Jump to content
Thank you for the great forum, Safe Driving over the weekend. Sincerely Jason
Youda

Youda's off-grid LAB

Recommended Posts

Directly, from the console port (front of battery to RPi), I believe it is as accurate as it can be, I have no other means to check it, unless you guys can assist. I use that cable they developed for ICC/Pylontech.

I am still struggling to have the data from both Kings being read. I have a 2nd Rpi available, but have to buy the software and then somehow consolidate the data.

For now I have the "correct" battery data with half of the setup data on the other side. Last night I plugged the cable into the other inverter to monitor that for a week or so.

I know my one array produce a bit less than the other, due to shadows in the morning and afternoon. My guess is that one array will get at least 1-2 hours less full sun than the other.

I am not willing to chop off trees, I am however willing to break down part of a wall though :)

  

Edited by Wilfred
added

Share this post


Link to post
Share on other sites

Youda, I followed your suggestion and changed my voltages to: 

C.C. = 52.5V
C.V. = 52.0V

Everything worked fine,, and the batteries did the usual "wait at 89%" and then settled at 100% SOC.  We are having a bit of cloudy weather at the moment, and as soon as the batteries had to recharge (after assisting the PV around 12:30) they went into a sort of an oscillation between CC and VC during the recharge after that.

Is this normal after lowering the voltages (they were both set to 53 V before) until the BMS sorts itself out?  I increased "float" up to 52.5 V and it got rid of the up & down bits right away!

797706923_SOCvsVwithYoudassettings.thumb.PNG.5b183f09ab0601301694cc3531526fb3.PNG

Share this post


Link to post
Share on other sites

@ChristoSnake

you're right - it looks like an oscilation between C.C. and C.V., which is strange as I'm seeing nothing like this on my system. And I have the same type of inverter, too.

Just an idea - don't you have some self-consumption optimization device in your installation? For example geyser that being turned on based the battery voltage, SOC, etc.?

Edited by Youda

Share this post


Link to post
Share on other sites

Nope - simple installation with no such fancy stuff!

In the mean time I'll retain the lower charge voltages, but keep "bulk" and "float" the same to stop the bouncing.  Just goes to show - no to installations are exactly alike?

Share this post


Link to post
Share on other sites

I do agree - moving both values to 52.5V will do no harm. Especially if you had C.C.=53V and C.V. = 53V before.

But there must be something going on, since 0.5V drop is not enough to instruct the Infini to repeat CC/CV cycle from the scratch. Just to be sure, pls check the settings of your Inifini for charging rules. Because there might be some setting that was valid for the old 53V setup and now it interferes.

Screenshots attached:

image.thumb.png.5176bcb4d8de604bfb3a7785c2ae4306.png

image.thumb.png.b2384c85369c73a084392300b169ee58.png

 

Share this post


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

Youda, I followed your suggestion and changed my voltages to: 

C.C. = 52.5V
C.V. = 52.0V

Everything worked fine,, and the batteries did the usual "wait at 89%" and then settled at 100% SOC.  We are having a bit of cloudy weather at the moment, and as soon as the batteries had to recharge (after assisting the PV around 12:30) they went into a sort of an oscillation between CC and VC during the recharge after that.

Is this normal after lowering the voltages (they were both set to 53 V before) until the BMS sorts itself out?  I increased "float" up to 52.5 V and it got rid of the up & down bits right away!

797706923_SOCvsVwithYoudassettings.thumb.PNG.5b183f09ab0601301694cc3531526fb3.PNG

How do your load and PV watts look like at the same time?

A bit of Ironing at that time maybe?

Share this post


Link to post
Share on other sites

I am running my 4x Pylontech 's at 52.5V for both CV an CC for the last few months without any issues. It goes to 89% for roughly 30 minutes and then 100%. Because of the inverter (Axpert) slow response on the SCC, the charge (battery) voltage does vary a bit when the sun comes and goes, but SOC stays on 100%. I also get that up and down graph when ironing. Also because of inverters slow response to load changing. 

Share this post


Link to post
Share on other sites

Thanks for all the comments!

Here's a graph plotting inverter load, battery & grid power during the same time.

The spike in the middle (12:30 - 13:10) was a 2kW hot water element turning on (via Geyserwise timer) to give our evening bath water a temperature boost.  Smaller spikes due fridge/freezer motors, and base load mostly swimming pool pump and a few computers.

I am not allowing AC charging, only PV.  The time and current based charge is set for <1 A for 60 minutes and <48.6 V, so it should not interfere.  I'll see what it does today with CC = CV, as reference, then try again with CC > CV tomorrow to see if I can reproduce the weird behaviour.

2118267084_Youdasbatterysettings-load.thumb.PNG.b61ab997c63cb605d5bab2eabbe44520.PNGse dangerous "iron" PS:

PS: I notice a lot of comments about "ironing".  I presume that it refers to the dangerous action of applying extreme heat to non-wrinkling fabric for the purpose of prematurely aging the fibers so that you have an excuse to buy new clothes?  I'm glad to report that we do not own any of those evil devices and would never consider taking part in such unnatural rituals :-)

Share this post


Link to post
Share on other sites
On 2019/05/08 at 5:27 PM, Youda said:

so, if you want to check the status of your Pylontech Phantom-S, US2000 or US3000, there's a great diagnostic tool for this called BatteryView.

thanks @Youda. I have recently changed from Lead Acid to Pylontech batteries. The diagnostic tool works perfectly and provides valuable information. There is only one issue you might be able to help me with. How do I display the Battery info (serial number, software version etc.) of all the batteries in the stack? In the window on the bottom left it only shows the info of one battery (I guess the master). 

Share this post


Link to post
Share on other sites

@Fuenkli

you're right - no matter which brick you highlight in the GUI, it always shows the SN + FW info for the 1st brick. BTW, the same is true for "Cycle Time" in the main window. All other values in the GUI are being read from the highlighted brick correctly:

image.png.57cbe894da050802162c6d9269b2fb42.png

In order to get detailed info, the best is to do it via CLI (described couple of posts earlier) and issue command: info 2 or info 3 and so on...
Some other usable commands are:
bat 2
soh 2
stat 2
pwrsys

 

pylon_debug>pwrsys
 Power System Information
 ---------------------------------
 System is discharging
 Total Num                : 8       
 Present Num              : 8       
 Sleep Num                : 0       
 System Volt              : 49756    mV
 System Curr              : -17724   mA
 System RC                : 558692   mAH
 System FCC               : 588892   mAH
 System SOC               : 94       %
 System SOH               : 100      %
 Highest voltage          : 3319     mV
 Average voltage          : 3317     mV
 Lowest voltage           : 3315     mV
 Highest temperature      : 22000    mC
 Average temperature      : 21500    mC
 Lowest temperature       : 20000    mC
 Recommend chg voltage    : 53250    mV
 Recommend dsg voltage    : 47000    mV
 Recommend chg current    : 118400   mA
 Recommend dsg current    : -296000  mA
Command completed successfully

 

pylon_debug>info
Device address      : 1
Manufacturer        : Pylon
Device name         : US3000A
Board version       : PHANTOMSAV10R03
Main Soft version   : B65.6
Soft  version       : V1.3
Boot  version       : V1.4
Comm version        : V2.0
Release Date        : 18-09-12
Barcode             : xxxxxxxxx
Specification       : 48V/74AH
Cell Number         : 15
Max Dischg Curr     : -100000mA
Max Charge Curr     : 102000mA
EPONPort rate       : 1200
Console Port rate   : 115200
Command completed successfully

 

Share this post


Link to post
Share on other sites

Does the Pylon report state of health or is this something that can be derived? I see my Venus is reporting 99% State of Health, batteries are less than 6 months in use.

Share this post


Link to post
Share on other sites
6 hours ago, PaBz0r said:

Does the Pylon report state of health or is this something that can be derived? I see my Venus is reporting 99% State of Health, batteries are less than 6 months in use.

Pylon BMS has a special counter for State of Healt for each brick. It's published via CAN bus, so I'm 99% sure that Venus is just reading the value.

Technically, whenever Pylon BMS encounters an alarm, it decreases SOH percentage counter a bit. For example, when you discharge some cell(s) too deep, overheat etc. 99% SOH does not mean that the cells were damaged nor that their capacity is lower than nominal. It just says that something dangerous happened in the past.

Share this post


Link to post
Share on other sites
15 hours ago, Youda said:

Pylon BMS has a special counter for State of Healt for each brick. It's published via CAN bus, so I'm 99% sure that Venus is just reading the value.

Technically, whenever Pylon BMS encounters an alarm, it decreases SOH percentage counter a bit. For example, when you discharge some cell(s) too deep, overheat etc. 99% SOH does not mean that the cells were damaged nor that their capacity is lower than nominal. It just says that something dangerous happened in the past.

Thanks, it is pretty strange that it did have an event.

Charge/Discharge limits governed by the BMS with coms to the Venus, batteries have never been lower than 30% and Temp has never been higher than 35°C. Oh well.. hopefully like you say it is not an indication that something is damaged or that I have less capacity. 

Share this post


Link to post
Share on other sites

An update from my solar Lab:

As some of you might remember, I'm currently running 3x InfiniSolar 5K combined into a single-phase AC source of 15kW. Most of the year, it's okay, since there's plenty of sunlight and the idle consumption of the three running inverters does not matter. But during winter, it's a completely different story and every watthour, that I'm able to save, is crucial. Therefore, I decided to code a feature that will allow me to run just a single inverter 24x7 and once there's need for a more power, start two other inverters automatically. So, here's the result:

  1. I created a short LUA script in Fibaro, that's constantly checking the AC load.
  2. Normally, just one 5kW inverter is running. Once the AC load goes above 4000W, the script starts other 2 inverters. It takes just a couple of seconds, and 15kW of power is ready to serve the loads.
  3. I've added a safety function, that will turn-off all three inverters once the SOC of the Pylontech stack falls bellow 25%. Once this happens, there's an ATS that will automatically switch all the house circuits to "ESKOM".
  4. Later, if all the inverters are OFF and SOC goes above 70%, then the script starts 1st InfiniSolar and the ATS will switch all the circuits back to the PV.

Here's the GUI:
image.thumb.png.360e8f875e38506fb30a8ce15f291f1f.png   image.thumb.png.ead1dff8371e1c484ed9a24ed3429c54.png

Here's the main part of the LUA script:
image.thumb.png.d80b1225ea8dd5a3ba7e82cc6f552a04.png

Since I have an EV charging wallbox installed too, I've also added a second script, that works like this:

  1. Normally, just one 5kW inverter is running.
  2. If there's no EV connected, the wallbox advertises 10A AC charging (2300W).
  3. Once the EV starts to draw power from the wallbox, the script will start another two inverters.
  4. Once all three inverters are running, the script will instruct wallbox to advertise 32A charging (7300W).

GUI:
image.thumb.png.b915525cd00ec9f15ba200bc774ba93a.png

I was too lazy to implement a function that will automatically turn-off two of the inverters once their power is not needed. But since I can turn them off via the mobile app, I'm okay with that. But I will add such power-off function later....maybe ;)

 

Edited by Youda

Share this post


Link to post
Share on other sites
On 2019/03/10 at 11:49 PM, Youda said:

The next part is the circuit with an SSR relay, that will direct "solar energy" into the geyser whenever one of these conditions will be met:

- the battery is full
- the charging current is higher than X amps

image.png.ca45ee1cc07e687db89ef5b22f73b2df.png

The great is, that you can tell the SSR how much power you want to use and it will take care of it using a standard 230V AC phase control. Basically, it's similar to a variable speed capability of an electric drill, for example. The circuit is made of SSR with 0-10V control, EMI filter and 0-10V module (Qubino) that can be wirellesly controlled from my home automation gateway (Fibaro). In the home automation gateway, I have a short script that monitors operating values of the battery and inverters and if there's some spare solar power, it basically adjusts the AC power flowing through the SSR:

 

 

Hi Youda

Thanks for sharing your experience on this forum.

I am planning on converting my geyser loads to variable loads too, the amount of power going to heat the water would vary based on current solar radiation, current household load, and battery bank SOC.

I have been looking into designing the equivalent of essentially a solid state variac so that I can ramp the AC voltage up and down to control the power going into the element.
I figured it would need to be done this way as I expected the inverter would not cope with a heavy phase angle switching load of 3kW on a 4.6kW inverter.
I then decided that designing a simple high power buck converter running directly off of the DC PV panel string voltage going into the inverter would be much simpler and lower cost and I was going to pursue this solution until I came across this topic. I wouldn't have to added in the complexity of MPPT into my buck converter either as I was going to rely on the inverter MPPT by using only a portion of what the inverter was taking from the panels, eg 50%

Can you confirm that your inverter has no problem driving a geyser element through a phase angle switched SSR to control a varying amount of power going into the element?
You have 2 inverters running in parallel for 10kVA and so your heating element is only 24% of you total inverter available power so perhaps this is making the solution viable for you?
Can you comment on using phase angle power control of a 3kW load on a 4.6kW inverter that is running off grid?

Thanks for your comments


 

 

 

Share this post


Link to post
Share on other sites

Hi @Power Me

the original heating element was designed for the 3-phase grid (3x 1333W) and I rewired it to 1x 1333W fed from ESKOM + 1x 2666W fed from the inverter. So it's not a full 3kW but only 2.6kW.

I tested that the 2666W heating + phase angle SSR runs okay, even when just one of my inverters (InfiniSolar 5KW) is running, while the other two inverters are switched off completely. Ironically, the loudest electric buzz is being produced when the SSR is giving roughly 0-500W of power. Once the SSR is delivering more power, up to the max, it runs very smoothly. From my experience, I would say that if your inverter is able to cope with 3kW load directly, it will be okay with the phase angle power control too. You might experience some buzz and flicker, but only in the initial phase of control curve, just like I am.

Personally, I went for 230V 25A Carlo Gavazzi RM1E48V50 with 0-10V control. Normally, I would choose a bit more expensive RM1E48V50 that's rated for 480V and 50A, but it was not on stock at that time.

https://www.gavazzionline.com/pdf/RM1EAAdatasheet.pdf

image.png.df1fcacab928eaaecca83cfd28b7713d.png

The SSR gets pretty hot during the operation, so don't forget to add a decent heatsink. I went for RHS100 heatsink, which comes together with a DIN rail clip. The only drawback is, that the bundle is too high to fit in a shallow distribution board.
http://www.productselection.net/PDF/UK/ssr_accessories.pdf

image.png.d503ae9bae54d459c8e81092678a4fb4.png

 

Mine last advice is: don't try to regulate SSR too quickly. Give MPPT at least a second to adjust it's power once you regulate the AC load up or down. Otherwise, your regulation will be faster the MPPT and the monitoring light on the geyser will flicker like a disco-ball :)

Good luck and please post your results on the forum!

Youda

 

 

 

Edited by Youda

Share this post


Link to post
Share on other sites

Thanks for your quick reply, @Youda

Very encouraging that you have successfully tested ~2.7kW phase angle switched load on a single 5kW inverter.
Can you confirm this single inverter was running off grid at the time of the test and not in hybrid mode (grid assisted)?

I personally would have guessed worst case would have been switching the load on mid cycle (peak inrush current) for 50% power delivery.

I am running a GoodWe 5048D-ES which practical testing has confirmed is good for 5.4kW with grid present and 4.6kW running off grid.
I want my geysers to heat even during load shedding taking place midday and so have to move the geysers to the backup output of the inverter.
I must however then give the geyser loads smarts to avoid overloading the backup phase as I can't always be around policing what other loads are being humanly activated in the home. 😉

I have tested the geyser and even the oven(4kW) on the backup phase and there is no issue driving the loads with no grid assist, obviously not at the same time.

Yip, heat sinking is a must for the SSR or Thyristors. (2V drop at 14A rms  = 28W heat dissipation in the SSR)
I will go with a custom discrete component design to reduce costs as a phase angle controlled SSR is very costly.

In the long run however I might still go and design the DC power circuit running directly off a portion of the PV DC supply, so as to avoid the conversion efficiency losses of the whole PV DC to 220V AC. Why place a 3kW resistive load on the inverter when it could just be supplied smartly from the PV array directly.

I am aware of existing products on the market with similar functionality, but in my opinion they are grossly overpriced and I don't think they can share existing PV panels with an inverter.

 

 

Edited by Power Me
corrected typo

Share this post


Link to post
Share on other sites

Well, if you are experienced with designing power electronics and it's a hobby of yours, then it's okay :)

And now I see that I made a mistake in my previous post. Correction: "I went for a cheaper RM1E23V25 (230V x 25A) since RM1E48V50 (480V x 50A) was not on stock."

 

12 minutes ago, Power Me said:

Very encouraging that you have successfully tested ~2.7kW phase angle switched load on a single 5kW inverter.
Can you confirm this single inverter was running off grid at the time of the test and not in hybrid mode (grid assisted)?

Yes I can.
While my inverters are hybrid models, I run them in pure off-grid mode. Even the grid MCB, that's before my inverter farm, is disconnected all the time.

Share this post


Link to post
Share on other sites
21 minutes ago, Youda said:

Yes I can.
While my inverters are hybrid models, I run them in pure off-grid mode. Even the grid MCB, that's before my inverter farm, is disconnected all the time.

Electronics design is more than a hobby it is my career too.

Thanks for confirming your test results.

Running completely off grid (incoming breaker in open position) is my end goal too. I am kind of tired of paying for the (1 to 2)kWh/day that leaks back and forth through the pre-paid meter as the GoodWe attempts to achieve 0 export.
The pre-paid meter obviously has been configured to meter feed into the grid as if one had fed the power off from the grid and so the only way to achieve true 0 consumption is to sever the connection completely, but obviously cloudy days happen and the connection must be very easily reconnected (even automated).

Edited by Power Me

Share this post


Link to post
Share on other sites

Feeling emboldened by your comments, Youda, I have just performed a quick basic test.
I grabbed a 2.2kW fined oil heater and stuck a 24A 400V diode in series with it.
Disconnected my inverter from the grid and ran the load without any complaints (odd sounds) or odd "DC offset" errors from the inverter.

I have no doubt an old school inverter with internal large magnet transformers will not like this kind of a load, but the modern high frequency switching inverters like the GoodWe don't seem to have any issue.

This kind of set up would provide a very simple and low cost 3 step (0%, 50%, 100%) of load control.

Share this post


Link to post
Share on other sites

Oh my! If it survived a diode without a single beep, then the SSR will be a piece of cake :D

BTW: For example Axpert offgrid inverters don't like when you consume just half of each sinewave. If you do, then they shutdown with an error message saying: "DC voltage detected on the AC output" or something like that...

2 hours ago, Power Me said:

The pre-paid meter obviously has been configured to meter feed into the grid as if one had fed the power off from the grid and so the only way to achieve true 0 consumption is to sever the connection completely, but obviously cloudy days happen and the connection must be very easily reconnected (even automated).

Well, if you can run your whole house on the backup side of GoodWe, then adding such an automation based on the SOC would be pretty easy task. Trouble is, if you need more than 4.6kW, as GoodWe inverters can't be connected in parallel on the backup side. And if you auto-connect the grid to assist the inverter, then there could be a short export spike once the huge load will be disconnected (for example an oven with a thermostat). If you don't care about such short export spikes, then it's okay. But if you'd like to eliminate them, it would need a bit more sophisticated solution, I suppose.

Share this post


Link to post
Share on other sites
Quote

BTW: For example Axpert offgrid inverters don't like when you consume just half of each sinewave. If you do, then they shutdown with an error message saying: "DC voltage detected on the AC output" or something like that...

That's exactly the kind of error I was concerned might be thrown up by the inverter. Haha
Do you perhaps know if the Axpert inverters use iron core magnetics in their inverter topology?

Quote

Well, if you can run your whole house on the backup side of GoodWe, then adding such an automation based on the SOC would be pretty easy task. Trouble is, if you need more than 4.6kW, as GoodWe inverters can't be connected in parallel on the backup side.

I generally can. I use "Home Assistant" for my home automation and currently use SonOff devices to make sure I don't run either of 2 geysers or pool pump or bore hole pump at the same time. I also use inverter aircons and fridges so no large power spike from those items. See attached typical load profile for a 24 hour period. One can clearly see the blocks of geyser consumption which I want to smooth out.
I have heard rumors that GoodWe are working on a FW upgrade that will cater for inverter paralleling in the future, but that means nothing until they actually do release a working update...

I have in the passed run the oven off grid during substation/cable fault extend power outages, but to make life easy I would probably automate connecting to grid for an hour during cooking in the evening, to allow for the odd short duration where the microwave runs while a pot on the stove boils and a chicken roasts in the oven.

Annotation 2020-02-23 184056.png

Share this post


Link to post
Share on other sites

Another option.

I have a single Axpert 5KVA with 4 Pylontech batteries and 12 x 300w panels. I use the ICC software and also Home assistant. HA/Nodered reads the battery SoC , and if it's 100% it enables a Sonoff relay/contacter to turn the geyser (3KW) on. It then leaves the geyser on until the SoC drops to 95% and then turns the geyser off.

It typically does +-30min cycles till the water is hot.

It makes the electronics a lot simpler than using a SSR.

image.thumb.png.da1841e9b18ae1d9dbd9b24275f0b7e5.png

Edited by 1Hallux

Share this post


Link to post
Share on other sites

Hi @1Hallux

I had the similar ON/OFF solution some time ago, when I started flirting with the solar power. It worked exactly as you described, but I was a bit scared by all the cycles that it introduced in my lithium battery. Therefore, I switched to the SSR and continuous regulation, so I can have a good sleep at night :)

Share this post


Link to post
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.


Topics

Contact Us - Power Forum South Africa

×
×
  • Create New...