SimpliPhi Full Charge and CPM

[raysun]

inverter was 54.4, CC 54.3 and system at 54.1

The active charger is going to read a bit higher. In bulk its always raising the target voltage above the battery voltage to keep charge flowing.

Was the FM80 calibrated with the firmware VBAT setting, or the trim pot on the circuit board in the FM80s wiring compartment?

I'm not sure about the FNDC. Mine drifts .1 or .2 in some circumstances. Usually it's in tight agreement though.

[provo]

Set up charging on a single inverter, with no PV.
Absorb=56.0V
CRA=9.0A
Charging was about 25ADC (4A per block) and proceeded slowly to 100% SoC, then to 56.0V Absorb, and termination.

The FNDC did not report CPM, though the Victron, using the same criteria, did.

I've never charged that slowly, but I do keep the bulk current to ~13A per block. Then I can get to CPM in 6-8m. At 15A per block it can take 9m or more. Seems very sensitive! This is with inverter at 28.0V and CC at 28.2V.

Fortunately I rarely pull more than 110Ah out of the batteries overnight so I can usually finish absorb with CPM by noon.

[JRHill]

inverter was 54.4, CC 54.3 and system at 54.1

The active charger is going to read a bit higher. In bulk its always raising the target voltage above the battery voltage to keep charge flowing.

Was the FM80 calibrated with the firmware VBAT setting, or the trim pot on the circuit board in the FM80s wiring compartment?

I'm not sure about the FNDC. Mine drifts .1 or .2 in some circumstances. Usually it's in tight agreement though.

I calibrated the inverter and the CC via software (from the Mate3).

[raysun]

I never found the CC VBAT calibration to be very precise. The trim pot method seems to be more so.

[raysun]

I've never charged that slowly, but I do keep the bulk current to ~13A per block. Then I can get to CPM in 6-8m. At 15A per block it can take 9m or more. Seems very sensitive! This is with inverter at 28.0V and CC at 28.2V.

The only thing I can think of is the BMS doing some late-stage "top balancing" of the cells. When I watch the Absorb cycle, I notice a fair amount of charge current "bounce", especially noticeable when focusing on the CRA target. The current may fall below CRA, only to "bounce" above it briefly. This happens repeatedly during the Absorb time until the net current falls well below CRA. I get the sense this is influenced by the charge current leading up to Absorb, and how the anode is "packed", but that is merely supposition on my part.

Watch for the "balancing bounce" and see if it may be affecting the time it takes to reach CPM.

[JRHill]

I've never charged that slowly, but I do keep the bulk current to ~13A per block. Then I can get to CPM in 6-8m. At 15A per block it can take 9m or more. Seems very sensitive! This is with inverter at 28.0V and CC at 28.2V.

The only thing I can think of is the BMS doing some late-stage "top balancing" of the cells. When I watch the Absorb cycle, I notice a fair amount of charge current "bounce", especially noticeable when focusing on the CRA target. The current may fall below CRA, only to "bounce" above it briefly. This happens repeatedly during the Absorb time until the net current falls well below CRA. I get the sense this is influenced by the charge current leading up to Absorb, and how the anode is "packed", but that is merely supposition on my part.

Watch for the "balancing bounce" and see if it may be affecting the time it takes to reach CPM.

Now this is interesting to me. With the max voltage of 56.4vdc and an Absorb time of .1 hr, what happens if the BMS isn't done 'balancing' in that .1 time frame? More so, if charging from the inverter to a lower voltage of 56.0 / .1 hr.... This happened last night:

I shut the data collection run down at 54.4 but the Wax On, Wax off thing was happening there.

[raysun]

I think the assumed "balancing routine" may hold the key to the fully charged but not CPM mystery.

I was looking through the Victron graph captures I've done, but they are too low resolution. I'll try for some higher resolution graphs under the various charging conditions as the opportunity presents itself.

Here's the best I have at the moment. The 9A CRA will have to be interpolated, as well as the current excursions.

The "top of the Voltage Mesa" is the Absorb stage. Small "jitter" in the charge current around the CRA value can be seen in that time interval.

[JRHill]

I never found the CC VBAT calibration to be very precise. The trim pot method seems to be more so.

The s/w adjustment seems good enough for my purposes. Esp since the FM works great with the s/w adjustment. The VFXR too, now. Its the divergence of the system voltage (FNDC) that is of interest.

Before I did the calibration it was suggested that because the system voltage was exceeding the absorb set point and therefore then terminated. Now, charging with the VFXR over 54.5, the system voltage is low by .2 to .3 and still terminating early. But now the VFXR is staying pretty close to the FM80 (.2)

I'm holdingoff of any re-flash of the VFXR firmware lest I inject some new oddity. The problems of the day are sufficient.

[provo]

Watch for the "balancing bounce" and see if it may be affecting the time it takes to reach CPM.

Oh definitely! I've always assumed it was just the GCC function on the Mate that couldn't regulate well, but the idea that it's the BMS working behind the scenes is interesting. I don't mind taking 11 or 12m to reach CPM, since Phi Tech Support doesn't seem to. I just need something to mess with, given how boring the charging of the batteries would be otherwise .

Edit -- actually GCC is out of the picture by absorb....I like your BMS explanation.

[raysun]

I just need something to mess with, given how boring the charging of the batteries would be otherwise .

I know right? Lithium takes a whole lot of fussing over battery charging out of the picture. We have to have SOMETHING to obsess over.

FWIW, when the CCs reach Absorb here, the system consistently reaches CPM (with rare exception).

On generator alone, I don't have enough cycles. Dual inverter chargers reach CPM, IIRC, but the single charger yesterday did not. More gas must be burned before that picture becomes clear.

[JRHill]

More gas must be burned before that picture becomes clear.

That's OK, Raysun. Fuel injection and a catalytic on your generator you are performing a valuable service and saving the planet at the same time.

[raysun]

My Honda sits inside the entrance to my connex, blowing exhaust out the open door. I have a CO detector right near it, and have never heard a peep from it.

I even work in there while the genny is running (but it does make me aware of the potential...)

[JRHill]

My Honda sits inside the entrance to my connex, blowing exhaust out the open door. I have a CO detector right near it, and have never heard a peep from it.

I even work in there while the genny is running (but it does make me aware of the potential...)

BTW, the exhaust extension I added a few weeks ago works like a champ. Sorry for the thread drift.

[raysun]

Charge complete and CPM via FM80s.

Absorb stage. Charge current around CRA (9A) note the oscillations around that line.

Wider view of charging up to and through the Absorb stage.

Charging events for the cycle.

[MJSullivan56]

BTW, the exhaust extension I added a few weeks ago works like a champ. Sorry for the thread drift.

Thx. Looks like an ideal addition to my outdoor shed.

[JRHill]

BTW, the exhaust extension I added a few weeks ago works like a champ. Sorry for the thread drift.

Thx. Looks like an ideal addition to my outdoor shed.

Price has gone up. What else is new:
https://www.amazon.com/gp/product/B07HB ... image&th=1

The SS manifold, adapter and 8' silicon hose are top quality machine work. There is nothing on the end of the hose so its your option. I removed the spark arrestor and the factory flange with the screen from the generator. Exhaust comes out luke warm. I plan to run the hose onto a buried steel 5g pail with the lid perforated and a rain cover. Doubt I'd hear the unit running - I barely can now with the hose pointed away from the house. No fire hazard.

[JRHill]

Charge complete and CPM via FM80s.

Screenshot_20220123-130719_Microsoft Remote Desktop.jpg
Absorb stage. Charge current around CRA (9A) note the oscillations around that line.

20220123_131252.jpg
Wider view of charging up to and through the Absorb stage.

20220123_133450.jpg
Charging events for the cycle.

I am happy for your set up and success. Last night was more of the same for me. From the inverter charging ended early and never came close to Absorb. That's OK, 94% and 54.6 is fine. Those numbers don't align with any charging parameters, on any device. It just ended. I'll let it go another day or so because I can. If lead batteries I'd be pulling my hair out.

I am sure I'll be doing a firmware 'reflash' in the next few days. At this point there isn't much left to try. I hope OB is following this stuff with the FXR series. But from their side if they can't identify a problem there is nothing to fix. It's a curious dilemma. But I bet they have some ideas....

[JRHill]

Here we go after updating the Mate3s correctly:

Starting the generator. Love the extension hose, BTW.... This test is for malfunctions related to Mate3 firmware programming.

[JRHill]

Gen mode off, Support on. A full 20aac charge:

Here we go. Two hours or so. Lets see what happens. Will I get an absorb f I can.

[raysun]

I did some configuration changes today too long to list right now, with inconclusive results, but in the process, ran into to your "spontaneous inverter charge termination" phenomenon.

Hmmm.... this needs more digging.

[JRHill]

Buddy, this is a problem. Some people have it and some not. There has to be a little tiny bit on its side instead of standing up. It affects some but not others. But believe me, if you're charging is screwed up it is an issue.

Really pisses me off. System voltage is 54.1.

[JRHill]

If there is any good news the its sun getting above some of the trees and last two days have gotten me close to half what is needed to recharge the batteries!

Sorry for the poor diction and typos.

[raysun]

Buddy, this is a problem. Some people have it and some not. There has to be a little tiny bit on its side instead of standing up. It affects some but not others. But believe me, if you're charging is screwed up it is an issue.

Really pisses me off. System voltage is 54.1.

System voltage at 54.1 and charger voltages higher than the system voltage is to be expected. The chargers will raise their voltage to a point high enough to allow full charge current to flow.

IME, what is depicted here is normal.

[JRHill]

No CPM. No difference in behavior.

All is not lost. Mary came home with some beautiful onion bubs. I have to get them out quickly.

Sucks to try to dig holes for bulbs in the frozen zone. Thread drift.

[raysun]

Jim, I don't know if this is good news or bad news. After charging today, I turned off the FM80s and let a little charge bleed off the battery. Then I started the generator and initiated another charge cycle using the FXRs. It was a short cycle that proceeded to Absorb in about 10 minutes. After a couple of Absorb Charging / Not Charging events, the charge current dropped below CRA, and charging terminated. A moment later a CPM record appeared in the event log.

Absorb Voltage = 56.0