Default LiFeP04 charge profile voltages too low?

[Al Thomason]

 

All,

 

Recently a question was posed on GitHub concerning the setpoint voltages used by default in CPE#8 ( LiFeP04).  The question was why the values were set to what seems to be low voltages.  

 

https://github.com/AlternatorRegulator/alt-Documentation/issues/2#issuecomment-438314529

 

 

Would be interested in anyone else’s insight into this.  When setting the default values, I try to be a bit conservative – but if indeed there is new info out there, would be interested in learning!

 

-al-

 

 

 

Viking Star

45' Monk Sr. / McQueen

mvVikingStar.blogspot.com

 

 

[Virjonen Antti-Pekka]

Hi Al,

 

I am using 300 Ah LiFeYPO4 cells by Winston and I think the use of Yttrium might change things slightly regarding the voltages. At least it does change the allowed temperature range for charging (down to -45 C), compared to some other LiFePO4 cell specs, which I have seen. The following thoughts only cover Winston LYP cells, other (LiFePO4) cells may and likely have different specs.

 

The enclosed charging curve picture shows the cell charge voltages for Winston LYP cells (at temperature 25 C) for different charge currents. How to detect the end of charge (battery full) situation? I think the only realistic and reliable way is to look at the point when the cell voltage starts to rise fast. I charge my 300 Ah house battery bank at 0.3 CA to 0.5 CA (90 to 150 A charge current). Following the curve, with this charge current at 3.5 V per cell (under normal temperature of 25 C) the battery pack is at 14 V and approx. 100 % full. But one can’t be sure of this, because the temperature changes and the charge current differs from the 0.5 CA in the curve. To be sure that the battery is full, I think we need to set the end of charge cell voltage a bit higher than 3.5 V.

 

In my setup I have set the regulator target to 14.4 V, but the charging will stop when the BMS detects cell voltage of 3.55 V is reached (battery pack at 14.2 V). This way I may be overcharging the battery a little, at least if going over the specified 100 % of capacity can be considered to be overcharging J. But when doing this I know the battery will be fully charged. If using lower end of charge voltage, the battery pack might end up at a much lower charge and I don’t want that. I rather have a full battery and slightly lower service life, than a lot less amp hours to use after the charging. After the charging has finished (by decision of the BMS), I let the regulator go to float with 13.6 V. This voltage is maintained indefinitely, until the engine is shut down. The charge current flowing into the battery pack is practically zero at this point (actually it discharges a little, then charges, then discharges again etc.), thanks to the close to perfect voltage regulation of the alternator regulator. I don’t see any problem letting the battery to float at 13.6 V.

 

After discharging my battery pack for example 100 Ah (200 Ah capacity remaining), the pack voltage is still around 13.2 V (at about 20 C temperatures). I have never seen it go under 13.0 V.

 

The other included picture shows the max. charge voltage of an individual LYP cell on a battery pack is 3.8 V at 25 C. The 3.55 V per cell, which I am currently using, is well below this limit.

 

I would say, at least for Winston LiFeYPO4, the mentioned 13.80 V end of charge is likely too low. If ending the charge at this voltage, the battery pack might be far from being fully charged, maybe only at 80 % capacity (depending on charge current and temperature). If the charge current is very high (like 1 CA), at 13.8 V end of charge, the charging will stop before it even really starts…

 

The most important thing is to have a quality BMS, which monitors every individual cell, and then stay within the manufacturer specs.

 

I hope this helps,

A-P

---

Antti-Pekka Virjonen

Tuotekehityspäällikkö | R&D Manager

Caverion Suomi Oy

 

Lähettäjä: smar...@googlegroups.com [mailto:smar...@googlegroups.com] Puolesta Al Thomason
Lähetetty: 13. marraskuuta 2018 17:51
Vastaanottaja: 'Smart Alternator Regulator'
Aihe: [smart-alt] Default LiFeP04 charge profile voltages too low?

--
You received this message because you are subscribed to the Google Groups "Smart Alternator Regulator" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smart-alt+...@googlegroups.com.
To post to this group, send email to smar...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/smart-alt/5beaf2f6.1c69fb81.aad22.be08%40mx.google.com.
For more options, visit https://groups.google.com/d/optout.

[Chris Jones]

Hi A-P

Some useful information from you especially since you seem to have some real world experience of using the Winston Cells. The charge curve from Winston is useful but maybe we should be a little careful - the Chinese have a reputation for recommending higher voltages than is really necessary. If we use the voltages from the chart then we should certainly get 100% charge and that is probably what Winston want us to do, however the voltage may be a bit high and might cause shorter life eventually?

What is the application you are using the cells for, obviously you are using an alternator for the main charge source but what is your discharge profile? Is it fairly constant or is it very variable?

The float at 13.6V seems to work well for you but I might set our value a little lower just to be safe. I have not yet installed our system so I do not have actual experience yet - just trying to make sure the design and voltage values are correct.

The only way to know if you are getting to 100% is to charge the batteries until the BMS indicates 100% and then run a capacity discharge test down to 11.5V at a discharge of about 0.1 - 0.2 CA. If you start at 100% then you should get close to the 300AH capacity from the batteries. Have you done any full discharge tests on your cells to see how they perform?

What do you think?

Chris

[Al Thomason]

A-P,

 

As always, thank you for your long time support as well as real-world testing  (I owe him a couple of beers – at least-  I am sure).

 

One comment your thoughts below:  

You are correct, that if one is using VOLTAGE ONLY decision points for charging, 13.8v will result in a less then fill charge.  

However, by using the VOLTAGE & CURRENT decision point capabilities and holding the 13.8v until acceptance current drops below 2-3% of battery capacity you will also achieve a full SOC condition.

 

Though a rather long and somewhat rambling read, if you scroll to the very bottom of this you can see an example of using the charge profile on 400Ah Winston cells: https://marinehowto.com/lifepo4-batteries-on-boats/   ref “Capacity Test #12 @ Cycle 550”

 

Do like that with your using the higher voltage set point you are having the BMS direct things, as it can watch out for an individual  cell out of balance.

 

 

-al-

 

 

 

Viking Star

45' Monk Sr. / McQueen

mvVikingStar.blogspot.com

 

To view this discussion on the web visit https://groups.google.com/d/msgid/smart-alt/HE1PR0301MB20114A898FC37CE28FD60A54A2DD0%40HE1PR0301MB2011.eurprd03.prod.outlook.com.

[ben]

Hello All,

As other people mention already in this forum, the lower settings is because experience from ""long time users"".
If you read for example this article from mainsail, he explains this lower settings.

(http://www.pbase.com/mainecruising/lifepo4_on_boats&page=1)

In short, on boats we charge LiFe with relatively low amps.
On boats we use the LiFePo4 completely different as the users for electrical cars.

The datasheets from the manufacture are more based for charging, as used for electrical cars.
But the newer datasheets from manufactures show also lower voltage as the datasheet from 8 years ago.

In this link: https://www.powerstream.com/lithium-phosphate-charge-voltage.htm
you see that at cell voltage off 3.4 (13.6V) the batteries are 96% full.

With the lower setting, yes, we miss a little capacity, but we will have maybe a longer lifetime.

LiFe is relatively new!

What will be the best charging voltage setting?

We will be know after another 10 years.
Laptop manufactures charge their battery also different as 6-8 years ago, because the learning curve.
Also wet lead acid batteries are charge with different voltage settings as 20 years ago, it is all a learning curve.

But personally I choice the advice from ""long time user"" that really did capacity tests and will use a lower voltage as the Chinese advice.

One note, If you use balancing boards and like that the boards doing his job, you should use the setting recommended by the supplier form the balancing board. If voltage settings are to low the balancing board are not working. This mains for different balancing boards manufactures a setting from 14.2V.

But even with balancing boards you can choice the lower 13.8V. In this case you can monitor the individually cell voltages and if you see a unbalance, you can raise the charge voltage temporally to 14.2V so that the boards start working.

Another test I read, the testers found out that, with a lower setting from 13.6V, even a existing unbalance was disappearing.

But this need more scientific tests.

Ben

To view this discussion on the web visit https://groups.google.com/d/msgid/smart-alt/5bef29f2.1c69fb81.705e6.03ab%40mx.google.com.

[Al Thomason]

“…..If voltage settings are to low the balancing board are not working. This mains for different balancing boards manufactures a setting from 14.2V….”

 

Ha, so maybe there IS a reason for having an Equalize phase enabled for LiFeP04!   IF one uses the lower terminal voltage, and ballancing will not work  -- go into the Equalize phase which has higher set-points to allow balancing.

[ben]

""IF one uses the lower terminal voltage, and ballancing will not work""

Balancing don't work below the 14.2V because off the design from the boards ,not because the chemistry.

Below the 14.2 the boards don't do bypassing current.

They could also let balancing starting at 14.3 or 14.4, it is just what the designer wants the the boards are doing.

It is not a equalizing phase, so far as I know equalizing fase don't exist with LiFePo4.

Ben

[Rick Bell]

Hey Ben,

I think what Al is suggesting is that the equalize stage could be used as a balance stage for lifepo4 batteries where it would set the voltage high enough to allow the top balancers to work.  My BMS system allows you to start balancing at any voltage, so I have set it to 3.4 but many are fixed at 3.55 or 3.6, which I don't like because it is too close to the top of the curve and can allow the cell to easily ramp up to an unsafe voltage if you don't have a large balancing resistor.  

Once I have finished my config tool the next step is to integrate the BMS, solar, wind, etc and add supervisory functions that will automatically limit the charging systems output keep this from happening and also add a function that sets the charge voltages high enough to start balancing on a predefined schedule, but that's still a ways off.

Rick

--

You received this message because you are subscribed to the Google Groups "Smart Alternator Regulator" group.
To unsubscribe from this group and stop receiving emails from it, send an email to smart-alt+...@googlegroups.com.
To post to this group, send email to smar...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/smart-alt/CAGd1JPCTLL%3DMY5mYWgu6CRFbVNaDEjd%3DuNx%2BsXjEW8hR1Y26pQ%40mail.gmail.com.

[Rick Gleason]

I think that is a great idea. 

BTW what is your BMS which allows adjustment of the balancing point?

[Rick Gleason]

[Rick Bell]

Hey Rick,

It is an Emus with the high current cell balancers.

Rick

To view this discussion on the web visit https://groups.google.com/d/msgid/smart-alt/e1269c07-4e2f-42d8-a545-a9892fd41e68%40googlegroups.com.

[ben]

Hello Rick,

How you describe, is exactly what I mean.

I gone read more about your BMS, looks like a design with great features.

I like the adjustable balancing setpoint voltage.

Ben