1 minute ago, plonkster said:

No, not your bad. Mine!

We keep each other on our toes. I like that!!! Victron Rules! 🤣

 

Looking at the Freedom Won range the comparison I look at is:

12v 200ah J200RE's (I have a watering kit) at +-R 5 300 per batt with a nearly guaranteed 1600 cycles to 80% life (I'm told).

versus

12v 100ah Freedom range at R 12 940.00 per batt and a "estimated" >2000 cycles - estimated as the companies have not been around since 1925.

And both the above including free delivery - T&C's apply.

So I can go and either:
Drive them hard for +-5 years and then buy another set, and still pay less, with no funny touchy voltages or no hard discharging.
Or I can nurture them for 8-10 years and then buy another set.

 

Then there is also this titbit: http://large.stanford.edu/courses/2010/ph240/eason2/

I could hazard a thought and wonder if lithiums will become more expensive than lead acid in due course, with lead acid pricing falling to compete, as Victron did with Voltronic. 

22 hours ago, Noobie said:

I am going to avoid using the kettle or micro when we have load shedding

Been my motto ever since I started. For us kettle and micro's have never been a need like in ever.

If we have outages daytime and I want to use the kettke / micro, solar is more than enough. Batts help a bit, re-charge moments later.

And if it is extended at night then we have the Cadac camping stove plates, have had them for yonks, re-using them now if we need to. 😉 

EDIT: The savings for us are in the grid tied solution, now THAT is awesome. Keeping stuff going daytime with panels when there are failures.

Edited February 18, 20197 yr by Guest

20 minutes ago, The Terrible Triplett said:

12v 200ah J200RE's (I have a watering kit) at +-R 5 300 per batt with a nearly guaranteed 1600 cycles to 80% life (I'm told).

versus 

12v 100ah Freedom range at R 12 940.00 per batt and a "estimated" >2000 cycles - estimated as the companies have not been around since 1925. 

Always check the spec sheet. That 2000 cycles is at 100% DoD and C/5 discharge rates. Your Trojans are good for around 700 cycles at the same DoD and certainly won't like a C/5 discharge every day. Scroll down to the bottom of the page and it has a nice chart showing life at other DoD levels, 3000 at 80%, 5000 at 50%.

Now keep in mind that LFP batteries are EOLed at 80% (20% capacity is irreversibly lost) while lead acids are EOLed when they can no longer do the job (so if you sized it at 50% DoD... it is EOL when it has lost 50% of its original capacity). So I don't think it is an exageration to say the LFP battery will last 5 times longer than the Pb if they play by the same rules AND if the claims of the manufacturer holds up... which is rightly something that we're only beginning to see data on (but it looks good).

It only has to last twice as long to make economic sense. Yeah, I know 5300 times two is not 13k... but when you replace your Pbs in 2 years time... the price will have gone up. Times two is close enough... no watering... no mess.

But you can't put those 12V batteries in series to make 24V. Or maybe you can, but don't just assume that you can. They usually come with a small BMS that internally holds the 4 cells inside in balance, but the moment you series two, there is no way to balance cross-battery.

11 minutes ago, plonkster said:

But you can't put those 12V batteries in series to make 24V.

You can. I quote them:

Freedom Won 12V Configuration Notice: Parallel Configuration: Can be connected in parallel to unlimited numbers of the same model provided that the voltages of all units are similar to each other (within a delta of 0,06V) prior to being connected.

Series Configuration: Can be connected two, three or four in series (maximum of four units) to provide up to 52V nominal (for 48V systems), provided that each unit is fully charged to 14,2V prior to connection in series. It is suggested that each unit is connected in parallel first, charged to 14,2V, and then reconfigured into series connection. If additional capacity is required multiple units can be connected in parallel to form a parallel group, then up to four of these groups can be connected in series. Each parallel group must contain the same number of units of the same model and each unit in a parallel group must be connected with all positive and negative poles connected to each other respectively (i.e. separate parallel strings that are only connected at the pack negative and positive ends are not recommended. 

Please reach out if you require more information on anything to enable your ordering process.
 

15 minutes ago, plonkster said:

... if they play by the same rules AND if the claims of the manufacturer holds up...

Ditto.

As I said before, as sound and absolutely right it is ito doing the maths ito cycles and cost, don't forget, giving the one side is not the full picture.

Other maths that apply: 2 x R 12 940.00 versus 2 x R 5 300 is a huge decision maker, no matter how long the more expensive ones may last.

And if one really want to or need to make ends meet right now, by keeping batts charged all the time bar power failures, then R 1 800 x 2 is a very attractive option to keep in mind.

Double the above for 48v obviously.

The specs, these titbits on the Freedom Light Mini:
10 years product warranty with 70% DoD average per cycle up to 3 500 cycles (whichever occurs first).
And they specify the max VA size of the inverters.

Still saying that 1600 guaranteed cycles out of Trojan's to 50% ot +-4 years (or 4000 at 20%), with price increases nowhere near what we expect over 5 years, that 2 sets of them giving me 3200 cycles at half the cost with AGM's a tiny bit more at 1700 x 2 = 3400 cycles with no mess, no funny voltages, is not something we must just throw out in favour of maths that may or may not turn out to be the nirvana we all are so looking forward to. 🙂 

Just saying that there are more options open than everyone forced to by Lithium banks.

1 hour ago, The Terrible Triplett said:

provided that each unit is fully charged to 14,2V prior to connection in series

You could probably use a battery balancer to make sure they are at similar voltages during charging.

In general, I actually buy your argument. It's the same reason I own so many Ryobi tools. I absolutely know that Bosch, Makita and Metabo is better and will last longer... but that was not the primary concern when I bought them. The goal was much more short-term (I need it now!) + intermittent use-case. Which sounds a bit like what you describe.

I even have a Schultz belt-sander I bought at Checkers. Paid R500 new. I was sanding down a big double garage door and figured if it lasts just long enough to finish this one job it's worth it... so far it has lasted 3 jobs. Amazing little thing, considering it is a complete POS from PRC (People's republic of China... ie not the more high-end Taiwan side we've learned to love and accept 🙂 ).

Edited February 18, 20197 yr by plonkster

1 hour ago, plonkster said:

I actually buy your argument.

Thank you.

Just hope others may also give it a moment of thought first too.

As you say, sometimes the el-cheepo's do the job far beyond our expectations, even like the "high-end Taiwan side we've learned to love and accept" yes. 😂

Edited February 18, 20197 yr by Guest

13 minutes ago, The Terrible Triplett said:

As you say, sometimes the el-cheepo's do the job far beyond our expectations, even like the "high-end Taiwan side we've learned to love and accept" yes. 😂

American components. Russian Components! All made in Taiwan!

(That movie had some funny parts in it. The kids look at you strangely when you use it... the wife smiles... just this weekend as one kid dropped something, my response was (in heavy russian accent): That's why I told you, touch nothing...but you bunch of cowboys...").

Edited February 18, 20197 yr by plonkster

Using a kettle or similar load and then measuring voltage of individual batteries in my 2 x 8 T105 in parallel seems like a cheap and easy way to to test my batteries. Does anyone have an opinion on the effect of the HA02 balancers. Will they supply enough current to equalise a faulty battery. Perhaps someone could suggest the load I should apply for the 450AH bank. It would just be such a lot of work to disconnect the HA02 balancers. If the amps drawn exceeds the capacity of the HA02's then that may prevent them from equalising a faulty battery so that the voltage is the same as the healthy ones.

The balancers will be able to delay the inevitable, but if a battery is faulty they will not be able to 'fix' the faulty one.

Do you suspect you have a dead battery?

@DeepBass9 thak you for your reply. The problem that I am concerned about is that the HA02 balancers may make it impossible to measure the voltage of the individual batteries in circuit with a large drain current of say 50A or more. I would expect that a faulty battery will show up as having a lower voltage than the good ones. But the question is whether the HA02's will supply enough current to a possible faulty battery, if there is one, to show a lower voltage than the rest. And all that presupposes that one could even try and find a fault battery in this manner as it is just such a tedious job to disconnect the individual batteries from one another.

I also have a strange problem with the way the BMV700 controls 100% charge. Charge voltage on the BMV700 is set at 58.2V which is 0.2V below the maximum charge voltage of the Axperts which is what @plonkster suggested on some past post on the subject . Tail current is set at 1% and detect time at 20 minutes for my 450AH battery bank. Even so, according to the battery graph of ICC, the charge voltage stays at 58.2V for only a minute or two before dropping to 54V which is the float voltage of T105's. Then during the night the charge percentage slowly drops to maybe 98.5% or so in the morning. With the present winter cloudy weather I run off grid during the night with the Axperts set to charge from solar before grid. So as soon as no solar power is available when it is sufficiently dark, the grid chargers will top up the batteries to 100%. To overcome the batteries being undercharged, I have set setting 32 to 60 minutes, thereby forcing the chargers to stay at 58.4V, the Axpert's maximum charge voltage, for that time. So the BMV does not actually control end of bulk charging but that seems a bit brute force to me and defeats the purpose of the BMV. I would appreciate any and all advice.

The balances won't balance such a large load, if there is indeed a faulty battery. If you think that is a problem though, just disconnect the balancer and measure.

On 2019/07/01 at 5:26 PM, ebrsa said:

way to to test my batteries.

Why are you wanting to do it, just for maintenance or because you suspect something?
Asking because maybe you are same boat as me, going on 4.5-5 years, we find the T105RE's are slowly showing usage.

16 hours ago, ebrsa said:

I also have a strange problem with the way the BMV700 controls 100% charge.

You are syncing the SOC manually every now and then right?
Checking that the zero volts is still zero volts?
You know, the maintenance that must be done every once in a while to ensure the BMV's input values are still the same.

Thanks a lot for your reply @The Terrible Triplett. Using Trojan T105s yourself, you are obviously in a position to give me first hand advice.

My T105 batteries are marked T105 with T2 Technology, whatever that means. They are almost 3 years old. I use them daily but not below 20% DOD. They are also recharged fully every day. My 2 x Axpert's grid chargers are set to charge at 20A each for 60 minutes so every evening the batteries are charged to 100% even if solar did do it during the day. At present with little solar power available, I frequently switch to grid manually using ICC on a Raspberry Pi so the batteries are rarely discharged below 90% DOD. I would prefer the BMV to feed the correct information to ICC-Pi to do the switching. Problem is that the Axpert maximum charge voltage is 58.4V and the T105 bulk charge voltage is 59.3V. Trojan USA advised that I charge for a longer time at 58.4V which is happening now with the 60 minutes brute force charge settting on the Axperts (setting 32). 

I have disconnected the BMV totally at times and have done manual syncing occasionally. Perhaps what is needed is an equalisation charge at 58.4V but for a full day or so. The SG of the cells all seems to be fine. What bothers me is that the batteries charge to 100% at night from grid and then slowly the SOC, as displayed by the battery graph of ICC, starts falling to about 98.5% in the morning. The voltage of the bank remains very close to 54V though. I have noticed that the Axperts draw as much as about 145W from the batteries when running off grid. This happens every monrning when the kettle of 2000W is turned on when still on grid. The battery graph shows a sudden drop to about 52V and the inverters draw 145W from the batteries. Perhaps @Coulomb would have an explanation for the sudden inverter consumption from batteries while still on grid.

I am just trying to find a way to ensure that I don't have a battery that is failing without having to disconnect all 16 batteries' connections as well as all the HA02 connections too. So I would be most interestd to hear how you maintain your T105s and also if you do occasional equalisation charging. Looking forward to your advice and that of anyone else with proposals for settings of the BMV700 and T105 maintenance.

11 hours ago, ebrsa said:

inverters draw 145W from the batteries

The Axpert inverters don't have a sensor for battery current. So they calculate it based on the difference between AC out and AC in, a fixed presumed efficiency, and a fixed presumed idle power. These figures aren't all that accurate, especially for unpatched firmware, but not much more can be done without an actual battery current sensor.

However, the BMV has a real shunt. Does it show this battery draw during kettle use?

19 hours ago, ebrsa said:

.. you are obviously in a position to give me first hand advice.

What Coulomb said. 🙂 

I don't think you have anything to worry yet. The batts ARE GOING TO GO one of these years. If they gave you 5 years, good. Anything more is a bonus I believe. I don't see Trojan's lasting 10 years on 20% DOD anymore no. Too many variables, the worst one being temperature and the death of lead acids, watering.

You do fill them regularly? If so, are you using more water?
Learn recently that when a bank starts using more and more water, that that is a sign of impending replacement.
Titbit: What annoyed me extensively recently is that I was not given distilled water, but deionised water, water that allows stuff to grow in it over time, and because I did not see that, the "growths" where pumped into the batteries for a while. I believe that that is why mine is using more and more water pre-5 years.

Trojan does say equalization is to be done yes.
I do mine every few months. As a matter of fact I'm doing it today after having filled my 4 batts with about +-2.5l of distilled water.
My batts are on wheels so I also shake them around a bit when I "service" them. Can hear them "talk" back.

 

I am not aware of any other method other than testing each battery individually to find out if anyone is getting weaker. But here is the thing. Say you found such a battery, what then? Is bliss not maybe preferred? I mean, if the batts do start going, it is EOL and that's that.

And then a thought popped to mind. If you start suspecting anything around the bank, there is a sweet spot where you could contemplate to sell them on to someone wanting to buy a 2nd hand set of T105's? You sell them at a price that is fair enough for you to off-set a new Pylontech bank and the new owner pays knowing full well they are not getting new ones.

Edited July 7, 20197 yr by Guest

Than you @Coulomb for your explanation. My Axperts are on grid at present and sun has just set so zero solar power. The batteries are at 94.1% but it is not dark enough for the Axpert grid chargers to kick in. I just checked the BMV700 and it shows -2.6A as does ICC and 145W drawn from the batteries. My firmware on both Axperts is the lastest version of yours and @weber firmware. I guess I will just have to accept that it is normal operation of the Axperts.

Thanks also @The Terrible Triplett for your advice. I regularly check the electrolyte level in my batteries and fill them about every 6 weeks. At that time the electrolyte is still reassuringly above the plates. I do not have a watering kit like you but have a filling jug which has a spout that fills to the correct level and shuts off when you lift it out of a cell. I guess an equalisation charge may be called for and could only be beneficial. I buy my distilled water from the local pharmacy (R60 for 5Litres) so hopefully it is the real thing.

At 81 it is a toss up who will last longest, me or the Trojans although I intend to my damnest to win the race. Lead acid works well for me as there is all that reserve power below 80% DOD down to 50% DOD which how far I have discharged before starting the generator on an occasion when Eskom blacked out for almost 2 days.

For how long do you do equalize charging or do you measure SG or have some other method of determining when to stop. Of course I have to find some sensible way that will work as my chargers can only reach 58.4V instead of the 59.3V that Trojan specifies.

 

6 minutes ago, ebrsa said:

For how long do you do equalize charging ...

I'm not sure, the kit I have does it for me. Maybe Google your model T's and see what they suggest?

On 2019/07/06 at 2:59 PM, ebrsa said:

I use them daily but not below 20% DOD

when have you last measured the actual capacity of your battery bank? I did a capacity test (0.1C down to 1.8V) last week on my 2 year old bank and was shocked to find that I have already lost 30% of the capacity ☹️ . I have never adjusted the BMV battery bank capacity settings and therefore used the wrong DOD values to manage the batteries. As a result I discharged my batteries way below the intended 40 DOD value 😱 .  

@Fuenkli the short answer is never. Perhaps it is time to do that. I don't have a situation that the batteries discharges noticibly faster that before but am just trying to establish the state of the batteries. Also I would like to know if a battery is failing rather sooner than later. Now to research how to actually do a capacity test. Can't be too complex I would expect.

Once upon a time I had Trojan T145 batteries. I remember when they started to go, they would appear to be okay, and then around 9PM one of them would just take a dive off a cliff. The overall voltage would hang around 24.5V (24V system) and just go down to 23V in a matter of minutes. When you see that... you know it is the end.

In the end, the battery that had the best SG values over time turned out to be one of the first ones to fail. Batteries are truly interesting beasts.

1 minute ago, ebrsa said:

Now to research how to actually do a capacity test

all did is basically discharge the batteries using the inverter. You have to find the right load to achieve +- C0.1. Our battery banks are almost the same size (mine is 20kW/h) I discharged with +- 1.5kW. Then measure the voltage on all batteries until they reach 1.8v per element and read the Ah value on the BMV. This test will also show you any differences between your batteries based on the voltage they show during the discharge. During my test all the batteries discharged evenly so my capacity loss is the same on all batteries.

@plonkster perhaps I should start with an equalisation charge of 12 hours at 58.4V and see if all the batteries are the same SG. No use trying to meaure voltages as the HA02 will defeat differences I guess unless a battery is truly dead. But overall voltage of the bank seems fine.

Thanks a lot for the advice  @Fuenkli. I will do an equalisation charge during the week and then try your test. Much appreciated.