WB jr. Q&A

[Resthome]

In case anyone if wondering, here my current classic settings. Absorb=14.4v / Float=13.4v / Max Absorb 4:30hrs / T-Comp -4.0mv / Ending Amps 5.5. Currently the Classic is wired to 2 strings of four Sopray SR-90s giving me approximately a VOC of 82v @ 10.12amps which the string calculator indicated a possible 50 amp output from the Classic. The Concorde Sunxtenders indicate they would like a 0.2C charge in bulk so I am slightly under the 60 amp requirement. Regardless, the Classic hasn't sent more than 34 amps to the batteries except once when there was 3 days of no sun and the battery voltage dropped to 11.8v. (11.8v while under a 2.5 amp load)

I am open to suggestions about what I am doing wrong.

Well from my experience you're about right on the money seeing that 35A. Depending on wiring size, distance, voltage drops, panels and controller efficiency you get about 70%-80% of the sticker rating of the panels when working with 12V battery systems. Your total PV appears to be 720W. At 70% you would get 504W out of the system and at 14.4v absorb you would see about 35A. And you have verified that 35A is the number you are seeing. That would indicate to me you only have a little over half the required PV to charge those batteries at .2C or 60A. What I really don't understand is you indicate your system gets to the absorb voltage pretty fast, not sure how long it is staying in bulk.

I agree that load testing the batteries may be in order just to see where the batteries really are.

I may be all wrong as I am no solar expert, especially with AGMs, but that has been my experience and I think others have mentioned the 70-80% numbers before.

The solar experts can weigh in.

[Tons001]

Thanks John.

I have seen the classic take in 718 watts and put out 49.9 amps to the batteries. It happened after 3 days of no sun (battery voltage was at 11.9v) and that was the peak before it went into absorb. Does that still indicate my PV is off by 50% what is needed for my batteries? 35a is just the average I see on a normal day peak before it goes into absorb.

On average the system is in bulk for 3 hours. If I go back over MyMidnite data, 10/19/13 ... bulk started 7:58am at 3 watts (.03a) with the battery at 12.2v then went to absorb at 11:13am at 472 watts (32.6a) at 14.5v. However only about 40 minutes of that is with the array in full sun which you can see in the attached screen grab.

Maybe it is time to get rid of the batteries with the .2C rating....

P.S. Sorry ... none of this has anything to do with the WB jr.

[Westbranch]

Maybe it is time to get rid of the batteries with the .2C rating....

Not yet, you do not indicate a charger???
Do a good bulk with that and then see what happens.

How old are the batteries?

[Tons001]

Based on the serial numbers about 2 years but only in use since August.

Given the .2C rating ... I should probably pick up the 75a Iota. Correct? NAWS has them at a decent price. I had a Noco G26000 but have not used it because at peak it only puts out 26a.

[Westbranch]

Use the Noco until you get a bigger one, something is better than nothing, it will complement your solar bringing you close to your target.

[Tons001]

Thanks. It is on the Noco now but the battery voltage is high enough that it isn't doing anything. LOL.

Iota is ordered and will be here Thursday.

[boB]

Still waiting to hear back about why VariMax didn't function like I thought it would.

Well, you only Absorbed for the minimum time because the VariMax parameters were not met.

If you look at the VMX setup, you will see a minimum amount of battery current required AND a
certain amount of voltage that the battery voltage must fall below the Absorb set point BEFORE the
minimum Absorb TIME (2 hours in your case) can start to INCREASE towards your 5 hour maximum.

Remember that minimum absorb time is the default absorb time UNLESS something makes it
increase (count up) towards the MAXimum absorb time (5 hours in your case).

With VariMax, the absorb time will NOT increase while in Bulk MPPT UNLESS the battery current
goes above the VariMax current  AND the battery voltage in Bulk-MPPT goes BELOW the VariMax
voltage set point.  The VariMax set point voltage is just a voltage below the absorb set point
voltage...   A negative relative voltage.  I think that VMX voltage starts out around -1.0 volt
relative to absorb voltage.

If you want a longer Absorb time, either increase your minimum absorb time OR adjust the
Varimax current to a fairly low value (just a few amps ?) AND adjust the VariMax voltage
to a fairly low relative voltage... say, maybe, 0.5 volts or so ?

boB

[phonetic]

I I have some 600 amp @ 75mV shunts, I'm thinking to use a 1kohm 10 turn pot as a voltage divider to scale down to 500 amps @ 50 mV in to the whizzbang input terminals  with using the 600amp 75mV shunt. & DC clamp current meter to calibrate the pot & classic display showing battery current.
Will my idea tweak work OK with wizzbang input circuit ?

[dgd]

Any tech info available for the WBjr for those of us who would like to interface it to a computer? - black box, ARM system or PC or whatever
I know I can find info the hard way by analysing what data etc is on that single wire interface but some info from MN would save a lot of messing about 

[Tons001]

Remember that minimum absorb time is the default absorb time UNLESS something makes it
increase (count up) towards the MAXimum absorb time (5 hours in your case).

That was the part I was not understanding. Thanks for the clarification Bob. I did see it count up past the 2 hour minimum when clouds rolled in. I assumed that if the 2 hours had counted down to 0 and EA had not been met, that the Classic would look at the max time. I will increase the minimum time a little higher since my vmax settings are already pretty low.

[zoneblue]

I I have some 600 amp @ 75mV shunts, I'm thinking to use a 1kohm 10 turn pot as a voltage divider to scale down to 500 amps @ 50 mV in to the whizzbang input terminals  with using the 600amp 75mV shunt. & DC clamp current meter to calibrate the pot & classic display showing battery current.
Will my idea tweak work OK with wizzbang input circuit ?

Im not sure that would be a good thing. The thing measures voltages in the single digit millivolt range, and the potential for induced noise in your proposed setup is high. The classic can scale the voltage easily enough. (The scale factor is set using one of the modbus registers.)

Anyway click on NAWs hit buy on a deca, they are only 20 bucks.

[mtdoc]

I I have some 600 amp @ 75mV shunts, I'm thinking to use a 1kohm 10 turn pot as a voltage divider to scale down to 500 amps @ 50 mV in to the whizzbang input terminals  with using the 600amp 75mV shunt. & DC clamp current meter to calibrate the pot & classic display showing battery current.
Will my idea tweak work OK with wizzbang input circuit ?

I don't think so because the ratio is different.  With a 500A/50 mv shunt, a  0.1mv potential change across the shunt=1 amp. With a 600A/75mv shunt, a 0.125mv change across the shunt = 1 amp.  I don't see how just a simple voltage divider can be used to change that ratio - only the absolute scale - but I very well may be wrong.....

There is also the issue of introducing noise as zoneblue states.

You could do it in software/firmware but easier to just buy the right shunt!

[boB]

I I have some 600 amp @ 75mV shunts, I'm thinking to use a 1kohm 10 turn pot as a voltage divider to scale down to 500 amps @ 50 mV in to the whizzbang input terminals  with using the 600amp 75mV shunt. & DC clamp current meter to calibrate the pot & classic display showing battery current.
Will my idea tweak work OK with wizzbang input circuit ?

I think this might be your lucky day !

600 amps at 75 mV  compares with 500 amps at 50 mV so that you need a gain reduction of 0.8 times.


There is a modbus register at address 4362, register 4363 that allows an adjustment of around  70% to 140% of the original gain. (+- 32,767)  The default of this register is 0 so that is a gain of 1.0

That ~should~ work if my calculations are correct and IF that is correct, a value of -15,625, or,  49911  should do the trick.
(that number is signed)

You should also be able to divide it down using resistors but instead of 1K Ohm, I would make it more like 100 Ohms to keep the accuracy.  1K  source resistance won't be as good as lower R.

boB

[mtdoc]

600 amps at 75 mV  compares with 500 amps at 50 mV so that you need a gain reduction of 0.8 times.

Doh! Of course, it's that simple isn't it. Not sure what my pea brain was thinking - that's what I get for trying to think about numbers in the middle of my day job.. 

Thanks for clearing the fog, boB

So it seems like it would be pretty easy to have the Classic software allow user selectable shunt types. But, I would guess you guys have bigger fish to fry...

[boB]

600 amps at 75 mV  compares with 500 amps at 50 mV so that you need a gain reduction of 0.8 times.

Doh! Of course, it's that simple isn't it. Not sure what my pea brain was thinking - that's what I get for trying to think about numbers in the middle of my day job.. 

Thanks for clearing the fog, boB

So it seems like it would be pretty easy to have the Classic software allow user selectable shunt types. But, I would guess you guys have bigger fish to fry...

Well, if those shunts fall under this particular ratio adjustment range, then it's pretty easy.

The gain can also be lowered by factors of 2 and 4 times less in addition to the modbus gain tweak.

Just gotta get the WB Jr. board to physically connect up to your shunt.

boB