Zener diode to limit voltage before MPPT ?

[Schtroumpf]

Hi,
I am in the process of installing a new solar system on my boat.
The main components are :
- 16 100 Ah prismatic LIFEPO4 cells in 4P4S ->12volts

http://en.calb.cn/UploadFiles/pdf/CA100.rar

- 4 solar panels sanyo hit VBHN214

http://www.ge.co.id/pdf/SPSanyopanasonic/SpecificationsVBHN214AE05_120316.pdf

- 2 MPPT regulators ZLPower 40A

http://www.zlpower.com/en/products_info.asp?id=592

I have also setup a security with relays to avoid over and under voltage at cell level on the bank.

I have set the MPPT to bulk charge at 14.4 and float at 13.7 V.
My problem is that the MPPT even if advertised at PV voltage 55v max is only taking 40 volts when connected to a 12v battery.
When I connect one panel in full sun (I am on the equator) I got close to 50 V (open max V = 51,7V) and the MPPT regulator goes in fault mode and do not charge the battery.
Nothing is destroyed because the regulator can handle 55V it just refuse to convert more than 40V on 12v batteries.
The solutions I see are :
-Shade 15 cells out of the 72 to decrease the max tension under 40V
-Add a zener diode at 39V and serial resistor in parallel to limit the tension
-Buy another MPPT. Unfortunately the ones that can handle over 50 volts and 70A are quite costly (over 500$ plus shipping to Malaysia) compared to the one I have (80$ each).
Between 1 and 2 what is the best option ?
I believe that because the max power of the panel is at 41,9 V the fact that the zener Diode will limit at 39V will have only little impact on the power I get. (les than 10% loss).
Most of the time the voltage of the MPPT will be under 39 V and then I don't lose any power.
How to size my diode and resistor ? Should I see the solar panel as a current generator and size the diode and resistor for (51,7 V - 39V)*5,12 A = 65w so this means a 2 ohms resistor ?
What would you do in my case ?
My problem is quite similar to this thread

http://www.solarpaneltalk.com/showthread.php?7800-Panel-volts-exceeding-MPPT-controller

[inetdog]

Hi,

-Add a zener diode at 39V and serial resistor in parallel to limit the tension
-Buy another MPPT. Unfortunately the ones that can handle over 50 volts and 70A are quite costly (over 500$ plus shipping to Malaysia) compared to the one I have (80$ each).
Between 1 and 2 what is the best option ?
I believe that because the max power of the panel is at 41,9 V the fact that the zener Diode will limit at 39V will have only little impact on the power I get. (les than 10% loss).
Most of the time the voltage of the MPPT will be under 39 V and then I don't lose any power.
How to size my diode and resistor ? Should I see the solar panel as a current generator and size the diode and resistor for (51,7 V - 39V)*5,12 A = 65w so this means a 2 ohms resistor?

The zener diode solution makes sense if you can find a zener much closer to the 50 volt Voc input limit on the CC, and well above the Vmp of the panels. The biggest problem with any parallel zener circuit is that the zener will be drawing current to create a voltage drop across the resistor (and the resistor will be wasting power all of the time if it is also in series with the connection from panel to CC.)

If you use a zener with a voltage lower than the panel MPPT point, then the zener will, as you say, be drawing current only based on the difference between Vmp and Vzener, which is about two volts. In that case, though, the zener would be stealing a full amp of your string output.
A more serious concern for me is that the presence of the zener will cause an unusual shape to the I versus V curve of the panel in the immediate neighborhood of the Max Power Point, which could confuse some types of MPPT controllers.

It does not take much current at all (compared to Imp or Isc) to drop the panel output a few volts below Voc, so a 48 volt zener diode with a 10 ohm series resistor, the combination placed in parallel with the CC at the panel output, would do the job a lot better and would not cost you any power after the CC has pulled the voltage below 48 volts. And if the charging current demand of your batteries causes the CC to let the panel voltage rise above 48 volts, you would not have been able to use the extra current anyway. Finally, the distortion to the shape of the I vs. V curve will be less of a problem.
In this case (48 volt zener), the maximum current will be much less than .4 amps, and you could use a zener with a 20 watt power rating on a good heat sink. The closer you get the zener voltage to 50 volts, the less current you will need it to draw, and the lower the power rating you will need for the zener.

If a 39 volt zener is all that you can get, you can add 13 normal rectifier diodes, forward biased, in series with the zener to raise the knee voltage to 48.1 volts.

[Mike90250]

you could use 2, 20V or 4, 10V (or whatever voltages make sense) zeners in series, and that also helps split the heat dissipation of the diodes. Not knowing how/when the controller loads down the panels, it's best to plan for the diodes & heat sinks to be able to handle the full power of the PV array.

Don't shade the PV panels, that forces the PV to use the internal bypass diodes, and will create hot spots on the PV, and eventually, the diode & PV dies.