Charge controller for each solar panel?

[socaldmax]

Originally Posted by tinkrman69

every body talks 'bout the power going into the controller and all the mathematical theory. but no one talks about the other side. the level of charge of a battery is tested by the controller and the output is adjusted to compensate. Hence if both controllers are hooked to the same bank they will get an erroneous state of charge reading cause by the other controller output, which could only be eliminated by a battery separator. I've had multiple controllers and now use a single. the only "real world" difference I can see is the extra add cost of materials and maintenance. the only thing that makes a decision on what is working is "do I have enough electricity to operate my stuff without running a engine?" I prefer to keep the cost down.

I think you misunderstand how these controllers work. They are simply buck (or boost) power supplies. They don't do some complex test of battery state of charge and compensate, they simply provide max. current in bulk mode until the battery bank reaches the voltage they were set to, then switches to constant voltage absorption mode and the internal resistance of the battery bank determines charge rate, not the solar controller.

If your 2 controllers didn't work well together, there was a problem with one of them. They should both charge to their setpoints, and those should obviously be the same. If they didn't do that, then there was a problem. I've never had a multi-controller system not work in harmony.

As far as costs go, we had a choice of 2 20A controllers for $100 ea, or a single 45A controller for $450 plus $120 for the display. $200 was a lot cheaper than $570.

[socaldmax]

Originally Posted by a64pilot

You won't have multiple power points, you will have only one, whether you have 2 or ten panels when you connect them to a single bus bar they in effect become one big panel.

I believe your answer you wanted was given to you, combine all MPPTs to one bus and read the one bus, it's the same as connecting them all to the bank



Sent from my iPad Pro using Cruisers Sailing Forum

I'm afraid you will. Each panel has it's own power point that varies minute by minute, based on irradiance, temp, etc. As soon as one panel gets shaded, it's MPP drops drastically, and 1 controller cannot possibly provide 2 separate MPPs on the same wire.

Based on the example, 1 panel is now in the 22v region, the other is in the 38v region. A single controller might set the MPP at 38v and this will completely negate any power from the 22v panel. A separate controller would give max power from the unshaded panel without a confusing voltage from the shaded panel, plus the shaded panel would put out whatever it can, since it's controller is still tracking it's MPP.

[Toys_with_time]

Originally Posted by socaldmax

I think you misunderstand how these controllers work. They are simply buck (or boost) power supplies. They don't do some complex test of battery state of charge and compensate, they simply provide max. current in bulk mode until the battery bank reaches the voltage they were set to, then switches to constant voltage absorption mode and the internal resistance of the battery bank determines charge rate, not the solar controller.

If your 2 controllers didn't work well together, there was a problem with one of them. They should both charge to their setpoints, and those should obviously be the same. If they didn't do that, then there was a problem. I've never had a multi-controller system not work in harmony.

As far as costs go, we had a choice of 2 20A controllers for $100 ea, or a single 45A controller for $450 plus $120 for the display. $200 was a lot cheaper than $570.

If you have more than one MPPT how do they read the true status of the battery as opposed to just seeing the voltage the other unit(s) is putting out?

TwT

[pcmm]

Originally Posted by socaldmax

I bought one Tracer controller, and I strongly recommend not buying them.

Their MPPT algorithm is lousy, even if the solar input is steady, the controller drifts up and down the voltage range, losing power all the while it's off from MPP. It's by far the worst performing MPPT controller I've seen.

Which Tracer did you buy? the RN, BN, A series?

[Privilege]

Originally Posted by tinkrman69

... Hence if both controllers are hooked to the same bank they will get an erroneous state of charge reading cause by the other controller output, which could only be eliminated by a battery separator.

I had read this somewhere else so prior to installation I spoke to the tech's at Morning Star corporation. They told me that the controllers don't recognize the voltage output from the other controller and misinterpret this to be a charged battery. It's the battery that controls the acceptance rate of the controller, not the controller itself.

[ramblinrod]

Originally Posted by epiic

How important is it to have a mppt solar charge controller for each solar panel.

I understand that it helps to make the system more efficient so that when one solar panel is shaded, it won't effect the others but it really complicates the installation.

I have seen plenty of installs and I almost never see people doing it this way.

Thoughts?

Not only is is not important, it simply shouldn't be done, except in rare circumstances. It adds complexity and more potential failure nodes for no real benefit.

The only reason to use more than one controller is if 2 x lesser amp controllers cost less than one controller capable of the full system current (also considering cable costs).

The argument that it protects the output of the other panels if one is shaded is bogus. Shaded panels connected in parallel will not draw down unshaded panels. If connected in series, a shaded panel will hurt the rest, so don't. Problem solved.

With one controller the integral or remote display represents the total charge coming into the batteries.

The argument for redundancy has little merit if one quality big controller costs the same as umpteen little China controllers that may be prone to poor performance or reliability.

A Morningstar Tristar MPPT is a great contolller. Quality and performance costs, but if it helps you get more out of the limited panel real estate one has on a sailboat, they are worth every penny.

[knockabout]

solar regulator for each panel works o.k / we ignore the voltage read out and have a separate voltage meter on the buss bar / small round waterproof voltage meter designed for motorbikes reads constantly and gives a true reading of voltage available at the buss bar / and good for warning signs that battery life may be shortening / if the solar panels are in full sun reading float charge on the regulators and after dark with very little load on the system the buss bar reads 11.6 it is time to load test the batteries and think about replacement

[socaldmax]

Originally Posted by Toys_with_time

If you have more than one MPPT how do they read the true status of the battery as opposed to just seeing the voltage the other unit(s) is putting out?

TwT

The voltage on the battery bus will always read what the battery is currently at. Both units should start and stop their charging in tandem, since they're set to the same charging parameters.

Let's say your solar controllers start charging at 830am and battery voltage is at 12.02v. The controllers might be set at 14.5v, but the voltage on the output of the solar controllers will always be battery V (plus line losses.) So you will see the V slowly rise from 12.02v to 12.1v, 12.2v, 12,3v and so on as the battery charges up. Both controllers will always put out max. charge as long as they haven't hit 14.5v. When both controllers see the battery V hit 14.5, they'll go into absorption mode and hold 14.5v for the set period and the battery internal resistance will dictate how much current it will take, tapering down to very little at full charge.

No charging device will ever be at a different voltage than the battery - they're connected to each other, hopefully by nice thick, low resistance cables.

[socaldmax]

Originally Posted by pcmm

Which Tracer did you buy? the RN, BN, A series?

I believe it was the A series. However, I don't have any indication that any of them have different firmware in them.

Have you heard that they do?

[pcmm]

Originally Posted by socaldmax

I believe it was the A series. However, I don't have any indication that any of them have different firmware in them.

Have you heard that they do?

The A series is the newest of the series and from what I understand ( though I can't verify) has the most up to date algorythms apparently much better then the BN which in turn is supposedly much better then the RN.

[socaldmax]

Originally Posted by ramblinrod

Not only is is not important, it simply shouldn't be done, except in rare circumstances. It adds complexity and more potential failure nodes for no real benefit.

The only reason to use more than one controller is if 2 x lesser amp controllers cost less than one controller capable of the full system current (also considering cable costs).

The argument that it protects the output of the other panels if one is shaded is bogus. Shaded panels connected in parallel will not draw down unshaded panels. If connected in series, a shaded panel will hurt the rest, so don't. Problem solved.

With one controller the integral or remote display represents the total charge coming into the batteries.

The argument for redundancy has little merit if one quality big controller costs the same as umpteen little China controllers that may be prone to poor performance or reliability.

A Morningstar Tristar MPPT is a great contolller. Quality and performance costs, but if it helps you get more out of the limited panel real estate one has on a sailboat, they are worth every penny.

Let me ask you a couple of questions.


First, about 2 panels in parallel, and one gets shaded. Panel #1 is unshaded and putting out max power at 38v. Panel #2 is partially shaded and putting out a small amount of power at 22v. How does the MPPT controller handle this situation? Track the 38v panel? Split the difference?

How is it supposed to extract max. power from 2 different output panels. That's like intentionally connecting a 300w panel in parallel with a 50w panel.

Would you install a system like that?


Second question. Since the TS-MPPT 45 costs in the neighborhood of $500+ with shipping, tax - possibly a lot more in some regions of the world, and a decent quality 20A MPPT controller (very fast tracking) only costs $100 ea, so 2 of them can handle 40A of output, this leaves $300 in savings, or buy 2 spare units and still pocket $100.

Who is going to be able to quickly replace his solar controller in the case of a failure, and who is going to have to spend $500 more to replace the single point failure unit that completely wiped out his solar harvest?

[ramblinrod]

Originally Posted by socaldmax

Let me ask you a couple of questions.


First, about 2 panels in parallel, and one gets shaded. Panel #1 is unshaded and putting out max power at 38v. Panel #2 is partially shaded and putting out a small amount of power at 22v. How does the MPPT controller handle this situation? Track the 38v panel? Split the difference?

How is it supposed to extract max. power from 2 different output panels. That's like intentionally connecting a 300w panel in parallel with a 50w panel.

Would you install a system like that?


Second question. Since the TS-MPPT 45 costs in the neighborhood of $500+ with shipping, tax - possibly a lot more in some regions of the world, and a decent quality 20A MPPT controller (very fast tracking) only costs $100 ea, so 2 of them can handle 40A of output, this leaves $300 in savings, or buy 2 spare units and still pocket $100.

Who is going to be able to quickly replace his solar controller in the case of a failure, and who is going to have to spend $500 more to replace the single point failure unit that completely wiped out his solar harvest?

OK, in answer to the first question, 2 or more panels connected in parallel are just one panel to the controller. It will operate at the maximum power point for THE panel connected to it. An individual panel s made up of numerous cells, following the contolller per logic, it would be even better to have a controller per cell. It sounds silly doesn't it? But it's really no different than the controller per panel argument.

Show me a reliable, exceptional performing "Made in USA"
MPPT controller for $100?

One good controller is far better than a dozen cheap ones. That one would need consider carrying spare cheap controllers is a testament why not to buy them.

[socaldmax]

Originally Posted by ramblinrod

OK, in answer to the first question, 2 or more panels connected in parallel are just one panel to the controller. It will operate at the maximum power point for THE panel connected to it. An individual panel s made up of numerous cells, following the contolller per logic, it would be even better to have a controller per cell. It sounds silly doesn't it? But it's really no different than the controller per panel argument.

Show me a reliable, exceptional performing "Made in USA"
MPPT controller for $100?

One good controller is far better than a dozen cheap ones. That one would need consider carrying spare cheap controllers is a testament why not to buy them.

No, 2 panels in parallel are still 2 panels.

In the example I gave, the unshaded panel is at 38v and max power. If you remove the shaded panel, the MPPT controller can easily determine the MPP of the panel. Conversely, if you removed the unshaded panel and left the shaded panel in the system, the MPPT controller would readjust to 22v - thus the panel does NOT appear to be the same as the other panel when shaded.

Sure, if neither panel is shaded and unicorns are shooting rainbows, everything is perfect and nothing is getting shaded. But this is the real world and panels get shaded. My experience has been each controller can optimize the output from it's one panel as dynamically as the sun's output.


If you feel that "made in USA" is some sort of proof of quality, you're entitled to your opinion. I don't share that opinion. BTW, Morningstar is mf'd in Taiwan, according to my sources. THe company is here in the US, but the units are assembled in Taiwan, for lower labor costs, which they do not pass on to the customers.

I happen to love TS controllers, I've used them on several larger installs, where we were using 4 or 6 very powerful panels, even then I used multiple controllers, 2 60 A units to be precise. Again, if one fails, the customer is still getting 50% harvest. To date, I've installed over 20 of the $100 controllers and 10 of the Tristar controllers. Not a single failure from either of them.

Based on cost alone, one could reasonably assume the $100 controllers would be less reliable, but if one disassembles one and actually looks at construction quality, they realize they're getting a very well built unit with surface mounted devices. That method of mfg drastically reduces costs, vs hand soldering or partial pick and place design.

[noelex 77]

Originally Posted by epiic

The install is significantly more complex and messy. I just want to make sure that if I do go this route, that its worth the trouble.

The gains of multiple controllers are not great, although there has been little practical testing.

The tracking of each panel with multiple controllers is better, but the self consumption rises, and generally the overall tracking is less good (because the circuitry is less sophisticated) so the net gain is only small unless the panels are likely to consistently experience very different conditions.

It is a complex subject, but if the panels are wired in parallel, in very simple terms the shading of one panel will not effect the output of the others even if you have a single MPPT controller, but individual controllers for each panel are likely to extract a little more out of the shaded panel. Against this, multiple controllers will have more self consumption if they have the same complexity of circuitry.

There are other drawbacks of multiple controllers. While the fear of multiple controllers poorly regulating the batteries is overrated, it can become a problem if you have 3 or more regulators, especially if the parameters are not easily adjustable, as is often the case with cheaper regulators. You can end up with a very long absorption time. The lack of easy adjustability of parameters also makes it hard to regulate the batteries perfectly especially with Gel and AGM batteries.

Overall, I think there is little to choose been a single expensive regulator and multiple cheaper regulators. Both are a valid approach. Two controllers for say port and starboard panels are another option that combines many of the advantages and drawbacks of both methods for larger arrays.

If it is significantly more complex and messy. I would certainly stick with a single controller.

[socaldmax]

I think the sweet spot is 2 controllers. Let's say you have 6 panels putting out 80 A total. I'd connect 3 each to a pair of TS-MPPT-60 solar controllers. Sure the 45 A model will work, but the 60A is only $100 more ea. Now you have a quality controller running at 2/3 power and a redundant system in case of a failure or lightning strike, you might still have the other controller to give you 1/2 of your power.

It just occurred to me that although I've been typing "one controller per panel" I've never installed a 6 panel, 6 controller system. I've installed a lot of 2 controller, 2 panel systems and almost as many 2 controller, 4 or 6 panel systems.