LA Batteries Failing on Victron System

[PhilandMena]

Looks like the test is not working, then. If the voltage of 12.56 represents 90% charge level, then the theory says that the 23Ah that you drew from the battery represents 10% of the capacity, so the actual capacity should in theory be 23Ah/10 x 100 = 230Ah. It's supposed to be 100Ah. No idea why it didn't work. The rate of 4.6A is very near the ideal 5A rate which would discharge it in 20 hours, so it's not related to the Peukert factor.

Considering how I have used these batteries since new in 2019 (a combination of wilding and hook up (including245 W of Solar), the loss of an estimated 14.5 % battery capacity feels about right. However, I have no knowledge of how battery capacity is lost over time and the fact the batteries more or less fully recovered after 5 hours with a 4.5 Amp drain on them gives me confidence they still have a long way to go. When I was younger I had national qualifications in statistics and even managed some high end advanced diplomas and degrees when 40 + . Today, I can hardly tie my shoe laces never mind getting to grips with electrical theory. I can only assume my batteries must have come out of the star ship Enterprise, if you maths calculation is correct. Live long and proper.

 

[PhilandMena]

Looks like the test is not working, then. If the voltage of 12.56 represents 90% charge level, then the theory says that the 23Ah that you drew from the battery represents 10% of the capacity, so the actual capacity should in theory be 23Ah/10 x 100 = 230Ah. It's supposed to be 100Ah. No idea why it didn't work. The rate of 4.6A is very near the ideal 5A rate which would discharge it in 20 hours, so it's not related to the Peukert factor.

Is it not the case that 23 Amps from a 100 AH battery is 23% and not 10% ?

 

[PhilandMena]

That chart seems a bit optimistic, I always use this one.

Another thing to consider for the wide chart voltages, is acid mix. Apart of the various dopings calcium, carbon etc., it’s the acid specific gravity mix. The higher the concentration is, the higher the voltage reading. A agm can have 1,80 specific gravity acid, while flooded can be as low as 1,24. A hoppecke stationary tubular cell uses a acid mix of 1,24, that will give a 12,65v rested voltage fully charged. Trojan, Rolls Surrette, flooded uses 1,27 acid and the 12,65v + is considered fully charged. Also the surface charge to disparate fully, for a true reading, can take as long as 24hr rest depending on temperature. Automotive and leisure flooded, use a 1,28 acid mix, that’s exide, Yuasa etc. and a 12,7v + is fully charged.
There are advantages and disadvantages on both sides, a high mix gives good voltage reading, more stable on voltage sag, less resistance but, at the expense of high rapid corrosion and shorter life. A weak mixture gives lower voltage, but less corrosion and long life. The later is mainly found In stationary applications with big cells, with design life of 15-20 years.

My thoughts exactly !

 

[autorouter]

Is it not the case that 23 Amps from a 100 AH battery is 23% and not 10% ?

Not in this case. We are trying to find what the Ah capacity of the battery actually is, thinking it might have reduced from its brand-new 100Ah value. So you look at the chart, and estimate what its charge level is. If the chart had shown for example that the battery was at 60%, then you could say that a 23Ah loss takes it to 60% (instead of the 77% you'd expect if it were brand new).

What's wrong is that using that chart, the charge level estimate is about 90%, when the expected level is 77%. So it seems to have increased, not decreased. I don't know why that is, so perhaps the chart is wrong, or there's another error somewhere.

 

[Raul]

Good point, the chart may be out for that battery indeed. Another work around it, is to follow a chart that corresponds to fully charged voltage. If charged tested voltage is near 13v or above, then that chart is not good reference. Bear in mind this is only a wild estimate, voltage can be influenced by many factors. The true test is total capacity discharge measured, or if access to electrolyte, a good temperature corrected hydrometer.

^{_{Subscribers  do not see these advertisements}}

 

[PhilandMena]

So! On the basis of my test, it may be a complete waste of time doing a capacity test on your leisure batteries!

 

[Raul]

I can’t see how can you estimate the remaining capacity by only doing a partial discharge test. You rely on voltage to guesstimate the remaining energy. Maybe if the partial discharge is taken deeper where voltage changes quite a bit, like 50% soc, it may yield a better indicator of remaining capacity.
Also, is the discharge done with a C/20 load? If the batt spec is 100ah at C/20, then it should be discharged with exactly 5a. On a bench you could do that, as it will adjust the voltage to maintain the 5a till end of test, regardless of voltage drop. With a 55w bulb, you will draw 4,4a at 12,5v under load, and will only increase the amps as the voltage drops. It’s close but this errors will build up and throw the calc of. I’ve been there and that’s why I stay away of any sealed lead, except motorbike and atv. Open flooded are the best and most robust lead money can buy, for me anyway.
The battery type with the indicator eye, they are actually flooded semi sealed, that you can open and service. The eye is a float in one cell and it sinks if discharge, or float when charged, making the green dot visible.

 

[weemalkie]

contact A and N caravan services. Describe your system and problem and Martin will give you an expert opinion free of charge. He is extremely knowledgeable about all things battery and charging.

 

[PhilandMena]

I can’t see how can you estimate the remaining capacity by only doing a partial discharge test. You rely on voltage to guesstimate the remaining energy. Maybe if the partial discharge is taken deeper where voltage changes quite a bit, like 50% soc, it may yield a better indicator of remaining capacity.
Also, is the discharge done with a C/20 load? If the batt spec is 100ah at C/20, then it should be discharged with exactly 5a. On a bench you could do that, as it will adjust the voltage to maintain the 5a till end of test, regardless of voltage drop. With a 55w bulb, you will draw 4,4a at 12,5v under load, and will only increase the amps as the voltage drops. It’s close but this errors will build up and throw the calc of. I’ve been there and that’s why I stay away of any sealed lead, except motorbike and atv. Open flooded are the best and most robust lead money can buy, for me anyway.
The battery type with the indicator eye, they are actually flooded semi sealed, that you can open and service. The eye is a float in one cell and it sinks if discharge, or float when charged, making the green dot visible.

When I was in the army we had a saying BBB Bullshit Baffles Brains.