Expert Power SCC with bms wakeup

[Substrate]

Have I been living under a rock all these years? How many SCC's have a bms-wakeup feature?

Ok, so this Expert Power CC is a small pwm type (although they have mppt as well) - with apparently scant specs other than what is reported by an Amazon reviewer as a bizarre pulsing action instead of a true CV. But maybe 14.4v at the top end with some sort of bizarre long pulsing?

I can't wait to find out - should be here in the next week or so to put my Fluke tracker all over it.

This ham radio guy did a video, so here's what it looks like. Seems like a few rebadges out there. Hopefully I can glean more info hands-on.

I don't really need it, but the bms wakeup should come in handy since all I have now are LFP with bms. And I could see in a power outage not looking forward to hooking up to my car battery with an inverter, and using an ac charger with bms reset to wake up the bms one can be in a panic in an emergency and not thinking straight leading to other issues...

I guess it could come in handy for those in the boonies with only one LFP battery that gets totally drained overnight - and have to wait for daylight for a bms reset from this controller.

We'll see when it gets here. I never really looked for a bms reset function in an SCC before.

 

[Substrate]

Got it in today. Will have to wait for sun tomorrow to totally flesh it out. Possibly a rebrand from Bateria Power or generic manufacturer.

Seems to default to LiFePO4. Even after a factory reset. Nice.

Which means 14.4v top end from factory. (Actual user setting is anywhere from 12.0 to 14.8v)

I'm really intrigued to find out about what they consider a CV / aka absorb. Hard to truly figure out what they mean by being able to disable pwm confusing some bms'.

Minor stuff - NOT going to use the usb ports. It's a single board design, and for me, I don't want to risk blowing out a usb port taking down the whole controller. Since there is no overload rating, I'm not going to push more than about 10A through this 20A rated one - this will allow for edge-of-cloud lensing affects to have some headroom. Also, the fets looks beefy, but they use the back metal plate as a heatsink, and use thermal stick on pads for the connection.

Oh, and the front panel voltage measurement is .1v higher than what is shown on my Fluke 87V. Whether that is a display shortcoming, or actual circuit measurement during charge will have to be seen.

So I'm not expecting a Victron or Midnite Classic here.

Dying to find out if it truly does have a bms-reset feature, and what that funky "pwm" option setting *really* does.

We'll see.

 

[Substrate]

DOH! Looks like these little white ones are positive-ground! The "upgraded" ones are negative ground. Just measured, mine is positive.

[Update] I just got schooled - maybe I won't freak out.

What is a charge controller with a "positive ground"?

Specifically, what is the difference between these two Renogy 10A charge controllers? They are both identical in appearance, but one is "positive ground".

diysolarforum.com

Yep, Expert Power, BougeRV, Bateria power, HQST etc - coming from the same parent company.

Thanks to BougeRV, also found out that you have to totally disconnect panel AND batt, and hook it all back up if you make any changes in the controller. Good to know.

Kind of losing interest. If it actually does a bms-reset, I might just call it a day and put in a utility drawer.

 

[Substrate]

YES - it has a bms-reset!

Took my little 12ah Talentcell down to where the bms disconnect kicked in. Only 0.2v at the terminals.

Hooked it up to the controller, attached a little 60 watt panel, and it woke right up. It appears that the controller got enough from the panel to get it's brain together, recognize a 0.2v battery, and take it out of bms disconnect.

While all the manufacturers still recommend hooking up in the proper sequence (batts first, panel second), some of my older controllers would not even wake up with what appears to be no battery connected.

This one does - at least in the LFP mode (b07) and did the reset. I didn't try the lead acid settings.

These things seem to be everywhere rebadged. HQST, you name it. Ternary and sometimes LTO battery support too, although I can't say if any other manufacturer defaults to LFP. My Expert-Power badged one did. Probably wise to doublecheck the settings.

But yeah, the "pwm" mode surprised me. Not what you'd expect. More later.

 

[Substrate]

PWM regulation - not fast, but seemingly random pulsing with long/short duty cycle(s)

With pwm on (default), the system was set for 14.4v. The panel was exposed to constant sun. I consulted the display, an inline Powerwerx power monitor, and deliberating overall was my Fluke 87V on the batt terminals capturing min/max/average voltage values. Long story short, this method resulted in an average of 14.32v as per the fluke during charge, and dropped to an average of 14.2v just before I lost the sun. So the CC is about 0.1v conservative as compared to the display. Ok.

This pwm mode seemed to turn from CC when the batt terminals reached 14.1v.

Fascinating Jim. This method of pwm is unlike the typical one where once you reach a set voltage after CC, the input is chopped at like 33khz or so (33,000 times a second to establish a net average), but waaaay slower with longer periods of intermittent charging with no current flowing at different rates.

The manufacturer provides a way to turn this pwm OFF, because they say that "pwm noise" may trigger a bms over-voltage disconnect. Maybe it's noise, but I suspect a proprietary algo-sauce.

What I witnessed was that the Fluke was catching super-fast intermittent voltage hits of up to 16.55v ! Not all the time, but the Fluke nabbed them. But NO current and so freaking fast that the inline powerwerx couldn't detect even 10ma if it was there. Thing is, these super-fast transients did NOT trip the bms in my Talentcell. I even witnessed a hit at 17.55v, but no current.

The longer pulses would change randomly from say full power out from the panel (about 3.2A), and the next one after a few seconds would be 1A, and then seconds later back to full power, and then seconds later maybe only 200 ma. But there was no correletion with all that voltage jumping from a low of 13.8v to the higher voltages. This went on for hours.

I did multiple iterations of allowing the Fluke to measure the average, and it *always* came up the same at 14.32v. I would have thought that "noise" as they state it, would be more random with my measurements.

Eagle-eyed users may notice that the display during some of these times will jump around too - exceeding their high voltage setting. One vendor had noted that this was "normal, fuggedabout it". Very true. The display wasn't as fast as my Fluke however.

While I can't confirm this, I have measured something like this before in the Tecmate-Optimate chargers during their final balance/optimize stage - albeit they never rise to such high voltages as this scc does. The theory being that during the leading edge of the pulse, it allows for laggard cells to get some energy before a runner starts to over-react.

Whether this was the case here, or maybe just my tinfoil-hat on that this noise is special-sauce algo time, I'm not sure. At any rate, I left the charge on all day till the sun set, and it didn't seem to overcharge. After 8 hours of rest, (no charge, no load) the talentcell is sitting at 13.57v, where it usually sits with my ac chargers.

Tomorrow, I'll do this again with the PWM turned OFF and see how it plays. Will it just stop at my target voltage, or act like a normal pwm cv ? Guess we'll find out!

 

[Substrate]

PWM turned off, acts like a normal CC/CV

That was quick. Purposely taking it out of the default to turn what they call pwm, turns it into a standard cc/cv routine.

And again, even though I set for 14.4v in the controller, the actual terminal voltage was 14.3v as per a Fluke 87V. Close enough for most, but good to know for those who set things at knife-edge for whatever reason.

Essentially a simple 2-stage cc/cv controller at this point. With no tail-current dropout setting, it dutifully brought my battery to full charge, (eventually zero current flowing) and holds the CV for hours on end. Not good long-term.

I'm sticking to the default mode of the long-duty cycle pulsed pwm, as that will be less harmful to the cells in the long run. As long as your bms isn't oversensitive to it, those milli-microsecond low energy "noise" spikes in this mode can be safely ignored. If it is, you can turn it off and go back to bog-standard cc/cv. Just be mindful that if you DO reach a full charge, it can actually be OVERcharged when the current under constant panel exposure is allowed to dribble all the way to zero amps.

Most of this is covered in other threads here, so I won't try to cover it here. Long story short - if you actually need bog-standard cc/cv, you'll know why. If you don't, leave the controller's "pwm mode" in it's default for long-duty pulsing to reduce the long-term exposure to full charge.

 

[Substrate]

Conclusion - I actually like this thing for small projects, even though I bought it only for one purpose - bms reset.

The lack of full specifications, such as overload rating (to handle edge-of-cloud lensing) makes me purposely derate the amperage specs by half just to be safe. Does my unit have ambient temp-comp if I decide to charge lead acid? Not exactly clear. Am I positive or negative ground? (though in many cases this will not matter, but for some it does).

Having to go to 5 different rebadgers sites to try and glean important info (such as if when you change a setting, afterwards you should disconnnect *everything*, and connect up again to make sure the changes are actually recognized) was maddening.

Marketing things like using a 32-bit cpu are fluff. I could give a rat's a** about that. Stick an 8-bit Z80 cpu in there - does the thing actually work?

There are a few different models / upgrades out there beyond just 10/20/30A models. Some "upgraded" models have a backlit LCD. Some will now toggle automatically between voltage and current, rather than having you go into the menus to see the current. Some have a different style of casing. No big deal, since I use my own monitoring equipment.

Thing is, once I put my Fluke on it to vet the default long-duty pulsed charge algo, instead of a bog standard CV, I was pleasantly surprised. That is healthier for the cells when reaching full charge. Whether that long-duty cycle pulsing actually helps passively balance cells instead of just relying on CV and bleeder resistors remains to be seen. They don't mention this, but there is that possibility.

Honestly, at this end of the spectrum, I was prepared to rip it apart. Sure the usb ports to me are superflous and won't be used, and the lack of full specs making ME do all the work to find out what is truly important is a bit maddening.

The one I got from Expert Power seems to default to the LFP settings. Not sure what the defaults are on the other rebadges. Best to check regardless.

Online, it looks like BougeRV has the latest models, although I have only tested these earlier square boxes. If they say it supports LTO or ternary batteries (which I don't have or use - just simple LFP), these are probably it. But, GAH - they say it can "jumpsart" your lithium battery - which is totally wrong and dangerous, but means bms-reset. Where do these marketing guys come from and how misleading certain statements are!

Is it the gold-standard? Certainly not. The bms reset was certainly cool. But that long-term duty cycle pulse surprised me, having seen something similar (although not exactly like it) in the Tecmate-Optimate chargers.

So instead of sitting in my utility drawer, I might just get another one.

 

[Substrate]

Minor note - 100ah Talentcell didn't trip either.

Bulk charged my 100Ah Talentcell to near full, and put this SCC (default LFP config) and a 100w panel to finish it off.

Same basic operations as before with the 12ah batt, but the highest milli-microsecond pwm spike caught by the Fluke, and fast and short enough not to trip the bms with their "pwm special sauce" was 15.6v Most of the time it never even went that high. Current levels jumps around slightly with many seconds in between. Standard cc/cv (their pwm off mode) works as expected, but I wanted to see how a larger sized LFP handles their default.

Conjecture time: Since I don't have interior access to the cells, I can't say for sure if their default long-duty pulsed pwm (as opposed to the bog-standard cc/cv) is actually helping to balance cells passively, or working in conjunction with the bleeders - or doing absolutely squat in that regard.

As promised, I got another. I guess I'm that weird.

 

[Substrate]

Inline monitor responsible for the .1V difference between display and Fluke 87V.

Ah, removed the PowerWerx inline monitor, and now the scc's display is only .02v different with nothing but a fuse inline. That's better.

[UPDATE] Ok, so two units operate exactly the same. B07 (LFP default at 14.4v) and Pon / Pof (pwm on/off default).

Man, I truly believe that what they call "pwm" is adopting the equalize/balance mode that is used during lead-acid settings. Thing is, the "pwm equalize" in lead acid, is only activated if you hit the lead-acid lvd and recover. That's good.

But for LFP, which is all I'm interested in, It almost appears that they may be adopting that equalize phase from the lead-acid algo to also serve in LFP for mild passive balancing with pulsing, delay times betwen pulses possibly helped along by the bms bleeder resistors.

But who knows - I'll need much better gear to track those pulse-widths. Instead, since none of my 3 LFP's I've tested it on so far have complained, and appears to be less aggressive than letting the bog-standard cc/cv (with Pof ) try to drive the batt to zero current, I think I'm good.

I normally don't buy at this low-end especially without a 20-page manual, but instead a small users-guide foldout. Thanks to that online reviewer who piqued my interest by revealing this weirdness. I dig it.

 

[Substrate]

Really odd - one of my LFP's doesn't seem to trigger the pulsing!

Is it perfectly balanced and not needing it, despite the fact that I have the default (Pon) enabled?

Driving me crazy, so I ordered some cells for a simple exposed 4S system to play with and see just what the heck is up with that!

 

[Substrate]

Bummer - the pwm-on seems inconsistent.

Put together a simple 4S (12v) with individual similar cells (top balanced first etc) with so I would have access to each cell's terminal voltage during charge in series, and for now, no balancing leads to see where this went.

Result - even with Pon, it seems to act like the standard cc/cv! None of the pulsing evident.

Driving me crazy not to know what is the trigger that makes this pulsing work on two of my batts, but NOT do it with two others. Two identical models of this scc tested with same result.

Conclusion: after all this, there is only one guy who wrote the firmware 6000 miles away behind a desk somewhere who knows. But not me. Seems like the only truth is that sometimes it will do it, and sometimes it won't!

My WAG guess: Maybe the trigger to go into it's own long-duration special pwm mode, is if it senses slight voltage drops from bleeder resistors kicking in on a battery's own bms. If it doesn't detect the slight voltage drops, as in the case of my non-bms batt, (or perhaps my sealed one that might be perfectly balanced) it doesn't do it. Probably wrong, but it entertains me to think so.....