Tapping into solar power before the MPPT charge controller...

Started by Onlooker, March 12, 2023, 07:52:16 AM

Previous topic - Next topic

Onlooker

I've often read that you shouldn't feed 1 array into 2 charge controllers; I presume it it because of power fluctuations and confusion that might be caused between them.

However, how 'bad' is it to tap into the solar  power while/before it fees into the charge controller.

Can I safely put a LM2596 step down on the voltage input for the charge controller to power a cooling fan(s)? Or maybe a 24-48V fan directly(skip the lm2596).  Looking to skim off the top and have it only run when the sun it out anyway, and save thermometer/timer other controller gear. (Aux doing something else).



boB


The regulator and fan could work as long as you didn't exceed the 40V maximum input voltage of the LM2596.

But why not just hook it up to the battery side in instead where it is already regulated and not on the input ?

K7IQ 🌛  He/She/Me

Onlooker

Quote from: boB on March 12, 2023, 04:30:32 PMThe regulator and fan could work as long as you didn't exceed the 40V maximum input voltage of the LM2596.

But why not just hook it up to the battery side in instead where it is already regulated and not on the input ?

If on the battery side, it is a load on battery that could/would drain through the night/cloudy week.. direct to panels; I only want it running when sun is out anyway.. 

I could put thermometer, timer, other things on the battery for fan; but those are a draw in themselves...  I am watt pinching for something I am building to leave unattended for 6-36 months. Running into all sorts of little problems that will likely screw it up hehe..

Example LifePo4 batteries when discharged to much BMS goes into a 'safe mode' showing ~2V;  Putting a big relay on an inverter to kill all power drain from battery except the charge controller once it falls below 20-30% SOC.  Otherwise I(might still) need to install a manual LifePo4 'wakeup' component ; maybe another LM2596 on panels set to 12.00V; that way once charge controller powers back on and detects 12V system to start charging, voltage will be quickly over 12.00 and the LM2596 will essentially sit there doing nothing; as voltage is already higher than the battery..

Haphazard, adding components and work around as I go.. ATM not using a midnite controller (250-CP's waiting), but that is planned for v2.. Not sure how the midnite I have responds to a 12v lifepo4 in 2v state; does it auto start charging or need a kick as well?

This is my first system with lifepo4 as well; keeping it cheap to minimize losses in mistakes. SLA are certainly more forgiving...


New to LifePo4's, I like their power and weight.. Lots to fiddle with BMS/longevity wise.. Seems like 20-80% charge range will extend their life, been fiddling with a 13.9V 30min absorb charge then falling back to 13.6v... It seems like the battery can get a full charge at 13.6 float

Another fun problem I was running into; When charging LifePo4's to spec 14.2-14.4V  the battery BMS itself would turn itself off to protect from overcharging; this caused a charge controller spike in voltage to above the Inverters high voltage protection shutoff; when inverter turned back on a load would start and open the battery BMS quicker than the charge controller would throttle up; and a few seconds of charging resuming from the charge controller would cause the battery BMS to turn off again... and there it would loop(ouch)..
Solution from that end is to set charge controller charge voltage/time to less than whatever would trigger the battery BMS to protect itself.  How these settings will effect the overall life of LifePO4.. guess I will have to test and see.

ClassicCrazy

not sure what battery bms you are using, but if battery bms shuts battery down , then the charge controller shuts down and won't turn back on to charge batteries.
Larry
system 1
Classic 150 , 5s3p  Kyocera 135watt , 12s Soneil 2v 540amp lead crystal for 24v pack , Outback 3524 inverter
system 2
 5s 135w Kyocero , 3s3p 270w Kyocera  to Classic 150 ,   8s Kyocera 225w to Hawkes Bay Jakiper 48v 15kwh LiFePO4 , Outback VFX 3648 inverter
system 3
KID / Brat portable

LilMT

Regarding the OP, when you do get the classics, I think you might be able to use the dawn/dusk setting on an aux port to fire a relay and feed those specific loads only during the dawn/dusk cycle from your battery.  I could still see a potential issue of having enough solar to feed your loads but still not having enough to charge the batteries.  Perhaps there are other functions in the AUX setting that would be more appropriate??  Not 100% sure on this as I am not overly familiar with the aux functions (yet).  Perhaps someone with more in-depth knowledge of them can chime in.
Thanks,
LilMT

Classic 250, 3s2p Trina Solar 405watt, 8 Renogy 200ah for 48V 400ah, Magnasine 4448 inverter.

Onlooker

Quote from: ClassicCrazy on March 13, 2023, 12:27:15 PMnot sure what battery bms you are using, but if battery bms shuts battery down , then the charge controller shuts down and won't turn back on to charge batteries.
Larry

Except with a load on the inverter, and the voltage spike when amps that were going into battery suddenly stop(causing inverter to cycle; as mentioned; then loop continues ). If it were just the battery and charge controller it wouldn't be a problem.  Weize brand LifePo4  100ah 12v.

qrper

Let me toss my hat into the ring.

From what I've read, you want to be able to run a fan (s) directly from the PV without using the charge controller's dry contacts. Is that right?

So, if you're PV voltage is 60ish volts, that's kinda high even for a LM2596. If you're fan current demand is less than an amp, then I'd build a linear preregulator to drop the PV voltage from 60 to 40 V, within the limit of the LM2596. There's some high voltage input linear regulators but I can't recall the part number.

Then from the preregulator, I'd opt for one of those cheap Chinese buck-buck converter modules from Ebay. They're dirt cheap, cheaper than getting the parts to build one.

When the PV is active, the fan runs, at night, it doesn't.

mike
System one: 7kWp w/ Trina 250 W panels @90 Vdc. Classic 150 to 16-6 V U.S batteries. Trace 5548 sine wave inverter.
System two: 6kWp grid tie with solaredge inverter.
System three: Midnite Brat, two 120 W Astropower modules, 100 Ah battery. Runs the LED streetlight in the back yard.

ClassicCrazy

Quote from: Onlooker on March 13, 2023, 01:42:52 PM
Quote from: ClassicCrazy on March 13, 2023, 12:27:15 PMnot sure what battery bms you are using, but if battery bms shuts battery down , then the charge controller shuts down and won't turn back on to charge batteries.
Larry

Except with a load on the inverter, and the voltage spike when amps that were going into battery suddenly stop(causing inverter to cycle; as mentioned; then loop continues ). If it were just the battery and charge controller it wouldn't be a problem.  Weize brand LifePo4  100ah 12v.

What kind of inverter is it ? Seems like the inverter should just switch the load to battery if the amps in go down. No different than a cloud coming over the pv is it ?
Larry
system 1
Classic 150 , 5s3p  Kyocera 135watt , 12s Soneil 2v 540amp lead crystal for 24v pack , Outback 3524 inverter
system 2
 5s 135w Kyocero , 3s3p 270w Kyocera  to Classic 150 ,   8s Kyocera 225w to Hawkes Bay Jakiper 48v 15kwh LiFePO4 , Outback VFX 3648 inverter
system 3
KID / Brat portable

boB

Quote from: ClassicCrazy on March 13, 2023, 12:27:15 PMnot sure what battery bms you are using, but if battery bms shuts battery down , then the charge controller shuts down and won't turn back on to charge batteries.
Larry

I am assuming that the BMS turns the battery back on again ?

K7IQ 🌛  He/She/Me

ClassicCrazy

My lithium battery shut down this winter at night on low voltage.
I was just getting to know this battery and took it too low ( the battery soc was wrong - the Classic soc was correct ).
But when the battery shut off, that shut off the Classic so it couldn't charge the next morning. And it shut off the Outback inverter too so I couldn't charge it from the generator.
I finally had the thought to take my ebike charger which was for 48v battery at 5 amps , and hook that to the battery and run the charger off of the generator. It only took a minute to two to wake up the bms and turn it back on so then the Classic and inverter could run too.
I have no idea how long the bms may have stayed in it's off state. I have a feeling it would have been off until it started to charge again.
Since then I have learned to not trust the bms reported soc . Most of the time it tracks the same as the Classic but it wasn't accurate when I would give an hour to two charge from generator ( via inverter) - then the bms reported too much SOC. 
But I am much wiser now and probably won't ever run the lithium batteries down that far again. When I did that it was on a week long overcast short days of winter period.
system 1
Classic 150 , 5s3p  Kyocera 135watt , 12s Soneil 2v 540amp lead crystal for 24v pack , Outback 3524 inverter
system 2
 5s 135w Kyocero , 3s3p 270w Kyocera  to Classic 150 ,   8s Kyocera 225w to Hawkes Bay Jakiper 48v 15kwh LiFePO4 , Outback VFX 3648 inverter
system 3
KID / Brat portable

boB

Larry, what amp-hour efficiency do you have the Classic set for your Li batteries ?

K7IQ 🌛  He/She/Me

ClassicCrazy

Quote from: boB on March 14, 2023, 01:16:23 AMLarry, what amp-hour efficiency do you have the Classic set for your Li batteries ?

I have the Classic set for 99% efficiency .
Here is snapshot of SOC from both  the batteries BMS and the Classic this morning.
The graph showing the capacity should say ah  not mah
Larry   
system 1
Classic 150 , 5s3p  Kyocera 135watt , 12s Soneil 2v 540amp lead crystal for 24v pack , Outback 3524 inverter
system 2
 5s 135w Kyocero , 3s3p 270w Kyocera  to Classic 150 ,   8s Kyocera 225w to Hawkes Bay Jakiper 48v 15kwh LiFePO4 , Outback VFX 3648 inverter
system 3
KID / Brat portable

Onlooker

Quote from: qrper on March 13, 2023, 02:36:47 PMFrom what I've read, you want to be able to run a fan (s) directly from the PV without using the charge controller's dry contacts. Is that right?
I just wanted to be sure an LM2596 attached to the same wires feeding the input on charge controller wouldn't #%$# with any MPPT tracking..

Quote from: qrper on March 13, 2023, 02:36:47 PMSo, if you're PV voltage is 60ish volts, that's kinda high even for a LM2596. If you're fan current demand is less than an amp, then I'd build a linear preregulator to drop the PV voltage from 60 to 40 V, within the limit of the LM2596. There's some high voltage input linear regulators but I can't recall the part number.

Then from the preregulator, I'd opt for one of those cheap Chinese buck-buck converter modules from Ebay. They're dirt cheap, cheaper than getting the parts to build one.

When the PV is active, the fan runs, at night, it doesn't.
My VOC is only 38V, so should be good there; I have some 35V max step downs I'm tempted to try on the Raw Panel voltage set to 12.0V as a wakeup should my BMS 'switch off' from over discharge.  Trying to set it just right to recover if overdischarged, and not quite 'fully charge' so the BMS doesn't switch off to prevent overcharging and cause that inverter/voltage spike loop dealy I attempted to explain hehe.

https://www.amazon.com/dp/B0922M5RWQ

Onlooker

Quote from: ClassicCrazy on March 13, 2023, 03:24:54 PMWhat kind of inverter is it ? Seems like the inverter should just switch the load to battery if the amps in go down. No different than a cloud coming over the pv is it ?
Larry

There's 2 kinds of shutdowns on the BMS, overly discharged/empty and from 'full' or 'overcharged'.. If I set it right, it never trips the overcharged top end.. There is a sweet spot.. like I could tell it to charge to 13.9v for 30minutes and then go to float at 13.6v;  So far that's working; but I can likely go higher(tweaking settings).. But If I go to 14.2v for 1 hour at some point BMS will say 'full' and shutdown(click); charge controller just sees 0amps or 0amps+inverter/other current draw, maintains whatever voltage charge controller is set to; except for voltage spike if current suddenly stops going into battery from the protection).

When in 13.6v float and a 30watt load on inverter; charge controller still pushes 125-180 Watt into the system for a ?couple of hours before it settles on only what is being consumed(never high enough for BMS to switch it's protection, but slow enough for cells to sit comfortably 'full; accepting no new current).

In this system, this inverter: https://www.amazon.com/gp/product/B083JYBW6K/

I am pretty happy with the inverter, bought it for ~$150;  PSW runs fans/fridges/AC/Computer-UPS's nice and smooth;  Fan only comes on when it is needed;and rarely does with a 400W load for 3 hours(AC unit capacity testing the lifepo4).

boB

Quote from: Onlooker on March 14, 2023, 02:07:13 PM
Quote from: qrper on March 13, 2023, 02:36:47 PMFrom what I've read, you want to be able to run a fan (s) directly from the PV without using the charge controller's dry contacts. Is that right?

I just wanted to be sure an LM2596 attached to the same wires feeding the input on charge controller wouldn't #%$# with any MPPT tracking..


It might screw with the tracking some during twilight early morning sunrise and around sundown but shouldn't hurt anything as long as your Voc doesn't exceed the max voltage of that regulator IC.

There isn't really any power available during those start and end times anyway.

boB
K7IQ 🌛  He/She/Me