyways.

You'll notice on my shunt, all the terminals were coated with a conductive electrical grease (NOT di-electric)

Could you include a brand name as I do have di-electric but not what your suggesting.

Does that make sense?  That should be enough info to pinpoint the culprit.  It's worth noting, Voltage drop only works with the circuit intact and functioning.  We cannot unplug anything to make measurements.

Yes it does make sense to me. I'll get back to you once I get more readings.

16 ft lbs on a 3/8's course bolt should be good

The electric grease I use is XG-12 by motorcraft https://www.amazon.com/gp/product/B000NUBB28?pf_rd_p=c2945051-950f-485c-b4df-15aac5223b10&pf_rd_r=DCY616X5SB174RA91PH5

The picture testing at the shunt, is that with the battery monitor plugged in?

Gerry Parent wrote:Double checked that all loads ARE connected to the side leaving the shunt.
This was the only day where we turned off the inverter and the BMK amp reading was not at   +1011.6 ADC  but crept down about 5 amps. Then when we unplugged the power from the BMK for about a minute then plugged back in it was down to around 980amps. Thought I was onto something but then when I tried it again, it went back up to +1010 ADC.

Also, "5G Total AH OUT" has been at 0 ever since all this started and never moves. Even when the inverter and charge controller was off it still didn't move.

Got a hold of Magnum on the phone today and they said to check the milliamp reading between the orange and blue wires. We checked them on the shunt (0mV), where they attach to the BMK (2.3mV) and then disconnecting the harness from the BMK and checking again (0 mV). To me they should have all been the same but weren't.  

Hey Gerry, sorry it wasn't something simple!  I'm a little confused by some of the readings you said, mostly because without pictures its not clear what your referring to.  Also, not to second guess you in any way, but I would like a picture of the meter screen when your make your measurements.  We are dealing with REALLY small amounts of voltage here and the measurements are critical.

First, you did install the shunt in the NEGATIVE battery cable correct?  Were the fasteners torqued with a torque wrench?

The following statements I'm about to say from an electrical engineer perspective is absolutely not true, however from layman's perspective like you and I, gives us a reference as to what must happen to get your scenario:

Your battery monitor is reading 1010 amps.  The shunt is rated for 500 amps at 50 mV and is calibrated to the battery monitor upon manufacture.  Given that information, we can assume in order to read 1000 amps, the battery monitor must see around 100 mV difference in voltage.

That's not much, and any poor connection could cause that easily.

My system is an outback, but a shunt is a shunt.  Here is how you can test if its a connection or the battery monitor.

You need to set your voltmeter up to read mV.  I highly recommend this specific meter for ease of operation and accurate results/ with cost in mind.  Your leads stay plugged into the common for black and volts for the red.  We must select the mV scale because we are going to read decimal places before 1 mV.  On this meter you place the dial at this postition, but you must press the yellow button at the top to select DC mV.  You'll notice there is enough static in the air get a reading with the leads not touching anything.



Leads touching each other should zero the meter.



Touch both of the leads to the each side of the shunt.  Polarity is irrelevant to us and we will disregard any minus sign.  You'll notice on my shunt, all the terminals were coated with a conductive electrical grease (NOT di-electric) and torqued, any resistance here make inaccurate amperage readings.



The reading across the shunt is .5 mV   or .0005 VDC



This equates on my battery monitor to about 5.4 amps  on a 24 v system


Follow those two wires back from the shunt and measure between them again were they connect to the battery monitor



The reading here HAS TO BE IDENTICAL to the first reading across the shunt


If the reading at the shunt is not identical to the reading at the battery monitor, we need to figure out the poor connection.

If the readings ARE identical, then we need to think about what reading we got.  Is the reading 50 mV? If so the battery monitor must read 500 amps.  If it reads higher, the monitor HAS to be bad

Is the reading 25 mV?  the monitor should be reading 250 amps and so on and so on.

If the amps on the battery monitor match what we are getting based on voltage drop of the shunt, IE, for your 1010 amp reading, we measure 100 mV or .1vDC across the shunt side to side, we are most likely looking at a poor connection at the shunt.

Does that make sense?  That should be enough info to pinpoint the culprit.  It's worth noting, Voltage drop only works with the circuit intact and functioning.  We cannot unplug anything to make measurements.

Hey Thomas, I rebuilt my feed tube with a refractory mortar.  It cracked a bunch but didnt get loose. Once the cob was around the bricks I was good. I suspect that the refractory mortar would have failed just like the clay did  without outside support.

You might brainstorm and see if you could pour  1 or two inch concrete / grout around the feed tube for support

I would think that the tank should be quite heavy.  Heck a 20 gallon tank at 6 lbs to a gallon should weigh 120 lbs full right? I would bump the tank pressure down a bit.  Those pencil pressure gauges arn't the most accurate things in the world.  When cars were mandated with tire pressure monitor systems in 2007-8 I was a firm believer in a digital pressure gauge.