Classic / Kid Boost mode : Stealing from the XW MPPT 60

[DMJ72]

Here is a screen shot from my Schneider SCP showing the charge controller setup for "boost mode" :
The first hour of absorption, the XW tries to take the batteries to 31V (t.comp'd) and then for the rest of the Absorption time it goes back to 29.2V. I have found this to be quite good at getting the batteries up to 1.275+ SG with less water usage than what I used to do before (with my morningstar TS-60 MPPT) which was to run Absorption @ 31v (t.comp'd) for the whole absorption period.

Something the Classic could benefit from?

Here is an excerpt from the XW Manual :

Boost charging allows for better utilization of flooded lead acid batteries under moderate cycling in off grid applications. Boost charging encourages a short duration charging voltage - above the gassing voltage - at the beginning of the absorption charge state. Testing has shown that boost charging improves battery performance by providing a regular mixing of the liquid electrolyte. Boost charging specifically discourages capacity-robbing acid stratification and plate sulfation.
Boost mode charging can be enabled by selecting 'custom' battery type and by setting the bulk voltage higher than the absorption voltage. The multi-stage charge algorithm then attempts to use the higher bulk voltage for the first hour of the absorption stage - unless interrupted by the max absorption timer or exit current threshold.

[Kent0]

You can do this with the Morningstar Tristar controllers by using the custom configuration tool to set up the absorption extension.

[ChrisOlson]

FYI, if your XW controllers have V01.05.00-0006 firmware you can also set the Bulk voltage to normal Absorb V and set Absorb to the finish stage voltage recommended by your battery manufacturer for two-stage absorb with finish.  US Battery and Trojan both recommend three-stage, bulk/absorb/finish charge profile.  The XW controller is currently one of only two controllers on the market that can do it - the other is Ralph Hiesey's new SC-2030 solar charger that communicates with and is controlled by the new TriMetric TM-2030 battery monitor.

[Robin]

There are other schools of thought about this. SMA and Surrette strongly disagree with this method.
Ryan or boB may want to chime in here. I will probably get it wrong, but this is a good place to discuss this.

[ChrisOlson]

There are other schools of thought about this. SMA and Surrette strongly disagree with this method.

Disagree with what?  Boost charging, or three-stage bulk/absorb/finish?  Just about all industrial chargers (forklift, floor maintenance machines, etc) use IUI charge profile.  I don't know about SMA, but Rolls' published charging specs are based on the use of a IUI constant-current charger, not voltage-based chargers like we have in the RE world.  Steve Higgins has been busy rewriting their tech support stuff to reflect the use of RE voltage-based chargers because too many people have been using the published Absorb V spec for IUI profile, and ending up with badly sulfated batteries.

Rolls doesn't actually publish the recommended profile, but US Battery does:

[vtmaps]

Just about all industrial chargers (forklift, floor maintenance machines, etc) use IUI charge profile.  I don't know about SMA, but Rolls' published charging specs are based on the use of a IUI constant-current charger, not voltage-based chargers like we have in the RE world.

Yes, but... I think these industrial IUI profiles are used in systems where there is NOT a lot of PSOC cycling (partial state of charge).  Some (most?) forklift battery manufacturers recommend against partial charging and partial discharging.  I think this has to do with the buildup of stratification (in tall batteries) during both partial charge and partial discharge.  Also, the cycle-life charts of these batteries often indicate that the best ratio of kwh throughput to dollars is at deep discharges.

It's not obvious to me that IUI profile is the best way to charge batteries in a RE system.
DISCLAIMER: very few things about batteries are obvious to me.

--vtMaps

[DMJ72]

Rolls has issued a bulletin for RE applications:

I have attached the pdf here:

[ChrisOlson]

It's not obvious to me that IUI profile is the best way to charge batteries in a RE system.

It's actually impossible to charge batteries with IUI in a RE system.  Well, I guess I should say darn near impossible without throwing excess solar capacity at the problem.  The Schneider Conext XW controllers do it in a sort of roundabout way that works.  But they use Bulk (maximum current available), Absorb (constant voltage), Finish (constant voltage).  We use 2.40 VPC for Absorb and 2.58 VPC Finish here and it works very well.

What the XW controller does is is charge during Bulk until the voltage hits Finish voltage, then the controller "throttles" the voltage back to the Absorb setting for one hour.  This limits the current to the battery when it is at lower SOC during gassing, which keeps its internal temperatures down.  After one hour it then increases the voltage to Finish stage voltage and exits at 2% C/20.

I tried for months to get the Classics to do this by using AutoEQ for the Finish stage.  But it didn't work because the Classic has a flaw in that it will not fully absorb the battery before it jumps into EQ.  So it got the batteries hotter than blazes when the AutoEQ kicked in because they were not ready for it yet.  So I finally gave up on it because it boiled way too much water off.

Ralph Hiesey has done one better than Schneider with his SC-2030.  It does not use constant current bulk, but instead uses all available power.  Then it does a constant voltage absorb and a constant current finish stage with current held at 2% C/20 until finish voltage is reached.  The controller itself is "dumb".  The charge profile is actually controlled and communicated to the charger by the TM-2030 TriMetric over serial link.  Ralph is a brilliant engineer.

[ChrisOlson]

Rolls has issued a bulletin for RE applications:

I have attached the pdf here:

That's some of Steve's work.  I've told people this very thing for years on various forums, that have ended up with sulfated Surrettes.  Invariably somebody always comes out of the woodwork with their published Absorb V spec that is way too low for RE systems and proclaims that if you use higher voltage you're going to wreck your batteries.  Well, Steve and I got together on this long ago when we initially had problems with ours and that's why I've been trying to tell other folks to crank those controllers up.  Glad to see Steve finally got around to getting this in print.

[vtmaps]

Rolls has issued a bulletin for RE applications:

I have attached the pdf here:

That's some of Steve's work.
<snip>
Glad to see Steve finally got around to getting this in print.

Huh? I don't think Steve can take credit for that document... he hasn't been at Rolls that long.  That document has been around since at least 2008.  I downloaded it a few years ago from here:
http://forum.solar-electric.com/showthread.php?p=21460#post21460

--vtMaps

[ChrisOlson]

They used to have a lot of PDF downloads of various bulletins and whatnot, and it appears they are going to more HTML based bulletins that you can search on their support pages to make it easier to find
http://support.rollsbattery.com/support/search

So if that WAS in print as early as 2008 very few people were paying attention to it, and still using the 2.4VPC Absorb V for RE systems.  Steve said it was his job to make sure this is better understood for folks using their batteries in RE applications, and to get the manuals rewritten.  That link to where that evidently used to be on Rolls' website is long gone.

[zoneblue]

Would a way to offer either/both of these two charge profiles on the classic, would be to have absorb1 and absorb 2, both with adjustable setpoints? Absorb 2 follows after absorb 1. Use both or disable one...

[DMJ72]

Would a way to offer either/both of these two charge profiles on the classic, would be to have absorb1 and absorb 2, both with adjustable setpoints? Absorb 2 follows after absorb 1. Use both or disable one...

Yep, I like that.

BULK-->>ABSORB1-Voltage---for time T>>ABSORB2-Voltage for rest of Absorb time or until End Amps>>Float.

[ChrisOlson]

BULK-->>ABSORB1-Voltage---for time T>>ABSORB2-Voltage for rest of Absorb time or until End Amps>>Float.

Actually, that's Schneider's method using a timed first Absorb stage.  And it's not ideal.  Ideally, the controller should stay in that first Absorb stage until ending amps is reached, THEN increase the voltage to finish voltage.  I would call the stages Bulk, Absorb, Finish as that is the terminology used in the industrial world.  And I believe that the terminology that Ralph Hiesey is using for his SC-2030 as well.

I don't know that a true constant current finish stage is absolutely necessary.  Ralph did it with the SC-2030.  But I really think the method that the Schneider Conext XW controller (and the XW inverters) use is just as good.  It simply bumps the voltage up to finish voltage and exits at 2% C/20.  In the RE world where you may be running out of power to complete the finish stage late in the day, the constant current method might not get the job done because it won't be able to hold the current at 2% C/20 for long enough to complete it.  That finish stage here can take up to an hour with a cycle every 2-3 days, and 3 hours if the cycle has gone 7 days or more.

[vtmaps]

So if that WAS in print as early as 2008 very few people were paying attention to it, and still using the 2.4VPC Absorb V for RE systems.

I think a lot of folks were influenced by Sandia research that was reported in this HP article from 2002:
http://www.homepower.com/view/?file=HP89_pg120_IPP

The Sandia report is very thorough. Four different brands of batteries were tested. They were all flooded, L-16 type batteries, the most common battery used in residential-scale RE systems.

The study has four conclusions:

• The finish voltage (sometimes called the absorption voltage) for a flooded lead-acid battery operating at 12 VDC nominal should be about 15.3 volts (2.55 per cell) rather than the customary 14.4 volts.
• Finish charge time should be at least 3 hours and often longer.
• The maximum interval between finish charges should be about five days.
• Not all brands of L-16s are the same (though the report names no names).

--vtMaps