Confusion over a few aspects with Conversol Invert ,Pytes Batteries and Watchpower

[nunrgguy]

Hi, hope it's OK writing about more than one thing in a topic.

Our PV array is 6of 550w panels - 3.3kW total, wired in 2 strings - 120V. North midlands UK. Mounted at less than the ideal angle (you deal with what you have) - approx 18deg. So I'd expect maybe a max of 2.2kW at peak conditions (approx 65% max output).
Pytes Ex box 48100R batteries (no they're not rebadged Pylontechs although I believe they use the same cells, the build, BMS, comms etc are all different) (2of 5.12kWh) and Conversol V7 inverter 5kW.

Today it's sunny.
The inverter says we're charging with 32A from PV only. There's a fair delay between what you see on the front panel and what you see on Watchpower but after a while, Watchpower correlates. However it says PV power (from the panels) is only 2.1kW. If the power is only 2.1kW how can we be charging at 32A when there's no current being fed in from Utility (SBU mode)?

It's quite difficult to see what's actually going on generally because of the delay but it never adds up. I'd generally expect what's reported as charging power to be less than what is possible from the array (being reported) at any time because the house is also running from the same load, it's not all going to the batteries?


On another topic completely. Last week the batteries were drawn down a lot due to all the cloudy weather we've been having and we kept getting an error 04 - low battery voltage. The lowest battery SOC is controlled wholly by the BMS, there's no control over that. I've done plenty of searching on this topic and error 04 comes up over and over. I assume that at the time of day (early morning) the batteries where sufficiently discharged (we were running in SBU mode and the batteries were showing 1 LED charge left) that minimal draw on the battery caused them to drop below the minimum voltage set in the inverter which was at the time the default (I think) 46V. Now I'm assuming what happens is that once the warning is triggered ,it stays in the warning state until the 'back to battery voltage' that has been set in the inverter has been reached or (I believe) 4V above the low voltage limit. I found that as solar for a few days was so low all I could do to get out of the warning state was to manually change mode to bypass, then USB (and left it) until the battery was quite well charged again.
Since then I've upped the lower battery voltage limit to 50V and the back to battery to 54V (actually I think 54V is the default anyway) but this doesn't seem to stop the batteries drawing right down the same, all it seems to do is ,is if you hit 50V due to inrush from an induction hob or washing machine (say) strobe the inverter into bypass until the situation is over when there's good charge in the batteries- which is what it should do, BUT, when the batteries are in a low state of charge you still get error 4, and in that situation it isn't putting the system in bypass? Why would that be?
Hope I've explained it well enough.

Other settings:
1 Output source priority: SBU (solar battery utility)
2 Max charge current: Set automatically by the BMS - 100A
05 Battery type: LIB protocol
10 Operation logic: Aut ( if selected & utility is avaiable inverter weorks in line mode. Once util freq is unstable inverter works in bypass)
11: Max charge current 30A
12 set voltage point back to utility source when in SBU: 50V
13 set voltage back to battery when in SBU: 54V
16 solar priority : SLB UCB (solar for battery first, allow utility to charge)
23: bye (bypass enabled)
26 bulk charge voltage: set by bms to 56.8
29 float charge voltage: set by bms to 56.8
20 low batt cutoff voltage: set by bms to 45.5


Just checked Watchpoweragain, it reckons 26A charging but only 1751W at the panel, 178w to house (errr....1751-178 = 1573/120 = 13A.....??????)

 

[houseofancients]

Conversol is made by voltronics ( just like mpp solar, eg4 ( some models), axpert is their own brand, voltacon , and many many more.

Watch power sucks

First get yourself a proper monitoring system, solar assistant ( solar-assistan.io) I found to be among the best

Based of proper monitoring, you should get a much better idea

 

[nunrgguy]

Cheers.
Yes, Conversol are made by Voltronics and I get that impression from what I've read elsewhere already that Watchpower isn't great.

Would you say the what the inverter itself displays is also inaccurate? I don't see how I can be charging with 32A from PV and that's what it reads? There's no draw from Utility.

Any ideas on the other issue (error 04)?
Thx

 

[houseofancients]

Error 4 = battery error (BMS) , which seem logical when completely empty

57,8 ( charge voltage) x 32 amps = 1849,6 watts, so doesn't seems off at all

And to quote an Australian youtuber.. watch power suck donkey balls.. yes it is that bad, and only updates once every 5 minutes

 

[nunrgguy]

Ah....so that's 1849 being drawn by the batteries.

So on the last reading it's 57.8x26 = 1502 plus what the house is drawing 178watts @ 230v = 1680, efficiency is what? 95% so 5% losses, so add those back in = 1764 so the apps round about right from that POV.

However, this is where it has me confused. The inverter from panel says current from solar (note that’s current from solar, not battery charging current) =32A, solar voltage is 120V, 32x120=3840???

It's a new install so I'm curious atm what I can get out of the panels etc...brain back in gear, the only way to see that is to temporarily increase the max charging current or switch everyone on in the house on to see what max current draw I can get off the panels is.

The error 04, true, but the point is I don't want them to draw down until completely empty. The BMS only allows them to draw down so far anyway but there's no control over the BMS max and min state of charge. Inverter Setting 12 should ensure that battery voltage doesn't draw down more than 50V, if 50V is hit when the batteries are asked to provide power it should allow Utility to join and it then shouldn't allow the batteries to be drawn upon until they're back up to 54V (setting 13), a low battery warning should never happen or, if it does ,the batteries shouldn't be allowed to draw until near empty? But the point is that even though I've got it set higher than recommended it's still allowed the batteries to draw down. Is that because it has happened on overnight 'trickle' draw where there's no significant power requirement until someone switches a kettle on in the morning i.e. there's not a huge draw so the voltage doesn't drop ,the batteries deplete, they're asked to provide power and it then hits the low voltage limit? Anyway around it, other than keep checking on the amount of charge and manually changing to SUB mode and back?

50A is higher than the recommended setting (48V from memory - only found it by googling and Schneider are using Pytes batteries, it's in one of their docs there's absolutely nothing about it in the documentation) by some margin btw.

 

[SeaGal]

So I'd expect maybe a max of 2.2kW at peak conditions (approx 65% max output).

why so?

I'd generally expect what's reported as charging power to be less than what is possible from the array (being reported) at any time because the house is also running from the same load, it's not all going to the batteries?

Does your inverter not show what PV is producing, what charge is going to batteries and what to house?

... that minimal draw on the battery caused them to drop below the minimum voltage set in the inverter which was at the time the default (I think) 46V.

46V is pretty low.
Voltage is not a good indicator of state of charge though.
It would be far better to connect the comms between battery pack and inverter so that when the batteries report a low SOC (like 15 or 20%) the inverter stops discharging from them. I guess that is possible with your configuration, but not familiar with those specific products myself.

 

[nunrgguy]

Hi SeaGal,
I'd probably expect to get 75% of rated capacity in semi-ideal conditions (guestimate) other than when it hits the 'perfect' sweet spot. The year has rolled on and my panels are at 18deg not at 26deg which would be nearer to ideal where I am, so thumb in the air I knock another 10% off potential power generating capability. But unless I'm actually drawing that much power I can't see what they're capable of generating, it would require load bank testing etc.: I've yet to stand over it midday, with the sun blaring with a fully charged or disconnected battery the get my wife to boil a kettle to see if it all comes from solar. Until 3 days ago we've had constant cloud cover for about the past 4-5 weeks, system only finally got commissioned 2 weeks ago anyway... I doubt it would.

..............You can only see what the PV is producing at a point in time. Current draw is load dependent, you can't see ultimate capacity just by looking at it. If current draw has been limited by battery max charging current (here 30A) it's not going to draw any more off the panels than 30A plus what the house is using at that time - assuming the panels are producing enough to cover it at that time.......

Scratch that last para...nah, that's wrong, other than current being load dependent, because I was playing around with charging currents and was setting one and the BMS kept changing it as I didn't realise it would auto set that. Utility max charge is set at 30A...ultimate max charging is set at 100A by the BMS so yes, 32A charging current is perfectly possible from solar. Doh!

The inverter does show what current is coming in from PV etc - my first post says so, but it doesn't match what Watchpower tells you. Also battery voltage etc - what the front panel tell you is not what Watchpower is reporting.

Yeah it's all running on comms, that's why certain inverter settings are automatically set by the BMS.

I agree 46 (I just checked, yes 46V is the default), the battery manual states a nominal voltage of 47.5 to 57.6 . I've seen recommendations online for 48V but I think that's still too low, 49.5 (as a voltage only) would give 20% of the range, which is why I've set my low voltage setting to 50V. But it displays exactly the same behaviour set for 50V, maybe that is still too low? I know voltage is a poor means of monitoring but this system allows nothing else and you have no control over the BMS. SOC is not reported to the user.

Re-edited again, hope it makes sense, typing so quickly I keep typing A instead of V

 

[SeaGal]

Why so? Because you'll usually only get about 75% of rated capacity in semi-ideal conditions. Year has rolled on and my panels are at 18deg not at 26deg which would be nearer to ideal where I am, so thumb in the air I knock another 10% off potential power generating capability. But unless I'm actually drawing that much power I don't think I can actually see what they're capable of generating. I've yet to stand over it with a fully charged battery the get my wife to boil a kettle to see if it all comes from solar, I doubt it would.

I'd be surprised if just the angle makes that much difference. But are you not set up to export any excess power? If so, then you'd see the generation without making a cuppa tea

Yeah it's all running on comms.

I agree 46, the battery manual states a nominal voltage of 47.5 to 57.6 . I've seen recommendations online for 48V but I think that's still too low, 49.5 (as a voltage only) would give 20% of the range, which is why I've set my low voltage setting to 50V. But it displays exactly the same behaviour set for 50V. I know voltage is a poor means of monitoring but this system allows nothing else and you have no control over the BMS. SOC is not reported.

OMG!

Redited again, hope ot makes sense, typing so quiclly I keep typing A instead of V

Yep don't get your A's and V's mixed up

 

[nunrgguy]

No, not exporting, it's an offgrid style inverter.
Re the SOC not being reported what I mean is there’s no real menu/display of it: there’s a bar display on the inverter that shows battery level and there’s the row of LEDs on the batteries but…what use is it really, unless you’re going to sit there 24/7 and watch it and manually change charging modes etc it’s pretty useless. There’s no setting in the inverter (other than the voltage setting) to say ‘when battery charge drops to x% prevent any further draw from the battery’. Afaik and it’s only from Googling that’s all handled purely by the Pytes BMS that lets the things drop to 10% and I do t want them dropping to 10%. There’s nothing in any of the documentation about this btw.

 

[ddmbr]

As others, I recommend you get solarassistant and a usb to rs232 cable. The cable runs from the solarassistant unit to the pytes console and will provide comprehensive metrics for the battery array. Also with solarassistant it provides an MQTT server which you can use to integrate with other components or build automated processes if you are technically inclined.

I have pytes type B batteries; below a screenshot of the solarassistant battery view:

 

[ddmbr]

I forgot to add, if you have a raspberry pi lying around you can order up the image and download it today to install for $55.

SolarAssistant

solar-assistant.io

You then just need a usb to rs232 cable from amazon.

https://www.amazon.com/gp/product/B075V1RGQK/ref=ox_sc_act_title_1?smid=A1IR1GJH6XB88L&th=1