Do shunts go bad?

[FlyWright]

I'm out here on Day 28 of my Quarantine Cruise, gunkholing in the California Delta. My 4 yr old batteries seem to be doing fine but it seems that all charging sources are providing about half of the normal charge for any given SOC and my big loads like microwave seems to discharge at about half the current than using the same appliance a month ago.

For example, my shore charger only seems to put out 25-28A vs. 50-55A, my Balmar alternator pumps in only 40-45A vs. the old 85A.

It seems to point forward a faulty shunt rather than a weak battery bank. I suppose it could also be a faulty Xantrex LinkPro SOC meter. I seriously doubt my independent charge sources all dropped by about 50% simultaneously. If my battery bank was getting weak, I wouldn't expect to see half loads on the current draw.

Any ideas before I start throwing parts at this?

Thanks much for your thoughts and assistance.

 

[boathealer]

Do you have a multimeter than can handle 55A? You could put it inline (or replace) the shunt for a temporary measurement of current.

Also can measure the voltage drop across it with a voltmeter - it is probably a 50mV shunt.

Those would at least be a couple of additional data points that are easy to get at.

 

[MYTraveler]

I doubt that they could wear out, but their resistence could be changed (making readings unreliable) from either exposure to corrosive materials (like battery gas) or the accumulation of conductive dust / soot / film. If you clean it up, it should be good as new (at least absent visible contamination).

 

[DDW]

The shunt is a pretty simple device, and any degradation in it is much more likely to raise resistance, rather than lower it. You reduction in indicated charge implies a lowered resistance. I can't think of a failure mechanism in the shunt (barring a piece of copper leaning against it) that would cause this. Much more likely the SOC meter has gone off calibration. It is comparatively far more complex with many failure modes that could cause this. First thing to check would be to see if there is a parameter setting for shunt resistance that has gotten changed.

If you have two batteries in parallel making up the bank, you might also want to check them individually and the connections. A dirty terminal on one would cause a 1/2 reduction in acceptance rate (does not explain the microwave though).

 

[kchace]

Shunts don't wear out per se, but the voltage readings from a shunt are very small - so ANY looseness or corrosion on the sense wires leading from the shunt to the meter could cause a huge error. I suggest checking the shunt and meter wiring very carefully.

Ken

 

[Xsbank]

Voltmeter on the batteries.

 

[High Wire]

Measure with a DC clamp-on ammeter. Only way to be sure. A DVM can't handle that much current.

Dropped by 50%. Could a setting have changed accidentally? Can you check the current settings against what it was?

 

[Olebird]

Check your batteries for an open or hi resistance connector between cells. An open battery would explain some of it.

 

[Peter B]

I agree. From my modest experience in these matters, but you sure remember them, when suddenly things electrical begin to misbehave where power is via batteries, and you've changed nothing, a cell dying in a battery is the first thing I'd check out, as suggested by others, by direct voltage measurement across each batt while isolated. How often I have come up with a reading of 10v, when the batt was operating perfectly just the day before. This, usually in a car, I might add. AGMs in the boat are more robust, but it can happen. Being in a bank as in a boat, it is a little less obvious than like a car, just refusing to kick over the engine.

 

[DDW]

It probably is not bad batteries, as his microwave indicates 1/2 draw. Bad batteries would indicate the same current, low voltage, and likely shut the inverter down. Best guess is something in the sensing circuit: wiring to the SOC meter from the shunt, or the SOC meter itself.

Always have a clamp ammeter on board. Then you don't have to guess about these things.

 

[richvanf]

Most likely is a hi-resistance connection from the shunt to the meter unit. The only time I ever saw a shunt go bad, one of the two resistance strips corroded so badly no current was passing through it, and the meter showed twice, not half the appropriate current.

 

[FlyWright]

I'll try this again...poor signal and lost my detailed reply.

Thanks for the help, guys. No clamp on meter aboard but 2 house bank voltmeters agree. I reached a float charge again last night so I know I'm charging sufficiently.

I'll take a much closer look at the batts and shunt today. I'll check back with my findings.

 

[Larry M]

Al: We’ve spent six months at anchor several times and have seen the same thing. We've had two different Xantrex SOC mentors and now we have a Link 10. My experience is that “out” does not always equal “in”. Periodically we’ve had to reset them. Motoring all day will sometimes do it or I will manually reset it. It sounds like your SOC monitor has lost it full point meaning your batteries are fuller than it indicates and that is why your alternator and charger are only giving you about 50% of their full output.

 

[WayneDana2]

Measure with a DC clamp-on ammeter. Only way to be sure. A DVM can't handle that much current.

+1
If you have a Fluke MM, you can buy a clamp shunt attchment for much less than a clamp ammeter. Handy things to have on board.
Anyways, I would buy, beg borrow or steal a meter before loading the parts cannon.

 

[DDW]

Regardless of the "full" point of the SOC meters, the amperage measurement from the shunt should be accurate. This does not depend on battery SOC, charge or drain source - only the shunt and the voltage measurement across it. Some meters have a parameter to specify the mv/A of the shunt (as these can vary). If that is set correctly then it is an electrical problem with the metering circuit.

 

[Ski in NC]

If a shunt has a problem, indicated amps fail high, not low. Issue is elsewhere.

 

[FlyWright]

You's guys are great! I'm starting to think that with your help, I've got this thing licked. (Not literally!)

I took a couple hour boat ride the other day and shortly after dinner with the genset, I was back to a float charge. Today, 2 days later after a battery and shunt cleaning with a small brush and CorrosionX and another float charge after a 2 hr boat ride, things have returned the old normal.

The indicated voltage in the SOC does not match the old analog volt meter or the Fluke. It appears to me now that the SOC indicated House V is varying between 0.2V and 0.4 V less than either the analog meter or the Fluke.

I think my Xantrex POCC (piece of Chinese crap) meter has failed again...my second one.

Maybe it's time for a Balmar SG 200, but I'll missed the A in/out. I like the idea of a Fluke attachment Ammeter. Got a link?

Thanks a million, gang!

 

[WayneDana2]

https://www.digikey.com/product-det...gFOKuT_wVXt76gjQYGtpPd9GqHCui20RoCLQ4QAvD_BwE


I'm sure you would be better off shopping this one out. There are others as well.

 

[bayview]

Is there an initial calibration that would be dome at installation. Try redoing that.

have you cleaned all the dedicated connections to the SOC meter?

 

[Larry M]

...Maybe it's time for a Balmar SG 200, but I'll missed the A in/out...!

The new SG200 does have a shunt and you can monitor amps in and out. It even has an optional Bluetooth smart phone app. I haven’t seen one yet but it sounds pretty good.

SG200 Battery Monitor | Balmar

 

[Stout]

Sg200

We put in a new SG200 and it worked great until the last firmware update. After that it took a lot of re-setting, pulling the fuse from the house bank etc. The factory Balmar tech was helpful but the time left function is still inoperative and was told by the factory tech that this is a “known issue”. As with other companies, most are working remotely so no time line was given for a fix. If you buy one with the older firmware, don’t update until they fix it.

 

[MYTraveler]

If a shunt has a problem, indicated amps fail high, not low. Issue is elsewhere.

SOC meters indicate amps in two contexts -- RATE of flow in or out, and Net Remaining Amp hours. I think you are talking about the later (even though, technically the indication is not amps). But if you are talking about the former, a corroded shunt (or its connections) will read low, not high. Conversely, if you are refering to amp hours, then I agree remaining amp hours will be overstated, but consumed amp hours will be understated.

 

[Ski in NC]

A shunt is a resistor. Very low resistance. The meter reads mV or mA across the shunt. If there is any corrosion on the shunt, resistance goes up. Higher R means the instrument (whether mV or mA) will read higher. I think most measure mV as that is less sensitive to line loss on the way to the instrument.

How this is processed by the SOC algorithm, I don't know.

 

[SBR]

Just a quick thought, you had mentioned that you had some old analog gauges those usually had the shunt on the + side of the battery and most digital meters require the shunt be on the - side, I have hade a couple of guages act "funny" with a positive shunt. Just a thought

 

[Olebird]

Not sure where you are on this project, but as i catch up on my reading - some of my experiences with shunts came back to me. While it's true if corrosion affects the current thru the shunt, the instrument reading it will likely see a high number, however If the corrosion is not the shunt proper but rather in the loop feeding the instrumentation reading across the shunt, it could be seeing a lower number. Because the output from the shunt is typically milli-volts, doesn't take much corrosion in that loop to cause some real problems.

My experience, shunts proper in an industrial setting almost never go "bad" but the mv feed loop is usually the source of the problem.


On the other hand, State of Charge (SOC) instruments I have worked with are almost never very accurate. They have to "integrate" (or accumulate) two variables - time and Watts (or volt/amps). Any noise or glitches in any input can send them off in never never land.

 

[FlyWright]

UPDATE:

I've had a bit of time with the boat after cleaning up the contacts on the shunt and getting the bank back to a float charge. The shunt is now indicating normally again. Thanks for the tip on that! CorrosionX to the rescue again!

Topping off the batts and resyncing the SOC also helped with the indications. Apparently, I still have an intermittently erroneous house batt voltage indication on the SOC but the analog voltmeter is indicating normal. That's something I can live with on this trip.

I've never been away from shore power for so long. Today is QC Day 44 (since I last left my marina...not counting my 7 days aboard before that) with only one night at a dock to reach float. My typical days start at 75-90% SOC and I go to bed at 85-97% SOC. My typical overnight draw down runs in the 10% range.

I don't run the generator at night but use it a few hours per day to make coffee, cook and make hot water if I haven't relocated that day. I also get a recharge during those activities which is typically enough to keep me in the +70% range. Before I diagnosed the malfunctioning volt meter, I mistakenly thought I was plunging much lower than 70%. (Yes, I admit I forgot all about that old analog meter down there! )

Thanks to all who have helped me sort out this mystery.