Jump to content
Haydn

Axperts and Bypass

Recommended Posts

Good Day All,

I have 2 x 5kVa axperts in parallel.

8 x 220ah Lead Acid batteries

18 x 300w Canadian solar panels ( 9 and 9 )

Lately the inverter keeps going to Bypass mode using only Eskom even though we have good solar production.

Have checked firmware and settings on units all same.

If I disengage Eskom then the system works as it should off grid.

Help Please

 

Haydn

Share this post


Link to post
Share on other sites

Sounds like wrong setting for Output Source Priority (setting 1, from memory). Or possibly settings 12 & 13. Sorry, away from keyboard at present. Study the manual and/or Chris' recommended settings documents. 

Share this post


Link to post
Share on other sites

Hi Haydn,

My guess is that your battery voltage is reaching [back to utility] setting in Menu 12. There can be lots of reasons for this, but most likely your State of Charge is low and battery voltage dips below this level when a load is applied.

  • How long have you been running the system?
  • What are you running on the system?
  • Are you using ICC software (or similar) to control your system?
  • Do you have a BMV-702 battery monitor to correctly monitor State of Charge?
  • Are you using HA-02 battery balance?
  • Have you flashed the latest firmware from Coulomb^^^ which fixes charging issues with Axpert inverters?

Share this post


Link to post
Share on other sites

Morning Thanks for the reply.

 

System been running for 2 years.

It is a small household with pool pump being the largest load. Average load +- 2kw per hour. peak can sometimes go to 6kw.

Firmware is all 72/40

in reality not much in the way of control. watch power on a dedicated laptop which is shared to other devices.

no battery monitoring in place as yet will investigate.

Cheers

Share this post


Link to post
Share on other sites

The Axpert system in parallel as you have it is very good value, but only if used in conjunction with the additional items listed in my email above. The problem is that Axpert battery management is very poor and will destroy your batteries very quickly unless you provide the proper support.

So if you are finding that your system is not functioning as well as this same time last year (we have had a bad couple of solar months in Joburg and I have been nursing my batteries through this period) then your batteries may well be looking over the edge of the abyss. You may be able to help them recover, and maybe too late.

I am not going to explain everything from basics because it will take forever - the info is all available on the forum, I will point you in the right direction and when you have done some research then by all means come back with questions.

  • Check the 8x individual battery voltages and 1x bank voltage in the early morning (before PV wakes up) while system is in grid mode to ensure no power in or out of batteries - if you find major discrepancies then post here for comment.
  • Search the forum for info on HA-02 battery balancer - they are essential in your type of battery setup to maintain battery balance without which you can lose the entire bank through 1 faulty battery. You will require 2 as each one connects to 4 batteries.
  • See attached instructions for @Coulomb and @weber firmware update which addresses an issue where Axpert often goes to float way to early which will result in undercharge leading to battery sulphation and damage.
  • Install ICC for Windows on your laptop which will give you vastly improved insight into functioning and performance of your PV system, and much better battery management if you add the BMV-702
  • Install a BMV-702 which gives vastly improved measurement of State of Charge of your batteries, and will work in conjunction with ICC to improve control.

The above seems like a long and expensive list but will actually probably only run you around R4k, about the same as ONE of your batteries.

 

Coulomb - Patch Instructions.docx

Share this post


Link to post
Share on other sites

Many thanks guys,

 

valued input.

Will do all the necessary investigations and will report back.

 

Share this post


Link to post
Share on other sites

I'm finding lately that my Axpert bypasses much more readily (with a much smaller load surge) when the battery state of charge is low. I have a failing lithium iron phosphate cell, which I suspect is exacerbating the bypassing. 

So what might be happening @Haydnis that your SOC could be low due to chronic under-charging due to the premature float charge bug, and this is causing your battery voltage to dip on even minor load surges. The Axperts are very willing to go to bypass when AC input is available, and the battery voltage dip is all the excuse it needs. 

Share this post


Link to post
Share on other sites
16 hours ago, Coulomb said:

I have a failing lithium iron phosphate cell, which I suspect is exacerbating the bypassing. 

Hi Coulomb could you expand on this. 

How big is your battery bank?

Is it 16S or 17S?

How old is the cell is it a recycled cell or something that was new when you bought it?

 

Share this post


Link to post
Share on other sites
On 2018/05/20 at 1:23 AM, Chris Hobson said:

How old is the cell is it a recycled cell or something that was new when you bought it?

My battery bank is an extreme DIY job. I bid on an EV kit on Ebay. It had the battery, motor, motor controller, battery charger, and other bits. I think I won it for about AU$2000, including 45 160 Ah LFP cells (nominally 144 V). These were older cells, from 2009, about 5 years old at the time. The EV kit was an impulse purchase by a business person on an overseas trip. 

EV Conversion Kit on AEVA has pictures and a full list of contents.

45 cells is a bit of an odd number, when 16 cells makes a nominal 48 V system. It's just a little short of 3 48 V strings. In retrospect, I should have been thinking 15 cell strings, but that wasn't on my mind at the time. Weber was interested in one string of 16, so that left 29 cells, enough for 14.5 buddy pairs. 

It happened that a colleague in Brisbane was converting a car, and had a BMS (?) accident that rendered some 24 odd cells useless for EV currents. I bought those for AU$200. His was a small car, and he used 100 Ah cells. So I had enough there to make an 8S2P 24 V pack to replace my ailing lead acid lighting system pack, and several cells left over to finish off my 48 V system. So I used two 100 Ah cells to buddy with one 160 Ah cell, and four 100 Ah cells to make the last "cell" of the 16S bank. So most "cells" of the 48 V bank were 320 Ah, the "hybrid" was 360 Ah, and the "quad" was 400 Ah. I figured that because the 100 Ah cells were abused and were unsuitable for EV work, I could not count on them being full capacity, but I thought I might end up with roughly 320 Ah each. So that makes about a 320 x 48 = 15,400 Ah or 15.4 kWh bank, or it would have been when new. LFP loses capacity of a percent or two per year, so realistically about a 14 kWh bank. 

Alas, the abused 100 Ah cells had a much higher internal resistance than the 160 Ah cells, which although now 9 years old, were unused before I put them in my energy system. Weber and I have our own battery management system, called LyteFyba, which can accommodate different internal resistances, so I put up with it for years. But now the quad "cell", which has by far the highest internal resistance, seems to be causing the whole pack to have low capacity. I work the pack pretty hard; it often drops overnight to 40%, sometimes 30% SOC. We have some of the most expensive power in the world here, so it's worth minimising mains power usage. 

So that's what I mean by one of my cells failing. I plan to swap out some of the bad 100 Ah cells for better ones in my 24 V system, hopefully in such a way that the 24 V system is still usable. The latter is convenient as a test pack for battery charger repair, and as an emergency backup for the main energy system. It still runs a few lights in the house as well. 

Edit: my latest figures for internal resistance are 490 μΩ for the 360 Ah buddy pairs, 850 μΩ for the hybrid, and a massive 1,800 μΩ for the quad.

[ Edit: it's the quad with the highest resistance, not the hybrid. ]

Edited by Coulomb
Fixed link; ruined -> useless for EV currents.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×