Solar - Series Or Parallel?

[noelex]

Or you need a more suitable controller perhaps?

Correct. This is not a problem with most controllers.

It is a pity. The Victron controllers are ideal for our sized installations. They are available in a large range of sizes at low prices (for a quality controller), which also makes them ideal for the “one controller per panel” approach. They are commonly used this way, but really need a means of disabling the tail current cut off function to work well in this role.

 

[Tim Good]

I might add to this conversation that an email to Victron is not ignored. They do come back with good solid answers to technical questions and varied installations. Whether the question is answered by an experience individual or ore determined answer .... well I don't know but they are helpful.

 

[lw395]

Correct. This is not a problem with most controllers.

It is a pity. The Victron controllers are ideal for our sized installations. They are available in a large range of sizes at low prices (for a quality controller), which also makes them ideal for the “one controller per panel” approach. They are commonly used this way, but really need a means of disabling the tail current cut off function to work well in this role.

The Victron controllers serve a market much bigger than leisure boats. So it's not surprising they don't have every feature which might be desired by a minority among marine users.
If you are not going to switch down to float voltage when the charge current tails off, how do you propose to enter float mode?
Or are you happy for your fully charged batteries to be at 14.8V for the whole daylight period, day after day?

Maybe there is a market for a device to supervise multiple VIctron controllers, but alas I suspect it's a pretty small market.
In the long run, the future is probably individual MPPT converters per cell or small group of cells, possibly several controllers integrated per panel. I was looking at chips suitable for this last year.

 

[noelex]

Whereas on a boat panels can be in very different locations. I have a pair on an arch at the transom and another pair on the coachroof, just ahead of the sprayhood. There will be days when the pair on the arch will be in a significant amount of sunshine before the pair on the coachroof are out of the shadow of the sprayhood.

Each pair is wired in series and the two pairs are then in parallel, connected to a single controller. Two smart controllers on a boat isn't a good combination.

On the contary the situation you descibe is exactly the sort of situation where multiple controllers have an advantage.

As you point out the panels on the arch and the panels on the coachroof are likely to experience very different conditions. It is likely therefore the MPP will also be significantly different. By connecting both sets of panels to one controller there can only be a single input voltage which will be a compromise between the MPP of the different panels.

Multiple controllers remove the compromise.

 

[noelex]

The Victron controllers serve a market much bigger than leisure boats. So it's not surprising they don't have every feature which might be desired by a minority among marine users.
If you are not going to switch down to float voltage when the charge current tails off, how do you propose to enter float mode?
Or are you happy for your fully charged batteries to be at 14.8V for the whole daylight period, day after day?

The primary means the Victron algorithm determines when to drop to to float mode is based on the time after the bulk voltage is reached. This is modified by the start voltage, and the user controlled setting of absorption time. This primary system works fine with multiple charge sources and does not need modification.

The “tail current cut off feature” I am suggesting should be disabled with multiple charge sources will never be used on a boat that is not in storage. If there is more than a 1A draw from the house load (from fridge instruments etc) the feature cannot function unless multiple charges sources are present and here it will work incorrectly.

 

[lw395]

The primary means the Victron algorithm determines when to drop to to float mode is based on the time after the bulk voltage is reached. This is modified by the start voltage, and the user controlled setting of absorption time. This primary system works fine with multiple charge sources.

The “tail current cut off feature” I am suggesting should be disabled with multiple charge sources will never be used on a boat that is not in storage. If there is more than a 1A draw from the house load (from fridge instruments etc) the feature cannot function unless multiple charges sources are present and here it will work incorrectly.

I'm not familiar with the finer workings of the Victron devices.
But generally, if you are pulling current from the battery in a variable fashion, it's going to be hard to determine the state of charge just by the charge current and voltage.
Whether that matters would need to be debated for an individual boat.
As more people get bigger panels and all that, some of these issues will become very real for some people.

 

[GHA]

It has been some time since I watched the video. Unfortunately, my internet connection is not sufficient to refresh my memory, but I think you are confusing Voc and Vmp.

The Voc has to be +5v for the Victron to start. The 16.5V is a Vmp measurement.

Once started, the criterion drops to +1v.

The victron will stay mode 0 (off) and seems to keep the panels open circuit until the panel voltage is more than 5v above battery voltage, then pulls the panel voltage down to Vmp. Below is data from a victron bluesolar with a mains charger on float so is a bit higher than just batteries would be.

The 16.5v panel voltage in the video is because the MPPT regulator is controlling the panel voltage, nothing to do with open circuit. All this thread is assuming 36 cell 12v panels, either in parallel or series a couple of panels will have no problem getting OC voltage high enough for the victron to start if there's enough sun for even just a little power. Just tried it on a 100w panel hanging vertically on a cloudy day - 20.55v open circuit, 0.6A closed circiut. Plenty to turn on the victron.

Another video in here > https://www.gonewiththewynns.com/Sailboat-Solar-Series-Parallel-Shading

 

[noelex]

I'm not familiar with the finer workings of the Victron devices.
But generally, if you are pulling current from the battery in a variable fashion, it's going to be hard to determine the state of charge just by the charge current and voltage.
Whether that matters would need to be debated for an individual boat.
As more people get bigger panels and all that, some of these issues will become very real for some people.

The challenge is to try and determine the batteries state of charge without a knowledge of the current that is entering the battery.

The Victron units use a clever and rather unique algorithim that is based on the battery voltage, the battery voltage at start up, the time since the battery reached bulk voltage and a user adjustable “absorption time”.

Note: Like most controllers current does not enter into the above algorithm. It is solely voltage and time based. This is because the current absorbed by the system is unrelated to the batteries state of charge. If the system is accepting a high current it could be because the batteries have a low state of charge or it could be because there is a high house load and the batteries are near full.

There are some (rare) solar controllers that also measure the current entering the battery separately from the current produced by the solar panels. This involves an extra shunt, but the combination of voltage and current enering the battery is the best way of determing the batteries state of charge, but this only applicable to a small number of expensive controllers.

However, the Victron has unusually for a solar controller has incorporated an extra feature that does use current. This “tail cutoff” assumes that if the batteries are at absorption voltage and the solar current is low (typically less than 1A) the battery must be full. This is a valid assumption if the solar controller is the only charge source, but is completely wrong if multiple charge sources are present. Hence if used in a system with multiple charge sources this small part of the algorithm should be disabled. This would not effect the normal drop down to float.

The Victron software already incorporates the ability to disable other parts of the algorithm such as temperature compensation and initiatiating an equalisation cycle. If they added the ability to disable the “tail current cut off feature” it would be a help, particularly for boats with many charge sources.

 

[lw395]

If some other charge source is reducing the Victron's current to a low value like 2A for a large bank, then unless the other source is much lower impedance or higher set point, the charge current is likely not much more than 4A anyway. With the size of bank that needs more than one 50A charger, you are probably not missing out very much. The other source, if it's solar will take uup the slack to some extent.
I think the tendency is to be cautious about overcharging, above trying to always get 100% charging.

The more serious problem is that even a small charge from another solar controller or other source will affect the start voltage enough to reduce the absorption time.

There are lots of smart chargers around which use the falling current at absorption voltage to determine when to switch to float. Some use the slope of that current, i.e. when the value is below a certain level and constant over some period. But these may be fooled by a load on the battery.

There are limits to what you can expect from a smart charger in terms of deducing the state of charge when there may be other loads and sources in play. It's probably better to err on the side of not blowing up the boat.

 

[noelex]

If some other charge source is reducing the Victron's current to a low value like 2A for a large bank, then unless the other source is much lower impedance or higher set point, the charge current is likely not much more than 4A anyway. With the size of bank that needs more than one 50A charger, you are probably not missing out very much.

The charge voltages will never be exactly the same.

So, for example, if my solar controller and alternator are set at 28.8v. The batteries are not fully charged, have just entered the absorption phase and are accepting 30A from the solar panels to maintain the absorption voltage of 28.8v. If I start my engine and the alternator controller is set even slightly higher at 28.85v, the alternator will deliver the just over the 30A to maintain the 28.85v. The solar controller will recognise that the voltage is above the set point and deliver zero current.

If I then shut the engine down, most solar controllers will recognise the voltage has not been held at 28.8v for the required absorption time (say 2 hours) and will go back to supplying the 30A the battery needs.

However, the tail current cut off feature of the Victron controllers will have incorrectly dropped the controller to float because the tail cutoff current criterion has been met.

This is a simple example of how the tail current cut off feature does not work with multiple charge sources.

The above is a hypothetical example, but this type of conflict will frequently occur with a charger that incorporates a tail current cut off function. It is a pity, as normally multiple charge sources do not create any significant conflict.

 

[lw395]

The charge voltages will never be exactly the same.

So, for example, if my solar controller and alternator are set at 28.8v. The batteries are not fully charged, have just entered the absorption phase and are accepting 30A from the solar panels to maintain the absorption voltage of 28.8v. If I start my engine and the alternator controller is set even slightly higher at 28.85v, the alternator will deliver the just over the 30A to maintain the 28.85v. The solar controller will recognise that the voltage is above the set point and deliver zero current.....

Consider the output impedances necessary for that to happen.
It's much more likely the alternator will contribute something like 20A and the charger 10A.
If the system had been designed, as opposed to put together with COTS units, you could ensure this was the case.


But at the end of the day, fairly cheap electronics are made with compromise specs to suit the greatest number of users. If it doesn't match what you want, you may have to either compromise or look elsewhere. You can get bespoke solutions but at a price.

 

[LadyInBed]

All this theorising using MPPT regulators, does the series / parallel argument change if using a dual battery PWM controller?

 

[noelex]

All this theorising using MPPT regulators, does the series / parallel argument change if using a dual battery PWM controller?

PWM (or non MPPT) controllers do not have any voltage conversion so the panels cannot be connected in series (assuming “12v” panels and a 12v battery bank) only in parallel.
This also means you cannot use high voltage panels.

 

[lw395]

All this theorising using MPPT regulators, does the series / parallel argument change if using a dual battery PWM controller?

If it's a typical 'PWM controller' which does not convert DC-DC voltages, then parallel is your only option.

 

[GHA]

PWM (or non MPPT) controllers do not have any voltage conversion so the panels cannot be connected in series (assuming “12v” panels and a 12v battery bank) only in parallel.
This also means you cannot use high voltage panels.

Wouldn't the panel voltage just get pulled down to battery voltage but still output current?

Though why would you, further away from Vmpp , no reduction in cable losses from higher voltage and still get both panels hammered with any shade over one cell...

 

[noelex]

Wouldn't the panel voltage just get pulled down to battery voltage but still output current?

Though why would you, further away from Vmpp , no reduction in cable losses from higher voltage and still get both panels hammered with any shade over one cell...

Yes, as an emergency measure you can connect a high voltage panel, or multiple low voltage panels wired in series to the battery directly, or via a PWM controller (although most PWM controllers will be damaged by high input voltages so you need to check).

The panel voltage will be pulled down by the battery and providing you do not let the battery voltage rise excessively there will be no damage to the panel or battery.

But this is an emergency measure only. It is very inefficient.

For example two identical 12v panels connected in parallel will give roughly double the potential output of a single panel. Two identical 12v panels connected in series via a PWM controller will only give slightly more power than a single panel.

 

[LadyInBed]

PWM (or non MPPT) controllers do not have any voltage conversion so the panels cannot be connected in series (assuming “12v” panels and a 12v battery bank) only in parallel.
This also means you cannot use high voltage panels.

I would have to say not so, my controller is OK for 12 or 24v o/p (self sensing). I have tried it with the panels in series connected to my 12 and 24v banks (not at the same time) without ill effect.

PS This one

 

[noelex]

I would have to say not so, my controller is OK for 12 or 24v o/p (self sensing). I have tried it with the panels in series connected to my 12 and 24v banks (not at the same time) without ill effect.

Have you a link to the controller manual?

If it is not a MPPT controller you will find it fine with a “12v panel” and a 12v battery bank or a 24v panel (or two 12v panels in series) and a 24v battery bank, but it is not designed to perform the voltage conversion that is necessary to efficiently use a different nominal panel voltage from the battery voltage.

Basically it is a controller designed to use with a 12 or 24v house bank.

If you connect two 12v panels in series and connect this to a 12v bank you will not cause any harm, but this is not the way the controller is designed to work. Without the required voltage conversion that is present in MPPT controllers the combination will be very inefficient.

Edit: I can now see the link to the controller. I will take a look at the documentation if my lousy internet connection lets me download the handbook.

 

[PaulRainbow]

I would have to say not so, my controller is OK for 12 or 24v o/p (self sensing). I have tried it with the panels in series connected to my 12 and 24v banks (not at the same time) without ill effect.

PS This one

The max input voltage for that is 30v, two panels in series will give around 40v at the panels.

The auto sensing 12/24v you mentioned is the boat system voltage.

 

[solent clown]

We are awaiting our 50watt panel with a PWM controller, and also an MPPT controller. When they arrive I will be testing with both controllers. First with just one panel, then next week with a second when we source it.
I am looking to test empirically differing set ups with these panels and controllers. I understand the theory well enough, but a pointer or two in method for the experimentation would help, what would people find useful during the test period. Any input gratefully received and will be acted upon.
I will set it up not on a boat, but on a shadow free outside rooftop environment.