Any epever experts here? Charge controllers tracking poorly and not responding well to loads.

[off.the.grid]

I have 3 arrays connected to 3 separate epever controllers.
The main array is connected to a large 60A model and 2 smaller arrays on 20A units.
All are tracer AN series.
Basically what happens is the controllers will shut off around the middle of the day with only one staying in float, meanwhile the bank is not taking in any more but when I switched in the genset the bank would still take in 10A.

The second issue is they don't respond to light loads even while there is plenty of sun out, the controllers are largely not putting out much power at all and leaving the bank drain.

I've been advised here that they don't track very well and the problems I experience are common.

Is there any newer firmware that improves this?
I'm wondering if anyone has hacked the firmware and made any modifications to it?
I have no idea what revision my controllers use, but I believe all AN series controllers take the same firmware.
I am interested to hear from others who have had the same issues I'm experiencing and if firmware updates addressed this for them.
I might get a USB cable to update them anyway before I consider upgrading to victron.
Not probably a big issue in the scheme of things, but nice to capture as much energy as possible over the winter months.

 

[chrisski]

With my Victron MPPT, I found if I set the float voltage to low, it would not keep the battery charged. I raised the float voltage to the same as the bulk voltage and that fixed my issue.

The battery I had was a assembled from Chinese cells 8S which had no guidance on float or absorb voltages. I decided to charge and float this battery at 3.475 volts per cell, or 27.8 for a 8S 24 volt battery.

 

[off.the.grid]

With my Victron MPPT, I found if I set the float voltage to low, it would not keep the battery charged. I raised the float voltage to the same as the bulk voltage and that fixed my issue.

The battery I had was a assembled from Chinese cells 8S which had no guidance on float or absorb voltages. I decided to charge and float this battery at 3.475 volts per cell, or 27.8 for a 8S 24 volt battery.

Funny you say that, because this was something I had considered myself.
It's not good to float the cells at bulk voltage for long periods once fully charged but in my case they still had a little way to go before fully charged.
My float voltage is 27.2V, I feel this is probably too low. The other issue with epever units is that you can only set the bulk charge to 3 hours max.
It's interesting to hear that you have had similar issues to me with your Victron unit.

 

[chrisski]

It's not good to float the cells at bulk voltage for long periods once fully charged but in my case they still had a little way to go before fully charged.

While not floating at bulk for long periods of time is not good, I also found that cells entering float mode never allowed a rebulk voltage to be hit until the sun went down. That meant going into the night with a battery not near full, which also meant not enough power to last the night.

I also fund that a low float voltage might replace 30% of the power I used. If I used 300 watts, float would pull 100 watts from my panels and my power continuously dropped.

This strategy might reduce the number of usable cycles. Who knows how much. Even if instead of 8000 cycles, I get 3000 cycles, that’s still 10 years.

 

[off.the.grid]

While not floating at bulk for long periods of time is not good, I also found that cells entering float mode never allowed a rebulk voltage to be hit until the sun went down. That meant going into the night with a battery not near full, which also meant not enough power to last the night.

I also fund that a low float voltage might replace 30% of the power I used. If I used 300 watts, float would pull 100 watts from my panels and my power continuously dropped.

This strategy might reduce the number of usable cycles. Who knows how much. Even if instead of 8000 cycles, I get 3000 cycles, that’s still 10 years.

Yeah, in my case i don't have a huge amount of PV capacity anyway and it's servicing a bank that is 306Ah capacity with 2500W max from my array which in the real world I don't see, 100A is the max current it would experience.
I was provided some figures closer to 3.5V per cell to float and in my situation i reckon the bank would barely be charged before the sun goes down so would not be floating too long at those voltages, probably only 2 or 3 hours a day in summer i reckon with my power use.
The last 5 percent or so of charge is only drawing about 10 amps and takes a wee while to get to a full SOC.

 

[off.the.grid]

Well today we have had a bit of cloud and less PV output, my main controller is still in boost phase, but my other controller shut off completely.
I increased the float level right up to 29V which is what my boost level is set to, and it ran for less than an hour before shutting off again.

 

[pollenface]

I have 3 arrays connected to 3 separate epever controllers.

Basically what happens is the controllers will shut off around the middle of the day with only one staying in float, meanwhile the bank is not taking in any more but when I switched in the genset the bank would still take in 10A.

The second issue is they don't respond to light loads even while there is plenty of sun out, the controllers are largely not putting out much power at all and leaving the bank drain.

Can you post your charging parameters for each unit?

It is not uncommon for one or two controllers to stop making power if another controller is generating sufficiently, I have observed this many times with different combinations of controllers.

I also observed poor mppt performance from my Tracer AN units. After jumping on to the Victron bandwagon with everyone else, all my problems disappeared.

 

[v_green57]

I have two 4210AN's, each with 1200W of array, connected to 22080 wH of 24V LFP batteries.

This has been performing quite well for some years now.

There was a learning curve.

Firstly, there may be small differences in how each controller reads the current battery voltage. Even .1 - .2 V will make a difference as to which controller will reach float first. You need to connect them one at a time to a known voltage and see is such a difference exists, and then program setpoints on one CC to take the error into account.

Secondly, you need to connect them to a busbar which is then connected to your battery bank so they "see" the same connect point. Stacking lugs or wires on a single battery terminal won't do.

You must also use identical lengths of wire of the same gauge to do this. Otherwise, small voltage differences will again arise, and one CC will drop into float before the other.

I don't know the effects of mixing different CC's such as you have. I've always used identical units when paralleling them. Just makes sense to me.

I assume you have already set the "Boost duration" to the same value on all three CC's.

 

[off.the.grid]

Can you post your charging parameters for each unit?

It is not uncommon for one or two controllers to stop making power if another controller is generating sufficiently, I have observed this many times with different combinations of controllers.

I also observed poor mppt performance from my Tracer AN units. After jumping on to the Victron bandwagon with everyone else, all my problems disappeared.

Here is the settings they are all set to.
Am using will prowse's recommended settings.

 

[mikefitz]

recommended settings.

The charge settings seem very high at 29 volts, perhaps you are getting cell overvolt protection in the BMS shutting down the charge path. The attachment you posted showed zero current, an indication of BMS shutdown. Epever calibration is not always accurate and its possible they are outputting a higher voltage than set.
The poor response to loads may be improved by raising the 'bulk recovery ' voltage to just below float volts.
There is no need to use high charge voltages, the battery will work quite well with 28.4 bulk and 27.0 float, bulk recovery 26.8.
Note the settings programmed into the units needs checking with a meter.

 

[v_green57]

Agree with Mike.

It is NOT necessary to chase the "100% charge" fairy tale.

Doing so just puts things on the knife-edge of shutting down due to spike overvoltages when the CC switches charge modes. The EPEVER and other CC's do this. It happened to me right after kicking my FLA batteries to the curb and switching to LFP, even when using the charge params supplied by my battery mgfr.. Backing off things a couple tenths of a volt stopped it.

After running at those setpoints for a few weeks, the new batteries had "settled in" and I was able to raise things about a tenth.

These posts helped me understand my "new" battery technology:

G

Thread 'Struggling with basic LiFePO4 settings in Epever Tracer'

Feb 3, 2021

Hello all. I am struggling to justify some basic LiFePO4 charging parameters, and relate them to how to set up my Epever Tracer 2215BN. The Epever has many charge parameters geared toward charging older chemistries. This is very confusing to apply to LiFePO4, and the fact that terminologies and recommendations for charging LiFePO4s are all over the place makes it even more difficult.

Can someone please help with some very fundamental questions?

The tracer needs settings for:

1) Charging limit voltage (is this the same as "bulk"?)

2) Equalize voltage (ok, I get that this is to be turned...

• gator
• Replies: 94
• Forum: Solar Charge Controllers

• 

T

Post in thread 'School me up on BMS in LiFePo4 please'

Mar 15, 2024

I recall an issue with Ampere Time batteries (before they changed the name to LiTime) but don't recall specifics. I did find this video:

Could be some issues have followed the rebrand.

• TacomaJoe

J

Post in thread 'Just getting started'

Feb 5, 2024

I'm using litime batteries and they worked fine accept that I can't set the charge passed 14v without having a spike. Do you have any battery recommendations. 24v and I'll prolly have 2 or 3 100ah for my load.

• jsemon

S

Post in thread 'MPPT Charge Controller Over-voltage'

May 7, 2024

So, I have the Litime LiPOE4 battery charger and had problems with charging it. I sent an e-mail to Litime and this was their response:

I have a LiTime 12V 230Ah battery with Low-Temp (200A BMS) and a LiTime
14.6V 40A LiFePO4 Battery Charger. Yesterday, when I tried to charge the
battery, the red light would turn on and then click and then the green
light would turn on. When the red light was on, it would show charging
around 14.4V and then various lower voltages with the green light. I
don't think it was charging the battery. Can you tell me how to
troubleshoot this?

Thank...

• sjkted

Especially the first one, Mike's post #6, where he takes the time to explain which of the plethora of settings affect CHARGING, and which only affect LOADS connected to the CC.

after running FLA for 20+ years, I had developed a serious case of "leaded brain" and needed to kick that to the curb. I would rather have 110% reliability than chase that last 5 - 10% of charge.

 

[off.the.grid]

The charge settings seem very high at 29 volts, perhaps you are getting cell overvolt protection in the BMS shutting down the charge path. The attachment you posted showed zero current, an indication of BMS shutdown. Epever calibration is not always accurate and its possible they are outputting a higher voltage than set.
The poor response to loads may be improved by raising the 'bulk recovery ' voltage to just below float volts.
There is no need to use high charge voltages, the battery will work quite well with 28.4 bulk and 27.0 float, bulk recovery 26.8.
Note the settings programmed into the units needs checking with a meter.

Well that's the thing, in reality it never applies 29V at the start of the charge cycle, it just slowly rises as the battery charges during the bulk phase.
The BMS is set to shut down charging at 29.2V and in reality I barely ever see the charge controller reach 29V.
29V is still a bit high i feel, so not too sure why will recommended that, but at 3.6V per cell it should be 28.8V and 28.4 as you say would be a bit safer at around 3.55V per cell.

Now to test this out, I switched on the genset, and my battery was still taking in 10A of current, so I know the BMS has not shut down.

What you are saying about the bulk recovery voltages could have a bearing on this, so might be a good place to start I'm thinking and I might lower the bulk voltage down as you say, it wont do any harm.
Others also recommended to raise the float voltage levels up a bit too.

 

[off.the.grid]

I just got a response from Epever and have been sent some firmware updates that they claim will fix these issues.
Still have to tweak the voltages as suggested here, but will do one step at a time.
Will provide an update once i receive my RS485 cable to upgrade the firmware.

 

[v_green57]

I just got a response from Epever and have been sent some firmware updates that they claim will fix these issues.
Still have to tweak the voltages as suggested here, but will do one step at a time.
Will provide an update once i receive my RS485 cable to upgrade the firmware.

Will be interested in how this goes. Did they explain what specifically the FW fixes are and the version numbers?

Added another 4210AN and some SanTan panels four days ago to add some additional capacity. The short winter days and wife running the bathroom space heater on cold mornings is making my float time shorter. The added capacity is almost exactly balancing the 1kWh daily use of the bathroom heater.

My 4210's have three different versions of firmware:

V01.56+V01.22

V02.01+V01.22P

V01.54+V01.22Xo


Everything is running fine in a 3P config, however.

 

[off.the.grid]

Will be interested in how this goes. Did they explain what specifically the FW fixes are and the version numbers?

Added another 4210AN and some SanTan panels four days ago to add some additional capacity. The short winter days and wife running the bathroom space heater on cold mornings is making my float time shorter. The added capacity is almost exactly balancing the 1kWh daily use of the bathroom heater.

My 4210's have three different versions of firmware:

V01.56+V01.22

V02.01+V01.22P

V01.54+V01.22Xo


Everything is running fine in a 3P config, however.

All they said is it "should" fix my problems.
Unsure of revision at this stage, but I can upload the files here.
For my 6420AN controller the file is named UP_TracerAN_DSP_V0245_20M.prg

The file for my smaller 2210AN controller is labelled ST_TracerAN_20min_V202.prg
Does that mean V02.45 and V2.02 respectively?

 

[v_green57]

All they said is it "should" fix my problems.
Unsure of revision at this stage, but I can upload the files here.
For my 6420AN controller the file is named UP_TracerAN_DSP_V0245_20M.prg

The file for my smaller 2210AN controller is labelled ST_TracerAN_20min_V202.prg
Does that mean V02.45 and V2.02 respectively?

Good question. You'll be able to see the actual versions with an EPEVER app after you upload them to the CCs.

I would need to see a detailed list of "fixes" before I would apply them as there's nothing wrong here.

"If it ain't broke, don't fix it."

Would also be interested in the current procedure for upgrading firmware.

 

[off.the.grid]

Good question. You'll be able to see the actual versions with an EPEVER app after you upload them to the CCs.

I would need to see a detailed list of "fixes" before I would apply them as there's nothing wrong here.

"If it ain't broke, don't fix it."

Would also be interested in the current procedure for upgrading firmware.

Here are the firmware files, let me know what you can find out, I dont see any published list of bugfixes, but hopefully the software can tell me what version my controllers have installed.
It should in theory be possible to backup the existing firmware and reflash it back should I run into issues?
You will see the instructions included in these files, along with the drivers and software to flash the units.

 

[v_green57]

Here are the firmware files, let me know what you can find out, I dont see any published list of bugfixes, but hopefully the software can tell me what version my controllers have installed.
It should in theory be possible to backup the existing firmware and reflash it back should I run into issues?
You will see the instructions included in these files, along with the drivers and software to flash the units.

*sigh*

Windows only. Jerks. I kicked Winblows to the curb 10 years ago.

I can probably run this in an Oracle VM after a bunch of horsing around granting permissions to various bits of USB hardware necessary to create the MODBUS connection. Had to do that with a Morningstar CC.

Not high on my list right now.

Never assume that once you upgrade firmware you can go back. I've had lots of hardware over the years where upgrading was a one-way trip.

Let us know what happens after you upgrade. Good luck.

 

[off.the.grid]

Yeah, not always possible to downgrade, but i have 2 of these controllers to test on.
Do you have a windows PE boot disc like hirens boot CD? That would be the easiest way to run this software and no need to install it on your computer. Could be handy for those times you ever need to use windows.
I too am sick of windows, but not ready to jump to linux just yet!

 

[v_green57]

My VM's already have W2K and XP already up and running, so that's covered as long as the EPEVER junk will run under a 32 bit OS.

I also have a stand-alone W10 laptop I can use if necessary.

Like I said, at this point it's just not a priority. And I don't get the warm fuzzies from a FW upgrade that doesn't give any details of what it "fixes". I usually don't apply those unless the hardware is very f'd up and I have nothing to lose.

You will have to be the guinea pig on this one, unless you can twist EPEVER's arm and get them to provide details.

 

[off.the.grid]

Im happy to be the guinea pig on this, will provide an update once I get the cable.
From memory the retailer said they performed a FW update on one of the controllers I sent them when I first reported issues, but I have no idea what revision it was, but it never fixed anything.
I will also communicate with Epever on this if I continue to have issues and see if I can get anything out of them on what their fixes are.

 

[off.the.grid]

My VM's already have W2K and XP already up and running, so that's covered as long as the EPEVER junk will run under a 32 bit OS.

I also have a stand-alone W10 laptop I can use if necessary.

Like I said, at this point it's just not a priority. And I don't get the warm fuzzies from a FW upgrade that doesn't give any details of what it "fixes". I usually don't apply those unless the hardware is very f'd up and I have nothing to lose.

You will have to be the guinea pig on this one, unless you can twist EPEVER's arm and get them to provide details.

I managed to update the firmware on my two smaller units.
The main controller does not power on when performing the update and i get a comms error on the app.
I have a few concerns, mainly if the controllers have actually updated, i dont see any way to check the installed firmware version, but it indicated it was successful on these two small units.
My cable uses the CH343 driver and there are no options in device manager to set the serial port to RS485, but it still appeared to update regardless, this was my main concern overall, but the controllers are all working and im not too concerned about the main controller at this stage as the small units are the ones giving me the most grief, but will be interesting if anything changes with these updates.

Would still be good to get the larger controller updated however.

 

[off.the.grid]

So this update appears to have done nothing to improve this.
The controller on the right is not putting out any current, meanwhile there is about 10 amps draining from the battery which is approaching full charge and the sun is right up and the array is more than capable of providing enough power.
The main controller was still providing about 300W from memory at the time, but should have been providing more power too.
I switched off all the controllers except for the one on the right in the photo and despite there now being more load on the battery, it still did not come on.
I turned them all back on and checked throughout the rest of the day and it appears it comes on and off at times but not at its maximum output, sometimes only a fraction of an amp.

 

[Kiwi2000]

So this update appears to have done nothing to improve this.

You are going through the stages of Epever MPPT ownership.

First stage is realizing you have a problem and are missing out on solar production.

Second stage is trying different settings to see if you can get them to work better.

Third stage is going through the headache of updating the firmware which may or may not fix the problems.

Since it didn't fix your problems you are about to enter the fourth stage of Epever ownership.

That is replacing them with Victrons and then kicking yourself for not doing it sooner.

 

[off.the.grid]

You are going through the stages of Epever MPPT ownership.

First stage is realizing you have a problem and are missing out on solar production.

Second stage is trying different settings to see if you can get them to work better.

Third stage is going through the headache of updating the firmware which may or may not fix the problems.

Since it didn't fix your problems you are about to enter the fourth stage of Epever ownership.

That is replacing them with Victrons and then kicking yourself for not doing it sooner.

Well I was recommended Epever here, so thats what I went with.
To be honest I cant complain too much, the battery is largely charged by this point, but feel I could be making more use of the sun during the day.
Im more concerned how it will perform over the winter.
I dont see how this could not be corrected in the software, but I guess its a waiting game.
I think i would not have any issues if all these controllers were able to communicate in some way with each other.
Victron is pretty good, I agree, but its pretty pricey and there are still countless options out there.
Im not going to spend much on this system until I plan on upgrading the whole system, as long as it ties me over in the meantime im not too bothered, I plan to eventually upgrade to 48V or possibly use a high voltage system off a tesla battery at some point.