OCV limiter - Shunt voltage controller

[ssond]

I am looking at an arrangement where the controller can go into hyper VOC when the temperature of the panels hits 0 degrees C. I am in the pacific SW of Canada where such low temperatures happen every year, though rarely. I could envision a system that uses high (ish)  pwm with a resistive load & capacitor (and a diode) to actively control the voltage to below 200V. Once the controller started to apply a load and further drop the voltage, the system would simply cut out.  The shunt would be arranged in parallel with the charger & battary as follows:

PV Pos - diode - Cap(to ground) - heater- FET - PV neg.

This would avoid losing the ability to charge the batteries during these relatively rare cold periods.

[boB]

Hi ssond.

You ~can~ add a diode to the input of the Classic and then add some kind of load to the
anode (PV) side of that diode to limit voltage but beware that when doing this, the Maximum
Power Point voltage will be very high compared to normal operating voltage and will make
the Classic heat up more than normal and may even limit its output because of the high
input voltage.

boB

[ssond]

The idea is that the load is applied in parallel with the controller and is only active at very high voltages. The purpose of the diode is to prevent the capacitance on the shunt side from changing delaying the response of the pv array to the controllers mppt scan. The capacitor is to provide a constant load to the panels during the FET off cycle of the PWM (the cap will charge during the off cycle and prevent the panels voltage from rising). The shunt will only be active when the max controller voltage is approached which is well above the max power point. Since the controller is seeking max power, it should quickly go below the region where the shunt is active since the shunt itself will be "stealing" power from the panels and lowering the output power.

PV+ ->diode-> cap->resistor(heater)->FET(pwm)->PV neg
PV+ ->Controller->battery->PV neg

The question is whether this load would fool the controller into incorrectly determining the mpp. If the whole voltage range is scanned then it should determine the correct voltage.

[Vic]

If your string Vmp is so high that the CC goes into Hyper Voc at just 0 degrees C (32 F),  then,  this will reduce the efficiency of the CC under all operating conditions.   This situation will increase the heating of the CC,  and COULD result the CC limiting output current below its maximum capability.

If the system that you are configuring is for off-grid,  when the CC is in Float,  with fairly light loads,  the input voltage to the CC will be well above the rated Vmp of the PVs,  and result in a surprising amount of heat dumped into any MPPT CC for even light loads when in Float.

It is unclear what your goal is for this for such an arrangement,   weather it is due to long PV cable runs,  or an attempt to use PVs that are already purchased,  or what,  but IMHO operating any CC continuously at the very high end of its capability is generally not good practice.

Am not a designer of CCs,   but believe that there is already a reasonably large bank of Caps on the CC's Vin terminal,  for what that might be worth.    Vic

[boB]

Now I see how you wanted to connect that load.
I thought you were going to put a diode in series with the controller
input (cathode side to the controller input) and then load down the
PV side with your PWM load.  That ~might~ have a chance of working
but would most likely interfere with solar mode MPPT tracking.
Legacy mode might have a chance of working if the diode was in series
with the input but the controller would still be operating its MPP V very
high.

It's not a good idea to "clip"  the PV by loading down its input because it will load down
the input of the Classic which is a Bi-Directional controller.  i.e.  it makes input
voltage from  the battery voltage as well as going the other direction from
PV to battery.  If your PWM happens to turn on just when the Classic turns
on, it might hurt the controller.

What is your PV array size in watts and what is your battery voltage ?
How far away is the PV array from the batteries ?  Do you need
high voltage ?  Can you wire for lower PV voltage ?

Would a Classic 200 or maybe even Classic 250 work for you if you
must use high PV voltage ?  That would depend on your array size
and battery voltage.

Also, as Vic said, the Classic already has quite an amount of input
capacitance.

boB

[ssond]

10  x 24 v 230 watt panels (36.7 ocv). I have 2 classic 200 which is cutting it close but I don't want to get a third. Winter is what the system is sized for so losing some capacity during the summer is really not of much concern. the run is around 40 ft but the series connection is convenient. Strings of 4 panels would be a better match.

[boB]

10  x 24 v 230 watt panels (36.7 ocv). I have 2 classic 200 which is cutting it close but I don't want to get a third. Winter is what the system is sized for so losing some capacity during the summer is really not of much concern. the run is around 40 ft but the series connection is convenient. Strings of 4 panels would be a better match.

4 modules of  36.7  Voc   should be just fine at even less than 0 degrees C.
That's only  146.8 Voc at 25 C  and around 163 Voc at 0 C.

Plenty of room with a 200.

boB

[ssond]

I have 10 modules and two controllers. I wold have liked to do series strings of five, hence the above discussion.

[RossW]

I have 10 modules and two controllers. I wold have liked to do series strings of five, hence the above discussion.

I'm running 6 arrays in parallel. Each array is a series string of 6, "nominally" 12V (17.2? Vmp) for system that runs "around 100V" into the classic.

[dgd]

10  x 24 v 230 watt panels (36.7 ocv). I have 2 classic 200 which is cutting it close but I don't want to get a third. Winter is what the system is sized for so losing some capacity during the summer is really not of much concern. the run is around 40 ft but the series connection is convenient. Strings of 4 panels would be a better match.

4 modules of  36.7  Voc   should be just fine at even less than 0 degrees C.
That's only  146.8 Voc at 25 C  and around 163 Voc at 0 C.

Plenty of room with a 200.

36.7V open circuit?  so what is the mpp voltage? somewhere in the 28v to 30v range?
if this is the case then I would string five of these panels in series and connect each string to a C200.
The 0c Voc will IMHO not be an issue as just one or two rays will get current moving and reduce pv voltage.
Better too use all 10 panels than just use 8 or have to get another 200 for just two panels.
(or you could put two strings of 3 on one C200 and one string of 4 on the other C200 but I wouldgo with the two strings of 5)

dgd

[ssond]

I changed my order t 1x 150 (2strings of 3) and 1x 200 (1 string of 4). This arrangement seems to utilize the cap of each respective unit quite well.

[boB]

I changed my order t 1x 150 (2strings of 3) and 1x 200 (1 string of 4). This arrangement seems to utilize the cap of each respective unit quite well.

Excellent !  That should work fine.

boB