Off grid PV with both thin film and monocrystalline questions

Hello, it's my firts post in this forum. I need your advice concerning a pv off grid system that is already working in Athens, Greece.
Its goal is to fulfil electric power demands of a small household (a small fridge, TV, laptop and lights), and support hot water demands.
Nominal Volatge is 24 V.
Modules. They are seperated in 2 parts. The first contains 10 modules monocrystalline, 190 W each, and the other 40 thin film Q cell UF Smart 105 Watt each. You may find more information in the pdf attached (not really a professionally draw but I think it works).
I was advised by people in Solibro which used to produce the Q cells, to ground the negative pole, using also a 2 A fuse, since the modules are mounted on a metal tilted construction.
My first concern is that since the grounding system of the whole system is common, as the country regulations indicate, there may be a problem with the grounding of the monocrystalline panels. Their aluminium frame, and the metal construction they are mounted on, use the same grounding system.
There are 2 seperate inverters, one is Powerstar 3000 24 V controlling the household loads, and a second modified sine wave HP 3000 W 24V that controls the resistance. Using a system that contains a relay and a voltage monitor, when the voltage of the batteries reaches 28.0 V, the relay is on and the heater works, until the voltage reaches 25.2 V and the relay goes off. This is to avoid drawning more power than used to cover the electric demands of the household. I have noticed that when the voltage of the batteries reaches 28.0 V and the relay "[email protected] theheating resistance to the system, the voltages drops to 26 V immediately. I think it's a huge voltage drop. When the relay cuts off the function of the heater, voltage at this exact time, raises to 26 V, from 25 V.
Is this normal?
I'd like to know if you agree with the whole grounding system and the total function and logic of the system. It's kind of experimental, and till now it seems to work.
Thank you
Vangelis
Attachment not found.
Its goal is to fulfil electric power demands of a small household (a small fridge, TV, laptop and lights), and support hot water demands.
Nominal Volatge is 24 V.
Modules. They are seperated in 2 parts. The first contains 10 modules monocrystalline, 190 W each, and the other 40 thin film Q cell UF Smart 105 Watt each. You may find more information in the pdf attached (not really a professionally draw but I think it works).
I was advised by people in Solibro which used to produce the Q cells, to ground the negative pole, using also a 2 A fuse, since the modules are mounted on a metal tilted construction.
My first concern is that since the grounding system of the whole system is common, as the country regulations indicate, there may be a problem with the grounding of the monocrystalline panels. Their aluminium frame, and the metal construction they are mounted on, use the same grounding system.
There are 2 seperate inverters, one is Powerstar 3000 24 V controlling the household loads, and a second modified sine wave HP 3000 W 24V that controls the resistance. Using a system that contains a relay and a voltage monitor, when the voltage of the batteries reaches 28.0 V, the relay is on and the heater works, until the voltage reaches 25.2 V and the relay goes off. This is to avoid drawning more power than used to cover the electric demands of the household. I have noticed that when the voltage of the batteries reaches 28.0 V and the relay "[email protected] theheating resistance to the system, the voltages drops to 26 V immediately. I think it's a huge voltage drop. When the relay cuts off the function of the heater, voltage at this exact time, raises to 26 V, from 25 V.
Is this normal?
I'd like to know if you agree with the whole grounding system and the total function and logic of the system. It's kind of experimental, and till now it seems to work.
Thank you
Vangelis
Attachment not found.
Comments
Welcome to the forum Vangelis!
Nice sized system there. A few comments...
First, do not ground any of the PV Panel negative power terminals/wiring with a 2 amp fuse to ground. If these panels have metal frames (not electrically connected to the PV +/-), then ground the frames to your system ground just like the other panels are. You want any short circuit to the framework to go to the safety ground (not make the frames/mounting racks "hot").
Because your battery bank main negative bus is already safety grounded (that is fine), you really do not need (or even want) any fuses in any of the negative/return wires. The fuses are (if everything is properly wired) in the negative leads are redundant. And worst case, if you have a return fuse blow, it still leaves the device powered from the "hot" + fused circuit. Ideally, you want the positive fuse to blow only... Then everything on that branch circuit is now "dead".
I have to be a little careful here--Generally in the US, we used "RED" for "Hot" or + wiring--Yours is reversed (Red=negative). So any of the panels that have fusing in negative should be moved to the positive lead.
Also, you show a bunch of wiring going to your battery bank and no fuses/breakers. With a negative grounded battery bank, you do not need any fuses/breaker in any leads to the - battery bus.
On the positive bus, every wire that leaves the positive bus should have a fuse/circuit breaker that is rated at no more than the current capacity of that branch circuit wire leaving the bus. To charge controllers, to loads, to AC inverter, etc. Fuses+Breakers are there to protect the wiring from too much current and starting a fire.
Note that the array string to charge controller does not really need any fuses/breakers.
Next issue... For arrays with (typically) 3 or more parallel panels, each panel should have its own series fuse. If, for example the panel has a 15 amp series fuse rating, then each panel should have its own 15 amp fuse/breaker.
Because you are using MPPT charge controllers, you can put two 36 volt panels in series and make for a 72 volt Vmp string--That will allow you to use smaller diameter wire with less voltage drop. And you will have fewer parallel panel connections--Fewer fuses/breakers per array.
I cannot tell from your drawing--But charge controllers should be close to the battery bank (low voltage drop for accurate battery charging). Long wire runs should be from the array to the charge controller.
AC inverter DC wiring from the battery bank to the AC inverter should also be short/heavy wire to limit voltage drop and save on the price of heavy copper cables.
And regarding the ground wire from the battery bank negative bus to your Ground rod/ground bus--The battery ground wire should be able to carry the maximum amount of short circuit current that you would expect. Nominally, it should be as heavy as the largest DC circuit in your power system (typically the DC input to your AC inverter).
That is a quick start--Your thoughts?
-Bill
First of all thank you for your time and kind advice.
1. It was Solibro's advice to ground the thin film modules this way. The modules are frameless. They indicated "Q.SMART UF L modules are technically identical to Solibro’s SL2 modules and should be negatively grounded to prevent potential induced degradation (PID)". In their technical datasheet is claimed that I should use a 2A fuse for each module string. Every string has Isc = 1.6 A
Let me see if I understand your advice right.
Do you mean that I should remove the ANL fuses from the negative wires and leave the ones in the positive?
Fuses connected to the wires from the negative pole of the panels to the charge controllers must be placed in the positive wires between panels and charge controllers?
Agree. Concerning the wires from the batteries to the inverter, the cable is 35 mm2, so the appropriate fuse is 125 A. The fuse is chosen according to the wire diameter, right?
Each panel is rated with Isc = 1.6 A. The technician from Solibro advised me to use a diode if I connect more than 3 panels in series, but in case of one by one parallel connection, it's not obligatory to use fuses or diodes. Would you agree?
These wires are 2 metres at most.
There you have a point. I decided to replace the 16 mm2 wire with a "heavier" one, just in case.
Well, it's the first time I ground the negative pole of the batteries, and the first time I use thin film panels of which the negative pole must be grounded. All these, in addition to metal parts etc. are connected to the same grounding system. This confuses me a little. But the grounding system must be common according to our regulations. Is it ok?
Also, what if someone touches the metal part of the tilted mounting construction of the panels (which is also grounded) and the positive pole of a battery?
What about my question about the voltage drop of the batteries when the immersion heating resistance goes on? Is it normal?
Your help is valuable, thank you
Vangelis
Partial answer: If your panels are in parallel, then the common point of the negative leads from all of the panels need to be grounded and the panel frames must also be grounded.
If you put two panels in series, the negative lead of the first panel in the string will be grounded and both of the panel frames must be grounded.
If you touch a panel frame and a positive battery lead at the same time, you will feel the corresponding voltage. If you bridge them with a wire, you will blow the fuse in the positive battery lead. An additional fuse in the negative battery lead will not provide any additional protection.
The rule is that all ungrounded conductors need a fuse or other overcurrent protection device (OCPD). If neither panel lead is grounded (as with some grid tie inverters) then you need OCPD in both leads.
I disagree on the series fuse question.
If the panel is rated for an Isc of 1.6A, the panel nameplate probably specified a maximum series fuse size of 5A or less.
The situation is that with only one panel in parallel, the maximum back current through a damaged panel would be ~1.6A, and it should handle that fine.
Add just one more panel in parallel and the back current becomes 3.2A, which may be more than the damaged panel could take. Hence the general rule that with more than three panels or strings in parallel they must all be fused.
If the nameplate data says that you can use a 10A series fuse (what does it say?) then you might think that you could group as many as say 5 panels on a fuse, but if you have a total of 10 panels, in two groups of 5, then one damaged panel can see 5 x 1.6 + 4 x 1.6 without a fuse blowing. Not a good thing, as the total current will be 14.4A which is more than 10A.
This is a rare occasion where I disagree with what I think that Bill said.
I was using a general example (for crystalline and some of the older thin film panels).
In the above case, I looked up the manual from the mfg. for these panels and quoted them.
The part I am not sure about is if these panels in parallel do they need blocking diodes to prevent damage if one (or more) panels are in shade.
These panels are suseptable to PID (Potential Induced Degradation, PID) -- And for this mfg., the panels may be/are permenently damaged if they are not negative grounded (via the PV - lead):
http://solibro-solar.com/en/news-downloads/documents/
http://solibro-solar.com/uploads/media/Solibro_Product_Information_EN_02.pdf
-Bill
Technicians from Solibro :
1."Please secure this grounding of the negative pole with a fuse as described in our installation manual on page 17. You are using frameless modules – if your installation is ground-mounted you can use a maximum 2A fuse."
2."As a reverse current protection, the use of diodes as described by Mr. Matthes is usually required. However, since the “strings” are only composed of one module each, this point may be neglected. Please make sure, however, that you fulfill all national safety and electrical regulations."