HELP - Another strange voltage plus "unexplained charging current" question

[smokey73]

HELP!
I was out sailing awhile back and had a "low voltage alarm" 11.5V on my chartplotter. I immediately started the engine but got to thinking about why it went to 11.5 V during a few hour sail. So I got out my Volt-Ohm Meter and did some checking. Here are my readings. I have a Victron battery monitor, was using a Digital Volt/Ohmeter. The Starting Battery is connected to the system via a Blue Seas ACR. Starting Battery remained at 2.35V and never changed. All charging sources had been removed for about 1 hour before I started this test so the voltage had not settled down. Outside temperature was about 85 degrees. I did not measure battery temperature but will next time. Batteries are 2 6V Duracell Golf Cart Batteries about 5 years old. I forgot to check battery fill, but did so about 1 month ago and added a few tablespoons per cell then.

Solar charging fuse removed. No charging source on. Bilge pump disconnected. DC main breaker off. No shore power connected
Voltmeter at Battery 12.95V
Victron Voltage reading 12.89V Surprised my Victron reading was off by this much?
Voltmeter at main DC Bkr 12,94V
Voltmeter on panel meter 12.9V (analog meter)
Victron Amp reading +0.28 A With no charging source on - where did this come from

Solar charging fuse removed, No charging source on. Bilge pump disconnected. DC main breaker on with about 13 Amp load . No shore Power Connected.
Voltmeter at Battery 12.32V (The battery voltage dropped fairly quickly when under load and continued visably dropping)
Victron Voltage reading 12.31
Voltmeter at panel DC Bkr 11.76
Voltmeter on panel meter 11.6 (analog meter)
Victron Amp reading -13.3A
I immediately turned off the DC main bkr at the panel and the Victron again showed +.28A with no charging source present. What would cause this?

Here are my thoughts.
1. I might need new Batteries but will only know for sure once I get this straightened out and can do a battery load test per Mainesail's instructions
2. I likely have some voltage drop due to resistance in connections, either the positive or negative side and will have to start measuring resistance coupled with cleaning all connections.
3. Since this happens between the Battery terminals and the DC power lead at the panel, it is likely in the cables/connections between the Battery main terminals and the DC breaker at the panel or the negative side.

Any help on narrowing down the source of the reduced voltage would be appreciated.

Just as importantly though is why my Victron Battery Monitor is reading +0.28 Amps positive current flowing into the battery if there are no charging sources installed? If its true then I have discovered a perptual self charing battery

 

[Rich Stidger]

Dan,

The apparent +0.28A is likely an offset in the Victron. Your owners manual should declare this. I don't have a Victron, but rather a Heart Interface Link 20 battery monitor. It basically provides the same function as the Victron monitor. My Link 20 has a +- 0.1A current resolution. So at actually zero amps, the current display could read +- 0.1 A. I actually have seen some accumulation of AH with no load or charge due to this resolution. I think 0.28 A is a bit excessive, but I don't know the Victron.

If the reading were negative (mine is) then the likely cause could be the current draw of the Victron itself. My Link 20 draws 20-170 mA depending upon how and if the display is lit- Brightness, LEDs only, etc.

You could test this theory by disconnecting one of the sense wires on the current shunt and connecting it to the other sense wire. That would provide a actual zero millivolt input to the Victron. Then see what the current meter reads.

All of your measured voltages look fine to me under the conditions that you measured. I would not expect the Victron to agree perfectly with your DVM, and definitely not agree with your analog panel meter.

 

[smokey73]

Rich, I'll research my Victron manual to see what it says about current draw and any +/- possibilities due to precision/accuracy of the measured current. I'm surprised by the magnitude of the "charging current" shown. Just can't understand. As far as voltage reduction when under load, I expect some since there is a cable run of 20 feet or so (one way) and several connections along the way but not that much.. It just seems excessive. I never had the warnings on my chartplotter of low voltage before so something has changed.

Since the voltages are about the same at the Battery, Victron, and Voltmeter at the DC on/off switch but off (not load) by .55 V or 4.5% I am pretty certain that is not normal. Remember that the voltage of 11.76 is actual measured voltage by an accurate Volt-Ohm meter. It may be even lower at the actual load depending on the drop from the panel to the load itself. I think there is some excess voltage loss due to resistance so I'll be checking all my connections just in case.

 

[smokey73]

@Maine Sail are you out there for a comment? Sure would like to hear your thoughts?

 

[smokey73]

@Rich Stidger the manual lists a voltage accuracy as +/- .4% and the current accuracy as +/- .3%. The voltage reading is okay at no load but is outside the spec under load. The current accuracy seems to be way less than the advertised amount. I may try to call the victron tech reps to see what they say.

Hope you are doing well during this trying time.

 

[Rich Stidger]

@Rich Stidger the manual lists a voltage accuracy as +/- .4% and the current accuracy as +/- .3%. The voltage reading is okay at no load but is outside the spec under load. The current accuracy seems to be way less than the advertised amount. I may try to call the victron tech reps to see what they say.

Hope you are doing well during this trying time.

It's not the current accuracy. It is the resolution. The meter reads a value with some accuracy and then a +- digit count of resolution.

 

[Davidasailor26]

I agree with you that it seems like you have a lot of voltage drop with a 13A load. I don’t know the gauge or length of your cables, but even 40’ of 6 gauge should only be 0.21V drop, so I’d guess you should see less than that.

For the mysterious charge source - you could try measuring the voltage across the shunt itself to see if it reads 0.0mV or some low mV. Or, you could disconnect the wire from one end of the shunt, put your meter on Amps, and put it inline in the circuit to see if it agrees with the Victron.

 

[dlochner]

Victron Voltage reading 12.89V Surprised my Victron reading was off by this much?

Where are the sense wires connected? That could account for some of the voltage drop along with simple measurement error or instrument precision.

I immediately turned off the DC main bkr at the panel and the Victron again showed +.28A with no charging source present. What would cause this?

How long does the positive charge last, i.e., does it fade away over time or is it pretty consistent? It might be residual current draining from the battery charger and the various electronics. These all have capacitors which may hold a charge for a while even when turned off. It's a long shot answer, but maybe?

 

[SycloneDriver]

I think I would start by removing and cleaning all the connections at the shunt and batteries.

 

[jviss]

Holy cow, RTFM!

Which model do you have?

I just looked at the manual for the 702, searched for "calibration" and found this, under battery settings:

09. Zero current calibration If the BMV reads a non-zero current even when there is no load and the battery is not being charged, this option can be used to calibrate the zero reading. Ensure that there really is no current flowing into or out of the battery (disconnect the cable between the load and the shunt), then press SELECT.

 

[jviss]

About the Victron, and battery monitors in general: I had a Victron on my C36, and I could never trust it. I spent more time trying to calibrate it than use it, and I never trusted it as a result. It was fooled by a couple of common events, too, like if I decided to stop the engine even though the batts weren't fully charged, I'd idle back to cool it before turning it off, the current would drop, and the Victron would say "fully charged!" Same with solar, when the sun went down, the Victron thought this was accept current tapering off and declared the batteries fully charged. Useless P.O.S.

My intuition about the house state of charge was more reliable and accurate than the Victron.

I don't think the Balmar device is much better, take a look at the comment thread on Panbo.

 

[Maine Sail]

This may help:

Making Your Battery Monitor More Accurate (LINK)


.

 

[sesmith]

Concerning your voltage drop issue. Resistance readings won't tell you much. Do voltage drop readings under load between each point in the circuit and corresponding battery post to find the points of high resistance / corroded contacts. Work back to the battery and when the high voltage drop disappears, the problem is between the good spot and the last one you tested (you can also test across each connection). Do this on the negative side as well as the positive side. You show about a 5% drop between the battery and panel with a 13A load. Seems excessive to me, but depending on the cable used and length of run...suggest using a wiring size calculator and running the numbers to see if this is within spec for the load, length, voltage and wire size. This will tell you for sure if you have a problem.

When you tested voltage at the battery, where did you put your meter probe? If you had it on the battery post (ie, not the cable where it is on the battery post) your problem could very well just be a corroded or loose battery cable connection. Do a voltage drop test between the battery post and the cable under load. I don't remember exactly what you should expect, but my hazy memory seems to think that anything over .1 v (100 mv) per connection is excessive.

The fact that your battery voltage continued to drop "quickly" with a 13 A load on it suggests to me that they are likely no good. Read some of Mainsail's writeups that show how to do a battery discharge test. You should be able to find discharge rate info on your batteries and be able to at least ball park it to good or bad with a discharge test over time.

 

[jviss]

@Maine Sail , I've read that article, nicely done. For me, it is far simpler to just "know" when I last charged and for how long, what loads I've been running, and to check my no-load or low-load battery voltage as a check on SOC. For me, a battery monitor is more trouble than it's worth. If I had to second guess the gas gauge in my car like that, I'd just put a piece of tape over it and watch my mileage between fills; which is what I do now on the boat. I charge all the way uless there's some compelling reason I can't, charge 'til the accept current into my 230AH bank is 4.5A, and call it a day.

 

[smokey73]

Holy cow, RTFM!

Which model do you have?

I just looked at the manual for the 702, searched for "calibration" and found this, under battery settings:

09. Zero current calibration If the BMV reads a non-zero current even when there is no load and the battery is not being charged, this option can be used to calibrate the zero reading. Ensure that there really is no current flowing into or out of the battery (disconnect the cable between the load and the shunt), then press SELECT.

@jviss Agreed RTFM. I couldn't find my manual and I wasn't where I could get the internet to download a copy. However, (though I failed to mention it) the Victron current always read 0 before when the charging sources were removed if my recollection is correct. Maybe I'm making a mountain out of a mole hill. If its just drift of the calibration I'd be happy with that. I'll recalibrate the current reading on the Victron and be done with that. I've read the Compass Marine articles and will study them some more. I know everything is hooked up with all the negative loads going through the shunt.

I was getting ready to set up for a 20 hour battery capacity test when all this happened. Unfortunately, it is not a very accurate test (but probably better than nothing) since I don't have a good way to hold current constant. Does anyone have a decent way to get "relatively constant" current draw by some DIY DC load testing device?

Assuming the unexplained "charging current" is just a calibration issue, I'll proceed on with measuring voltage drops and see if I can pin down where the voltage drop is occurring.

 

[jviss]

The only thing that's an issue is the 12.31/11.76 voltage measurement, that means that you probably have a poor connection somewhere along the line, or a too-small conductor.

 

[smokey73]

The only thing that's an issue is the 12.31/11.76 voltage measurement, that means that you probably have a poor connection somewhere along the line, or a too-small conductor.

Agreed, I am pretty sure the conductor is reasonably sized since I never had this problem before and the voltages on the Victron and my installed analog voltmeter (which reads the same as my handheld Volt-Ohm meter) always tracked pretty close to each other under load (subject to analog limitations). So the voltage I read of 11.76 is "real" and the 12.31 is "real" indicating some change in the resistance, most likely a corroded or loose connection. Its tedious but I'm pretty sure its solvable with a little perseverance.