diy solar

diy solar

Epever 6415AN doesn't respect my settings

... so if you can't beat 'em, join 'em
default settings thinks i have a 12S configuration. It was the first thing I changed when customizing the configuration. THIS time, I just went with it and have been able to adjust boost, float and boost reconnect voltages to more tolerant than default settings. Though it went far too high for my taste, the controller performed as expected, ramping down the wattage shortly after boost voltage was achieved. It has floated all day since varying between 54.85 and 55.05v, staying in float. Boost recon is being honored starting around 5pm. Assuming all is now kosher, I will let it go for another whole day to confirm, then attempt to get these voltages into the position I can live with.
 

Attachments

  • 2860w-21.png
    2860w-21.png
    218.2 KB · Views: 0
adjusted boost, float and boost recon at the end of the charging day and the software responded immediately. More confirmation that the 12S default change to 16S was to blame.
 

Attachments

  • 2860w-22.png
    2860w-22.png
    196 KB · Views: 3
Finally!
This is what I should have been seeing all along
 

Attachments

  • 2860w-23.png
    2860w-23.png
    249.1 KB · Views: 4
Yesterday, I experienced what you have been talking about with the controller not using your changed settings. I had bumped up both the boost and float values. The controllers went into float at the old setpoint. I went down and rebooted them and they switched back into boost mode. That's an odd story too I'll mention later.

I had updated my parameters using the RS485 interface and custom software I wrote myself (not the EPEver PC software). I confirmed the values using the MT50 and they showed as I set them. There must be an extra level in there between the set parameters and what the controller is actually using. I guess this could make sense because changing things during the middle of the charge cycle could mess things up and would require a lot of extra code to prevent doing something stupid.

Question for you on your "reboot". What I did was disconnect the solar panel breakers first. Then open the switches from the controllers to the batteries. I waited a minute, then closed the breakers. About two minutes later, I closed the panel breakers. Is this how you do it?

Now the other interesting thing. The reboot reset the controller and it went back into boost mode. The batteries were almost full but the boost time is still two hours. They pushed the battery voltage all the way up to the charge limit (28.5) plus about two tenths 28.7v. The amps dropped from around 50 to right around zero. After the two hours the controllers did shift to float and the voltage dropped to 27.6v. After some draining the voltage dropped to the float setting of 27.2v.

Thankfully the controllers behaved. After your experience, I was watching them closely to make sure the voltage didn't climb above the disconnect (29v).
 
yes, that's how i've been doing it. i would guess your boost duration is 120 minutes. i think the "reboot" starts the duration timer over again. i've decreased mine to 10 so that wouldn't happen. I've seen that behavior as well. I'm pretty convinced that the default lithium cell count of 12S for 48v operation is the culprit.By leaving it there, instead of adjusting to the correct cell count for 48v (16S) my lifepo4 cells, i've been able to see "normal" behavior, with no more white knuckle voltages.
 
Last edited:
yes, that's how i've been doing it. i would guess your boost duration is 120 minutes. i think the "reboot" starts the duration timer over again. i've decreased mine to 10 so that wouldn't happen. I've seen that behavior as well. I'm pretty convinced that the default lithium cell count of 12S for 48v operation is the culprit.By leaving it there, instead of adjusting to the correct cell count for 48v (16S) my lifepo4 cells, i've been able to see "normal" behavior, with no more white knuckle voltages.

This is quite a saga you both have been involved in.

I was getting interested in an Epever MPPT based on the generally-positive reviews here on the Forum, but reading of your ordeal is giving me second thoughts.

Seems like Epever for LiFePO4 is on the bleeding edge.

In addition, looks like you are both using 150V models (6415AN) while I’ve been looking at the equivalent 200V model (6420AN).

It appears your 6415ANs both have settings for ‘Lithium’ while according to this Epever presentation: https://www.epsolarpv.com/upload/file/1812/EPEVER Tracer AN_Main_Pres.pdf

There is this statement:

“High current series doesn’t have self-activation function for lithium-ion battery, thus it cannot be used with lithium-ion battery.

So I am having second thoughts about using a 200V 60A Epever to charge my 24V LiFePO4 battery - sounds like it is much more complicated (and risky) than it should be...
 
My pain is self-inflicted from going ahead and changing the configuration to what I thought it should be. I have a 16S lifepo4 bank and naturally set my series to 16(circled in blue). This pushed the voltage up way farther than it should have been. It took days to figure that out. I need to work on my chinglish...

The LiBattery protection (circled in red) is integrated into my software and performs the function you mention.

Don't discount Epever just because I'm not too bright though. My pain is your wisdom. I've had this since 9/2019 initially using sealed lead acid, 24v50ah, then 24v100ah, then 24v155ah, then 24v210ah, then 24v320ah. Now it's 48v280ah and my arrogance and ignorance, which I thought was confidence, caused my problems. I have thought this was a very good controller throughout my variations, even now, when running max solar watts, and rising battery amps and volts.
 

Attachments

  • Image41.png
    Image41.png
    87.6 KB · Views: 15
My pain is self-inflicted from going ahead and changing the configuration to what I thought it should be. I have a 16S lifepo4 bank and naturally set my series to 16(circled in blue). This pushed the voltage up way farther than it should have been. It took days to figure that out. I need to work on my chinglish...

The LiBattery protection (circled in red) is integrated into my software and performs the function you mention.

Don't discount Epever just because I'm not too bright though. My pain is your wisdom. I've had this since 9/2019 initially using sealed lead acid, 24v50ah, then 24v100ah, then 24v155ah, then 24v210ah, then 24v320ah. Now it's 48v280ah and my arrogance and ignorance, which I thought was confidence, caused my problems. I have thought this was a very good controller throughout my variations, even now, when running max solar watts, and rising battery amps and volts.

Don’t worry, I’ll be slow and deliberate in my final decision. The two things causing me greatest pause are:

-Epever’s statement that their ‘low current’ AN-Series MPPT SCCs support LiFePO4 but their ‘high-current’ AN-Series MPPT SCS cannot be used with Lithium ion battery. I’m still confused about whether the menu you attached a pic of was from one of the newer high-current MPPTs that supposedly have no support for Lithium batteries...

-the struggle I sense you are both having with getting the Epever MPPT to charge whenever to battery is drained enough to take charge - I want to run an inverter off of my LiFePO4 battery all day long and expect my MPPT to convert available solar energy to charge current whenever the battery is drained enough to accept more charge. One charge cycle per day is not going to cut it for my application...
 
I can't say for sure as who knows what their delineation between low and high current is. I would think 60 amps would be high. but my chinglish isn't fluent. JMH1783 and I both run constant loads and I have just connected up my full size frig yesterday after extensive testing, (so far, so good) so will begin to see how a surge load affects things. I have a 24v500w inverter running my WISP tower and a 24v1500w inverter running everything else, both are being fed through 48v-24v buck converters and supported by parts of my old sealed lead acid bank mounted nearby the inverters. I just ordered a 48v3000w today so i can begin trying to run AC and/or hot water tank. 60 amps though isn't nearly as much as you would think once you put large loads on it. I've been running led light strings, laptops and wireless radios which use less than 30W each. The frig has a 55amp draw at 24v to start and the system has managed it fine so far. Run load is around 5 amps so more in line with other items. When you start getting near max draw, the system tends to stay in boost (bulk) and hammers in the juice all day. I've set another all-time high generation today at 14.01kw with about 2 hours left in the day. I was at about 20%SOC this am and had good sun for most of the day. My controller stayed in bulk all day. I did get close to float, but caught a few clouds and it just missed float. My struggle has been more about fine-tuning the voltage/saturation at the top end of the voltage. When the battery is low, the charge controller hammers in the juice at 50+ amps for as long as the sun will give.
 

Attachments

  • 2860w-24.png
    2860w-24.png
    247 KB · Views: 5
I can't say for sure as who knows what their delineation between low and high current is. I would think 60 amps would be high. but my chinglish isn't fluent.

I would have thought so as well. Did that menu with reference to ‘Lithium’ come stock on your 150V 60A AN-Series or was it associated with some software or utility you downloaded?

JMH1783 and I both run constant loads and I have just connected up my full size frig yesterday after extensive testing, (so far, so good) so will begin to see how a surge load affects things. I have a 24v500w inverter running my WISP tower and a 24v1500w inverter running everything else, both are being fed through 48v-24v buck converters and supported by parts of my old sealed lead acid bank mounted nearby the inverters. I just ordered a 48v3000w today so i can begin trying to run AC and/or hot water tank. 60 amps though isn't nearly as much as you would think once you put large loads on it. I've been running led light strings, laptops and wireless radios which use less than 30W each. The frig has a 55amp draw at 24v to start and the system has managed it fine so far. Run load is around 5 amps so more in line with other items. When you start getting near max draw, the system tends to stay in boost (bulk) and hammers in the juice all day. I've set another all-time high generation today at 14.01kw with about 2 hours left in the day. I was at about 20%SOC this am and had good sun for most of the day. My controller stayed in bulk all day. I did get close to float, but caught a few clouds and it just missed float. My struggle has been more about fine-tuning the voltage/saturation at the top end of the voltage. When the battery is low, the charge controller hammers in the juice at 50+ amps for as long as the sun will give.

If the battery reaches float but then gets discharged below that later in the day, does it go back into bulk for a second time in the same day?
 
Lithium is an option screen in the software that I hadn't actually expected. I had planned to use the lead acid user settings which can accommodate the settings for a wide variety of battery chemistries. It is the stock software downloaded directly from Epever for use with the usb interface. Yes, if you charge to boost voltage, transition to float, then use enough to fall below boost reconnect, then it will resume boost charge. I see it occasionally if I get a thickly cloudy spot mid-afternoon. It will fall below boost reconnect voltage and switch back to boost. JMH1783 has had some experience with this as well.
 
Lithium is an option screen in the software that I hadn't actually expected. I had planned to use the lead acid user settings which can accommodate the settings for a wide variety of battery chemistries. It is the stock software downloaded directly from Epever for use with the usb interface.

OK, so that ‘Lithium’ screen didn’t exist on the stock charger, it came along when you did the download.

Epever explicitly states that their ‘low-current’ MPPTs support LiFePO4, so I’ll need to understand what it means that they explicitly state that their ‘high-current’ models do not...

Yes, if you charge to boost voltage, transition to float, then use enough to fall below boost reconnect, then it will resume boost charge. I see it occasionally if I get a thickly cloudy spot mid-afternoon. It will fall below boost reconnect voltage and switch back to boost. JMH1783 has had some experience with this as well.

And you have control over where ‘boost reconnect is set, correct?

I believe one of you posted in the thread I just started this morning that you had been in contact with Epever - what is the best way to reach them?
 
Honestly, I only used the controller screen on the initial hookup to confirm I had battery power flowing to it. My next step was to connect the MT50 and use that for about an hour when I ordered the usb cable. Installed that the day I got it and have been in the software ever since.

Yes, boost reconnect voltage is programmable.

support@epsolarpv.com
 
Hi Fafrd. I would recommend the EPEver controllers. We have finally learned how to live with one another. I just replaced my cheap MakeSkyBlue third controller with a 5420AN 200V model. I don't have the third array up yet, but the Blue produced so much interference my monitoring circuits kept getting errors.

The lithium screen is in the setup software. I found it from this forum but had mine setup just fine without it. EPEver has added some niceties for lithium that I don't need. It has low and high temp protection. I don't need them. You might. I moved my battery bank into a temperature controlled space when I switched from AGM to LiFePO4. You can configure everything you need via the MT50 or plug in a PC to see everything at once. It has a clock that keeps drifting. Not sure how it uses it. It has no battery back up and seems to lose time. It reminds me of the real time clocks in PCs back in the 90s. The harder the CPU worked, the less accurate the clock was. :) In any case, having to reboot the controller by disconnecting from the batteries seems to cause it to lose its date and time completely.

jnrhome and I both run continuous loads and therefore we need to look at the charge setting much differently than someone who is say using it in an RV while camping. I need my batteries to last the night and on cloudy days therefore I want my batteries to be as full as possible when the solar kicks out. Therefore boost/bulk is the fullest they should get most of the time. See my post about resetting the controller during the day. The float is the setting you want the batteries to be at when the sun goes down. Note please that if you don't have a continuous load, use the conventional wisdom not what I'm telling you.

Playing with the settings over the last view days has made a tremendous difference in the performance of my system. Cloudy days aside I've gone from ending my charge day at 26v to ending it at 26.7v. That's about 35% capacity vs 90% capacity. On the 35% day the transfer switch flipped at 24v to building power, my software kicked the inverter off at about 1am (24w load), the controllers followed but by sunrise I was still down near 21v. I've since fixed the timing of things to prevent that from happening again but you get the point. You can tune the parameters to make the most sense for your needs.

In case it helps, I have an 8S LiFePO4 200ah battery bank. I have two arrays of 1200w each. The first is 100w panels. The second is 200w panels. I have a third but its not setup. I'm working on the ground mount now. It will be 1320w of 330w panels. I have two EPEver 5415AN controllers running the first two arrays. I have an AIMS 2000w low frequency inverter. If you want to start motors and compressors (a dehumidifier in my case), look for low frequency inverters. At odd times I've run a 12 inch cutoff saw and a tank compressor...not at the same time of course.

Good luck with your project.
 
Hi Fafrd. I would recommend the EPEver controllers. We have finally learned how to live with one another. I just replaced my cheap MakeSkyBlue third controller with a 5420AN 200V model. I don't have the third array up yet, but the Blue produced so much interference my monitoring circuits kept getting errors.

The lithium screen is in the setup software. I found it from this forum but had mine setup just fine without it. EPEver has added some niceties for lithium that I don't need. It has low and high temp protection. I don't need them. You might. I moved my battery bank into a temperature controlled space when I switched from AGM to LiFePO4. You can configure everything you need via the MT50 or plug in a PC to see everything at once. It has a clock that keeps drifting. Not sure how it uses it. It has no battery back up and seems to lose time. It reminds me of the real time clocks in PCs back in the 90s. The harder the CPU worked, the less accurate the clock was. :) In any case, having to reboot the controller by disconnecting from the batteries seems to cause it to lose its date and time completely.
I’m not terribly worried about temperature protection or clock accuracy, but I do want to understand why Epever has stated that their ‘high current’ AN-Series chargers will not ‘work with’ lithium batteries. Will contact them about that and will report back what they say.

jnrhome and I both run continuous loads and therefore we need to look at the charge setting much differently than someone who is say using it in an RV while camping. I need my batteries to last the night and on cloudy days therefore I want my batteries to be as full as possible when the solar kicks out.

Mine is for residential as well (not RV).

I want to start charging when the sun comes up in the morning and to begin powering an inverter once the battery reaches a minimum SOC (~30%) at which point I will generally have enough solar coming in to power the inverter while continuing to charge the battery.

As the sun sets, inverter will slightly outweigh solar power coming in, but I should enter the evening with a pretty full battery at which time I will drain it powering loads in the evening and overnight.

So 75% of what I am doing is time-shifting while the remaining 25% is dynamically powering loads during the day (primarily fridges and freezer).

Therefore boost/bulk is the fullest they should get most of the time. See my post about resetting the controller during the day. The float is the setting you want the batteries to be at when the sun goes down. Note please that if you don't have a continuous load, use the conventional wisdom not what I'm telling you.

Appreciate the advice but I’m still in the charger-selection phase.

Playing with the settings over the last view days has made a tremendous difference in the performance of my system. Cloudy days aside I've gone from ending my charge day at 26v to ending it at 26.7v. That's about 35% capacity vs 90% capacity. On the 35% day the transfer switch flipped at 24v to building power, my software kicked the inverter off at about 1am (24w load), the controllers followed but by sunrise I was still down near 21v. I've since fixed the timing of things to prevent that from happening again but you get the point. You can tune the parameters to make the most sense for your needs.

My main two concerns are:

1/ avoiding a charge overshoot (perhaps that was just not using the correct settings)

2/ avoiding wasting available solar energy when the battery becomes fully-charted but then more loads start using some pf that power.

But of course, this is all after confirming with Epever that the 60A 200V charger has no limitation preventing it from working with LiFePO4 batteries...

In case it helps, I have an 8S LiFePO4 200ah battery bank. I have two arrays of 1200w each. The first is 100w panels. The second is 200w panels. I have a third but its not setup. I'm working on the ground mount now. It will be 1320w of 330w panels. I have two EPEver 5415AN controllers running the first two arrays. I have an AIMS 2000w low frequency inverter. If you want to start motors and compressors (a dehumidifier in my case), look for low frequency inverters. At odd times I've run a 12 inch cutoff saw and a tank compressor...not at the same time of course.

Good luck with your project.

This is helpful. I’ve also got a 24V LiFePO4 (2P8S). Biggest difference between my setup and yours will be that I use GTIL inverters to partially-compensate for loads during daylight hours and while the batteries last (no transfer switches needed and no concerns about startup surge).

I’m putting in 10 450W panels but will treat them as 4 separate arrays due to shading issues.

So I’ll have 3 60A 24V MPPTs and a lone 20A MPPT and I’m happy to hear you’ve had no issues running multiple Epever MPPTs in parallel.
 
Hey jnrhome. I just noticed something that might help. I reconnected an MT50 to controller #1 and tried to change a parameter. It gave me a parameter error no matter what I did. EPEver has a list of rules in the manual about what parameters have to be greater than what other parameters. In short I managed to set the under voltage warning the same as the under voltage reconnect. The controller accepted it via software but wasn't happy with it. Recall my having to reboot the controller. Just something to consider.
 
I’m sorry to trouble you both again, but I just wrote my question regarding ‘High-current Epever MPPTs working with LiFePO4’ and would appreciate a quick recap from you both:

-which specific Epever model are you using?

-did you successfully get your ‘High Current’ Epever MPPT to charge your LiFePO4 battery using User Settings ‘out of the box’ or did you need to download some software or firmware first?

-the Epever document I found states that:

‘ High current series doesn’t have self-activation function for lithium-ion battery, thus it cannot be used with lithium-ion battery’

Do you have any idea what this ‘self-activation function’ is?

And finally, a separate, unrelated question for you both. I’ll have a total of 4 MPPTs charging the same battery. Do either of you have multiple MPPTs charging the same LiFePO4 battery and if so, have you had any issues or is it pretty straightforward as long as the different SCCs are all programmed to the same settings?

Thanks.
 
fafrd,

After submitting a pic of the asset tag for my Tracer 6415AN, Epever support responded:

"Hi Jeff,
Unfortunately, the version of your controller does not support lithium battery. It lacks a part control circuit and some protections. But this isn't mean that you cannot use it on the lithium battery. It's just on customers own risk. If you still to use it on lithium battery, please refer to the attached file in last email."

Those files are attached.

You should keep in mind that almost any chemistry battery can be used with almost any charge provided the charger settings conform to the battery charge settings. "user" settings have a min/max range that can be set to match the needs of the battery, regardless of chemistry. Most (if not all) Epever charge controllers have a user setting that ranges anywhere within 9-17v on a 12 volt battery. The settings are in .1v increments allowing you to use any rechargeable battery that accepts a charge within this range. Even lead acid settings can be adjusted to service lithium cells by adjusting voltage and disabling equalization. There isn't any special charge controller for a specific chemistry. It's totally about the boost (bulk), float, absorption, equalization voltages being set for the specific battery you have. I'm sure the marketing department can come up with some doohicky feature that they can use to attract users of certain chemistries. In the end, it's all about the voltages being sent to the battery in a range they want. "Lithium settings" refers to a default template preloaded with proper settings for that chemistry and for factor. Nothing says you can't manually configure the controller with the proper settings yourself. Just saying...

JMH1783 uses multiple controllers so he would have insight on that.
 

Attachments

  • 2-6. How to modify the parameters for lithium battery via PC software.pdf
    463.8 KB · Views: 36
  • Lithium Battery parameter setting rules.pdf
    88.3 KB · Views: 32
fafrd,

After submitting a pic of the asset tag for my Tracer 6415AN, Epever support responded:

"Hi Jeff,
Unfortunately, the version of your controller does not support lithium battery. It lacks a part control circuit and some protections. But this isn't mean that you cannot use it on the lithium battery. It's just on customers own risk. If you still to use it on lithium battery, please refer to the attached file in last email."

Those files are attached.

You should keep in mind that almost any chemistry battery can be used with almost any charge provided the charger settings conform to the battery charge settings. "user" settings have a min/max range that can be set to match the needs of the battery, regardless of chemistry. Most (if not all) Epever charge controllers have a user setting that ranges anywhere within 9-17v on a 12 volt battery. The settings are in .1v increments allowing you to use any rechargeable battery that accepts a charge within this range. Even lead acid settings can be adjusted to service lithium cells by adjusting voltage and disabling equalization. There isn't any special charge controller for a specific chemistry. It's totally about the boost (bulk), float, absorption, equalization voltages being set for the specific battery you have. I'm sure the marketing department can come up with some doohicky feature that they can use to attract users of certain chemistries. In the end, it's all about the voltages being sent to the battery in a range they want. "Lithium settings" refers to a default template preloaded with proper settings for that chemistry and for factor. Nothing says you can't manually configure the controller with the proper settings yourself. Just saying...

JMH1783 uses multiple controllers so he would have insight on that.

Thanks, this is more or less exactly what I thought. So by setting user settings correctly, you can successfully charge a LiFePO4 battery and having the flexibility to disable Equalization is probably the single most important feature.

After that, it is just being certain the controller can’t ‘wake up’ in some confused state where is mistakenly goes to default settings for the incorrect battery type.

If that can’t happen (after battery shuts off from BMS LVD protection, for example), there’s probably not much to worry about.

Have you stressed your Epever by simulating BMS battery disconnect and reconnect to confirm it ‘wakes up’ in the correct state (your custom user settings)?
 
Hi fafrd. Do not use a BMS between your charge controller and your batteries. Every charge controller I've ever seen out there tells you to never disconnect the controller from the batteries while the array is connected to the controller. Will mentions this in his videos. Buy new batteries not used. Set your charge controller to cycle about 80% of the capacity of the LiFePO4s. You won't get anywhere near overcharging your cells. You can check your cells periodically to make sure they stay in alignment with one another. The greatest difference I've seen with mine is 80mv but most of the time its about 10mv.

The statement from EPEver is pretty typical. They don't have an idiot button that says lithium. You can find starting settings for lithium on this forum for any charge controller as long as it has a user mode. I was using user mode with my AGMs too.

Yes, I have two charge controllers as mentioned in my post to you. I just bought another that will be going live in about a month. Be aware that solar isn't plug and play. Bad things can happen no matter who's equipment you buy. You have to periodically check to make sure things are running properly. I wouldn't worry specifically about a failed controller. It can happen. Not much you can do about it.

What happens with more than one controller is they both do their thing. The voltage of both is constrained by the battery. The current is dependent on what they can push to the batteries. As the sun changes, array #2 which is more westerly facing increases while array #1 decreases. If the batteries are charged early in the day one will take the lead in supporting the load while the other backs off to almost zero. Occasionally the split it. There is no coordination between them. They just figure it out on their own.

One thing to be careful about with multiple controllers is total charging amps. Make sure you can't push more than the batteries are designed to take.
 
here here.
If you disconnect your battery during charge, you WILL destroy your charge controller instantly. The current from the panels has nowhere to go and lets the smoke out every time. I know of no charge controllers that can take solar charge without having a battery connected.
 
Hi fafrd. Do not use a BMS between your charge controller and your batteries. Every charge controller I've ever seen out there tells you to never disconnect the controller from the batteries while the array is connected to the controller. Will mentions this in his videos. Buy new batteries not used. Set your charge controller to cycle about 80% of the capacity of the LiFePO4s. You won't get anywhere near overcharging your cells. You can check your cells periodically to make sure they stay in alignment with one another. The greatest difference I've seen with mine is 80mv but most of the time its about 10mv.

Appreciate the input but believe there is some controversy about that (that disconnect will destroy the SCC).

In any case, if I’m truly worried about that I will use two batteries in parallel (which will never both be disconnected by their BMSes.

The statement from EPEver is pretty typical. They don't have an idiot button that says lithium. You can find starting settings for lithium on this forum for any charge controller as long as it has a user mode. I was using user mode with my AGMs too.
Good to know - thanks.

Yes, I have two charge controllers as mentioned in my post to you. I just bought another that will be going live in about a month. Be aware that solar isn't plug and play. Bad things can happen no matter who's equipment you buy. You have to periodically check to make sure things are running properly. I wouldn't worry specifically about a failed controller. It can happen. Not much you can do about it.

AC-coupled Microinverter-based solar is pretty much plug and play. This is my first time trying a DC-coupled array and it’s a much trickier landscape.

At some point, I may decide to just jump the DC ship and just go with an Enphase smart AC-coupled battery solution (though it’s still unclear to me if that will work with a user LiFePO4 battery...).

What happens with more than one controller is they both do their thing. The voltage of both is constrained by the battery. The current is dependent on what they can push to the batteries. As the sun changes, array #2 which is more westerly facing increases while array #1 decreases. If the batteries are charged early in the day one will take the lead in supporting the load while the other backs off to almost zero. Occasionally the split it. There is no coordination between them. They just figure it out on their own.
That’s pretty much exactly what I’ve been counting on.

Some posters on my ‘best MPPT’ thread are raving about the value of Victron’s or Magnum’s smart capability with direct-to-battery SOC sensors.

Do either of you see any attraction to that capability based on the Epever-based solutions you are using now?

One thing to be careful about with multiple controllers is total charging amps. Make sure you can't push more than the batteries are designed to take.

I’ve got a 300A BMS and a 2P 280Ah LiFePO4,so overcharging the battery current-wise should not be a problem.

I’ll have a 4.5kW array which will never peak close to rating, but even if it did, that would only be ~180A (0.32C).
 
EPEver has a unit to coordinate the controllers; two actually: PT-ADP-PORT and PAL-ADP-50AN. I think the second one is a replacement. The nice thing about the first is it has one central connection for the battery sensor. They are also supposed to use the same settings. Not sure how that works but I think you just set them while connected to the ADP unit and it syncs with the controllers. Don't really know. I don't own one.

After watching the behavior of mine closely and logging all the data. I don't see a problem with them acting independently. As mentioned they seem to figure things out like the controller with more sun gets to push more power. I've seen this morning and evening. Sometimes they will switch during the day (as mentioned with regards to full batteries). I actually didn't have any issues with the MakeSkyBlue unit working in tandem with the 5415AN when I was having issues.

An accurate SOC sensor would be a good thing for keeping track of how full the batteries really are. I'm calculating it in software using the shunt I have on the battery negative terminal. Btw, that's what those coulomb counters use...a shunt. Victron calls its model a smart shunt. I'm not so convinced. It's a 500A shunt!!! If you don't know anything about shunts then let me explain. A shunt is a very very very small resistor that you place on the negative side of whatever you want to measure. You can put a meter across it and measure the voltage drop from one side to another.

Because this thing is a resistor and you are passing a lot of amps it has to be extremely small so as to not generate a lot of heat and of course reduce the flow of current (by any significant degree). Typically shunts are designed to produce a voltage of 0-75mV for the shunts full range of amps. For example I'm using a 75A shunt on each of my controllers (controller to battery). The math is easy 0-75A can be read as 0-75mV. Current = 1A * (75mV / 75A) = 1mV. A 1 amp flow will show up as 1mV (approximately of course). Now lets take that huge honking 500A shunt. Current = 1A * (75mV / 500A) = 0.15mV. Now you need a pretty darn good meter to read 0.15mV. Consider of course this is also the difference between say 53A and 54A all the way up the scale. At that small a voltage you really need an opAmp circuit to boost the voltage to a readable level and at that point you start introducing noise. So best practice is to chose a shunt as close to your maximum range as possible. The one I chose for the battery is a 175A shunt.
 
EPEver has a unit to coordinate the controllers; two actually: PT-ADP-PORT and PAL-ADP-50AN. I think the second one is a replacement. The nice thing about the first is it has one central connection for the battery sensor. They are also supposed to use the same settings. Not sure how that works but I think you just set them while connected to the ADP unit and it syncs with the controllers. Don't really know. I don't own one.

After watching the behavior of mine closely and logging all the data. I don't see a problem with them acting independently. As mentioned they seem to figure things out like the controller with more sun gets to push more power. I've seen this morning and evening. Sometimes they will switch during the day (as mentioned with regards to full batteries). I actually didn't have any issues with the MakeSkyBlue unit working in tandem with the 5415AN when I was having issues.

An accurate SOC sensor would be a good thing for keeping track of how full the batteries really are. I'm calculating it in software using the shunt I have on the battery negative terminal. Btw, that's what those coulomb counters use...a shunt. Victron calls its model a smart shunt. I'm not so convinced. It's a 500A shunt!!! If you don't know anything about shunts then let me explain. A shunt is a very very very small resistor that you place on the negative side of whatever you want to measure. You can put a meter across it and measure the voltage drop from one side to another.

Because this thing is a resistor and you are passing a lot of amps it has to be extremely small so as to not generate a lot of heat and of course reduce the flow of current (by any significant degree). Typically shunts are designed to produce a voltage of 0-75mV for the shunts full range of amps. For example I'm using a 75A shunt on each of my controllers (controller to battery). The math is easy 0-75A can be read as 0-75mV. Current = 1A * (75mV / 75A) = 1mV. A 1 amp flow will show up as 1mV (approximately of course). Now lets take that huge honking 500A shunt. Current = 1A * (75mV / 500A) = 0.15mV. Now you need a pretty darn good meter to read 0.15mV. Consider of course this is also the difference between say 53A and 54A all the way up the scale. At that small a voltage you really need an opAmp circuit to boost the voltage to a readable level and at that point you start introducing noise. So best practice is to chose a shunt as close to your maximum range as possible. The one I chose for the battery is a 175A shunt.

Appreciate this explanation. At least while I’m trying to make my system work at 24V, I can’t afford the voltage loss associated with a shunt (too close to the lower voltage limit of my inverters). So that’s that then. No shunt and no ‘Smart’ charger capability for me (at east until I jump ship for 48V).

Someone posted in my other thread about how his Epever MOPT would occasionally drop custom settings in favor of default settings. So that’s an issue I want to run to ground - have you ever experienced that?
 
I haven't experienced random controller loss of settings. They even stay when i disconnect panels and batteries.
 

diy solar

diy solar
Back
Top