solar charge controller analogies

I was in a weird mood today:

PWM is like a turtle on its back when battery voltage is low. It struggles until it can get back on its feet, after which everything is fine. Adding in isolator charging turns the turtle over and onto his feet.

Example: When the battery is 10.5v a 100w Renogy poly {on PWM} would make a maximum of 59.91w (5.63A x 10.5v). At 12.1v it would make max 68W. At 13v it makes max 73w.

MPPT is a weeble that can’t get flipped onto its back, but costs 3x as much.


backchannel: island of misfit [solar] toys

Our unfortunate fellow has already spent hundreds on some stuff that won’t play well together.  Each component is fine on its own, interestingly enough.

good news first

The setup (other than the oversized inverter) is sized rather well.  200w of panel, 20A of b2b charging, and 100Ah of battery.   Good.

the less good news

Renogy needs to be horsewhipped Continue reading “backchannel: island of misfit [solar] toys”

the problem[s] with 1:1 solar charging

this is a draft

1:1 rule of thumb

There is a rule of thumb that says a 1:1 ratio is the minimum amount of solar needed to charge a deep cycle bank.  The long form is 1w:1ah, which means 1 watt of solar panel for every amp-hour of lead chemistry deep cycle battery capacity under good insolation.  So if you had 150w of panel and 100ah of battery it would be 150w:100ah, or 1.5-to-1.  Yes, I know the case on the unit abbreviations is wrong but it’s easier to read.

As with all rules of thumb, it serves as Continue reading “the problem[s] with 1:1 solar charging”

backchannel restraint

So proud of myself.  Didn’t post the following:

It’s not “hidden in the manual”, the manual doesn’t say “it’s bad”  and it’s not a “bad controller”.  This situation derives from a failure to understand product specs.

To paraphrase Dave Chappelle, “that’s how PWM works;  that’s what PWM is for.”   I encourage you to slow down on buying components until you have a fuller understanding of how they work and how they support whatever it is you want to accomplish.


To give you an idea what he’s up against,  in the same post he says “Keep in mind that for now I’m only feeding the one single Optima Blue Top AMG battery“.  This is an idiot trifecta (quadrafecta?):

  1. it’s a “marine hybrid”, not a deep cycle in any normal sense
  2. AGM is a poor choice for solar-only charging
  3. it’s AGM not AMG
  4. Optima is hype, overpriced for people who thrive on marketing pap. The Optima he bought is roughly 6x more expensive than a normal solution, and will underperform.

An Optima battery icon could be the avatar assigned to offgrid folks who make counterproductive choices.


backchannel: new guy nails small solar setup

Most of my backchannel comments are lamentations about bad decisions and techncal ignorance.  But this newcomer did a great job designing a small system: and it’s worth talking about.

> Just need enough power to charge a phone and occasionally a laptop, power the roof vent fan, and a small rv fridge would also be nice.

It’s nice to see someone have a reasonable expectation of what can be powered from a small setup.  No electric cooking, no A/C, etc.   Carry on.

Here are the parts he picked out, and my comments on them: Continue reading “backchannel: new guy nails small solar setup”

backchannel: 300w panel making 20w

from this reddit post:

I pointed to the RVwiki Is My Solar Working? article, but let’s flesh this  one out a bit.

> I have a 300 watt solar panel. I have it wired into a MPPT charge controller. On a super bright sunny day I get about 40V coming in, but I’ll be lucky to get more than 0.5 amps. That’s only 20 watts. Much less than I should be getting. Anyone have any idea why this is happening?

The most likely scenario is the bank is fully charged and it requires 20w to maintan Vfloat (or late Absorption).

A 300w panel is likely nominal 24v, with Vmp around 36v and Voc a little over 40v.  OP’s “40v and 20w” exactly describes what it looks like when MPPT is limiting current to keep battery voltage at a given setpoint.

Note:  OP elsewhere describes the bank as 200Ah AGM,   Trailing amps at the end of Absorption would be C/200-C/100 (1-2A, likely 15w-30w).    Look familiar?  🙂

Continue reading “backchannel: 300w panel making 20w”

backchannel: “my [200w] solar refills my batteries by 10:30 am nearly every day!”

I think it’s worth talking about so onlooker solar newbies don’t get confused about the abilties of their solar config, worry about whether or not it’s working, and/or murder their lead batteries.

From this post on the Promaster forum.  This is the system in question.  The author and I are running the same battery bank.  He charges it with 200w of panel and I charge with 570w;  I still don’t get my bank fully charged “by 10:30am nearly every day”.   It can happen (a night of shallow cycling, early AM driving, etc)  but it’s not normal.   I am not calling BS;  rather, I suggest critical information is either being left out or not understood.   Either way it’s bad for solar newbie onlookers.


I think 200w of panel is a real sweet spot for vandwellers. I like that amount and recommend it as a sane default.

> my solar refills my batteries by 10:30 am nearly every day!

Depends on what “refilll” means.  It’s pretty easy to hit (an incorrectly measured) “100% SoC” by voltage ~12.8v or even Vfloat.  It’s pretty tough to get deep-cycled Pb fully charged (ie, completed Absorption down to battery manufacturer spec — endAmps) with solar-only charging by 10:30am.

Given that lead-chemistry batteries take 2-3 hours to finish Absorption after deep cycling, Vabs, usually in the mid-14v range, would have to be attained by 7:30am-8:30am. Not too many solar-only systems can do that, even significantly overpaneled ones charged by MPPT under excellent ouput conditions (temperature and insolation).

Caveat to onlookers: the controller switching to Float does not mean that Absorption stage was successfully completed. It means the controller switched to Float.  The way to know for sure that Absorption is being completed is to watch your meters during late Absorption and verify that the battery is accepting something like a typical  C/200-C/100 cutoff (or whatever your battery mfg recommends). For a typical 200Ah lead bank this would be something like 1-2A. If your controller allows you to set endAmps of C/200 or whatever, go ahead and set it. Most consumer controllers don’t do this (or can’t separate out battery charging and loads to measure accurately) so you may have to just max out any Absorption duration setting that is available. Once you know your system is finishing Absorption correctly you can let it fly on autopilot. If you are offgrid and your controller refuses to finish Absorption, you might choose the nuclear option and set Vfloat == Vabs (check water levels more often).

Example with data: I am putting ~175w into a 220Ah bank this partially-cloudy morning in NV and at 10:30am local I still hadn’t kissed my 14.6v Vabs. This puts me at a fully charged (“filled”, IMO) between noon and 1pm. My system has more panel but because I am running loads on my system (~200w worth of laptop and crypto mining) this is a pretty good approximation of how a 200w:215Ah system with no loads would charge the bank on a good day.

If I stop being an idiot and don’t run the crypto rig until later in the day I can typically get fully charged by noon assuming my preferred level of snowbird insolation. 🙂 This is with an overpaneled system and not the usual experience with solar.

Having said all that, banks cycled lightly to ~80% or assisted by alternator (or other) charging in the early morning would have a decent chance of being fully charged by 10:30am.

backchannel: PWM makes more power at 80% state of charge than at 50%

a continuation of this reddit post:

Because I drank more coffee than usual this morning:

PWM power output fluctuates with voltage because amps * volts = watts, and panel current stays pretty stable under normal insolation. Using the Renogy 100w mono at 70F ambient again, and some crude assumptions about voltage/SoC:

  • MPPT output: 82w. Much like the honey badger, “MPPT don’t GAF about Vbatt”.
  • PWM output at a high-ish Vabs (14.8v): 5.29A x 14.8v = 78w <– very close to mppt! Hmmm….
  • PWM output at Vfloat (13.6v): 5.29A x 13.6v = 72w
  • PWM output at 100% state of charge: 5.29A x 12.8v = 68w
  • PWM output at 80% state of charge: 5.29A x 12.5v = 66w
  • PWM output at 50% state of charge: 5.29A x 12.1v = 64w <– PWM is hating life down here

Corollary 1: one could tweak PWM to get a bit more power by configuring setpoint voltages, while still staying within battery manufacturer recommendations. With PWM and similar controllers, higher battery voltages directly relates to higher overall output.

Corollary 2: This one isn’t obvious, but other things being equal PWM will typically get more power from poly panels than from mono due to the latter’s higher voltage (and necessarily lower current for the same 100w rating). In the case of the Renogy 100w examples PWM makes about 6% more power on poly than mono.

Article: Bogart on PWM controllers

Here are some thoughts on the PWM article from Bogart’s FAQ, with a focus on solar power for boondockers.  Ralph (owner of Bogart and author of the text below) has a background in off-grid housing.  He is a good engineer and is well-respected in the solar community.  Below I will suggest his housing-centric observations are not completely in line with the needs of vandwellers and other mobile off-grid folks.

I’ve added emphasis in some of the quotes.

…Why did you choose PWM technology instead of MPPT for your SC-2030 Solar Charger?”

The “good” for PWM: It is simpler and lower cost technology.

Absolutely.  I use PWM, MPPT, and shunt controllers where each is best suited.

Under some common circumstances–it can actually deliver more amps to the battery. That could be when:

(1)days are moderate or warm, with few clouds.

Continue reading “Article: Bogart on PWM controllers”

PWM loves poly

[this is an old draft I’m going to publish]

I was thinking about how common 100W Renogy panels are (both mono and poly) and how to extract maximum power from them with PWM controllers.

The question is on my mind because no folks will link to one of Renogy’s mono/pwm kits and ask for feedback.  I am annoyed that they offer mono/pwm at all;  it’s an inefficient Continue reading “PWM loves poly”