maximum smoke for low end charge controllers

**PNjunction** · 11-21-2015, 05:28 PM

Originally posted by MikeCanuck

So should I be going back to using my cheapo on/off controller instead of PWM?
The cheapest garbage kit charge controller is actually best one to use for an undersized panel?

You are on the right track, but NO, do not go back to the 1970's ping-pong controller. Not to mention the fact that it is not temperature compensated!

It all has to do with "duty cycle". Yeah, you could consider the ping-pong cheapo as one that has a duty cycle of about 10 seconds or more ramping up and down. The pwm controllers do the same, but at a MUCH faster rate - so fast that your voltmeter won't be able to detect the pulses. WAAAY more efficient. These ping-pong controllers were best known as hysteresis chargers, not pwm per se.

Us old timers used to try and emulate a more efficient pwm by making the low and high voltage points REAL close to each other - say 14.6 on the low end, and 14.7 for the high. LOTS of relay chatter, and the changing hysteresis of the battery made doing this properly a chore. Think bell-bottoms, happy days, love boat, feathered-hair. bye bye. Morningstar used to have some good docs on the history of it all. Maybe it is still there.

The next step up is MPPT. This keeps the "power point" in the bulk stage up higher during charge, much better than pwm which follows the battery's lead. BUT do note that this mppt controllers actually use pwm! The most simple explanation is that "pwm" controllers do not do buck/boost to keep inside the most efficient power-point during the bulk stage. Mppt controllers do, AND use pwm pulsing. Many threads on that here. Generally, while MPPT is considered the best option, one has to weigh the additional cost vs possible negligable gains when the panel is producing less than about 200 watts, and is in a very warm climate. Thus, an additional panel may be the cheaper solution than going mppt. But this has been rehashed many time in other threads. In the end, a *quality* pwm controller will beat an unknown POS gotten from a disreputable online el cheapo source, so be careful.

Most charge controllers simply go to float too early - as they are really doing the CYA thing for standby/float applications. For cyclic users, you want to run as long and as fast as you can during your solar insolation period. In your case, even doing this won't help - you've GOT to get well into absorb daily which means more panel.

And absorb is the point where the controller merely holds the voltage at an upper level, and the battery absorbs as much as it can - when it does so, the *current* drops well below the max output of the panel, because now the battery terminal voltage and the controller output voltage are starting to get closer and closer to each other. Less voltage difference = less current flowing. Thus the battery is the one that regulates the current flow once the absorb stage (commonly known as "CV") is reached.

So the question is are you reaching absorb and actually FINISHING it? Like C/50 before the sun sets? If not, beef up that panel array as long as the maximum current during bulk doesn't exceed the battery's rating (C/8 for flooded, C/4 for standard agm etc)

**Sunking** · 11-21-2015, 06:23 PM

Mike there is not much difference between a PWM and old relay type controllers. They both do the exact same thing, connect the panel directly to the battery. If your panel is undersized or not getting your battery fully charged, you are better off getting rid of the controller and connect panel directly to the battery through a diode.

The downside to PWM, Series Controllers, or direct connection is they are inefficient because INPUT CURRENT = OUTPUT CURRENT. You must also use extremely expensive Battery Panels vs inexpensive Grid Tied Panels.

Example all battery panels are 36 cell panels that operate at roughly 18 volts A 100 watt battery panel produces 5.5 amps. Remember Input Current = Output Current so you have 5.5 charging into a 12 volt battery is 5.5 amps x 12 volts = 66 watts. Best efficiency can get with a PWM or direct connection is 66%.

OK back to max smoke. Does not matter what controller type you use or do not use. The point is you set the voltage as such it never stops and passes maximum smoke to the battery as long as there is sun light. The absorb stage of charging is the most inefficient charging algorithm there is and takes 4 to 6 hours to complete. There is not enough hours in a day to complete. You do not have the luxury of time to go slow in most solar systems because just about every off-grid system is undersized.

But like I always say forget voltages on a solar system to evaluate battery health. Other than a relative indicator is completely worthless on a operating system. Only way to set your controller to the correct voltage is with a damn hydrometer. Keep cranking the voltage up until you see 100% SOC at the end of the day. What you will most likely notice is you never get to 100% unless your system is over sized or summer time. It is a full time job that never ends and must be checked and corrected at least once a week.

FWIW Absorb i snot a timed event which every charge controller uses. It is a CV event and ends when charge current reaches 2 to 3% of C. Great if you are using a commercial power AC charger. Almost impossible with solar.

I am not suggesting you use your crap controller or no controller, just change the way you use it. FWIW I will let you in on a little secret. When you see a traffic control light like a blinking light that is solar powered. Most of them have no controller. They use a special panel. Well not so special as it is a very old design made by by Shell and BP. a 30 cell 15 volt panel. Requires no charge controller. Only requirement is using a AGM or Gel battery.

**MikeCanuck** · 11-21-2015, 10:59 PM

Originally posted by Sunking

I am not suggesting you use your crap controller or no controller, just change the way you use it. FWIW I will let you in on a little secret. When you see a traffic control light like a blinking light that is solar powered. Most of them have no controller. They use a special panel. Well not so special as it is a very old design made by by Shell and BP. a 30 cell 15 volt panel. Requires no charge controller. Only requirement is using a AGM or Gel battery.

It almost sounds like given the current conditions I'm getting up here (today had totally no sun, it was raining/snowing all day, and most days now are getting very little sun, ie i'd say lower published supposed avg 2 sun hours a day solar insolation in November for where I am), it would be best to just connect my junky 40w panel directly to my junky battery and grab as much as I can get...?

Originally posted by Sunking

Remember Input Current = Output Current so you have 5.5 charging into a 12 volt battery is 5.5 amps x 12 volts = 66 watts. Best efficiency can get with a PWM or direct connection is 66%.

I know my newbie math is completely utterly wrong and I'm just writing below so I can understand... but given the example above, if I connect my 100watt panel directly to the battery instead of using my pwm controller wouldn't this be true?

pwm controller bulk stage is 14v, so Input Current = Output Current if it does 5.5a charging into 12 volt battery at 14v, wouldn't it be 14v x 5.5a = 77 watts under ideal conditions with a pwm controller?

then if I directly connect the same 100 watt panel to the battery with no charge controller, wouldn't it be whatever the operating voltage of the panel across the day? ie max noon full sun 18v x 5.5 = 99 watts? I know it won't be 18v all the time, and as PNjunction was I think explaining, that's why mppt is the best with the buck/boost thing etc, but in *theory*, it seems to me that if I don't have an mppt controller, a pwm charge controller is just limiting things and it would just be better to directly connect the dang panel to the battery and get more into the battery, or just use that 1970's crap controller just to make sure battery doesn't overcharge? Would directly connecting it actually get more into the battery like what my incorrect math is suggesting? Sunking was saying direct or pwm is same? But batteries accept higher than 12v, on my grid smart charger experiment learning how a 3 stage charger works in another thread, I was seeing battery accepting 14.8v, not limited to 12v...

Just trying to re-understand why the need for any solar charge controllers other than mppt again given this thing about maximum smoke and during the winter up here, and if directly connecting the panel to the battery would actually get me more...

thanks again guys...

**Sunking** · 11-22-2015, 12:43 AM

Originally posted by MikeCanuck

I know my newbie math is completely utterly wrong and I'm just writing below so I can understand... but given the example above, if I connect my 100watt panel directly to the battery instead of using my pwm controller wouldn't this be true?

pwm controller bulk stage is 14v, so Input Current = Output Current if it does 5.5a charging into 12 volt battery at 14v, wouldn't it be 14v x 5.5a = 77 watts under ideal conditions with a pwm controller?

Nope does not work that way. Once you get into the Abosrb phase charge current drop significantly. The only time any controller produces full power is the Bulk phase (constant current or constant power), or sometimes in Float mode if the load exceeds the panels output.

That is the point I was trying to get across, you force the Controller to remain in BULK mode from sun up to sun down. Solar panels are current sources, not voltage.

**MikeCanuck** · 11-22-2015, 12:51 AM

Originally posted by Sunking

Nope does not work that way. Once you get into the Abosrb phase charge current drop significantly. The only time any controller produces full power is the Bulk phase (constant current or constant power), or sometimes in Float mode if the load exceeds the panels output.

That is the point I was trying to get across, you force the Controller to remain in BULK mode from sun up to sun down. Solar panels are current sources, not voltage.

Right, so if I cannot change the charge profile of the pwm controller, the next best thing would be to connect the panel direct to battery or use that crappy 1970s technology dumb controller, yes?

ps - about your hint about traffic lights and agm:
during summer peak this year, my 100watt mono panel with pwm controller setup was getting at best 4.5A@13.5v=60.75w bulk stage to the battery according to what the LCD on my charge controller was showing. This is why I was thinking the battery accepts more than 12v. What I'm not sure of is why only 13.5v when according to the controller manual, bulk stage is supposed to be 14v. Is the charge controller or battery limiting it to 13.5v? I think you are saying it is the battery right? Does this have to do with the thing about the internal resistence of a FLA versus AGM battery, so the FLA battery will never accept 18v, and that's what an mppt basically does - take that extra the battery can't accept and change to to what the battery can accept (higher amperage at lower voltage that the battery can accept)?

**Sunking** · 11-22-2015, 01:12 AM

OK you just do not understand how batteries are charged and the current limits of the power source. The voltage of a battery under charge is:

Vb = [Charge Current x Battery Internal Resistance] + Battery Open Circuit Voltage.

So lets say you have 10 amps of charge current into a battery that with an OCV of 12 volts and an internal resistance of .010 Ohms. [ 10 amps x .010] + 12 volts = 12.1 volts. Makes no difference what you have the CC voltage set to. It can be 100 volts, but if all it can supply is 10 amps under the above battery condition is 12.1 volts at the battery.

The only time your controller output voltage equals the voltage set point is when no current is flowing. That is why voltage does not mean a lot on a battery under charge or load.

**sensij** · 11-22-2015, 02:47 AM

I think what the OP is misunderstanding is how the solar panel operating voltage is set. In an on/off or PWM controller, the panel voltage = the battery voltage. The current the panels are capable of delivering at a particular voltage is illustrated by the panel's IV curve. "Battery panels" are designed to give something close to maximum power at typical battery voltages, but given the variables involved, they will inevitably be operating at some sub-optimal condition.

The idea behind max smoke is to hold the highest safe voltage you can and deliver as much current as possible, mostly assuming that there isn't enough energy coming from the panels to really get through absorb. An on/off controller will not do that... it will hit the max voltage and shut off, unless rigged with tight setpoints as Sunking suggests (until the chatter kills the relay). A PWM controller will hold the max voltage in absorb for some period of time before dropping into float... this should be better than on/off, because the PWM cycle can be fast and the voltage can be held in a tighter band.

Hooking the panels straight to the battery will be a bad thing if you do ever actually hit absorb. The idea behind absorb is that the voltage is held constant as the current drawn by the battery gradually decreases to a cut-off point. Without a controller, if the panels are capable of delivering more current than the battery wants, the voltage will continue to rise, since there is nothing to limit it. Rise high enough and the battery can be hurt.

**PNjunction** · 11-22-2015, 06:41 AM

And here is what happens when you connect your dinky 40w panel to a battery bank without any controller:

https://www.youtube.com/watch?v=z5RtZe9AW2E

Once again just say no to that ping-pong hysteresis controller. I think it got covered adequately above.

Sunking simplified it all by just saying bulk=absorb=float to one value. On cheaper controllers, you may be able to disable float, or if you can't disable it, you can cheat by setting the float to the absorb value.

For instance, my Scheider/Xantrex C12 controller allows for custom settings for absorb and float via internal pots. I set the absorb to 14.7v for my agm's. BUT after an hour or so of reaching absorb, the Xantrex will time out, and go to float. That is a bit too early for my usage. Well, there I cheat by setting float pot to 14.7v as well, so it makes no difference. I just absorb and then float at the same value until the sun sets for my cyclic operations.

**MikeCanuck** · 11-22-2015, 09:07 AM

ok i'm starting to scratch the surface of getting it now...

1) panel direct to battery = bad, after battery is mostly full it will boil the battery and kill it or worse, so need controller to limit things

2) on/off controller = bad, it just does turning on and off aka ping-pong with the voltage and current without the smarts to charge the battery properly, good only for to preventing boiling in 1) but not much else

3) pwm, like what a 3-grid smart charger does, it's purpose is to regulate/limit voltage and current at each stage.
The max smoke thing is to fine tuning this to hold it at bulk stage as long as possible with max *safe* voltage.

I think I'll need to spend the rest of the winter to try to fully understand:
-how batteries are charged and the current limits of the power source
-how the solar panel operating voltage is set / IV curves
- then finally i may scratch at understanding mppt

thanks all so far for pointing me in the right direction on all this stuff, it just needs to get into "absorb" phase of my low capacity little brain

**PNjunction** · 11-22-2015, 04:52 PM

Yup almost.

Originally posted by MikeCanuck

1) panel direct to battery = bad, after battery is mostly full it will boil the battery and kill it or worse, so need controller to limit things

While the current isn't usually bad enough to send up a red-flag to the owner, positive plate corrosion from holding the battery at a VERY high voltage for too long does the damage. Hydrogen can build up internally if the vents are clogged, and one spark internally sets it off. Very unsafe.

2) on/off controller = bad, it just does turning on and off aka ping-pong with the voltage and current without the smarts to charge the battery properly, good only for to preventing boiling in 1) but not much else

It spends too much time in the "off" or disconnected position watching the battery's voltage fall before ramping up again. You don't have the luxury of time in solar. Very inefficient.
You could emulate it yourself with your own hands on a switch and a voltmeter. If you were stuck on Gilligan's Island, and this is all you had, then I'd use it.

3) pwm, like what a 3-grid smart charger does, it's purpose is to regulate/limit voltage and current at each stage.
The max smoke thing is to fine tuning this to hold it at bulk stage as long as possible with max *safe* voltage.

Well, actually to hold ALL the voltages in bulk/absorb/float to the same value. Now it is only a 2-stage charger really. Bulk/Absorb. In absorb, the *battery* does the regulating naturally by it's own terminal voltage rising in an attempt to be equal to the absorb voltage setting on the controller. As they get closer in voltage, less voltage difference, and less current flows.

And "PWM" is really just a charging waveform that is different from a steady-state output such as from a typical linear power supply. The pulses (on/off) are so fast, you usually can't see it with average consumer voltmeter gear. In reality, the pulse varies from a very low state (such as in bulk when there is essentially NO pulse, and then it ramps up the pulsing very very rapidly for absorb.

thanks all so far for pointing me in the right direction on all this stuff, it just needs to get into "absorb" phase of my low capacity little brain

Don't sweat it - plenty of material to read. When you understand what's going on, it can be a lot of fun - and most importantly, save you money from buying into what's called a "deficit charge" scenario. In the bell-bottom days, one just slammed in the very largest battery they could afford, and murdered it prematurely. Not cost effective, and a whole lotta' lead going on.

maximum smoke for low end charge controllers

maximum smoke for low end charge controllers

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment