Is open loop charging of the Powerflo16 permitted? I have a customer with a Outback GFX3048 that needs new batteries. The Powerflo16 would be an easy substitution for 8 L16 AGM batteries.
Yes Open loop is acceptable as well
I made this chart over time that includes the average of most LiFePO4 batteries and I have used these settings found in the LiFePO4 Column on the Powerflo16 on both Outback Radian and XW6048 in open loop.
Let me know how you have setup your Outback and other open loop systems.
You can't really tell the SOC of a lithium battery by it's voltage.
It is way too flat of a curve and is going to be under load or charging and not just resting voltage.
The only way to know it's SOC is using some device with a shunt like the Whizbang or another that will count everything going in and out and reset to 100% at the top for whatever voltage you think is best. I use 55.2v , others go higher up to 56v , etc. It really depends on the time you want it to take and balance settings of a particular bms.
Larry
I absolutely agree that a chart that correlates SOC to voltage isn't accurate (just like it isn't on Lead batteries) and I also agree that the system should have a current meter like WhizBang, or the case of Outback, the FlexNET DC. In the end, You still equate 100% SOC to a voltage and I also prefer 55.2 as 100% full.
The charge controller and inverter all use Voltage for many settings and the chart above was just a way to enter some value to have the system operate in Open Loop with some appropriate levels. Even in open loop, it's good to have another device that can display the batteries internal SOC data, like an old laptop or a CAN / RS485 meter or Cerbo GX, etc.
A future project is buy a JBD BMS Smart LCD RS485 Display to see if I can provide a basic display for a more detailed status for some customers that still have older inverters.
Quote from: mvrck on January 23, 2025, 10:41:43 AMI absolutely agree that a chart that correlates SOC to voltage isn't accurate (just like it isn't on Lead batteries) and I also agree that the system should have a current meter like WhizBang, or the case of Outback, the FlexNET DC. In the end, You still equate 100% SOC to a voltage and I also prefer 55.2 as 100% full.
The charge controller and inverter all use Voltage for many settings and the chart above was just a way to enter some value to have the system operate in Open Loop with some appropriate levels. Even in open loop, it's good to have another device that can display the batteries internal SOC data, like an old laptop or a CAN / RS485 meter or Cerbo GX, etc.
A future project is buy a JBD BMS Smart LCD RS485 Display to see if I can provide a basic display for a more detailed status for some customers that still have older inverters.
If you have batteries with Pace BMS type you can use an esp32 or raspberry pi to get all the batteries in series with just one battery connected. It will show all 16 cells in every battery .
I also use a victron smart shunt which is easy to bring into Home Assistant .
The other monitoring I have been using Shelly EM ( just got the EM3 Pro ) and also I use Peacefair PZEM -017 for DC or PZEM - 004T for AC . Both of those peacefair are supported by ESP32 flashed with Tasmota .
Here is the pace bms project which I have used for years ( and also the Classic project )
https://github.com/ClassicDIY/PylonToMQTT
Larry
does the powerflo 16 have an inverter pre-charge built in or do I need to do a resistor for the Radian or other outbacks?
also, is there a source for the settings for the powerflo 16 for using older non communicating CC's (FM-80, FM-100, Classics)? bulk V, absorption V and Time, float V, refloat V?
Other Thread started.
Quick add: they don't seem to have a pre-charge circuit in them.
I just saw this on Midnite page for open loop charging specs.
It may have been recently added
Absorb Voltage 56.4V, Absorb Time 5 minutes, Float Voltage 54V
https://www.midnitesolar.com/productPhoto.php?product_ID=768&productCat_ID=61&sortOrder=1
Larry
I did see that recently as well.
56.4 / 16 is 3.525 V per cell. That seems like normal if not on the high end of other LiFePO4 battery makers. The 5 minute Absorb seems short, but I don't know the actual voltage that the MidNite Active Balancers start working. Since Float is 54, maybe that is the magic number. I wonder what the time needed is for proper active balancing?
Speaking of open loop, I made a video adding 3 Powerflo16s to a Radian. I made comments about voltage settings, etc. Let me know what you have set for your Outback equipment with MidNite batteries.
https://youtu.be/AwoaTZtHMjU?feature=shared
Quote from: mvrck on January 28, 2025, 10:55:00 PMI did see that recently as well.
56.4 / 16 is 3.525 V per cell. That seems like normal if not on the high end of other LiFePO4 battery makers. The 5 minute Absorb seems short, but I don't know the actual voltage that the MidNite Active Balancers start working. Since Float is 54, maybe that is the magic number. I wonder what the time needed is for proper active balancing?
Speaking of open loop, I made a video adding 3 Powerflo16s to a Radian. I made comments about voltage settings, etc. Let me know what you have set for your Outback equipment with MidNite batteries.
https://youtu.be/AwoaTZtHMjU?feature=shared
if you mean that the balancing starts at 54v ? No it would never be that low - it could make the balance worse. Usually balance starts at anywhere between 3.4 to 3.5 - seems like every battery manufactuer does it a bit different. If you want to see what Gobel Power 280 ah has their settings at for Eve cells, here is what they are for the main bms and for the additional 4 amp active balancer.
What you see below is that the bms will balance at 3.40 but only with a 30mv differential . The active balancer will start to balance at 3.45v until a 3mv differential and will stop balancing at 3.4 . I think I have that interpreted correctly . Also it is set to show 100% SOC at 55.2v and 5 amps .
Larry
Great information! Thank you.ill fold this into the big spreadsheet of batteries.
I connected the Powerflo 16 to a Schneider XW system with the InsightHome internet gateway. The battery reported as a PolyonTech battery with an operating range of 46.5 to 57.6V, 200A charge and discharge. Seems like the Powerflo fits in the usual guidelines for 16-cell LiFePO4. That's good news.
So far they have been working great.
Does float time matter for Powerflow 16 in open loop?
Using the settings now posted on the website:
Absorb Voltage 56.4V, Absorb Time 5 minutes, Float Voltage 54V
thanks
Quote from: brett on February 01, 2025, 05:01:33 PMDoes float time matter for Powerflow 16 in open loop?
Using the settings now posted on the website:
Absorb Voltage 56.4V, Absorb Time 5 minutes, Float Voltage 54V
thanks
Once the batteries are at their absorb setpoint, they stay at that voltage for the absorb time of 5 minutes. Then they go to float for the rest of the day . The next day the bulk to absorb to float will start again. The float setpoint is at a voltage that the batteries won't charge or discharge but the controller will be able to make power directly for the loads. So float time is whatever is left after the Absorb cycle is complete.
Larry
Quote from: ClassicCrazy on January 24, 2025, 08:34:26 PMI just saw this on Midnite page for open loop charging specs.
It may have been recently added
Absorb Voltage 56.4V, Absorb Time 5 minutes, Float Voltage 54V
https://www.midnitesolar.com/productPhoto.php?product_ID=768&productCat_ID=61&sortOrder=1
Larry
I'm not seeing that particular information at that link. The label on the side of the battery says that Absorb is 56.16 volts. I just use 56.2 since I'm not aware of any equipment that let's you set voltage to the hundredth of a volt.
I guessed at the absorb time and put in 15 minutes. I guess I was close but that's certainly more than 5. The problem was that with the system charging the battery 56.2 volts was reached very quickly and then the charge controllers went to float at 54 volts. The battery still only read about 60% charged. I guess it will eventually charge up at 54 volts. I guess this is one of the difficulties with integrating to old equipment that does not support closed loop.
Quote from: Wirenut on April 08, 2025, 05:43:36 PMQuote from: ClassicCrazy on January 24, 2025, 08:34:26 PMI just saw this on Midnite page for open loop charging specs.
It may have been recently added
Absorb Voltage 56.4V, Absorb Time 5 minutes, Float Voltage 54V
https://www.midnitesolar.com/productPhoto.php?product_ID=768&productCat_ID=61&sortOrder=1
Larry
I'm not seeing that particular information at that link. The label on the side of the battery says that Absorb is 56.16 volts. I just use 56.2 since I'm not aware of any equipment that let's you set voltage to the hundredth of a volt.
I guessed at the absorb time and put in 15 minutes. I guess I was close but that's certainly more than 5. The problem was that with the system charging the battery 56.2 volts was reached very quickly and then the charge controllers went to float at 54 volts. The battery still only read about 60% charged. I guess it will eventually charge up at 54 volts. I guess this is one of the difficulties with integrating to old equipment that does not support closed loop.
The battery will be almost full at 55.2v so certainly full at 56.2v . Some of what is recommended for the charge voltages has to do what the bms is programmed for - at what voltage the bms resets the SOC to 100%. 5 minutes or 15 minutes absorb time doesn't make too much difference unless there was some issue with how the bms balances the cells and needs a longer time if some cells were way off. I have been using open loop and it is okay. I still don't understand how the closed loop works exactly - I mean what information does the battery send to the controller to decide when to quit charging ? I think in some cases it is just going by the SOC getting to 100% , but there may be other criteria ? I haven't seen any manufacturer ever write up what exactly the charging order of operations are for closed loop.
The bms reported SOC can get really out of whack sometimes from what I have seen on other than Midnite batteries.
You will know when they are full when you see the current input to the batteries drop way down at whatever absorb voltage you end up using.
Larry
New to forum, first post. Thank you for your engagement here! We could not have achieved what we have so far without you all taking the time to write and exchange ideas in this forum.
Thank you especially to @mvrck for your very informative video shared in this thread.
We are running the XW PRO in open loop with the PowerFlo16 after upgrading from 6 x PHI-3.8-48-M (no communication). So far relatively good performance. Batteries communicating with the Schneider Insight using CANbus again using a separate video from @mvrck. Thank you!
We've seen the notes on Absorb and Float voltage. Any thoughts on Bulk/Boost voltage settings? We are running with both SOC and Voltage control enabled in case we lose communication with batteries. My understanding is that we can still run on voltage control in this case instead of SOC.
I also have an interesting warning we received a few days ago after flipping on a rarely used exhaust fan. The XW PRO shut down with a Fault 71 and posted the following message in the log: Cause: Battery discharge current exceeds the discharge current limits sent by the external BMS.n.
Surprised we exceeded max discharge current for batteries without causing trouble somewhere else. Batteries in the system (see signature) do not seem to be the limiting factor.
Any thoughts on these topics appreciated!
I think there is a setting on InsightHome for the inverter to issue a Warning or an Error based on BMS activity. I set that for Warning. It could be that the battery was full or near full and was instructing the system to taper charge. Or the battery was low and also was trying to lower current draw? Not sure either way, but it was set to warning, at least you would know about without losing power.