absorption stage for lithium

[exscott]

Hi everyone,
I have been playing with a lithium bank of 160a/hr winston cells and are trying to fine tune the charging process. On the load side they are unbelievable how long they can run the 2 waeco portable fridges but they seem to be spending all there time almost full with solar just topping them up each day. This has got me thinking that they are spending too much time in the top of their charge so wanted to try and reduce the time spent in absorption with the solar. I have a victron bmv700 victron, mppt75/15 and a victron 50a buck boost that is set to charge at 14.4v but is blocked by the bmv once the battery reaches 14v. This allows it to charge at 20amp (reduced amps due to crappy supply wire causing volt drop) right up till full charge and this seems to work great with no absorption but when solar charging the reg spends a long time in the upper voltages during absorption.
So my question is do you think I am doing excessive damage while the solar is in absorption and if so how are others using solar without this period of absorption.
cheers
Scott

[Firegirl54]

Hi Scott,
The Victron mppts are fully programmable. Its all explained on the Victron website, but it is a bit cryptic to navigate. If you get a victron VE-direct bluetooth dongle (about $70 off fleabay) and install the victron connect app (free) on your phone you can set up a custom profile. You can set absorb volts and duration (zero if you want), float volts (as low as you like which can effectively turn off the controller), and cancel temperature compensation. Then in live view you can see what the controller is up to while you drive along! Quite fun really... No doubt others will contribute re preferred voltage settings, but I think the default Victron ones are somewhat high for a DIY lithium set up.
We use a Victron 100/30 unit and it performs as programmed, currently we use 13.9V absorb, 30 min max ( but to get that you have to select 3 hr max, because if bat voltage at start is over 12.5, and this is almost always the case with lithiums, the controllers apply a x1/6 correction), float 13.4 (which just basically offsets usage till sundown if the bats reach full charge, and they only do that on a good sunny day if parked in full sun...)
Hope that makes sense!
Cheers, Leslie

[exscott]

Thanks firegirl yep got the app and seem to spend most of my time checking it pretty surprised at the range of it too. What I am finding though is the controller is reducing output as it gets close to the absorb voltage cutoff of 13.9v atm. So for your absorb time what does your terminal voltage get to? My controller seems to go from bulk to absorb when terminal voltage is 13.7 I think. When I entered a shorter absorb time the terminal voltage didn't get to 13.7v where the bmv resets to 100%.
Maybe I could set a higher cutoff voltage and cut back the absorb time as I presume it's all about the terminal voltage while the tail current has only a small effect on SOC

[T1 Terry]

Thanks firegirl yep got the app and seem to spend most of my time checking it pretty surprised at the range of it too. What I am finding though is the controller is reducing output as it gets close to the absorb voltage cutoff of 13.9v atm. So for your absorb time what does your terminal voltage get to? My controller seems to go from bulk to absorb when terminal voltage is 13.7 I think. When I entered a shorter absorb time the terminal voltage didn't get to 13.7v where the bmv resets to 100%.
Maybe I could set a higher cutoff voltage and cut back the absorb time as I presume it's all about the terminal voltage while the tail current has only a small effect on SOC

What settings are you using for the Victron 700BMV? It is quite likely the 100% is not actually occurring when the battery reaches fully charged but rather when the recalculated Ah (not actual Ah) has been replaced causing 100% to show well before actual 100% is reached.
100$ SOC is when all 4 cells or cell groups hold better than 3.4v when the charging stops but no high loads applied, any cell reading less than that is not actually fully charged.
With lithium batteries the only problems this creates is a 100% SOC shown where there is not actually all the advertised capacity available because 100% was not actually reached, and over a long period, 12mths plus, a memory charge voltage will occur and this will also rob useable capacity. Both are a temporary problem and not permanent damage as both problems can be rectified with controlled charging constantly monitoring the cell voltages till all cells reach 3.6v and hold better than 3.5v over night with no load. You will be quite surprised by just how many more Ah go into the battery before this can be achieved and apart from the indications on logging graph it shows just how much effect memory charge voltage has in regards to capacity. Many reports of lost capacity are due to nothing more than memory charge voltage effect, yet very few understand the phenomenon or have even spent the time in research testing to see it actually exists.

T1 Terry

[Firegirl54]

Agree 100% with what Terry said!
Plus.. the controller does not have a separate voltage sense wire, so any resistance in the connection to the batteries will cause the controller to see absorb voltage well before the battery terminal voltage has, and this effect is greater the higher the amps flowing. So if you set a zero or very short absorb you will stop charging before full, you miss out on a little capacity but it does the battery no harm. If you compensate by setting a higher voltage, you will put more into the battery, but as battery volts go up amps come down and the differnce between what the controller sees and the actual battery voltage drops...you risk overcharge unless you have appropriate protections in place.

[exscott]

A separate sense wire would be handy but even then if the regulator is set at 14v the amps will still slow down as it gets near 14v. I am waiting for the battery to run down atm but the next charge will be from the dc/dc set at 20amps and 14.4 volts but I have the bmv relay blocking the dc/dc when terminal volts reach 14v. I will see if it puts back the same amp/hr as the solar set on 13.9v.
Also Terry I guess once a year carefully charging to 14.4 should control any memory loss although between 13.9 and 12.5v I am getting 160amp/hr atm which is above what I was haoping for anyway. When I saw some of your graphs Terry in another thread you seem to get full amps to full and then stop are you using a higher voltage to push the amps in and a separate controller to stop it at a set terminal voltage?
Sorry about all the questions but I guess I am in the early stages with lithium where I am playing with settings every cycle and watching what happens like a hawk.

[BruceS]

Most of 'us' simply use a 'full charge' till required voltage (battery full) & then halt charge for set time and then full charge again etc etc etc.
The old idea of lead/acid charging is NOT advised. (reducing charge as voltage climbs to full)

[Firegirl54]

Actually I find the lack of a dedicated voltage sense wire to be a major shortcoming of the Victron controllers. When we first set up the lithium with the victron we used the existing van wiring from the controller to the battery and BMS. Problems with voltage drop between controller and battery were enormous. With over 10 amps flowing, it could hit 14V at controller while battery was only 13.4! That triggered the change from bulk (when the controller gives all it can) to absorb (where controller gives only enough so that the voltage it sees does not exceed the then 14V set point – this is why the amps drops off during absorb). Initially I had set zero absorb time and float to 13.2 (effectively turning off the controller). Got quite a shock to find it going to float at less than 70% full. We only had 100 Ah of lithium then, so that was a problem. Replaced and rerouted the connections to battery with 16mm sq wire and got voltage diff down to a more respectable 0.2V, usually drops to absorb at about 90%, but still not ideal. Seems like the controller just has to see the set voltage for an instant to drop out of bulk...little spikes like sun coming out behind cloud, or fridge cycling off are enough to do it, and thus kill optimum solar charging, regardless whether you try to compensate by setting a non-zero absorb time (and of course then you can risk overcharge without other controls in place). You could always kick it back to bulk by turning the solar off and on again, but that wears a bit thin after a while!
Anyway, now we have 200Ah, so no problem if it drops out early. And as you suggest, a controlled full charge now and again should sort out any memory effects.
cheers, Leslie

[Firegirl54]

Just noticed the time on the post. I don't really sit up past 1am typing....its actually 5.30 pm Sunday here...We are on a cycling holiday in Germany, and just relaxing in the hotel after a pleasant 55 km trip today along the Elbe. The lithiums are at home at 50% and disconnected.....

[Newcastle George]

Leslie,
As you are cycling in Germany , you may be interested in this blog from a couple of German cyclists we met in Victoria. They are just about to leave Australia heading home. It is a very interesting and informative blog.

George