AIMS charger: observations and quiescent power draw

[BobCollins]

I am finally putting my van power system together and have some observations on charging from the AIMS mains charger.

The system:

100 A-h LiFePO4 12.8 V battery​

battery switch​

AIMS 12V/24V charger (switches set to 00, LiFePO4 mode)​

Victron Energy SmartShunt​

multimeter to confirm voltages​

​

1) The absorption stage (constant voltage) was indistinct and quick. After taking hours to reach about 14.1 V (in the bulk stage, constant current), the voltage quickly climbed to 14.4 V in a couple of minutes, at which point the current dropped (sorry, I didn't record the bottom current value at this point).

2) The transition from absorption to float involved cycling between the two several times with a period of several minutes. The cycling was observable due to the voltage changing from about 14.4 V and 13.6 V and back again (see note 3). Eventually, the system settled into "float" mode which seemed persistent. Having a "float" mode at all for a LiFePO4 battery is, of course, not ideal, but it was expected from this charger.

3) The "float" voltage is about 13.6 V, in spite of the manual on page 7 indicating the float is 14.4 V. This is a good thing, though I don't understand the discrepancy with the manual.

4) The AIMS charger consumes about 2 W with the AC mains disconnected and the power switch in either position. This may be part of the voltage detection (12 V vs 24 V) function. As battery power is applied to or disconnected from the charger, the click of a relay is heard from the unit.

Note: the AIMS charger appears to be mostly (or completely) and analog system. Monitoring the voltage and current using the digital sampling of the SmartShunt provided a confusing display as the values jumped between the quantization levels of the meter. The result is what looks like a significant oscillation on the "Trends" display, but is likely a minor ripple in the analog electronics of the AIMS charger.

I am interested in other user's thoughts on my observations and conclusions. I am not happy with 2 W of wasted power, but I am not particularly interested in another switch either.

Cheers!

Edit (2022-06-27): I fixed the errors in my description here where I left off the tens digit several voltage values. Thanks sunshine_eggo!

 

[sunshine_eggo]

I am finally putting my van power system together and have some observations on charging from the AIMS mains charger.

The system:

100 A-h LiFePO4 12.8 V battery​

battery switch​

AIMS 12V/24V charger (switches set to 00, LiFePO4 mode)​

Victron Energy SmartShunt​

multimeter to confirm voltages​

​

1) The absorption stage (constant voltage) was indistinct and quick. After taking hours to reach about 4.1 V (in the bulk stage, constant current), the voltage quickly climbed to 4.4 V in a couple of minutes, at which point the current dropped (sorry, I didn't record the bottom current value at this point).

2) The transition from absorption to float involved cycling between the two several times with a period of several minutes. The cycling was observable due to the voltage changing from about 4.4 V and 3.6 V and back again (see note 3). Eventually, the system settled into "float" mode which seemed persistent. Having a "float" mode at all for a LiFePO4 battery is, of course, not ideal, but it was expected from this charger.

3) The "float" voltage is about 3.6 V, in spite of the manual on page 7 indicating the float is 4.4 V. This is a good thing, though I don't understand the discrepancy with the manual.

Please confirm that you have omitted a "1" in front of 4.1V, 4.4V, 3.6V, etc.

Concerning 2), this cycling behavior may indicate a LFP battery that's out of balance.

4) The AIMS charger consumes about 2 W with the AC mains disconnected and the power switch in either position. This may be part of the voltage detection (12 V vs 24 V) function. As battery power is applied to or disconnected from the charger, the click of a relay is heard from the unit.

Note: the AIMS charger appears to be mostly (or completely) and analog system. Monitoring the voltage and current using the digital sampling of the SmartShunt provided a confusing display as the values jumped between the quantization levels of the meter. The result is what looks like a significant oscillation on the "Trends" display, but is likely a minor ripple in the analog electronics of the AIMS charger.

I am interested in other user's thoughts on my observations and conclusions. I am not happy with 2 W of wasted power, but I am not particularly interested in another switch either.

2W could be worse. It's also such a low number, it might not be accurate. If accurate, I would put a disconnect in place for when the system may be idle for extended periods.

 

[BobCollins]

Please confirm that you have omitted a "1" in front of 4.1V, 4.4V, 3.6V, etc.

Fixed. Thank you.

Concerning 2), this cycling behavior may indicate a LFP battery that's out of balance.

The battery is an off the shelf SOK. It had been sitting unused since last October.

1) Is it reasonable that the cells weren't balanced at the factory? Or do they just drift over time?
2) Assuming they need to be re-balanced, is this a manual process, or does it happen auto-magically when brought to full charge?
3) Would the cycling behavior indicate that the BMS is balancing the cells? If so, I should expect less or no cycling in future charging?

Thank you. The behavior of LiFePO4 batteries in new to me. Unfortunately, there seems to be little authoritative information on them available.

2W could be worse. It's also such a low number, it might not be accurate. If accurate, I would put a disconnect in place for when the system may be idle for extended periods.

I suspect that it is accurate. Whether it is significant is up to the user.

 

[sunshine_eggo]

Fixed. Thank you.

The battery is an off the shelf SOK. It had been sitting unused since last October.

1) Is it reasonable that the cells weren't balanced at the factory? Or do they just drift over time?

Both are possible and likely, particularly drift over time - Sitting for eight months? I would be shocked if balance wasn't lost.

Even with spendy battleborns, their imbalance can grow to the point that only 50% of the battery is available if not allowed to balance regularly.

2) Assuming they need to be re-balanced, is this a manual process, or does it happen auto-magically when brought to full charge?

This, but you can influence it somewhat by holding the battery at full charge to force it. I recommend you leave it on the charger for 24 hours and see if the behavior changes.

If you were on solar, I would recommend an absorption of 13.8V and a float of 13.6V. These elevated voltages and daily available charge current can help balance batteries over time.

3) Would the cycling behavior indicate that the BMS is balancing the cells?

To some degree, but the primary driver is BMS cut-off at high voltage, then the battery voltage drops to the point that it reconnects - charge, trip, repeat.

If so, I should expect less or no cycling in future charging?

Yes.

Thank you. The behavior of LiFePO4 batteries in new to me. Unfortunately, there seems to be little authoritative information on them available.

Oh it's everywhere... just hard to sift through.

I suspect that it is accurate. Whether it is significant is up to the user.

Significant or not, one has to plan for it. If there is ever a time when a 2W draw could flatten the battery due to lack of use, a disconnect should be installed. Technically, the charging wires should be fused/breakered, so a suitable breaker on the (+) lead could pull double duty.

 

[BobCollins]

Both are possible and likely, particularly drift over time - Sitting for eight months? I would be shocked if balance wasn't lost.

Even with spendy battleborns, their imbalance can grow to the point that only 50% of the battery is available if not allowed to balance regularly.



This, but you can influence it somewhat by holding the battery at full charge to force it. I recommend you leave it on the charger for 24 hours and see if the behavior changes.

The cycling behavior only lasted for a few cycles, after which the system settled into a constant float value of 13.6 V. Would leaving it in this float mode for 24 hours add anything to the balancing?

Note: as I reported above, in spite of the AIMS charger's documentation, LiFePO4 mode (00) appears to provide an absorption voltage of ~14.4 V and a float voltage of ~13.6 V.

If you were on solar, I would recommend an absorption of 13.8V and a float of 13.6V. These elevated voltages and daily available charge current can help balance batteries over time.

I don't understand what you mean here.

To some degree, but the primary driver is BMS cut-off at high voltage, then the battery voltage drops to the point that it reconnects - charge, trip, repeat.

This makes sense to me.

Yes.




Oh it's everywhere... just hard to sift through.

I find a lot of people sharing their experiences and what appears to be conclusions based on their experiences. To me, this seems like reverse-engineering a system for which none of us really understand the details. I guess I am used to design where there are accepted and documented models of behavior. Battery University seems to be the closest I can find.

Significant or not, one has to plan for it. If there is ever a time when a 2W draw could flatten the battery due to lack of use, a disconnect should be installed. Technically, the charging wires should be fused/breakered, so a suitable breaker on the (+) lead could pull double duty.

I have avoided circuit breakers due to their complexity and reports of unreliability. I do have fuses on all circuits and a big red battery switch. Removing a bolt-down fuse regularly would be fussy. I guess really need a switch (or switchable circuit breaker) in this circuit. I hadn't expected a quiescent load from the output of the charger. More of this reverse engineering stuff.

Again, thank you for you comments!

 

[sunshine_eggo]

The cycling behavior only lasted for a few cycles, after which the system settled into a constant float value of 13.6 V. Would leaving it in this float mode for 24 hours add anything to the balancing?

Note: as I reported above, in spite of the AIMS charger's documentation, LiFePO4 mode (00) appears to provide an absorption voltage of ~14.4 V and a float voltage of ~13.6 V.

I misinterpreted that. Simply leave it on the charger for a few days. 13.6V should encourage balancing. After a few days, load it up to discharge it a little bit and then recharge it. See if the cycling repeats.

I find a lot of people sharing their experiences and what appears to be conclusions based on their experiences. To me, this seems like reverse-engineering a system for which none of us really understand the details. I guess I am used to design where there are accepted and documented models of behavior. Battery University seems to be the closest I can find.

That's a fair assessment. BU is a good resource.

 

[Roswell Bob]

Have you taken a look at the input current waveform of the charger? I am curious how close it resembles a sinewave. There should be a fast acting fuse between charger and cells. Battery University has some obnoxious engineers who think they know more than they do

 

[BobCollins]

Have you taken a look at the input current waveform of the charger? I am curious how close it resembles a sinewave. There should be a fast acting fuse between charger and cells. Battery University has some obnoxious engineers who think they know more than they do

I'm not completely sure I know what you are referring to.

The input to the charger is 120 VAC from my local mains. I did not monitor the current waveform on the input. It is likely a switcher, so it is not likely to be a pretty sine wave. Why would it matter to you?

Please tell me about your claim for a need for a fast acting fuse. I am using a pre-assembled battery from SOK that includes a BMS with a high voltage cutoff.

Also, what are you in disagreement with what is presented by Battery University?