question

solvan avatar image

How to remove SmartSolar automatic absorption time?

Hi everyone,

I've recently switch from a Bogart Engineering Trimetric + SC2030 to a SmartSolar 100/50 MPPT.

Aside from underwhelming bulk gains and inaccurate metering, I'm also having problems with the absorption termination.

The manual states this is done based on either a (fixed!) 2A current limit, or a very crude 4-option time calculation based on voltage at first light - whichever comes first.

The problem I am having is that auto time calculation is consistently off, always skipping to float prematurely, far before the 2A cutoff or battery recommended termination current. Consequently, this unit is consistently failing to achieve 100% SoC (accurately measured with the Trimetric monitor) .


Is there any way to remove the time limit from this unit entirely, so it holds absorb to 2A termination? Or better yet, an adjustable amp value? I've set the 'max absorption' to 6h but it seems to just ignore this setting, and I see no other parameters to tweak.

It's very disappointing - right now this unit is failing to achieve what the far cheaper PWM Bogart unit has been doing reliably for years.

MPPT - Solar Charge Controllerabsorption
1 comment
2 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

4 Answers
solvan avatar image
solvan answered ·

So, the solution to this turned out to be simple:

Return the highly inaccurate Victron unit, and get a proper charge controller.

https://community.victronenergy.com/answers/7061/view.html

10 comments
2 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

We just installed the 100/50 smartsolar which we purchased specifically to replace our pwm controller where we could not control the absorption time. This new controller stays in absorption for only 30 minutes which is less than our old controller. The battery manufacturer recommends absorption of 2-3 hours. We would be happy if it stayed in absorption until 2 amps. We checked that we could control the absorption before buying this but apparently that's not true? We set the volts manually with a max absorption of 3 hrs which we thought was overriding the algorithm. So frustrated to be going through this issue AGAIN. And your answer indicates you basically had to give up?

Hi Inanne. What your Victron is doing is applying a factor to your time setting which is dependent on how low a V your batts start the day. So if you take no loads from your batts overnight and they start the day pretty much fully charged, even your batt maker wouldn't recommend a long Absorb. It's explained briefly here: https://www.victronenergy.com/live/victronconnect:mppt-solarchargers See 'Absorption time limit'. You could set up to a 12hr absorb and it will be reduced by 'the factor'.

Take care too with the 2A tail. The mppt can't see the loads directly, so if you have a 3A (say) load it can never reach it.. Higher end Victron kit can manage this of course, but you won't have that.



Hi John, thanks for answering this. Sent a long reply but it doesn't appear here so not sure where it went. I'll try to come back later and rewrite my reply and give an update.

The Victron showed 30 minutes of absorption which means 1/6 of the 3 hrs max we set so it thought we were already charged but the low volts it shows was 11.9 which should have been 2 hrs (2/3rds of 3 max or 2 hrs). Also it shows it got up to the 14.8 max volts but our trimetric showed 14.2 as the max for the day with 24 amps still coming in. However, we did run the generator so perhaps it looked at the volts then to come up with the 3o minutes. Not sure how to deal with that.

On your last comment, that's a good point (basically if we have any load >2 which we always do, it won't reach the minimum amps going to the battery). Just wish we could control the absorption time (or minimum amps) like we thought. Manufacturer says 2-3 hours which was our experience with the old controller (which we had to force manually to keep going since it was hard coded for one hour only).

Thanks again for your thoughtful reply.

Yeh, I appreciate your concern for your batts, but don't be foxed by Vmin or Vmax readings coughed by any kit. Outliers can be 'inadvertently' generated by a spike from a load coming on or off, like say even a fridge motor. The working kit may take even milliseconds to respond, but the V recorded will remain - and may not be representative of the truth. Take care looking at those instantaneous V's.

If you're using 'standard' Pb's, it's hard to go past the stock Victron charge algorithms. They really need to be graphed to appreciate, just little things like dropping back into absorb later in the day after load application or cloud interference.

Dare I say it? - probably better than the batt maker's spec sheet.. I won't enter an argument re this, but I'll keep using the Victron ones.

Your choice, but getting to know your batts and understanding them is a great asset to have, even multimeter in hand. Good luck, 'Stay Charged' mate.

With the trimetric it records "filtered" high/low volts meaning they must be at that voltage for over 4 minutes. With a clear day and no generator it didn't get up to 14.8 volts under either measure of volts! So no absorption time to test!

Yeh ok, with spikes under control, look at what's happening. Your Smart app will say when it's doing Absorb and it's V. If it's not what you expect, then ask why. Where I live it's been often over 40C for the last month, and if it stayed at 14.4V (equiv) setting I'd be peed. It does as I expect though, and temp-compensates the target down to ~14.2V.

Lots of other little things can impinge too, like wire length/diameter and where the reading is being taken from. So best installed correctly or compensated for.

No more useful kit exists in my toolbox than a $20 multimeter. I choose the cheaps because I break em, flood em, drop em overboard, etc. And frankly, they usually can be relied upon for an accurate V.


@lnanne

Solvan has been banned from the forums for his posts about these problems.

He says: what you describe is almost identical to what he's observed. Dont trust the numbers on the Victron, they are all over the shop depending on the unit. The Trimetric is better set up to take accurate measurements and is a far higher qulity unit - trust thos numbers.

Dont listen to the waffle about the "magic Victron algorithm", thats pure marketing handwaving, like all the other excuses being made here...

The simple fact is that if your Victron isn't holding Absorbtion high enough or long enough, it's not working properly. 30 minutes is pathetic, considering the voltages you state. Trust the Trimetric, look at the Amp-hours deficit it reports

Return the unit. It's just not worth the gamble.

@lnanne


Also - see how many amps are still going in once it switches to 'float' - if the trimetric is still reporting the bank is absorbing 2-3% of your rated capacity when at float voltage, the Victron isn't fully charging your bank. This works (even with loads on, remember the trimetric reports the net current flow to/from the battery.

JohnC avatar image
JohnC answered ·

Hi Solvan. I'll have a go, but it'll mostly be via explanation, or an attempt thereat..

Presuming Pb's. Some points for consideration:

* No, I don't believe Max Absorption Time is configurable beyond 12hrs, and then seems subject to the 'multiplier'. Set Custom Charge profile and what you want. For 99+% of users this would be adequate, and likely unnecessary to even go that far.

* Set float to absorb level if you must continue beyond there. And ask yourself why you need to. Those Victron algorithms I consider well thought out, so much so that I've never changed mine. I have the full portal graphs to look at too, and can't fault what mine does.

* The 2A trip point I treat as an 'emergency' level. The mppt alone can't tell where it's going, batts and/or loads. Mine never gets that low, tis just the loads..

* SoC is a calculated figure. There's quite a bit involved in doing that correctly, and if you're relying on your existing kit as a 'standard' to refer to, then I'd encourage you not to treat it as gospel, because it likely has flaws too.

* You haven't bought a dud. Bit like stepping from a Jeep into your new Porsche, and not asking the dealer what the buttons are for.. :)

Offered in good faith. No need to flame me, I'm incombustible anyway.. :)


8 comments
2 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

I wouldn't flame you - this is actually a decent, relevant answer, unlike most of what ive received here. Though I don't necessarily agree fully. :)


I've seen the 'float as absorb' trick before and I more consider it a fairly dodgy workaround, not a long term solution. On days (or weeks...) when I'm away, or just use less power, I dont want to have AGMs sitting at absorb for 8+ hrs a day. So I'd have renember to manually set float any time my usage varies, or risk gassing them. which is a step backwards, and I didn't spend $500 to go backwards


I agree that it's a dodgy way, but nor did I recommend it. I guess it highlights the difference in what what each of us are prepared to do and expect on a daily basis.

For me, happy to see float reached by midday and continue on accepting charge there until sundown. That's starting with say 12.4V (equivalent, mine are 48V) at sunup. Nice healthy bulk curve from ~7am, <2 hrs in absorb, tis a good day.

And, surprise surprise, later on will return to absorb for a short time if something like clouds or a load drops me below float V. I think that's clever.




I am using the Custom profile, and have it set to the same voltages ( drop compensated) as the SC2030. The only difference is the SC2030 is fully programmable, and I manually specify it hold absorb until below 2% of capacity. It reliably reaches this, switching to float at ~98%.


The SmartSolar has no such setting as far as I can tell, and so I'm at the mercy of its own "algorithms" which for all.their advancement switch to float always ~90%. Tweaking the only thing I can, Absorb V, doesn't change this. The auto timer is simply premature.

Holding absorb until 2% capacity is something I'm not prepared to do. A vague target anyway when you have loads interfering. Eg today @ 11:48, just before dropping into float, my batt loads were 2.3A, my mppt 6.1A. So 3.8A into batts.

By 15:44 in 13.8V float, that's now 0.3A into batts. I like it that way.

I know the complications in SoC (peukerts, charging inefficiency etc), but the Trimetric I have can compensate, and Ive dialed this unit in over two years. Its gives a good SoC estimate, but more importanly its based off an ACCURATE Ah meter. Ive cross referenced it with other metrics to verify. So when it says the SmartSolar drops to float 40Ah early, I trust it.

That said, a simple sanity check of OCV (taken 24 hrs after isolating batts at switch to float) shows that the SC2030 switches to a higher SoC (13.0v as per man spec for ~100%) than the SmartSolar (12.9v).



I don't use SoC at all. I need a firmware update on my Multi to do that, and haven't bothered. But I do have those magical portal graphs to go by. Shrug that off as you may, but they're special. Showed me how bad my 'old & sluggish' batts were, so I replaced them.

I'll probably add a BMV to my system, but largely for the accuracy and my dc loads. SoC I don't care, because I know it will show what I set it to show. Re the accuracy, Victron have always said that's a 'best effort' thing, same as competitors. Why you already own that Trimetric?



So with all that said @JohnC, Id have to dispute your anaolgy - this SmaetSolar may not be a 'dud', but if its a Porsche, its a Cayenne, full of 'hi-tech' auto-gadgets that promise a lot but fail to perform. The SC2030 may be the clunky old jeep, but at least it has the manual functions to get the job done.


It baffles me that a $500 controller has NO ability to set a custom termination for absorb... If so, this unit is going back.


Or maybe this is another of those things where I have to then spend more on other Victron gear to get the original job done.


Yeh, the character limit on these replies is killing me. :) And my analogy was for fun, not too serious.

But I do encourage you to continue with an open mind. Victron have been doing this for a long time.

The product range is immense, and the options diverse. Sure they cost, but for you might I suggest a GX box. Even a RaspPi if you want to cheap-it. Victron will even help you do that, and the free portal access is for dreaming of.

I'm sorry you're disappointed so far. But you're still on the start-line. Hang loose..

redrover avatar image
redrover answered ·

Hi Solvan. I too was/am having a similar problem. But I asked the wrong question and/or tried my own solution on this forum and got pounced on. Basically my lithium battery wasn't getting to 100% before the mppt 100/30 switched from bulk-absoption-float. So each day battery was going backwards. I did find the abs time limit was configurable upto 24hrs and even tried this on one occasion which caused absoption to run for 4hrs (24/6). Now I have it set to 12hrs and get 2hrs of absoption (12/6). Even so I reckcon it switches from bulk to absoption too soon. Now I find that after being on float for an hour or two it switches back to bulk for a few hrs then absoption for 2 hours and float again! all in one day. It's got a mind of its own. I wish I knew what it was doing, or how it's algorithms work so that I can then know what to tweak. cheers Red

3 comments
2 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Where did you get pounced on? If you expect your problem in being solved in just fine tuning the absorption Voltage, that might not be it. This issue could be more complex and involving further measurements and diagnosis. As you stated you are not a battery person. Maybe it should do the trick, to get some expert and give him money for solving this. Sure, the community is always willing to help and support, but you cannot expect to be trained in just questioning here.

Sure, there is a possibility that you have a faulty charger somehow, a Pro will tell you in minutes, if you have he/she on site

It's probably me being considered the 'pouncer' there, and certainly wasn't intended to offend. Bump it Red, and we'll have another go. The issue there seems to be too short a Bulk phase, which you see as an issue.


Hey John, I wasnt offended, not easily offended. Not by you. I also didnt get any useful advise on helping my apparent problem. Just got told my tweaking was wrong and to revert.

mvader (Victron Energy) avatar image
mvader (Victron Energy) answered ·

Hi Solvan,

Thank you for raising this. And I'm sorry to hear that you were not happy with the charger.

You're not the first to have questions about the absorption time in the MPPT Solar Chargers; and looking back, it perhaps took me too long to be `wake up`; and realise that something needs changing there. My apologies for that.

First thing we'll do is change the text in VictronConnect to better explain how the current mechanism works; see picture below. Perhaps with a link to the manual and explanation or something else; I'm not sure; yet.

Then secondly I'll discuss and think about adding an option to have a real fixed absorption time; besides the existing automatic calculation. As you know, our reasoning behind having a calculation is to not want to charge the battery too long in case it hasn't been discharged over night. To do that; it uses battery voltage at startup in the morning. See the product manual for details; (there is no magic, secrecy or marketing involved there).

Best regards, Matthijs Vader


Ps. picture of the setting how it looks now:


1550001748312.png (286.3 KiB)
6 comments
2 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

ps. the nicest solution is perhaps be to have the MPPT work together with a BMV; so it can see actual battery current; and decide on that to stop charging. Could probably be done with VE.Smart Networking.

(this won't solve issues for installations that don't have a BMV of course)

UPDATE: this is not an existing feature. Its just an idea of something to improve on the products.

lnanne avatar image lnanne mvader (Victron Energy) ♦♦ ·

I like those ideas and I can work with (and mostly like) the theory of the current algorithm but mine doesn't seem to be working that way (i.e. base the absorb time on volts in the morning - see my latest question about why it only went 45 minutes in absorb when my low volts were 11.9 in the morning and max absorb time was set for 4.5 hrs.

The MPPT references the battery voltage just before it 'wakes up' & then calculate's the max absorption time. To 'wake up' the solar input MUST exceed the battery voltage by >5v. If the 'rated' open circuit voltage of your solar panel array is not significantly above your system voltage, then this can take a fair while on an overcast day. My bet is that the engine was running & charging when the MPPT 'woke up' in the morning, and acordingly saw a higher battery voltage which led to the calculation of Abs time max/6. There is no magic to this logic & it's there to optimise battery charging & life under normal usage scenarios. It is NOT a good idea to use the same lenth of absorption time every day, regardless of the depth of discharge overnight. That said, it would not be a bad option to allow the user/installer to fully configure all these parameters themself (with a clear warning that it's at their own risk).

I think that's a good theory.

@lnanne what time did you start the alternator, and what was the cloud condition like then?

It was a clear day and the controller woke up and went into bulk by 7:30 am with 11.9 as low volts. Started driving around 9:00 am.

ok; thanks for the feedback and alerting me to that question; I'll read that.