Solar Panels - Series or Parallel ?

[nigel1]

Thanks for all the research, about convinced me to go for series.
I contacted a UK supplier for additional info on the Kyocera 135 panels, and then they suggested a panel made by "Cleversolar" which are mono-crystelline, and rated at about 21% efficiency. So I contacted them with regard to how many cell strings/bypass diodes, and got this reply
Thank you for your email. The CS-135 panels consist of three in-panel strings, hence the cells are split into three parts. There are no blocking diodes fitted to the panels as the monocrystalline cells have a very high electrical ‘back flow’ resistance and hence do not need the diode to prevent battery charge leakage at night. Also, I think that the Blue Sky controller has a night-time disconnect feature built-in to further prevent charge back flow. If you would like to mak triple sure that this is not a potential issue, you could of course fit external blocking diodes, which are readily available online.


This a link to the panel
135 Watt Solar Panel CS-130 | Monocrystalline | Cleversolar UK

They cost more than the Kyocera, but a big advantage is the size reduction.
Would any experts on solar care to put forward an opinion on this panel

Cheers

[Cotemar]

nigel1,

True.

Blue Sky controller has a night-time disconnect feature built-in to further prevent charge back flow.

[s/v Jedi]

Okay:

- you need a controller (MPPT) that can handle the input voltage of the series connection.

- the blocking diode is something different than the bypass diodes. If the dealer mixes that up you better stop listening to him!

I was under the impression that everybody knew you need an MPPT controller for series connection, but now think Cotemar is right to put that assumption back into my face. The poster that I was replying to does know this I think, but many don't.

ciao!
Nick.

[GeoPowers]

Originally Posted by s/v Jedi

Exactly this is the reason that the myth of parallel vs series connection stays alive... there is tests and research out there that must be from decades ago because that is how long all the panels have bypass diodes.

The research is dated 2009, not from a decade ago, but as Chuck points out it looks like they really only account for blocking diodes and not bypass in their model (even thought they reference work written on bypass diodes!). Perhaps that would have made the computer model too COMPLICATED...

Originally Posted by s/v Jedi

Just to make sure: yes, it is as simple as demonstrated and as simple as calculated before demonstrating it: series outperforms parallel by such a huge amount that it is absolutely weird to even contemplate parallel connection. A 25% difference... !!

ciao!
Nick.

Well, here I disagree with you, both in the claim of a 25% greater efficiency in series systems vs. parallel connections, and the supposed simplicity in figuring out which is better for a given scenario (which according to you series connections always are superior). I would like to think that if it was that easy to determine- the solar industry, with all their technical expertise, would be able to figure out the above on their own without help from a Dutch cruiser on a boat in Panama.

I'm not saying that series might not be the right answer, but I'm not convinced myself. And I'm not here to debate, just share info, so if this is just about debating which we like better then I'm out of the conversation because I haven't figured that out for myself yet.

Frank

[s/v Jedi]

Originally Posted by GeoPowers

The research is dated 2009, not from a decade ago, but as Chuck points out it looks like they really only account for blocking diodes and not bypass in their model (even thought they reference work written on bypass diodes!). Perhaps that would have made the computer model too COMPLICATED...

Well, here I disagree with you, both in the claim of a 25% greater efficiency in series systems vs. parallel connections, and the supposed simplicity in figuring out which is better for a given scenario (which according to you series connections always are superior). I would like to think that if it was that easy to determine- the solar industry, with all their technical expertise, would be able to figure out the above on their own without help from a Dutch cruiser on a boat in Panama.

I'm not saying that series might not be the right answer, but I'm not convinced myself. And I'm not here to debate, just share info, so if this is just about debating which we like better then I'm out of the conversation because I haven't figured that out for myself yet.

Frank

Frank,

If you share info, you should be open to some debate on it... this ain't grafiti

About the research: it is based on panels as they were sold decades ago. I do not believe that any panel of any brand that was sold in 2009 did not have bypass diodes. So these guys are either clueless or they got their hands on some ancient panels.

You don't have to believe my claim.. I don't even own it, others have come forward with it before me. It also isn't 25% more efficient in every situation.. if you never have shading, it is much less, only a bit better efficiency of the controller. But that was clearly indicated, and the 25% was an average for conditions where shading occurs... which is about every boat.

The solar industry assumes that installers are electrical engineers. They assume we know that series is better. It is not rocket science, diodes are very simple and predate the transistor. No, the source of the myths is that the first panels did not have bypass diodes. Engineers added those to installations; most cruisers found that too difficult. Freaks bought individual cells and made their own arrays incl. diodes.

Read back this thread with the theory of which shading causes which output. It's solid (and understandable for non-engineers) and tests confirmed it. You can't keep denying with clear evidence presented here by CF members.

This isn't about which we like better... I do not believe some would choose parallel because they like to have less power generation.

ciao!
Nick.

[GeoPowers]

Originally Posted by s/v Jedi

Frank,

If you share info, you should be open to some debate on it... this ain't grafiti

About the research: it is based on panels as they were sold decades ago. I do not believe that any panel of any brand that was sold in 2009 did not have bypass diodes. So these guys are either clueless or they got their hands on some ancient panels.

You don't have to believe my claim.. I don't even own it, others have come forward with it before me. It also isn't 25% more efficient in every situation.. if you never have shading, it is much less, only a bit better efficiency of the controller. But that was clearly indicated, and the 25% was an average for conditions where shading occurs... which is about every boat.

The solar industry assumes that installers are electrical engineers. They assume we know that series is better. It is not rocket science, diodes are very simple and predate the transistor. No, the source of the myths is that the first panels did not have bypass diodes. Engineers added those to installations; most cruisers found that too difficult. Freaks bought individual cells and made their own arrays incl. diodes.

Read back this thread with the theory of which shading causes which output. It's solid (and understandable for non-engineers) and tests confirmed it. You can't keep denying with clear evidence presented here by CF members.

This isn't about which we like better... I do not believe some would choose parallel because they like to have less power generation.

ciao!
Nick.

Well, my diploma says I'm also an engineer, but just because I know a little about pipe piles, haul rates, and soil tests doesn't mean I know everything about electrical.

I'm finally getting ready to buy my own panels now that my arch is installed, and had no idea all the drama..err...discussion about series vs. parallel. Last I read a year or two ago, it seemed that everyone thought parallel was the way to go due to shading. I read several recent threads, informative reading.

One thing I didn't see answered (but might have missed): How do you know the capacity of the bypass diodes used in a panel so that you don't exceed the number of panels in series, or is this even a concern? I'm guessing most panels would have a higher max "system" voltage for their bypass diodes than we would ever see in a boating application, simply because we only have room for a few panels? I don't see this rating on spec sheets, or can you assume a very high number?

Frank

[s/v Jedi]

Originally Posted by GeoPowers

Well, my diploma says I'm also an engineer, but just because I know a little about pipe piles, haul rates, and soil tests doesn't mean I know everything about electrical.

I'm finally getting ready to buy my own panels now that my arch is installed, and had no idea all the drama..err...discussion about series vs. parallel. Last I read a year or two ago, it seemed that everyone thought parallel was the way to go due to shading. I read several recent threads, informative reading.

One thing I didn't see answered (but might have missed): How do you know the capacity of the bypass diodes used in a panel so that you don't exceed the number of panels in series, or is this even a concern? I'm guessing most panels would have a higher max "system" voltage for their bypass diodes than we would ever see in a boating application, simply because we only have room for a few panels? I don't see this rating on spec sheets, or can you assume a very high number?

Frank

That is the beauty of series connection: the current stays the same so diodes are always good. The diodes don't care about voltage other than the 0.67V that they keep for themselves when conducting. Their power rating is that voltage times the number of amps that flow through the series string.

I have been telling people to put panels in series for ever, but it was only recently that enough woke up to start supporting that. People have become Comfortably Numb

ciao!
Nick.

[Cotemar]

So in summary, if you are using an MPPT controller than:

Parallel wiring = Serial wiring power output


Use parallel wiring if your panels are in two locations, far apart.


Use serial wiring if your panels are butt together or very close to each other

Use 10 gage or larger wire from Solar panels to MPPT controller
Put fuses on + just before wires go into controller

Put MPPT controller as close to batteries as possible.
Put fuse on + just before connection to battery.

[s/v Jedi]

Originally Posted by Cotemar

So in summary, if you are using an MPPT controller than:

Parallel wiring = Serial wiring power output

Use parallel wiring if your panels are in two locations, far apart.

Use serial wiring if your panels are butt together or very close to each other

Use 10 gage or larger wire from Solar panels to MPPT controller
Put fuses on + just before wires go into controller

Put MPPT controller as close to batteries as possible.
Put fuse on + just before connection to battery.

Ehmmmm.....

We didn't calculate anything about loss in cabling for parallel vs series for different locations of the panels. The wire gauge is completely dependent on the length of the cabling and the amperage flowing through. Fuses are not required in the array cabling, but I would strongly suggest a miniature battery switch there. There must be a fuse to every wire that connects to a battery except if it goes to an alternator so yes to that fuse.

If you end up with 2 panels far apart and you get a positive and negative wire from each panel together at the controller, you can easily choose either series or parallel connection. The loss in the cabling will be exactly the same for either method because the wire is the same length and diameter and the same current flows through it. This means that on a boat, the series connection still outperforms the parallel connection in that case. You will find it is very hard to find a case for parallel connection on a boat, except when the controller does not support it.

ciao!
Nick.

[nigel1]

Originally Posted by s/v Jedi

Okay:


- the blocking diode is something different than the bypass diodes. If the dealer mixes that up you better stop listening to him!


ciao!
Nick.

I was just about to email the dealer regarding this when I got this in the inbox

I had a sudden realisation that I had misread your email yesterday - I read it as 'blocking diodes', not 'bypass' as you had written. There are three bypass diodes in the CS135; one for each string. Apologies if I have confused matters.


So, thats cleared up, but has anyone used or is using these Mono-crystalline panels

[stevensuf]

jedi, the losses in cables will be less in series as the current flow is much smaller, eg in parallel say you have 2 panels oc 20v ic 5a in full sun you will have 10a at say 20v, in series you have 5a at 40v, from my limited testing in the back garden parallel worked better with shading.

[mrm]

Originally Posted by stevensuf

jedi, the losses in cables will be less in series as the current flow is much smaller, eg in parallel say you have 2 panels oc 20v ic 5a in full sun you will have 10a at say 20v, in series you have 5a at 40v, from my limited testing in the back garden parallel worked better with shading.

Nope, in this specific scenario:

If you end up with 2 panels far apart and you get a positive and negative wire from each panel together at the controller,

he is right.

Think about it. Each panel has its own set of wires, so currents do not add up in the wiring. They add up at the charge controller terminals.

Marius

[nigel1]

Sent an email to Cleversolar, who also market Kyocera, to ask how many bypass diodes were fitted to the latest Kyocera 135 panels, to which they replied - Two, one per 36 cells.
The Cleversolar panels have 3 bypass diodes. As I was asking this is regards to shading, and only because this thread had prompted me to do so, the guys at Cleversiolar told me they had not thought about it in this regard, and obviously there was an advantage in having three diodes as opposed to two, and as a thanks for getting them to think about it, they have given me a 10% discount voucher.
So, thanks to all who have contributed tio this thread, it really did help

[s/v Jedi]

Originally Posted by mrm

Think about it. Each panel has its own set of wires, so currents do not add up in the wiring. They add up at the charge controller terminals.

Exactly. When panels are very close together, you can also wire them efficiently in parallel just like batteries, but then it requires a bigger diameter cable to connect that parallel array to the controller. In series connection, you can use the same wire throughout.

For back-yard testing with shading: the only way that works is with controller and half-empty battery attached, and measuring both amps and volts to the battery to calculate the power transfer in watts.

ciao!
Nick.

[stevensuf]

that is what i did, took 2 panels connected in parallel to the mppt controller, measured current/voltage then shaded half of one with a towel, battery was pretty dead, then did parallel comparison, much more output in parallel with one half shaded and the panels have 3 bypass diodes . try it yourself its easy to do.

as others have said if your panels are never shaded, series is better, but if partial shading then parallel is better.