Jump to content

RCT/Axpert 5kW (3x) not switching over to Solar power in the mornings


Elysium5057
 Share

Recommended Posts

Hi guys

Very new to the forum and to the world of solar, having taken over an installation with my new home.

My inverters have been misbehaving this last week where they aren't switching over to Solar, and just keep on running on utility.

> Previous to this, they take a LONG while to switch over to solar in the mornings (sometimes only at 9:30 or 10am, when there has been a LOT of sunshine already on the PV).

I have now resorted to intervene these last few days by restarting them, whereby the immediately they go over to solar feed as if nothing was wrong.

Not having vast knowledge of anything solar, I would really appreciate if there is some obvious things that might be the issue here.

The low-down on my installation and settings:

  • 3 x RCT (Axpert) MKS-MKS+ 1-5KVA (15 kW) "hybrid" inverters, running in parallel 
  • 40 x Shoto 12V 100AH battery bank -- already 3/4 years old and already giving issues. I do know that this is my Achilles heel currently, but still, inverters should still be feeding solar to the house if they provide enough power, which they do if it works correctly.
  • 36 x  ILB HELIOS 255W solar panels
    • As you can see, daytime shouldn't be a issue at all. My base load is minimal and I manage all major loads very close.
    • On current readings, around 12:30 I get from the best placed sets (2 x 12 panels) 2.1kW and 1.9kW respectively. The other 12-set around 1.6kW.
    • So power throughout the day if properly managed, is totally off-grid.

Inverter settings rundown - firmware 72.4:

I do acknowledge that I have to update firmware with Columb's extensive work here (on my to-do list):

1 SOL 13 54V
2 70A 16 OSO
3 APL 23 byd
4 SdS 25 FdS
5 AGn 26 56,4V
6 Lf(?)E 27 54V
7 tfE 28 PAL
9 50Hz 29 42V
11 30A 30 ONE
12 46V 31 SbE

 

When I checked my Master this morning, this is what I am seeing. Note the PV input vs zero wattage at 8:30am in the morning (full sun exposure on panels) and MINIMAL load on the system and batteries at 48.5V (yes, crude estimation as I have come to learn).

This is not close to any major event that might've trigged the utility switch-over (geyser / kettle etc). It stayed like this until I manually intervened at 11:30!

Image.thumb.jpeg.d8d1c1380cc1d1325a1d47db9b43bd9d.jpeg

1295569499_Image(1).thumb.jpeg.4ccc921a21a0f73fa98c71125ff6a4c8.jpeg

888572089_Image(2).thumb.jpeg.22e20fb41b8e31c0fbbbbcf1c7c9fe50.jpeg

 

Would really appreciate any expert guidance here with my limited knowledge. Obviously there must be something that isn't right that is invoking this behaviour. 

I don't know if the bad apples in my battery bank (there are quite a few already tested with manual meter) is causing something like this?

TIA!

 

Edited by Elysium5057
Changed title
Link to comment
Share on other sites

  • Elysium5057 changed the title to RCT/Axpert 5kW (3x) not switching over to Solar power in the mornings

Please post the sticker on the side of one of your inverters. 

Off the bat, if these are 145 V pv models, your strings are providing too high a voltage. As the temp rises in the morning, the voltage starts dropping, and when it goes below 145V, the Unit/s switches to solar.

Link to comment
Share on other sites

28 minutes ago, Sidewinder said:

Please post the sticker on the side of one of your inverters. 

Off the bat, if these are 145 V pv models, your strings are providing too high a voltage. As the temp rises in the morning, the voltage starts dropping, and when it goes below 145V, the Unit/s switches to solar.

Thanks. That being said, am I just stuck with this situation, or can I do something about it?

If the above is the case, why does "restarting" immediately brings solar into the mix and solve my problem? 😕

Label as requested:

 

3.jpg

For context, early morning sun exposure on 24 of the 36 panels (I would guess feeding 2 of the 3 inverters):

Image.jpeg

 

Edited by Elysium5057
Link to comment
Share on other sites

Thanks Sidewinder.

Guessing the spec of the panels as I haven't had time to jump up on the roof to see what the sticker (eludes to model for specs I presume) says.

This seems to align with the information I currently have: http://www.solarsolved.co.za/datasheets/solarpanels/ILBHelios/255Watt.pdf

Unfortunately, currently, I have NO way of providing the latter panel arrangement. I am literally figuring this whole thing out with my father (luckily much more clued up than me) as we go along. Previous owner didn't provide (or probably know) all of this detail, and at the time we didn't think of it.

Having read up on string calculations, we are guessing that the installer might have got it wrong, adding more panels in series than what he should have? Hence overreaching the max voltage rating of the inverters.

Doing the math, the setup should be max 3s4p (for 12 panels per inverter), if I understand the terminology correct. Max 3 panels per string, 4 in parallel.

It might be that they did something other than this (switched around) which is causing the issues, especially with the COLD weather last two days.

We are going to start hunting this weekend to get more answers on the setup which would infer our assumptions and I'll report back as I know more. I take it we have to look at wiring and what was done in the 3 isolator boxes to get to an answer.

Thanks though for giving the breadcrumb, it probably put us on the right path!

Link to comment
Share on other sites

Actually, I can confirm the setup now - just jumped on a ladder to look at one of the isolator boxes close to the panels (yes, they're not close to the inverters :-/)

If I look at this correctly, it is 4s3p (as suspected), which means my fears were substantiated. This is for 12 panels.

Something that has exasperated the problem - I have removed two palm trees, thinking smartly that it would give me much better wattage in the morning, and that is why this issue has now started earlier in the day and takes longer to get to 145V! The irony. 

At 7:45am this morning the voltage on the PV read 155/153/154V for the 3 inverters, so, too much 😕

Solution? Change over to 3s4p?

Image.jpeg

 

 

Edited by Elysium5057
Link to comment
Share on other sites

6 hours ago, Elysium5057 said:

Solution? Change over to 3s4p?

Yes. We can't see your latest images; you can't paste a link from your email. Save the images to a file, and drag and drop from a file explorer into the post editing window. You may have to use something like Windows Paint to cut down the size for posting; phones take ginormous photos these days.

Also consider reducing setting 13 (back to battery voltage). At 54 V, the battery will have to be nearly full before you switch away from using utility power. I would also increase setting 12 (back to grid battery voltage), so the battery doesn't discharge too far before getting rescued by utility. I would put it at 48 V, with setting 13 at perhaps 50 V, or 51 V if you find it switches to and from utility too often on a poor solar day.

You have a big system (1000 Ah or around 48 kWh battery when new) for a 78 W load 🙂  That's about 0.65% of your system's rated capacity. Your three inverters draw about twice that 78 W in idle power.

Your three inverters are similar to mine, which also came with main firmware version 72.40. They will do well with patched firmware version 73.00e, as mine do, except you'll need the lead acid flavour to my LFP flavour.

Link to comment
Share on other sites

Thanks so much @Coulomb and @Sidewinder for your very helpful advice.

1 hour ago, Coulomb said:

Yes. We can't see your latest images; you can't paste a link from your email. Save the images to a file, and drag and drop from a file explorer into the post editing window. You may have to use something like Windows Paint to cut down the size for posting; phones take ginormous photos these days.

Noted. For reiteration purposes, see below image of isolator box for 12 panels. Guessing from this its clear that the installer did the installation using 4s3p, and now I'm sitting with the hefty work of fixing it. 3 times.

241544773_unnamed(1).thumb.jpg.3864064f3536cbcc56a195a953576573.jpg

 

Ample sun at 7:30am, which I cannot use:

113942270_unnamed(2).thumb.jpg.879d3dbe7cf0b508567b4135305b27e6.jpg

1 hour ago, Coulomb said:

Also consider reducing setting 13 (back to battery voltage). At 54 V, the battery will have to be nearly full before you switch away from using utility power. I would also increase setting 12 (back to grid battery voltage), so the battery doesn't discharge too far before getting rescued by utility. I would put it at 48 V, with setting 13 at perhaps 50 V, or 51 V if you find it switches to and from utility too often on a poor solar day.

Noted. Thanks for the suggestion!

1 hour ago, Coulomb said:

You have a big system (1000 Ah or around 48 kWh battery when new) for a 78 W load 🙂  That's about 0.65% of your system's rated capacity. Your three inverters draw about twice that 78 W in idle power.

You are correct! The omie I bought the house from, in my opinion, just threw money on the issue and the installer didn't really question or suggest otherwise. Granted, he wasn't really load-conscious and had a lot going at home when he was living here (3 aircons / underfloor heating possibly as well / no Geyser timer for load management / 15 year old fridge and freezer etc etc).

My family is much more conscious about this and I don't unnecessarily run things without thinking of the utility bill - hence me really trying to solve our issues with the solar input throughout the day, to be off-grid. I don't want to wait until 11am for my MASSIVE solar availability to come online.

Already implemented a wireless CBI switch on my geyser to handle warming cycles. So I can comfortably warm a 200L to 70c without worrying about the consumption throughout the day. The wife loves the hot water compared to my previous 55c solution at the flat!

1 hour ago, Coulomb said:

Your three inverters are similar to mine, which also came with main firmware version 72.40. They will do well with patched firmware version 73.00e, as mine do, except you'll need the lead acid flavour to my LFP flavour.

Great. Your extensive work here and my extensive reading quickly made me aware that this is a must and it is on my to-do list. Thanks for also confirming the correct one, you can imagine for someone new this is a deep rabbit hole to jump into 🙂 

 

Yet again, I'm already hooked and value all the knowledge everyone is providing.

Link to comment
Share on other sites

Agreed first order of business must be to reduce the string voltage to 3s. I suggest you disconnect PV just after dark (never open that breaker under load) and start work the next morning. Even if you just do 3s3p for now, until you can pull in a 4th set of pv cables, or just parallel up with a set of mc4 Y connectors.

If the photo is the complete combiner, then there is several items missing. Like fuses per string and type 2 SPD's, Earthing and metal trunking for DC cables, to name but a few. 

Edited by Sidewinder
Link to comment
Share on other sites

9 hours ago, Sidewinder said:

Even if you just do 3s3p for now,

Good idea.

Quote

until you can pull in a 4th set of pv cables, or just parallel up with a set of mc4 Y connectors.

I'd much prefer to see another cable run, and put the fuses in the combiner box. I think @Elysium5057 you will need bigger combiner boxes for the missing DC rated fuses, etc.

In retrospect, AC-DC is a terrible name for an electrical supply company. Just glancing at that breaker, you might see the "DC" part of the logo (unhelpfully in a different colour to the "AC" part) and have it register subconsciously as "ah, DC rated, good"...

Edit: zooming in, I see that it's actually only AC rated, so those will have to be replaced. Noark make suitable DC rated breakers; I don't know about their availability in South Africa.

Edited by Coulomb
Link to comment
Share on other sites

Agree with @Coulomb, never a AC breaker in a DC circuit, and yes, if you can get another set of cables from roof to combiner box, the better, but it looks like you may be short on conduit/s. If the distance from panel to box is very long (>20m?), you should upgrade to 6mm2 PV cable, perhaps. 

Link to comment
Share on other sites

@Elysium5057, this is more or less what a 2 in 1 combiner should look like. From (string 1)left: - fuse, SPD, + fuse, (-&+ isolator), and inverse for the second string. Repeat for string 3&4. So you may end up with quite a few of these, as you have 3 x 3sp4 (after changes). So in my view, i would suggest(do so at your own risk!) removing 1 Axpert(keeping it as a spare), and re-org. your panels into 2 x 3s5p. Each 3s5p connected to one inverter. You will end up with 6 panels spare, the reason is that 3s5p = 255x15=3825W, as i suspect the inverter's pv max is 3000W. I streched mine to 3600 for 4 years now. If yours is a different model, adjust your pvconfig to suit. 

IMG_20210717_082532.thumb.jpg.fa628469df5b2452ec6702bf3d4a08ee.jpg

Link to comment
Share on other sites

16 hours ago, Sidewinder said:

3s5p = 255x15=3825W, as i suspect the inverter's pv max is 3000W.

3825 is 27.5% over rated power, which is getting a bit too high. I think 20% is a good limit, and 25% is really the upper limit. The problem is the overshoots; the SCC control algorithm isn't expecting such capable PV at its input. 3S4P (3060 W) is a good match for each inverter, so I'd be inclined to keep all 3 if possible, and not have any spare panels.

Link to comment
Share on other sites

  • 3 weeks later...

Hi everyone,

 

Managed to upgrade all 3 inverters to 73.00e, as suggested.

Also, plan in the works to get someone to come fix my string config - rudimentary solution of covering a portion of each panel in a string worked like a charm, albeit sore on the eye 🙂

 

@Coulomb, I’m in the process of fine tuning the Kettlekomp feature as per the manual.

Just like to clarify the following portion: 

If the final cut-off voltage differs from its original value by more than 0.2 V in either direction, reduce or increase it by 0.5 V so that it is within +-0.2 V of its original value.

 

I noted my setting 29 at 42V, and increased the bar to show full (settled at 42.4V). Upon this, with the kettle test it worked 100% with no adjustment needed.

When I read the above though, I was a bit unsure on what to execute given the instruction to “increase by 0.5V” if my change of set 29 was by more than 0.2V (which it was). Do I increase my setting 29 by a further 0.5V or what do I do?

 

Would appreciate clarification 🙂

 

 

Link to comment
Share on other sites

1 hour ago, Elysium5057 said:

I noted my setting 29 at 42V, and increased the bar to show full (settled at 42.4V). Upon this, with the kettle test it worked 100% with no adjustment needed.

When I read the above though, I was a bit unsure on what to execute given the instruction to “increase by 0.5V” if my change of set 29 was by more than 0.2V (which it was).

Assuming that 42 V is a "realistic" setting for the cutoff voltage, the instructions actually tell you to now reduce it by 0.5 V to 41.9 V. Adding or subtracting 0.5 V (or multiples thereof) will make no difference to the KettleKomp™ operation, but this setting is still active as a cutoff voltage. I think that 42 V is a shockingly low cutoff voltage as it is, so I would not decrease it. Peruse your battery documentation, and decide on a setting, and adjust it to the nearest voltage ending in .4 or .9 V, and you're done. Or just leave it at 42.4 V. 42.0 V is the default value for the cutoff voltage setting. It's that low as an "if everything else fails" kind of value. In South Africa, with the frequent load shedding, I'd be inclined to use 47.9 V, so you don't discharge your (presumed) lead acid battery below about 50% SOC when there is no utility or solar power available.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...