Jump to content
Rick C

Lights Flickering when on Battery Mode Axpert

Recommended Posts

So if I understand correctly, you only have the issue when in battery/solar mode and not when the inverter is in bypass mode (loads are being supplied by council)

I know when I first installed my Axpert I was getting some strange voltage readings between earth and neutral and I was getting about 120volts between live and neutral when running in battery/solar mode. 

When running in bypass mode (council supply ) the voltages were then corrected.

A grounding relay sorted this out and I have good voltage output when in battery mode and bypass mode.

Try this as a fault finding exercise...I have no idea if it will help, but its worth a shot?

Disconnect the input AC supply to your inverter (Eskom/City Power) but ensure that the input earth connection is still connected.

Bridge the inverters neutral output and earth output (manual grounding relay)

As plonkster says this should not be done as standard practice (a grounding box should be used)

Turn on the inductive load and see if this makes any difference?

Share this post


Link to post
Share on other sites

Just some info . Our. Washingmachine Whirlpool toploader 10.1 kg  . Newgeneration no gearbox . All clockwise - anticlockwise movement by stopping and starting of motor . This is the display readings while washing . Machine running on batteries utility switched of .

1. Volts.   49.9 tot 50.6 volts.        2. Load. 80 to 990 watts

We checked this early this morning to be sure the findings is correct . Checked different lights and types of lights no flickering .

Share this post


Link to post
Share on other sites
2 hours ago, Noobie said:

So if I understand correctly, you only have the issue when in battery/solar mode and not when the inverter is in bypass mode (loads are being supplied by council)

I know when I first installed my Axpert I was getting some strange voltage readings between earth and neutral and I was getting about 120volts between live and neutral when running in battery/solar mode. 

When running in bypass mode (council supply ) the voltages were then corrected.

A grounding relay sorted this out and I have good voltage output when in battery mode and bypass mode.

Try this as a fault finding exercise...I have no idea if it will help, but its worth a shot?

Disconnect the input AC supply to your inverter (Eskom/City Power) but ensure that the input earth connection is still connected.

Bridge the inverters neutral output and earth output (manual grounding relay)

As plonkster says this should not be done as standard practice (a grounding box should be used)

Turn on the inductive load and see if this makes any difference?

I think @Noobie is onto something here I think you have a floating neutral while you are in solar battery mode.

Share this post


Link to post
Share on other sites
2 hours ago, Chris Hobson said:

I think @Noobie is onto something here I think you have a floating neutral while you are in solar battery mode. 

If you don't have that grounding box wired in, then yes, you have a floating neutral in battery mode.

Share this post


Link to post
Share on other sites

Finally, I will call an electrician to solve it. I will let you know then. Thanks you all for the help. This forum is amazing.

Rick C

Share this post


Link to post
Share on other sites

Hi again,

 

Next week an electrician will come to see the inverter and the installation. Also, today I made a couple of measurements that may be interesting.
   Using a multimeter in battery mode, I verified that the installation has no floating neutral. The phase at 230V and the neutral less than 1V, . Probably, the inverter makes the internal bonding.


I turned on the washing machine and first using the utility, the voltage varied very little. Instead, in battery mode the voltage had  variations of almost 8 volts. Probably more, because the multimeter is a cheap and digital one, not a pro. Incandescent bulbs have no capacitors and things like that. Thats the reason why they are more sensitive to flicker

Well, I found out why the lights flicker, but I don't know why the voltage varies.

I let you know what's next.

Share this post


Link to post
Share on other sites

Rick the morning I ran the washing on battery mode I did not give you the voltages readings . It was up and down 215 to 230 V on Axpert display . This is no worry to me , its been running for more than 3 years with no problems to any electrcal equipment . Infact somethimes we run 2 machines at once no problems . Hope everything ends well .

Share this post


Link to post
Share on other sites
11 hours ago, Rick C said:

almost 8 volts

That is technically acceptable. It is within 5%. Power is proportional to the square of the voltage, so the reduction in power on resistive loads (such as incandescent lamps) would be 215^2/230^2,  or around 13%. Perhaps this should have been no surprise then :-)

Share this post


Link to post
Share on other sites

20m isn't a long run for AC, and 4mm wire is sufficient for that. 4mm wire, at 30 degrees celcius, on 20m should give about 1.61v drop. Why yours drop by 8V is a mystery but it could either be a faulty connection / isolator ?

Share this post


Link to post
Share on other sites
Posted (edited)
56 minutes ago, SilverNodashi said:

Why yours drop by 8V is a mystery

It actually correlates well with the voltage drop on the DC side. He said "49.9 tot 50.6 volts", around 0.7V, let's say 1V because he was using a DMM that's probably sampling around 3 times a second. The boost stage in the inverter multiplies that by 8, so you'll get an 8V fluctuation on the high voltage DC bus. This particular inverter than bucks that down again to keep the AC side more or less stable (around 325V is what you want, 230 * sqrt(2)). Now you're talking about a 900W load coming and going, so this buck converter has to adjust backwards and forwards continually. It likely uses a closed control loop. so when the voltage drops it adjusts the mark/space ratio upwards, and then as the voltage rises again it adjusts that ration downwards. So the inverter is essentially playing catch-up all the time. A tiny bit of fluctuation is inevitable. One would expect it to react so quickly that this is merely a volt or two. I don't know enough about the internal design to say anything about this, I'm really just thinking out loud.

Edited by plonkster

Share this post


Link to post
Share on other sites

This topic has actually set my thinking about the design choices for inverters. I need to do a lot more study, unfortunately (towards this endeavour) I majored in software and not in hardware :-)

Last night I was looking at the relative weight of the inverter in question here compared to a similarly sized Multiplus. The Axpert 5kva weights around 13kg. The Multiplus is more than double that at 30kg. You need two men to hang that thing on a wall. Now some of that is down to other packaging decisions as well, for example, the Voltronic has a very light-weight steel case. So a better comparison would be the new Multiplus II which is in a steel case as well, but since that is a 3kw model, you need to compare it to the smaller Voltronic. That again yields a number a little larger than double, 7kg vs around 18kg.

Now the one item that I know makes by far the most difference in weight is the very heavy toroidal transformer used in the Multiplus, compared to a much smaller E-core transformer used in the Axpert. The difference is so stark that untrained salesmen will harp on about "transformerless design" forgetting that these are both full-bridge high-frequency designs utilising a transformer.

The marketing documentation says this is done for high startup power, but also for efficiency (so you will note a 93% efficiency for Axpert, 95% for Victron). But what I am looking for right now is a more scholarly treatment as to why one would use a toroidal transformer over the smaller alternatives. Unsurprisingly, the toroid is more expensive (more difficult to wind, more material to make). What I found so far is that it is much harder to saturate a toroidal transformer, and if that is true I would expect a much better throughput in terms of magnetic flux, which should result in better energy transfer.

Somewhere an electronic engineer is likely reading this and laughing.... :-)

Share this post


Link to post
Share on other sites
Posted (edited)
1 hour ago, plonkster said:

Last night I was looking at the relative weight of the inverter in question here compared to a similarly sized Multiplus. The Axpert 5kva weights around 13kg. The Multiplus is more than double that at 30kg.

If I'm completely honest, I don't quite get how Victron actually sells any of their products.

It is significantly more expensive (double to triple the price) than the top end brands like SMA and Fronius.

Based on the specifications it looks like a mid-range inverter, for example the 5kVA model is down-rated to 3700 at 40 degrees celsius. 5000kVA is only possible at 25 degrees which is typically something companies with lower quality products do to make the product look better.

Compared to SMA and Fronius that are either passively cooled or has minimal fan cooling (+ IP65 vs IP21 rating of Victron) and provide their rated output to 50 degrees. It has lower efficiency than the models by SMA and Fronius (well obviously because they make passively cooled devices).

They also just don't have much market share, in fact their market share looks tiny (looking online). They may be a big deal in SA (likely due to marketing), but in the US (the biggest market in the world), they practically don't exist.

Reminds me a lot of brands like Bose and such that sell decent products priced way out of the segment they should be competing in.

Voltronic (makers of Axpert) are tough competition for companies like Victron because you just don't compete with China on commodity devices. And the amount of Voltronic devices sold is quite impressive. They are constantly bringing out new models and the latest versions really do have quite a number of improvements (both internally and feature list).

If you don't use the solar of the Axpert and get the PWM versions you can actually get away with around 8k for a 5kW inverter (now with ground relay). The Victron 5kW version is like R60k or something crazy like that. I mean I could buy 7 Voltronic inverters for that so longevity is obviously a non-starter. Efficiency isn't that much better, IP rating sucks on both.

Issues like this also don't really seem worth the expense because I could just get two Voltronics and run it exclusively for the bad device.

Anyway TL;DR: How is Victron still operating? What makes their products unique?

Edited by Gnome

Share this post


Link to post
Share on other sites
2 minutes ago, Gnome said:

It is significantly more expensive (double to triple the price) than the top end brands like SMA and Fronius.

Where do you get SMA and Fronius that cheap?

Share this post


Link to post
Share on other sites
21 hours ago, Gnome said:

If I'm completely honest, I don't quite get how Victron actually sells any of their products.

Victron started off by and still supplies the international marine industry. 

When I bought, there where no Axperts, but their was Victron. Axpert is the new kid on the block in other words.

Axpert, when they came out first in SA was via IT companies supplying the shortage of UPS'es. Some had a small PWM controller with lots of marketing hype. They are good UPS'es for a damn good price.

Since then they have upped their offering with like MPPT's etc. with a huge following internationally. Yes, they are kicking everyone else arses ito price.

Negative: Voltronic does not give after-sales services though, nor do not fix their firmware charge bug/s.
 

21 hours ago, Gnome said:

The Victron 5kW version is like R60k or something crazy like that.

Here are the latest prices from Current Automation:
Quattro 48v 5kw is Ex VAT R24 264 + controller of your choice, add 3-4k = +-R30k
Multiplus 48v is ex VAT R21 819.00 + controller of your choice, add 3-4k = +- R25k

Negative is price.
But they have excellent after-sales service / support and all bugs are sorted ASAP, if you can find one.
And 5 year warranty if ever you need it.

Will I buy Axpert? No. I like after-sales support too much. 
Will I buy a 5kva inverter in the city? No.
The loads I wanted to kill where  all below 1000w. So 1200VA is perfect  = R7855 Ex VAT for 1200VA Multiplus - if I have to start over.
Or if I really want to treat myself, i.e. and my Phoenix gave up the smoke, Multiplus II for R13 960.00 - as it is Multigrid.

My point: We cannot ask why people are buying expensive stuff, for expensive for you is cheap for me and vice versa. :-) 

Share this post


Link to post
Share on other sites
Posted (edited)
22 hours ago, Gnome said:

SMA and Fronius 

O yes, SMA and Victron is like chalk and cheese, off the shelf SMA is purely grid tied and very expensive to add battery backup, where Victron is not off the shelf grid-tied. T&C's apply.

Once you have added up all the bits and pieces to have SMA installed, it gets pretty impressive in price. 

Fronius. Fronius and Victron are "buddies, seeing that Victron's software interfaces quite nicely with Fronius inverters. :-) 

Edited by The Terrible Triplett

Share this post


Link to post
Share on other sites

I've said it before... when you look at the whole spectrum of battery inverters, there's the Axpert right at the bottom end, then there is a big massive gap, then you have the infini's... then a whole lot of makers bunched together from around 25k upwards... and if you look carefully, of this bunch there are plenty more expensive options.

In terms of a brand I don't understand, Outback is definitely it. Proprietary protocols, significantly more expensive, I'm told somewhat old tech too... but rock solid and low no-load power use. It has a market somewhere... even if I don't get it.

Share this post


Link to post
Share on other sites
3 hours ago, The Terrible Triplett said:

Victron started off by and still supplies the international marine industry. 

That's likely why they aren't that big in America: They compete against Mastervolt. Also... exchange rates. The dollar is weaker than the Euro. A few years ago it was heading towards unity, but the latest political situation has pretty much pegged it around 1.20.

Share this post


Link to post
Share on other sites
On 2018/06/08 at 10:05 AM, Chris Hobson said:

I think @Noobie is onto something here I think you have a floating neutral while you are in solar battery mode.

Yes Chris, exactly my point when I asked the question in the beginning, does the neutral remain earthed when in inverter mode? It should be.

Share this post


Link to post
Share on other sites
6 hours ago, SOLARWIND said:

Yes Chris, exactly my point when I asked the question in the beginning, does the neutral remain earthed when in inverter mode? It should be.

The situation where neutral is not referenced to earth is dangerous, but unlikely to cause flickering. The neutral to live voltage will remain stable at around 230V from the Axpert inverter, it is neutral to earth that will be floating.

Share this post


Link to post
Share on other sites
1 hour ago, pilotfish said:

dangerous, but unlikely to cause flickering

I was thinking the same thing but decided not to say it again :-) I think the flickering is simply because there is a voltage drop AC side (from 230V down to 215V) and when using incandescent lamps (not the fancy SMPS stuff that compensate for this), there will be a visible reduction in light. How much? Well, about 13% as I calculated above, enough for the human eye to notice.

Even newer lamps with an SMPS might show some flickering as well, not a clear dimming, just a little bit of a flutter. They are probably designed to be as cheap as possible and with the bulk of your customers on stable AC voltages there would likely be no reason to spend money on that!

I actually want to go test this on my test rig. I have a 3kva 12V inverter in a test rig for development purposes. It would be interesting to see what kind of voltage drop it shows on the AC side, because some of this might not even be manufacturer specific: It might just be the nature of small generators. If you had a petrol engine driving a generator, guaranteed you'd get the same thing in addition to a petrol engine spluttering up and down.

Years ago we were on a farm where the Diesel generator had this weird behaviour, where upon a load the voltage would sag and the engine would lose speed... and then suddenly recover and carry the load. Turns out the fuel governor physically had a faint groove worn in the place where it normally operates, and it took some force to get it out of there!

Share this post


Link to post
Share on other sites

Okay then... I did the test. It is not completely representative, since my incandescent lamp is connected to the same end of a 10m extension cable (that's pretty much running at its max capacity) as the large heating element (aka kettle), but I thought it remains a good demonstration: A mere 6V, or around 5% in power terms, causes a visible flicker!

Of course, when in inverter mode, the flicker is significantly more noticeable. The battery was receiving a healthy charge at the same time (to eliminate some of the voltage sag on that end), but the result was pretty similar: Drops to around 215V for about a second, and a clear dimming of the incandescent lamp.

The cell-phone camera does exaggerate the effect a bit too, it isn't quite as bad as it looks, but definitely noticeable.

It's not your inverter. My significantly more expensive inverter does the same.

My advice would be to find an LED lamp that's less susceptible to this. Should cost less money.

Share this post


Link to post
Share on other sites
2 hours ago, pilotfish said:

I guess that is a definite proof that it is the voltage regulation delay in the inverter and nothing to do with cabling.

Agreed.

But I am still at a loss why the computer at the end of a long extension lead reboots, whereas the same computer plugged in close to inverter, does not.

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

×