Feature requests

[stephendv]

As if you guys didn't have enough on your plate, I thought I'd throw some ideas out there for some nice-to-have features in the classic.  Some of these features come from a newly released MPPT charger (I'll call it "St") from a European company, others from the SMA Sunny Island (SI) and some others from various battery chargers.

• End amps setting for EQ charge (St)
• Option of doing EQ after absorb (St)
• Option of going straight to float if the batteries have not dropped below a preset voltage for a preset time (St and SI) - the idea is to reduce positive plate wear by not going through bulk absorb if it's not needed.
• Option of using an additional set of charging parameters (absorb V and absorb Time) every X days. (SI)  Idea is to allow a separate Full charge every X days in addition to an EQ charge every Y days, so that you can separate the functions of fully charging the batt and mixing the electrolyte/EQ'ing the cells.  E.g. you can have: daily charge to bring SoC to > 90% every day.  Then a full charge to make sure SoC gets to 100% every X days, and an EQ charge to mix and EQ cells every Y days.
• Set a current limit for EQ.  This is recommended by most batt manufacturers (5A per 100Ah), but I don't know of any charge controller that implements it.
• An additional EQ setting would also be nice, e.g. a short EQ weekly, and a full one, once every 3 months
• I know you're working with Sunny Island integration, but not sure what your objectives are... from my point of view it would be really nice if the Classic could act as a slave to the SI's charge profile, i.e. follow the charge pattern set out by the SI.

[boB]

>>As if you guys didn't have enough on your plate,

That's OK...  We just eat what we have and go back for seconds.  OR, get a bigger plate !!

>>>1 End amps setting for EQ charge

Does that really work  to tell when the end of EQ should be ?  If the batteries have already
gone through an Absorb charge cycle before starting the EQ, is the current going to drop
that much ?  Seems that it will drop but I would think it would be a very small amount.
Ideas welcome on that.

>>2 Option of doing EQ after absorb
I Like it !!

>>3 Option of going straight to float if the batteries have not dropped below a preset voltage for a preset time. The idea is to reduce positive plate wear by not going through bulk absorb if it's not needed.

I am trying to understand the first sentence about voltage not dropping after a time.  If the controller is in bulk and/or absorb,
why would the voltage drop ?  I would think it would continue to rise.

The second sentence  is what VariMax is supposed to do (I ~think~)  by starting the absorb time at a short time and raising that time only if it took a long time to bulk.  This idea sounds a bit different.  More info or where in the SI manual might it talk about this ??

>>4 Option of using an additional set of charging parameters (absorb V and absorb Time) every X days.

Not sure why you would want to do this unless maybe you are not there to use the energy every day ?  Maybe sort of like "Vacation Mode" we were looking at implementing except that VariMax or your number 3 recommendation would maybe have the same effect ??

>>5 Set a current limit for EQ.  This is recommended by most batt manufacturers (5A per 100Ah), but I don't know of any charge controller that implements it.

My first thought here is that if you limit the current, then the battery won't be at the EQ voltage and therefore won't be in the Equalize stage.
Is that for some kind of protection for defective batteries ?   Or ?.....   There is already a current limit in the Classic.  You mean and additional but lower current limit for EQ ?

>>6 An additional EQ setting would also be nice, e.g. a short EQ weekly, and a full one, once every 3 months

OK...   But is there any studies or experiments to show this adds life to batteries ?  Maybe you have more ideas on the subject ?

>>7 from my point of view it would be really nice if the Classic could act as a slave to the SI's charge profile, i.e. follow the charge pattern set out by the SI.

Yes.  That's exactly what it would and should do.

Thanks for the input, Stephen !
boB

[stephendv]

>>>1 End amps setting for EQ charge

Does that really work  to tell when the end of EQ should be ?  If the batteries have already
gone through an Absorb charge cycle before starting the EQ, is the current going to drop
that much ?  Seems that it will drop but I would think it would be a very small amount.
Ideas welcome on that.

I find that the EQ current curve looks very similar to the absorb curve, just at higher values.  Just after the battery had reached EQ voltage, the current was 32A, which then quickly dropped down to 19A in an hour, after another 3 hours it had reached 16A and didn't really dip below this value in a reasonable amount of time.  So it would be handy to be able to set my end of EQ amps at 16.5A for example.

>>3 Option of going straight to float if the batteries have not dropped below a preset voltage for a preset time. The idea is to reduce positive plate wear by not going through bulk absorb if it's not needed.

I am trying to understand the first sentence about voltage not dropping after a time.  If the controller is in bulk and/or absorb,
why would the voltage drop ?  I would think it would continue to rise.

The second sentence  is what VariMax is supposed to do (I ~think~)  by starting the absorb time at a short time and raising that time only if it took a long time to bulk.  This idea sounds a bit different.  More info or where in the SI manual might it talk about this ??

What the SI does is that if the battery is above 80% SoC then it won't go through bulk-absorb at all, it will just go straight to float.  There is an additional time limit on this too which I havent' worked out yet, so if the battery has been going to directly to float for X days in the row, then at X+1 days it will go through the usual bulk-absorb phase.
That other MPPT controller implements the same idea, but because it can't measure SoC it uses battery voltage.  So if the battery voltage has not dipped below 50V during the night, then when the day starts, don't go into bulk-absorb, but go directly to float.

I guess the Varimax could achieve the same thing, if it could be tweaked to spend very little time in absorb.

>>4 Option of using an additional set of charging parameters (absorb V and absorb Time) every X days.

Not sure why you would want to do this unless maybe you are not there to use the energy every day ?  Maybe sort of like "Vacation Mode" we were looking at implementing except that VariMax or your number 3 recommendation would maybe have the same effect ??

Yep, varimax + end amps or number 3 would make this irrelevant.

The SI has 2 sets of charge settings, one for "boost" mode which is the daily setting, and one for "full" charging which is as I described above.  This SI uses a fixed absorb time for both those settings.  So I think the reason they did this was because you'd then have to choose between either aiming to overcharge or undercharge the batts daily.  With the additional "full" charge setting you can choose to undercharge every day and prevent positive plate corrosion and then once every week/month do a full charge to fight off the sulfation effects.

>>5 Set a current limit for EQ.  This is recommended by most batt manufacturers (5A per 100Ah), but I don't know of any charge controller that implements it.

My first thought here is that if you limit the current, then the battery won't be at the EQ voltage and therefore won't be in the Equalize stage.
Is that for some kind of protection for defective batteries ?   Or ?.....   There is already a current limit in the Classic.  You mean and additional but lower current limit for EQ ?

Yeah, an additional lower current limit only once the battery is above the absorb set point.  This is a limit I've seen in many battery spec sheets: bulk current below 2.4V/cell is unlimited or limited to 20A for every 100Ah, but when the voltage is above 2.4V then current should be limited to 5A/100Ah.  E.g.: Section 2.6 here: http://www.hoppecke.com/content/download/277/1625/file/Operating_Instructions_vented_lead-acid_batteries_en0712.pdf
And "The charging process" here: http://www.crownbattery.com/?page=support_chargingprocedures  which limits it to 3.5A/100Ah (C5).

>>6 An additional EQ setting would also be nice, e.g. a short EQ weekly, and a full one, once every 3 months

OK...   But is there any studies or experiments to show this adds life to batteries ?  Maybe you have more ideas on the subject ?

Yeah, this is kind of a personal hunch    With an EQ we're trying to do 3 things in one:
1. Make sure the battery is full charged and reverse sulfation effects
2. Stir the electrolyte through bubbling
3. Equalise the cells

With a short EQ, I'd like to _just_ stir the electrolyte to fight stratification and allow for more efficient charging without spending too much time at the EQ voltage. 

This is partly inspired by industrial battery chargers which do a short EQ as part of every charge cycle.  Instead of Bulk-Absorb-Float, they use a Bulk-Absorb-ShortEQ cycle (IUI), like this: http://www.zivanusa.com/ChargeCurve.htm  that final phase is actually done at constant current (which they usually call the finish current) until the batt reaches 100% SoC.   
More info here: http://www.railwayinverter.com/sheets/Battery%20Charger_Nov11.pdf
Here's a rather lengthy Phd thesis on a variation on this idea of doing a "destratification" step as part of the normal charge, the conclusion on page 83 is interesting (bare in mind that their Phd topic is a variation on the "standard" IUI charge): http://tel.archives-ouvertes.fr/docs/00/44/36/15/PDF/these_Phuong_Nguyen_bis.pdf

Since RE systems don't typically go through daily deep discharges, this type of charging is not necessary, so a compromise where we can do a short weekly EQ would have a similar effect IMO.
 

>>7 from my point of view it would be really nice if the Classic could act as a slave to the SI's charge profile, i.e. follow the charge pattern set out by the SI.

Yes.  That's exactly what it would and should do.

Awesome!

[stephendv]

...here's the same IUI charging curve that Trojan recommends, but which no RE charger can perform:

[Vic]

Hi Stephen,  boB,

Several of these items I have been wishing for;  In particular,  the EQ Current Limit -- different from the global CC Current Limit.  This is because by the time that the CC finishes Absorb,  there is often more sun and therefore more charge current available,  and it seems easy to exceed the Max recommended charge current recommended for EQ speced by some battery mfgs  (although,  in a recent conversation with Surrette Tech,  the tech person was unfamiliar with ANY suggested I limit).

That IUI Trojan charge curve is quite a joke for most off grid PV arrays.  One would need to wait for Solar Noon to begin Bulk, and many systems simply cannot supply that amount of current.   The most-used system here usually outputs about 4.5% of real 20 Hr Capacity).

I,  too,  have wished for a recharge mode that would wait for N-days between recharges,   if the SOC at the begin of recharge was at or above a certain amount.  Since voltage is such a poor indicator of SOC,  it is possible that this and some other functions may really belong in the MN Bat Mon.   There is room for a very programmable Charging Strategy Unit,  perhaps,  as well.   But  this would porbably be for folks like myself who really want to go off the very deep end of OCD.

Will study Stephens's ideas some more,  and read the links.  That Crown Batt charging link has only the main page for me,  and nothing else accessible ...  perhaps I am missing some  doodad in the computer.

Comprehensive data of the behavior of LA batteries vs recharge processes is quite lacking.  Particularly from manufacturers of the batteries.

Thanks Stephen for broaching this topic.  Many of us are always wanting more,  and thanks boB and MN for addressing a number of these desires/"needs".

Thanks,  Needy Vic

[Robin]

Remember that you guys here are in the less than 1% class as far as understanding this stuff. Most people would shy away from a controller that had all these features because it would just be too complicated for them. I am one of them. I'm sure that some of the things that thr Studer does is really nice, but I want something to use that doesn't make my head hurt. I will let boB and all you smart guys figure out what is needed and what is just stuff that only the tweaks want. We do not want to be adding more tech service personnel just to answer questions about stuff that the average consumer has no interest in. I say if we are going to be adding some of these things, please hide the menu away so regular people like me won't be scared off.
By the way, I looked up the Studer. They made a nice box to put it in. Here are some quick observations from the American point of view:
1. You need to purchase an additional piece of gear just to program it.
2. Not easy to hook up conduit to this piece. It was designed for Europe where they do not use conduit.
3. No UL or ETL
4. If European prices held here, the 65 amp unit retails for $1084 and the 80 amp unit retails for $1219.
Not cheap considering you still need to buy more stuff to make it work!

Thanks,
Robin

[Vic]

Hi Robin,

Of course you are correct about the complexity overhead.  Stephen is far ahead of me in many respects,  and is now using some of the ModBus hooks that you MN folks have conveniently placed in the Classic's FirmWare.

It is quite possible that folks like me would be satisfied with those hooks,  and perhaps a few more (know of none at this moment).

The "Vacation"  charge mode would be convenient.  At this moment,  I often switch the CC Mode to OFF,  when am going to be away for some number of days.  Hate to turn my back on the charging of the bank,  but probably better to avoid frequent recharge from 97% SOC,  but I still do get nervous wondering while away with the CC=OFF.

Do believe that an EQ Current Limit,  would be good,  but one could just reset the CC global Current Limit  before starting EQ,  as IMHO,  EQ should be monitored by a person.

The OPTION to EQ after end of Absorb would also be nice,  but not make an additional Charge cycle a requirement.  (the brand S CC requires a new charge cycle to do an EQ,  if you do not initiate the EQ early enough in the day,  which is kinna stupid),

Also realize that added complexity and options/features require more programming and DEBUG time,  which can delay things that 80% of customers are waiting for,  more Tech Support,   and so on.

A very flexible ModBus BatMon,  or GP controller could help address some of the more esoteric functions.

Perhaps Stephan  can make that as a product?  Thanks,   Vic     

[stephendv]

Hi Vic,

Sorry must have posted the wrong crown link, page 11 here: http://www.crownbattery.com/wordpress/wp-content/uploads/2012/10/IndustrialBattServiceManual1.pdf
Another bit of evidence about the need for a short EQ comes from the battery technology handbook, attached is a comparison of the disadvantages of the various charging profile, notice what they say about the Bulk-Absorb (IU) profile.

You could go through the Trojan IUI profile on a normal solar day, just ignore the recommended minimum current, it would be the same as regular bulk-absorb with the addition of another current limited stage after absorb. 

Robin,

Completely understand about adding too many "techy" features.  Some of these could be hidden from the user using Apple's approach: automatically choose defaults that will work for 95% of users; and still make it possible for the other 5% to make changes if they choose.   You could even hide some of the more obscure settings and only make them available in the local pc app so as not to pollute the MNDC with too many settings.

Some of the features might make nice selling points too, the classic would be the only controller on the market that offered a specific destratification EQ charge to lengthen battery life and optimise charging.  Of course you could call it something like "BatteryExtender(TM) technology" instead.   

[mtdoc]

Agreed about complexity and  the need to have more esoteric features in more hidden menus (as long as they ALWAYS default to a benign value that does not interfere with basic functions). 

IMHO the last thing the Classic needs is to be loaded down with bloatware that interferes with the basic functions.

A very flexible ModBus BatMon,  or GP controller could help address some of the more esoteric functions.

I agree.  ONE feature that the many users of Classics with Outback systems would be thrilled about is the ability to interface with the Flexnet DC (via the Hub?) and terminate absorb based on shunt measured end-amps. 

[Vic]

Stephan,   Thanks for the updated Crown Batt Maint  Manual.  Learned some things about EQ Frequency vs DOD vs Cycle Frequency,  etc.

mtdoc,  of course,  as I do not have an FNDC or Mate,  my preference is for a capable MN Bat Mon,  which could start out with simple features,  and perhaps become more comprehensive.   But both would be better.   The MNGP should mean that  a MN Bat Mon could be simpler  and perhaps somewhat less costly vs the  FNDC and Mate3,  for example ...   whatta I know ...  OH yea,    time to get back to work.   

Would love to try to roll my own,  but there are sooooo many unfinished projects here ...

Thanks,   Vic

[philb]

I agree.  ONE feature that the many users of Classics with Outback systems would be thrilled about is the ability to interface with the Flexnet DC (via the Hub?) and terminate absorb based on shunt measured end-amps.

That would be a nice feature for me also.

[mtdoc]

mtdoc,  of course,  as I do not have an FNDC or Mate,  my preference is for a capable MN Bat Mon,  which could start out with simple features,  and perhaps become more comprehensive.   But both would be better.   The MNGP should mean that  a MN Bat Mon could be simpler  and perhaps somewhat less costly vs the  FNDC and Mate3,  for example ...   whatta I know ...  OH yea,    time to get back to work.   

Yes I agree Vic. Both would be nice 

One thing that is great about having a Classic with an Outback system - I see no reason to spend $450 on a Mate 3.  I can control my CC with a nice easy to use LCD or a computer. And I can remotely monitor and control my system using the local app (and Wattplot) anywhere I have internet access. 

[stephendv]

Do believe that an EQ Current Limit,  would be good,  but one could just reset the CC global Current Limit  before starting EQ,  as IMHO,  EQ should be monitored by a person.

Just been reading one of Victron's papers and they've fudged a solution for this which they call "batterysafe".  Since they can't measure the actual current going to the battery and neither can most charge controllers, what they do is monitor how fast the battery voltage changes above gassing voltage.  If it changes too fast then they know they're pumping too much current into it, and they back it off a bit.

Been thinking more about this refreshing EQ, EQ current limit and the IUI profile, and I think you could effectively do all of those by just adding one feature to the classic: EQ after absorb. 
By tweaking the absorb time you could effectively get that last "I" phase of the IUI late in the day when full current isn't available.  When you want a full EQ just make sure you hit bulk very early in the day, or disable the "EQ after absorb" option and boB's your mother's brother.

[mtdoc]

The manual states that Aux 2 can be programmed for input or output. I can't find any documentation of any input functions. Are any available?

One feature that would be really nice would to be able to send an input signal into Aux 2 that would end absorb.   I've been toying with the idea of setting up a multifunction DAQ board to measure several things including a shunt based current measurement.  If I could end absorb with a signal sent to Aux2 then I could get around the limitations of the Classics end amp function.

[boB]

The manual states that Aux 2 can be programmed for input or output. I can't find any documentation of any input functions. Are any available?

One feature that would be really nice would to be able to send an input signal into Aux 2 that would end absorb.   I've been toying with the idea of setting up a multifunction DAQ board to measure several things including a shunt based current measurement.  If I could end absorb with a signal sent to Aux2 then I could get around the limitations of the Classics end amp function.

Input for Aux 2 is a very recent addition to our software.   What it does is to enable and disable the Classic from charging.
This was mainly for compatibility with LiFePo battery management systems.

I'm sure that we can add a function that sets it to Float.  This would most likely have to be "edge triggered" so to
re-trigger a Float from Bulk/Absorb, you would have to bring the logic low and then high again (I think)...

The minimum voltage for a logic high is around 6 volts and the maximum voltage for a logic low is around 3 volts with
present and previous hardware.  So a 12V logic signal is perfect for this.

boB