SmartGauge -- Real Life Experience

[Dockhead]

Originally Posted by a64pilot

One issue with it just interpreting voltage to a SOC is it doesn't know battery chemistry, and that does change the voltage / SOC doesn't it?

You should be able to get a table not just for every battery chemistry, but for every specific model of battery. I used the official Trojan table for my batteries.

Probably obvious to everyone but worth noting -- this approach is completely useless with LiFePo batteries. This whole conversation is specifically about lead-acid batteries.

[MarkJ]

I agree with Dockhead!

When I started the cruising caper i only had my multimeter and stuffed it in the cigarette socket when the fridge wasnt on.

That gives about the same value readings as the lickety-spilt monitoring thingies i have now.

Their best value is the nice digital readouts. A voltmeter digital readout is $200 so you may as well get a whole monitor.

But its still volts thats my guide.
Amos in and out is nice but its guestimate work by the meter to work out % charge.

Just keeping my eye on it all day every day is the key. So the readouts are in the galley just above the beer fridge... Easy to see

[S/V Alchemy]

Originally Posted by Dockhead

The problem is not installing the shunts (although that is a minor PITA). The problem is I don't have a single cable through which all the DC power passes, and can't create one without creating a long detour which would be harmful to system performance.

What I really need is an ammeter, or even better a counting ammeter, which can AGGREGATE the data from two different shunts or hall effect sensors.

I have never been able to find such a device. If someone has any hints, I would be very interested.

I am shortly going to be installing one of these units. I have several charge sources (shore, alternator, solar, wind and Honda 2000 x 2, if needed) and I wanted to monitor inputs and outputs in one device. I'll report in my blog on my findings: PentaMetric Components - Bogart Engineering

This short video shows the Pentametric in action. Many of your parameters can be seen here. The options are daunting, and it's clearly an "engineer-designed" device, but I'm hoping it will give me the information I need for battery bank management offshore.

https://www.youtube.com/watch?v=rABhk70NCbs

[Paul L]

I don't understand the bad mouthing of the amp-counters like a simple Link. They give you the most useful info simply: current volts, current amps going in or out. Then they give you an estimate of net amp-hours that the battery is down. The estimate is close enough much of the time. Sometimes it gets off and will eventually reset. It seems far more useful than the Smartgauge that also gives an estimate of charge state that is possibly more accurate when it is reading correctly and less accurate when the battery state is changing -- which it does for most of the time on a cruising boat.

[a64pilot]

Originally Posted by Dockhead

You should be able to get a table not just for every battery chemistry, but for every specific model of battery. I used the official Trojan table for my batteries.

Probably obvious to everyone but worth noting -- this approach is completely useless with LiFePo batteries. This whole conversation is specifically about lead-acid batteries.

Got that, for my Lifelines, but assuming the Smart Gauge tracks voltage and just displays that as SOC, wouldn't you have to set the thing for at least the type of LA battery you have?
Course a high amp drain really kills bank voltage while the load is on, but the Smart Gauge doesn't know load, can't, so is it just heavily dampened to keep it's SOC display from crashing when a high amp load is put on the bank.
I think it's operating principle is simple, just PFM is all

[a64pilot]

Beauty of the thing is its simplicity, and that it would be real hard to bugger it up.
Plug it in, and forget about it, it supposedly gets more accurate if left alone and ignored unlike others that require constant calibration.

[Paul L]

Originally Posted by a64pilot

Beauty of the thing is its simplicity, and that it would be real hard to bugger it up.
Plug it in, and forget about it, it supposedly gets more accurate if left alone and ignored unlike others that require constant calibration.

I don't do any calibration on my Link. It resets itself after a full charging followed by some discharge. As far as the Smartgauge getting more accurate overtime, is that what you see in actual cruising practice?

[Dockhead]

Originally Posted by Paul L

I don't understand the bad mouthing of the amp-counters like a simple Link. They give you the most useful info simply: current volts, current amps going in or out. Then they give you an estimate of net amp-hours that the battery is down. The estimate is close enough much of the time. Sometimes it gets off and will eventually reset. It seems far more useful than the Smartgauge that also gives an estimate of charge state that is possibly more accurate when it is reading correctly and less accurate when the battery state is changing -- which it does for most of the time on a cruising boat.

Well, sort of. The Smart Gauge, to the same extent as an intelligent interpretation of simple system voltage, gives a very accurate picture of the state of charge when the battery is discharging. Far more accurate than amp-counting meters in my experience and also according to MaineSail's testing.

It's only when the battery is being charged, and the charge is not carried on to a full charge, that it gets confused.

But I do agree that an amp-counting meter is very useful for seeing the load on the system, charge rate, and cumulative amount of power used -- just ignore then SOC calculation, which is much less useful.

[S/V Alchemy]

A second video by the previously cited technically minded Australian shows data output to a PC. This seems more than sufficient information to me. In a boat context, you could learn the SoC of your house and start banks, which is not usually an issue with ACR/Echo Charger, as the start battery is customarily used as a dump for the house bank. Handy if you have to use the start bank as an emergency house bank, however, via a big red switch:

https://www.youtube.com/watch?v=XePNBydCtTY

[Dockhead]

Originally Posted by Paul L

I don't do any calibration on my Link. It resets itself after a full charging followed by some discharge. As far as the Smartgauge getting more accurate overtime, is that what you see in actual cruising practice?

The problem with your Link is that it needs to be recalibrated for real capacity of the battery bank, which is changing all the time. It can't do that automatically. And you can't do it in any straightforward way. You would have to do some laborious testing of capacity.

And besides that, there is not a linear relationship between amp/hours used and state of charge -- because of Peukert.

So the Link, like other amp-counting meters, does reset itself to 100% SOC when there's a full charge, but figures below that are a vague guess, and besides that, unlike reading system voltage, the errors in this method tend to go the wrong way -- they tend to overstate the real SOC of the batteries which may lead you to discharging them more deeply than you intend.


Concerning the SmartGauge -- it is supposed to "learn" the bank over time. I have not personally noticed any learning going on. It has tracked very closely my simple SOC vs voltage chart, since day 1.

[Paul L]

Originally Posted by Dockhead

Well, sort of. The Smart Gauge, to the same extent as an intelligent interpretation of simple system voltage, gives a very accurate picture of the state of charge when the battery is discharging. Far more accurate than amp-counting meters in my experience and also according to MaineSail's testing.

It's only when the battery is being charged, and the charge is not carried on to a full charge, that it gets confused.

But I do agree that an amp-counting meter is very useful for seeing the load on the system, charge rate, and cumulative amount of power used -- just ignore then SOC calculation, which is much less useful.

I haven't used one. Does the bolded sentence above then mean that all day long when the solar is charging that the device is inaccurate? Daytime is usually when I decide to additionally charge or not.

[conachair]

Originally Posted by Dockhead

- just ignore then SOC calculation, which is much less useful.

The SoC reading does seem to drift if you aren't getting back to 100% every day and using the autoreset function, but then if you aren't getting back to 100% very regularly then your batteries will hate you

[a64pilot]

I only rarely don't fully recharge my batteries as I'm spooked as they weren't cheap and are AGM's. If anything I'm overcharging them slightly. I watch for the .5% of capacity charge rate at absorption voltage that indicates full charge, for me a 660 AH bank, that is only 3.3 amps at 14.3 VDC, and that takes quite a few hours to get to.
Supposedly the Smart gauge does not indicate well on charging, but mine seems to, my charge sources will kick out to float voltage usually around 93% or so and I have to reset them to get them back to float voltage and that 3.3 amps with parasitic loads happens right at about when the Smart Gauge indicates 100% SOC.

I guess the thing is learning, can't really tell as it seems accurate right out of the box and I don't have anyway of knowing actual SOC, I get an idea from voltage, but that's just a ballpark is all.

[Rapanui]

I have a SmartGauge and it does not exhibit the the behaviour Dockhead has observed. Whether charging or discharging with light or heavy loads and charging, it's SOC indication seems from all my volt and current clamp measurements spot on to what I would expect and exhibits no anomolous behaviour.
If the SG pos and neg connections are not directly on the battery terminals, SG state the readings will be eroneous. So for accurate SOC you should have the batt 1 positive and common negative directly on the same house bank battery. Another thought I have had is you may get erroneous readings if the house bank is not wired up to ensure the charge and discharge currents are not equal for each battery in the bank. A quick check running the inverter on load with a clamp on current meter could verify that. SG and others publish papers on wiring banks. Apologies if this is teaching you to suck eggs.
Failing that, contact Merlin

[zboss]

Originally Posted by Rapanui

I have a SmartGauge and it does not exhibit the the behaviour Dockhead has observed. Whether charging or discharging with light or heavy loads and charging, it's SOC indication seems from all my volt and current clamp measurements spot on to what I would expect and exhibits no anomolous behaviour.
If the SG pos and neg connections are not directly on the battery terminals, SG state the readings will be eroneous. So for accurate SOC you should have the batt 1 positive and common negative directly on the same house bank battery. Another thought I have had is you may get erroneous readings if the house bank is not wired up to ensure the charge and discharge currents are not equal for each battery in the bank. A quick check running the inverter on load with a clamp on current meter could verify that. SG and others publish papers on wiring banks. Apologies if this is teaching you to suck eggs.
Failing that, contact Merlin

Then your device is behaving differently than even the manufacturer says it should... its specifically says that the device does not read correctly when system is being charged.