Victron Disrepancy

[leedav]

I really like the Victron, as I’ve said in other posts it’s a complete game changer. But... camping over the past few days, woke up and the Victron said “battery 86%.”

It also said I used 33 Ah. That calculates to 80% remaining of my two 81 Ah batteries. I checked the settings on the Victron and it’s set at 162Ah.

Any thoughts on how the Victron arrives at the 86% number?

[pcskier]

I have given up on the percentage readout. 80 percent calculated vs 86 displayed is actually pretty good compared to my results. I’m sure Bob will weigh in here with a technical answer on how it works and how it’s really just a guess at any moment in time. I’ve learned to ignore it and just monitor voltage instead. That’s the number that matters to me. Same way I look down the toilet to see black tank status rather than rely on the tank monitor :-)

[leedav]

Right now if I had to choose I'd rely on the Ah reading. The voltage, in my view, is just more precise but no more accurate

[wulfraat]

Without getting too crazy on the math, this is the basic gist of it:

% Available = 1-(AH Usage / Total AH Available)

This value is calculated and updated every second or so, time-wise, and the resulting value is made available on the BMV monitor and VE.Bus devices such as CCGX.

AH Usage value is dynamic based on energy flowing in and out of the batteries as measured by the shunt. AH Usage will increase if you are discharging, or decrease if you are charging. Current (amps) flow in/out is measured at the shunt, and then calculated based on a 60 minutes to get amp hours. Can go further on the math but not needed for explanation purposes.

For example, if you have discharged 25AH of energy from a battery bank over some period of time (5 minutes, 30 minutes, 2 hours, whatever) with 200AH total (as set in the BMV), usage % would be:

1-25/200 = 87.5%

AH is a much better and more precise measure of battery usage vs. voltage. Many elements can affect voltage to give you false measurement of how much energy has actually been depleted from the batteries.

For instance you may read 12.6v for a fully charged battery in the day, sitting on the ground with nothing attached to it. At night, with temperature drop, voltage might read 12.4v, even through no energy has been depleted from the battery.

[wulfraat]

Originally Posted by leedav

I really like the Victron, as I’ve said in other posts it’s a complete game changer. But... camping over the past few days, woke up and the Victron said “battery 86%.”

It also said I used 33 Ah. That calculates to 80% remaining of my two 81 Ah batteries. I checked the settings on the Victron and it’s set at 162Ah.

Any thoughts on how the Victron arrives at the 86% number?

One of two things:

1. You did not set 162Ah on your BMV as battery capacity and so BMV is not displaying correct %.
2. You have set battery bank capacity correctly, but there was some energy in-flow into the batteries from solar or what have you that slightly offset the 33ah used.

Victron will not show you NET in the AH used value - they will show you GROSS AH used.... AH added back via charging sources is a different value. You need to know both to obtain NET AH.

In your example, your numbers tell me that your NET was likely ~23AH used. (33AH depleted - 10AH charged). (1-23/162) = ~86%

[AirMiles]

I am not a fan of battery monitors. They precisely measure the AH output from an unknowable pool of AHs. This article goes in depth on why I feel this way about battery monitors: https://marinehowto.com/programming-a-battery-monitor/ . Its a great read for those with battery monitors.

Its not that battery monitors are inaccurate, its just that nobody knows the actual number of Amphours in their battery bank. Just like voltage is inaccurate when the temperature changes, your battery bank's available AH change too. Sure we know the lab rated, 77 degree number of Amp hours. Are your batteries new? Is it 77 degrees outside? How abused were your batteries in the past . . .

I use my batteries like propane exchange tanks, when they run out of juice I'll just replace them. I buy a cheap pair ($250 total) of golf cart batteries, run the furnace all night, fantastic fans all day, use the inverter to watch Nascar races and Dancing with the Stars, play the radio all day, use lights and pumps as needed. When my batteries can no longer do these things, I'll just run over to the nearest Sam's Club that sells 6V golf cart batteries with $250 and install a new set.

I get it if you have an expensive bank of Lithiums. I'd sweat the details too if my replacement cost was thousands of dollars. That's why I choose to use a $250 battery bank. A senior moment that kills my batteries and costs me $250 is just a bad day.

[rmkrum]

Yeah to that. Missed a switch setting on my MPPT solar controller and boiled out a set of GC-2 batteries in a few months from excess charging voltage. Lesson learned. Cleaned up the mess, installed new batteries and drove on.

I will use a shunt-based monitor, but voltage is my fast check.

Installed a battery watering system and just make sure I schedule a visit to the trailer once a month in storage to add a few ounces of distilled water to top them off.

[leedav]

Wulfraat more math please. The formula you cite is the same as I used to get the 80 percent. The part I don't understand is the 86 percent. I double checked the battery size setting, I have no solar, and nothing is connected outside the shunt except the unused Zamp connector. I would expect the meter to do a simple integral of Idi/dt and plug it into your formula above. But there's apparently some spin on it.
Oops edit minus peukert losses

[leedav]

Maybe some other setting is wrong?

[sgschwend]

Not sure if anyone has suggested that AH is not a accurate way of describing energy. Which is often defined as one amp delivered for one hour (the actual battery voltage is not taken into consideration).

If possible try using watt hours which is actual voltage times current delivered over time. I would guess that Victron graphs both voltage and current then multiplies the two to determine wattage per a short time period, for example one minute. The sums after 60 minute and divides by 60 to come up with the energy in watt hours.

[leedav]

Originally Posted by sgschwend

Not sure if anyone has suggested that AH is not a accurate way of describing energy. I
f possible try using watt hours which is actual voltage times current delivered over time.

Maybe... But power is also i²R across the shunt, and R does not change appreciably. BTW please I'm not trying to be a butt. (wife says I don't have to).

[sgschwend]

It is still related to the voltage and in a lead acid battery that can range nearly 8 percent.

Good luck to you.

[wulfraat]

Originally Posted by leedav

Wulfraat more math please. The formula you cite is the same as I used to get the 80 percent. The part I don't understand is the 86 percent. I double checked the battery size setting, I have no solar, and nothing is connected outside the shunt except the unused Zamp connector. I would expect the meter to do a simple integral of Idi/dt and plug it into your formula above. But there's apparently some spin on it.
Oops edit minus peukert losses

Peukert Exponent is coming into play in terms of efficiency of energy delivery to the load as it relates to stored capacity. Given that you have the correct battery AH entered in BMV and no source of charging, assuming other items are not coming into play, it sounds like you had a relatively low current draw down of those 33ah, which helps to preserve battery capacity, vs. a heavy current, shorter timespan of the draw-down.

Here is a good thread:

https://community.victronenergy.com/...sumed-off.html

If you care about the formula that BMV uses in addition to other calcs:

T = C / In (or C = T * In), where C is theoretical capacity (in amp-hours, equal to actual capacity at one amp), I is current (in amps), T is time (in hours), and n is the Peukert number for the battery.

[leedav]

Thank you! That's even better news for the Victron!

Originally Posted by wulfraat

Peukert Exponent is coming into play in terms of efficiency of energy delivery to the load as it relates to stored capacity. Given that you have the correct battery AH entered in BMV and no source of charging, assuming other items are not coming into play, it sounds like you had a relatively low current draw down of those 33ah, which helps to preserve battery capacity, vs. a heavy current, shorter timespan of the draw-down.

Here is a good thread:

https://community.victronenergy.com/...sumed-off.html

If you care about the formula that BMV uses in addition to other calcs:

T = C / In (or C = T * In), where C is theoretical capacity (in amp-hours, equal to actual capacity at one amp), I is current (in amps), T is time (in hours), and n is the Peukert number for the battery.

[GSansoucie]

Originally Posted by wulfraat

Here is a good thread:

https://community.victronenergy.com/...sumed-off.html

That thread was very helpful. I have the same setup as the OP and am testing it right now, the % value didn’t seem to be dropping fast enough, but now I know why (I don’t have a huge load on it for my test).

[uncle_bob]

Hi

Everything on the 712 is programable. If you want as accurate a readout as possible, you need to carefully match up the settings on the 712 to *your* battery. It also helps (a lot) if you have a temperature probe on your battery (hooked to the 712 .... ).

The default settings on the 712 (just like on the MPPT) make some assumptions based on Victron made batteries. Often these are a bit wide of the mark for something you just got from Costco. Key settings are:

1) Temperature coefficient - ( under "misc" on the menu ). This is a pretty big deal. Your batteries can change capacity by +/- 30% (or more ...) over a normal camping temperature range. For +/- 40 degree (F) change from 70F that would be 0.75%/F (if the 30% applied to your batteries).

2) Charged Voltage - (under "battery settings"). This is the voltage at which the monitor resets to 100%. What this is set to depends both on your battery and your charger. 14.2V is not much use if your converter / charger only gets to 13.6V. (See the next to as well, they *also* must be true to reset).

3) Charged Detection time. ( under "battery settings" ). How long does the battery have to be at or above the magic voltage (and below magic current) to be fully charged? A spike in the voltage (or fast dip in current) should not reset things to zero. Anything under 10 minutes is probably nonsense, but it does depend on your setup and settings.

4) Tail Current (under "battery voltage"). (under "battery settings"). If the current into the battery falls below this level the battery is considered full. What you set this to very much depends on your charger and the various loads in your system. All of the conditions need to be met (time, voltage *and* current) in order for "full" to happen. When all three come true, the past current history is cleared and the display goes to 100%.

5) Charge efficiency (same) How much of the current put into the battery *actually* turns into stored energy? It's never going to be 100%. The number is going to depend on battery type, battery resistance, battery condition, and charge current. If indeed things seem to be heading out of whack, this and 8 below are what you play with).

6) Current threshold (same) If the current is below this level, ignore it. The shunt is only just so accurate. As you get to very low currents, errors mount up. Ignoring small currents is a way to get these out of the calculation.

7) Time to go averaging: (same) How many minutes does it average over to come up with your "hours left". I don't see much use for this function ....

8) Peukert Exponent: (same) The ever popular elf in your battery As you discharge a battery heavily, it may not have the same capacity as at lower discharge rates. Lead acid's have a lot of this, lithium's not so much. To be even more nutty, different electrode alloys / constructions / chemistries all get into the act as well. You can set this from the data in your battery spec sheet (if you bought a good battery brand) or you can guess ....

9) Zero Current Calibration (same) The magic brain in the unit can have a slight offset in it's current measurement. For the highest accuracy, make sure there is no current in the shunt and do a zero cal.

And no, that's not everything but it's plenty to deal with.

=====

So how does this work in practice?

Assuming you charge up to full every day, the reset stuff happily puts you back to 100% each day. Welcome to living on shore power.

If you put a big hit on the battery (lots of current for a short while) number 8 comes in and fiddles things. The amp hours and the percentage now are out a bit. Same thing with very low currents, except it's out the other direction. It's not *wrong* it simply does not add up in terms of current used vs percentage left.

If the temperature changes from 31 degrees this morning to 72 degrees right now, the capacity is going to change a bit. The basic amp hours of battery capacity has changed ( from maybe 70AH to 100AH). 10AH of usage this morning took down a bigger percentage than 10AH of usage this afternoon. It shows 20AH used, but half of that took a bigger bite.
If it had been hot this morning, it would have taken a smaller bite.

To the degree the numbers match up with your battery, the results can be very accurate. Even with a few guesses here or there you will do well. Even with the defaults, you will be doing much better than the stock display that you get with the solar package. The only exception to all that is if you set 2 to a voltage the charge system does not get to (so it never resets to 100%).

Bob

[pcskier]

Outstanding post as always Bob!

[uncle_bob]

Originally Posted by pcskier

Outstanding post as always Bob!

Hi

Now lets hope I didn't make yet another typo and get something backwards

Bottom line still is, there are a couple ways the percentage and amps get so you don't think they are right (but they are indeed correct).

Bob

[pcskier]

My 712 is showing 88 percent SOC this morning in rested state, 12.4V. That’s at 39 degrees but I do have the temp monitor. Clearly not accurate. I’ll play with settings as per your post but really, I’m worrying less and less about all this. It’s been a good ‘learning’ trip. I mainly like the monitor to keep track of powers usage ‘as’ I’m using it, particularly after the panels have gone to sleep for the night. Helps make decisions on how liberal to be with furnace, inverter tv/dvd) etc.

[uncle_bob]

Hi

That does indeed seem a bit weird.

Bob