SimpliPhi Full Charge and CPM

[raysun]

My guess is the battery voltage is rising above Absorb voltage.

[JRHill]

My guess is the battery voltage is rising above Absorb voltage.

I know that to be true and this has been happening all along. With Absorb set at 56.0 the Absorb voltage can get as high as 56.3 vdc.

[raysun]

My guess is the battery voltage is rising above Absorb voltage.

I know that to be true and this has been happening all along. With Absorb set at 56.0 the Absorb voltage can get as high as 56.3 vdc.

I bet that's cutting the Absorb stage short.

(I am growing less than in love with certain aspects of the FXR behavior.)

Has the inverter Calibrate function been used to get the battery voltage in line? I'd consider setting it to meet the Absorb voltage.

[JRHill]

Has the inverter Calibrate function been used to get the battery voltage in line? I'd consider setting it to meet the Absorb voltage.

Nope. Its set to where it came from the factory. I just changed the battery cal from, ironically, +0.4 to 0. Surprisingly the calibration only goes in 0.4vdc steps.

[raysun]

Has the inverter Calibrate function been used to get the battery voltage in line? I'd consider setting it to meet the Absorb voltage.

Nope. Its set to where it came from the factory. I just changed the battery cal from, ironically, +0.4 to 0. Surprisingly the calibration only goes in 0.4vdc steps.

That "CF" (Calibration Factor, I guess) is stepped to the inverter's resolution, it seems. Try setting it when the inverter charger hits the Absorb stage.

[JRHill]

That "CF" (Calibration Factor, I guess) is stepped to the inverter's resolution, it seems. Try setting it when the inverter charger hits the Absorb stage.

I could set it back. I don't doubt that you've addressed this in a different thread but: Why just a change just before Absorb? I already know if I let the VFXR inverter run its full charge cycle the battery voltage will hit 56.3 consistently with a 56.0 absorb setting. So if it stays below that tonight by about the offset amount, so what? Is there a risk of an oops?

[raysun]

That "CF" (Calibration Factor, I guess) is stepped to the inverter's resolution, it seems. Try setting it when the inverter charger hits the Absorb stage.

I could set it back. I don't doubt that you've addressed this in a different thread but: Why just a change just before Absorb? I already know if I let the VFXR inverter run its full charge cycle the battery voltage will hit 56.3 consistently with a 56.0 absorb setting. So if it stays below that tonight by about the offset amount, so what? Is there a risk of an oops?

This thread is exploring the settings necessary to reach battery full charge and set CPM. The purpose of adjusting the inverter's battery voltage sense is to achieve those goals.

The ideal would be for the inverter to track the battery voltage exactly. However, in practice it doesn't, in my experience. The 0.4V steps it appears to resolve are somewhat "coarse" for the purpose. The objective, as I see it, is to have the sensed voltage agree with the Absorb Voltage setting (56.0V in this case), while the actual battery voltage may, ideally, be in the 56.0-56.4V range.

[raysun]

A cloudy, rainy day with just enough sun to keep the charge controllers in Bulk and leaking current into the battery. The battery got to about 98% SoC at one point, but no Absorb or CPM

It's been 32 hours since the last full charge. The FNDC and the Victron are in pretty good agreement.

[JRHill]

Last night after getting the voltages as close as possible to the batteries' voltage it occurred to me that there's still a missing piece. The battery's BMS, although there's nothing to be done with the subject. I have to wonder what the 'trueness' is of the SimpliPHI BMS. I just hope I never find out if one ever goes back for warranty, they D/L the data and come back saying the battery was constantly charged to too high a voltage.

[raysun]

The black boxes are indeed Black Boxes.

[JRHill]

A few after thoughts of the voltage calibration process:

- The procedure says to fully charge the batteries before hand. It says nothing about letting them rest - for however long, for your style of battery. Apparently if one is to do this project you should apparently know these basics. The batteries will settle. Trying to make any adjustments even after 15 minutes is still a moving target. It takes a while and esp depending on your type of battery, age, etc. Bottom line is you let them rest until the battery voltage is consistent. With no loads and batteries settling, the system will be down.

-Energize the system. That means the inverter and charge controller be powered up so you can see the Mate3x system voltage. It doesn't say to make sure all inverter loads are disconnected (all or main breaker, esp after a while and a bunch of stuff may want to fire up) and you don't do it during sunlight which adds the CC charging. The missing part is everything is powered up and doing mostly nothing.

- Adjust the trim pot on the FNDC to match the display on the Mate3x with your volt meter. Hopefully you are using a commercial quality DVM with good leads. (hint, switch your DVM to ohms and short the leads. The reading should be close to zero ohms. If not, find another meter and test it. It needs to be in the appropriate ohm range. If the ohms are off the voltage may be as well).

- Then calibrate the inverter to the system voltage and then the CC to the same. Go back to the system voltage and check it against the batteries with the DVM. If off, dial it in as close as possible. Go back after a few minutes and recheck the inverter and CC. Go sip some coffee, etc. Do it again. Go sip more of the same. If all is to your tolerance you are done. I had 3x through this.

Today:

Well, isn't that cool. The Mate3s fluctuated between the inverter and CC -.1 vdc. Generator off, a bit of CC incoming.

TBD: the inverter may still be overshooting Absorb. We'll see over the next few days.

Lastly: I should've done this years ago. But with the lead it didn't seem a problem. Even with the lithium it wasn't really a problem. But lithium and generator seem to be a picky problem.

[raysun]

Testing a lithium battery at full charge is difficult, as the "voltage bubble" above the LiFePo4 fully charged cell voltage will interfere.

Resting the battery for an extended period is an exercise in patience I can't afford. After 24 hours, the voltage may still be artificially inflated.

Its best to bleed a little charge off the battery, then measure.

[JRHill]

Its best to bleed a little charge off the battery, then measure.

Good point. That would help. I have other thoughts on this but am still gathering info.

[raysun]

Yes, it would be.

[JRHill]

Two days ago I crawled into the appropriately named crawl space. Its not so bad - actually 4+'. I cal'd the FNDC and components. The first night needed no gen run as I had mostly filled the batteries. Last night on a programmed run everything was spot on: The inverter, the CC, and the FNDC within .1vdc. Then it happened. After reaching about 54.4 the inverter and CC grew to .2 off, bouncing off .3. At the same time the system voltage increased to .4 over the inverter and CC. WTF? Everything worked in concert until 54 1/2 and then gets weird QUICKLY. And the charging cycle is ended way early before the 56vdc Absorb. And when it ended the inverter stops feed through with the generator running but I'm on batteries.

Lacking anything reasonable I went to Generator mode (from Support) this morning. Yeah, I'll lose some generator input. I am just curious as to different behavior. There are some wonky things happening. I'm limited to inverter charging right now due to low sunlight. I can't wait to see what changes when I can once again get a full recharge from the FM.

[raysun]

That's puzzling to be sure. If I get an opportunity in the next few days, I'll try a single-inverter charging session and report the results.

I can't put anything concrete to it, but the various behaviors cropping up may suggest some issues in the FXR firmware.

[JRHill]

That's puzzling to be sure. If I get an opportunity in the next few days, I'll try a single-inverter charging session and report the results.

I can't put anything concrete to it, but the various behaviors cropping up may suggest some issues in the FXR firmware.

The FW is up to date. But we've seen the need to reload Mate3 FW that fixed a problem. I'll see what the Gen Mode change does on tonight's run. Besides, I want a few more cycles to pass before doing anything other than a minor config change for comparison.

[raysun]

I wasn't going to say it out loud, but I wonder if we may be dealing with bugs in the system programming (firmware) in certain circumstances.

Too many features that seemed to work without incident in the FX series seem to behave differently in the FXR series for it to be simply configuration issues, IMO.

While FXRs have been out a good while, the FXR A-01s are relative newcomers.

[JRHill]

Shhhh. Yeah, we'll keep this between us.

[JRHill]

Today by the skin of my chinney-chin-chin at 1:57, for just a moment long enough:
Inverter 55.6
CC 55.7
System 55.8

(55.6, 0.1, 6a, 98%)

CPM Met. Notice the alignment of the voltages. NOT like the alignment when charging from the inverter. I certainly didn't expect the solar today. Woohoo! the first of the season.

But the inverter is not behaving like the good 'ol FM. The FM, just like last year and in years past stepped right up to the plate. Charge Parameters Met. Obviously I won't be running any tests tonight going from Support to Generator. But I'm getting short of patience. Why am I doing all this troubleshooting on the VFXR when the FM has the same parameters, less 0.4vdc,and works perfectly at the first instance this season?

Everything is as it should be and there are several threads beating this VFXR programming stuff to death.

[raysun]

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.

[raysun]

Next, the PV arrays were reconnected to the FM80s. They started a Bulk, Absorb, and terminate.

This time, the FNDC reported CPM.

[JRHill]

I can some times hit CPM by hitting Rebulk. Flipping from your inverter to the CC in short order isn't a whole lot different is it? It's like a rebulk but from a different device.

[raysun]

Yes, similar.

Its interesting that the low and very slow charge rate definitely reached a full charge, but the FNDC didn't register it.

Had it not been for the incidental 2nd charge prompted by reattaching the PV arrays, I wouldn't have bothered with CPM. The FNDC metering was effectively synced at 100%. That's all I'm after.

[JRHill]

What I really meant to say was the inverter charge is one thing - with the discussed problems. Sometimes I can hit CPM with a rebulk from the inverter. Maybe not.

Today I had my first of the season full charge from the FM80. Inverter, FM and system voltage stayed within 0.2 the whole time and CPM was hit. No screwing around, no attempts at retuning. It just worked. Everything was in concert and close voltages from everything.

Tonight at 92% and 52.8v I did a gen run. Inverter 52.8, CC at 52.9 and system at 52,8 before it connected. I had changed to Generator Mode. I didn't do a full run, just 3/4 hour. When I shut it down the inverter was 54.4, CC 54.3 and system at 54.1. What the heck? But this is is much better than it was before calibrating. With the solar at this time of the year the charge to the PHIs wasn't much different than from the generator so they were comparable.

5 minutes later with a short settle down the inverter was 53.2, CC at 53.4 and system at 53.3. Like I said I'm gonna keep going for a bit. Doesn't seem Support to Gen Mode made much difference. We'll see after a few more cycles.

I can't imagine trying to explain this in a support call.