ghatikar
-
Posts
94 -
Joined
-
Last visited
Content Type
Profiles
Forums
Gallery
Blogs
Events
Store
Downloads
Posts posted by ghatikar
-
-
On 2018/07/06 at 11:57 PM, Coulomb said:
I've often wondered about that. Yes, there is no transfer time, but there may be a very similar effect. When the grid goes down, depending on where the disconnection happens, there could be a hundred houses of load at the AC input connection. So the AC input, which is initially still connected to the AC output, is close to a short circuit. So the output will go to zero volts until the inverter can disconnect the AC input. Disconnecting a relay might be a little faster than connecting, but it will still be of the order of 10 ms, I would think. But now, different to the non-hybrid setup, the inverter is working into a short circuit for that time, and the input relay opens under that overload situation. That has to be hard on those components. Meanwhile the load sees a similar drop in voltage.
Have I got that wrong?
Despite asking repeatedly Victron has not satisfactorily replied to my queries on duty cycle of those contacts. I suspect the 100 Amp is nominal which means they may be rated different for different types of loads
https://en.wikipedia.org/wiki/Utilization_categories
e.g. For example, the Schneider LC1D18xx contactor is designed to control motors with a nominal current rating of 18 A (AC-3).
This contractor is capable of withstanding current peaks (~ 100 A) on starting, such as are typical on inductive machines like motors, without sustaining any damage.
It can also control resistance heaters for current ratings up to 32 A (AC-1). This load is resistive, hence there are no current peaks -
24 minutes ago, plonkster said:
There is also something I realised that really should be said about hybrid inverters, and that is that the (less than) 20ms transfer time really doesn't apply. A Multi running grid-parallel has the transfer switch closed. When the power fails, all it has to do is open the transfer switch (to disconnect the grid side), the output is already connected to the inverter and there is no transfer time at all. So another reason why manufacturers might not be going the online-UPS route... is because they are way past it already.
It all boils down to price and application. e.g. Nigeria where grid-tied is not really a viable option due to missing grid regulation and infrastructure.
They are targeting mission-critical applications primarily like ATMs, digital signage where efficiency is not a real concern. Imagine if you have 70 % saving in deployment cost per ATM !!
The other battery-less 5KW inverter is what piques my interest. the biggest advantage is you can use crappy /secondhand battries for applications like airconditioner loads without bothering about AH and C10
Although I suspect even the Victron will keep on working after disconnecting the battery but I dont want to try it with 2000 Euro ineverter
-
On 2018/06/12 at 12:50 PM, pilotfish said:
This Inverter looks awesome, I WANT!
Had a quick glance at the user guide, 5kVA version adds the following;
- 5000VA / 5000W inverter
- 4000W 500V MPPT
- BMS comms for LiFe04
- 61V max charge voltage for the Royal's
- battery EQ function
- built in bluetooth with android app,
- detachable LCD control for remote up to 20m
- and for the battery-phobes.......
- you can run grid/PV only with NO Batteries
I think @plonkster got a tough job on his hands now
Guys I have a bad news.( and also good news)
I contacted MPPSolar who sell the models as retail . the BMS part will not be ready till next year. and the initial release will be only with OEM orders ( where battery and inverter is sold as bundle)
But in General
PIS - 5048GK - (Voltonics Axpert VM III) 5KW/48VDC, 80A MPPT, max 500V PV open circuit with all the above features except BMS
PIP- 3048 MK - ( voltronics King) - Zero changeover time ( PV input 145 V - ONLY 3KW model ) others features same ( no BMS yet)
If I look at the graphic of the display I guess they are using their knowledge of online inverters and applying to the inverter to achieve zero transfer time - essentialy using DC rectifier and then inverter as two-stage process. it is not the ongrid inverter ( maybe grid interactive is a better term ?)
Maybe that's why only 3KW model is available as they may not be able to handle the components and heat losses in the same chassis of 5KW
Should be shortly available on Aliexpress on their official store - https://www.aliexpress.com/store/629424
https://www.mppsolar.com/v3/pip-gk-series/
-
On 2018/06/25 at 9:02 AM, Richwost said:
Greetings...
This has absolutely nothing to do with solar. I need a very simple UPS system to run my LED TV, the Explorer and one led light. No more than that. I have a 20kva Generator, but If I am watching a Test and the power goes off, I can switch to the generator - either automatically or manually (which I far prefer) - but I don't have to tell you how disruptive that is. If I am out and recording the Boks, then I come back to fine absolutely nothing.
Now for the rub. Those of you who know me from previous correspondence on the Forum (more than a year ago - hell how time flies) know that I am passionate about Lithium batteries. I bought a 50Ah Lithium battery about 6 months as well as a special C-Tek lithium charger. The battery is about half the physical size of a car battery. I attach specs.
This battery cost me R10k and I bought it for a certain purpose which I've now decided against. So I have a brand new Lithium battery being kept perfect by the Lithium charger for the UPS.
Where do I go from here? I would really appreciate your help...
I am using a similar solution for my NAS / router/Network switch for quite a while now except I use a Victron Filax changeover which is fast ( https://www.victronenergy.com/transfer-switches/filax-2)
a 20 AH Lifepo4 with cheap Charge controller - https://www.ebay.co.uk/itm/6v-12v-Lead-Acid-3-6v-7-2v-NiMH-6-4v-12-LiFePO4-Li-Ion-solar-charge-controller/292584753349 with a 20 Watt panel
and a generally good quality inverter with minimum fan noise - https://www.ebay.co.uk/p/Inverter-Power-Bestek-300w-DC-12v-to-230v-AC-Car-Converter-Electronic-Dual-USB/5017022038
why the wired combo ?- most of the components were left over from my earlier projects except the charge controller
and I needed something to keep my internet/server going when I am playing around with various assistant setups in the Victron
-
Thanks, Guys,.
I contacted REC and they Confirmed the 16Q model is fully compatible with Victron over CABUS out of the BOX . The BMS on spec sheet looks wonderful but is rather expensive 400 Euro plus for basic ( software, cables, contactor, adding temp probe etc) take you well north of 900 Euro
Maybe I will get one of those cheap Chinese versions, USE the ATC/ATD contacts and keep the battery isolated with sandbags
-
Which combination to use with 48 V , 200 AH Lifepo4 cells ( keep in mind battery was less than 3000 Euro)
Following options
Option 1 - Costliest most reliable Victron Multi 5KVA /48 - with REC BMS ( cost with Contactor, software etc almost 900 Euro)
Option 2 - Axpert 5KVA - Coulomb's firmware patched - Which BMS ??? have not found any third party which will work with Axpert yet via serial port
Option 3 - use a generic BMS to switch off the battery on extreme low/high voltage with the contractor and use additional BMV 702 relay to switch off Axpert load through a power contactor on low SOC say 20 % ( my loads are backed by the generator in office so the soc rarely drops below 70 %)
The BMS I have in mind ( also works with victron with two wire assistant)- https://www.ev-power.eu/Battery-Management/BMS123-Smart/123-Smart-BMS-Complete-Set-4-cells-with-Bluetooth-4-0.html
http://gwl-power.tumblr.com/tagged/BMS123Smart
========================
I recently got hold of a cheap Chinese Lifepo4 battery with BMS - The BMS has allowed to charge and allowed to discharge contacts
I found out that the BMS has no equalization !! ( active or passive) and reported Current and voltages my flukes did not agree to at all
closer inspection Revealed why it was cheap - The BMS looks to be fried although batteries report almost similar voltages
-
1 hour ago, plonkster said:
The first is that the GTI must preferably support GFPR, grid frequency dependent power reduction, and then you must install the relevant assistant on the Quattro. When in off-grid mode, the Quattro can then shift the frequency to signal the GTI to back off production. That way you don't need the limiter CT. -
1GFPR specially tested one like Fronius is expensive and I am sure the Chinese have no product with that. it does disconnect at 51 HZ So my Generator logic of SOC based contractor does that - the GTI only gets connected if the SOC is below 95% . or there is enough load on output ( was 100 watt for 500-watt GTI for 1000 watt GTI I can even set 500-watt minimum load needed before GTI kicks in
I agree with the second caveat - but a man can dream
to prevent battery voltI have a contactor disconnecting GTI ( SOC above 99 %) plus AC OUT2 disconnects at 57V as a backup. ( can be set lower but then AC out2 flip flops during normal charge cycles )
-
I was wondering if my thinking is correct or I may end up with a big paperweight on my hand.
Recently I came across this product - https://www.aliexpress.com/item/1000W-Battery-Discharge-Power-Mode-MPPT-Solar-Grid-Tie-Inverter-with-Limiter-Sensor-DC22-65V-45/32762489820.html
It is essentially a 1000W Grid-tie inverter with built-in export current limiter which can work with lead acid battery bank instead of PV panels as DC input
I was thinking to use it on the output of 5 KW Quattro with Current limiter CT inserted between Grid-tie and Quattro ( see crude sketch attached)
So in normal working, the Grid tie will supply the house power essentially stoping Quattro from discharging its battery( BYD lithium in my case) or drawing from the grid . ( using islanded mode) The grid tie inverter can be set to stop working on any voltage setpoint so as an example set it to use lead acid upto 48V or even insert a power contactor driven by BMV to disconnect SOC below 50 %)
The grid tie maker claims 3 second reaction time on the power control ( from Full 1000 watt to Zero ) .
If the combination work can give many advantages
1. Use Large /OLD lead acid bank fed from solar charge controller - remotely located ( SAY in a different shade 50 meters away) and tie to the main inverter
2. Reuse the old lead acid bank instead of discarding when changing to lithium using an el Cheapo charge controller
3. tie two different banks together of different battery types /chemistries
4. Increase the total load capacity e.g. 5KW Victron can handle 6 KW load at 40 DEG ambient with one 2000 watt grid tie attached for the temporary peaks of washing machines or vacuum cleaners
I am aware that the power conversions from DC to AC do not make it the most efficient systems
In real life case
My lead acid of 400 AH is being fed from 5.5 KW solar panels through 2 No.s of Axperts ( total 10 KVA) and gets used only on weekends, public holidays as it supplies only airconditioning and washing machines, water heating etc. So essentially the 5.5. KW panels can feed the Quatro through the grid tie but also keep the lead acid bank topped up for nighttime AC operation in case of power failures. The Axeprets are set to charge bank only using AC and set at default solar mode so use grid only in the night
T
I have successfully setup a APS systems microgrid tie to work with Quattro AC/OUT 2 for my office. as the only power required is during weekends to keep batteries charged when generators are not running and the grid is not available - see attached logic and scheme ( I am on ground floor of 4 story building and the 4 mm2 cables would have been costlier than the whole setup)
I am using PV inverter assistant and generator Logic to disconnect PV inverter when the grid is available in addition Quattro ACout will switch off using the assistant with battery voltage higher than 57V to avoid feedback into the main battery - see similar scheme in operation in my office ( using panels on 500-watt grid tie inverter)
-
15 hours ago, flamegrilled said:
Under Battery control tick or no tick . It does not display.
Thanks Plonkster : screen /dev/ttyUSB0 19200 works.Get the data streaming
Should be ticked. and BMV should say thread running with time indication
-
-
The meter issue was partially resolved - We narrowed it down to cable or the USB /RS485 converter - The whole USB bus shuts down after power is restored to ET112 . The ET340 does not exhibit same behaviour
I will replace the cables with CAT6 ( using telephone cable) over weekend and see if it goes away or alternatively only plug Et112 to a separate venus on PI and check
Izak tried alternative drivers etc before concluding it is a hardware issue
steps until now
--------------
. Both meters are plugged in directly to venus2. The problem is only with ET1123. even if the power of ET112 is removed the ET340 continues to function4. If the power comes back within few seconds the ET 112 returns and function normally5. If the power to ET112 restored after a delay of 1-2 minutes the ET340 also disappears along with ET1126. Both USB ports actually lose their power ( confirmed with USB doctor) and plugging in FAT32 8 GB drive7. It does not matter which USB port of Venus the EM112 is plugged in8. Both cables are extended by the same quality twisted pair Telephone cable9. EM112 is closer than EM340- approximately 5 meter away while EM340 is approx 15 meters awayafter reboot, the power to USB port returns and both meters or the meters function normallySome videos ( sorry for the crappy quality )I can try to switch the USB (RS485) cables to localize if the problem is cable or EM meterHowever, Can do that over weekend onlyIf you want any other feedback or changes Let me know1. General system - https://youtu.be/wTT5WC2PNnA2. Short power interruption meter works normal - https://youtu.be/wHcPGJjt6As3. Long power interruption - Both meters vanish - https://youtu.be/xw4Ee0bFgKw4. general status of the system when USB bus dead but Em112 has power - https://youtu.be/utBQC6aV1vc -
18 minutes ago, plonkster said:
You sure? I can't find where that was removed. Testing on the latest gui version indicates the password is stil ZZZ.
You making any progress with your meter issue? :-P
I never needed to use the password as far as I remember. May be it has to be locked first before you need password . - The assitant password was removed I guess
-
Mainly directed at @plonkster
With the distributed voltage and current control enabled in my Venus GX with 2.12.17 I am seeing a strange See Saw pattern of the battery terminal voltage - probably because BYD has some hysteresis built in for charge and do not charge values and the Venus + Quattro internal circuits slowly drains the BYD dopping terminal voltage over 3-4 hours from 57 V to 54.8 before the charger switches on briefly and brings it back to 57 V and stops charging
ON previous logic with Quattro in control the battery terminal voltage remained on 55.2Any method using dbus to monitor the Allow to charge contacts status to find out if the voltage is being controlled by contact or by Distributed voltage /Current logic needs a short name ( DVCC?)I have turned it off and gone back to conventional Hardwire1. Distrubuted voltage and current control - See Saw ./Sawtooth pattern Battery terminal voltage2. Before the upgrade stable voltage3. 7 days history -
The password is only needed on all firmwares. I think on 2.1.xxx it is removed.
-
1 hour ago, Chris Hobson said:
In an off-grid situation you could easily exceed the 25A limit of a single unit. 4 Pylontechs however could supply 100A an under those circumstances I think whether the system is off-grid or not is irrelevant. According to the Victron forum Pylontech are busy testing in off-grid situations.
BYD recommends 3 x 2.5 KWH for a 5 KW solution. my 10 KWH ( 4x 2.5 ) 48 V system regularly reports 140 Amps charge-discharge limits at 30 DEG .
This charge limit drops rapidly when the temperatures are lower than 12 DEG C
-
@plonksterThank you, your link put me on the correct path. I found a CANBUS whitepaper ( backtracked from MODBUS .XLS)
Canbus seems to be preferred choice for lithium vendors. I noticed from today's email from Bluenova that they went from NO communication to MODBUS to CANBUS
Hopefully one day it will be industry standard and you can plug in any battery to any inverter
-
On 10/11/2017 at 9:25 PM, plonkster said:
All you need to do is take those values and write them to /Bms/AllowToDischarge and /Bms/AllowToCharge on the vebus service on dbus. This is exactly what the redflow BMS does (except it uses modbus tcp). Writing those flags will cause the inverter to go into charger/bypass mode.
Is there a communication white paper apart from thishttps://www.victronenergy.com/upload/documents/Whitepaper-Data-communication-with-Victron-Energy-products_EN.pdf)( where it explains which Dbus values to address
IU have attached The Herewin Canbus protocol for another inverter ( i think Goodwe ) . They need spoon feeding on what registers to use and what canbus address as they could not figure out on their own. and Lithium batteries is not something I am willing to experiment my limited programming skills on
Otherwise, I am planning to do the scheme hardwired " allowed to charge and allowed to discharge contacts to multi ( Lynx Ion assistant)" with battery protector as additional safety ( remote on off from Battery BMS) and programme 0 ( Low voltage cut off at 42 V)
The herewin Battery has Pretty wide tolerance ( discharge 40 V and charge 58.4 Volts ) making it almost drop in for lead Acid applications
Manual-Connecting-other-lithium-battery-systems-to-Multis-and-Quattros-EN.pdf
Whitepaper-Data-communication-with-Victron-Energy-products_EN.pdf
48V 210AH(10.08kWh) datasheet for ref..pdf
Manual-Battery-Protect-48V-100-A-EN-NL-FR-SE-DE-PT-ES-IT.pdf
-
On 09/11/2017 at 12:52 PM, plonkster said:
There is a long history behind this. I will dig into this one of these days, but at this moment I don't know the full details. The protocol was originally designed by some German company that I cannot remember the name of, but it sort-of became a de facto standard (ie in practice, not so much by decree). It uses the CAN-bus. Batteries like LG Chem and similar all speak this protocol already, and so does our locally made FreedomWon batteries. I believe you can just ask Victron for the document and they'll give it to you.
T
@plonkster I have asked Victron Both to sales@ xxxxx , docal distributer and Mattijias directly so far no results. I am trying to get canbus of another Chinese battery (compliant. The chinese did not get hold of any document from Victron so far and are willing to spend development efforts to design it
the battery http://www.herewin.com/en/ProductShow.asp?ID=148 is less than 3000 USD for 10 KWh with 2000 cycles.
I have asked herewin to put allow to charge and allow to discharge and I will use it that way with a 3 KW multi for my office
-
On 19/10/2017 at 10:05 PM, Coulomb said:
The other problem is a severe selection bias. Those who had trouble post about it, looking for solutions. Those that have no problems forget about it, and don't tend to post about it.
But yes, I'd love to know the failure rate too, as long as it was reasonably accurate.
I just got a feedback from Ipower installation engineer ( Voltronic rebrand for Nigeria) - Apperantely after 2016 the 60 and 80 amp models have a very low failure rate apprentely less than 1 in 100. But most of these units are installed in banks with large battery banks with loads at less than 30 % except when the ATM starts counting notes. THE CC part is fully loaded with a 3 KWp panel setup
-
Peeps
Latest News from Tapeta street Home Project
A more mechanically talented guy from my office helped me install ( actually he did the install I just watched � ) over two days this weekend the first wind turbine
I was also in parallel installing the BYD LIfepo4 batteries – 10 KWh 48 V
See video for turbine in operation -
https://www.youtube.com/watch?v=mI84Xa_9LA4
It is a cheap Chinese make horizontal axis wind Turbine rated 200 Watt /24 V DC connected to the drivers /maids room through an off grid setup of 3 KWH . It is also augmented by 800 watt of Solar PV
Cut in Speed 2 m/s - rated 11 m/s - Installation location - 9.078 7.482 , 492 m+ 14 meters
weather history - https://www.wunderground.com/personal-weather-station/dashboard?ID=IABUJA3#history/tgraphs/s20171106/e20171106/mdaily
I have also on order a Vertical Axis 300 Watt wind turbine to compare. Will be installed next year
Now I have Four different RE models running in the same house
1. Pure Offgrid – 24 V DC Solar PV+ Wind – AGM batteries – 3KWH. 0.8 Kwp Solar
2. DC coupled Hybrid – Grid + Solar PV + Diesel, 48 V Flooded lead acid , 20 KWh( 10 KVA peak) – 2.5 Kwp solar
3. AC coupled Grid Tie ( normally running in islanding mode daytime) - Solar PV +Diesel Hybrid + Micro Grid inverter ,Lithium batteries – 10 KWh ( 5 KW peak) , 48 V DC – 3 Kwp Solar
4. Solar thermal – 120 Litters EV tube water heater with PV Solar assisted 1.5 Kwp heater and 100 watt water boost pressure pump
-
On 01/11/2017 at 8:11 AM, Chris Hobson said:
For the 3500 versus 1600 cycles I would add one more Pylontech and now you have just under 7kWh which will carry me through the night. For me the biggest advantage is you don't have to pony up the money in one go. I liken the purchase of Pylontechs like the purchase of 105Ah lead acids. First you have one string and then your second and so on without all the problems of lead acids of different ages.
Calculate your requirements or determine your budget and buy x number of Pylontechs. If you made a mistake/ your requirements increase or your budget allows further expansion buy another. With the Axpert where there is no BMS communication between the inverter(s) and battery bank your battery bank size is unlimited. You are however restricted to stacks of four but a busbar install will allow multiple stacks. So 1,2 3,4, (3x2) and (4x2) are possible combinations.
Based on BYD and Victron combination recommendations 5 KW requires at least 3 Bplus ( each 2.5 Kwh) ( Quattro with DG backup) or 5 units B-box ( multiplus) . Although Axpert can handle 4 KW max I would say 3 units is on safer side inase you switch inverters or to carry peak loads on regular basis. With two batteries I will stick to 60 amps x 50v = max 3 KW peak
Please remember the continuous current capacity is quoted at 25 DEG C . my BYD were reporting 130 Amps Max Continous discharge current at 30 DEG C instead of the theoretical 140
see attached for BYD specs and recommendations
-
1 hour ago, PurePower said:
After doing more and more reading on these Pylontech lithiums... a few questions came to mind.
The Pylontech inverter setup manual that installers use has no mention of the LifePO4 firmware that needs to be installed on the Axpert... It just has the following settings that need to be set:
2 questions that i have:
1. Do we need to install the LifePO4 firmware? and if yes, are there any benefits?
2. What float voltage needs to be set?
The absorb on Axpert is not ajustable in normal firmware. Normally lithium with active balancing need max 1 hour absorb . and with passive balancing absorb is not needed at all
-
23 hours ago, Chris Hobson said:
For R48K you can have 7.2 kWh Pylontech Lithium
Pylontech have an Axpert compatible firmware.
Pylontech May not work ( Axepert) inOffgrid Setup unless there is large no. the batteries are heavily derated for continuous operation. Please check the datasheets I think for 5 KW you need minimum 2 or more - e.g. BYD with similar setup recommends atleast 3 units of 2.5 KWh for 5 KW off grid but only one unit of 2.5 KWH for ongrid 5 KW
-
@plonkster Thank you . Most comprehensive answers I got. I will try UK settings over the weekend
Development on Raspberry Pi
in The Internet of Things
Posted
I will be also watching the development with interest Although don't have enough time and inclination to write it myself I do have a Modbus energy meter which I would not mind monitoring to separate loads between AC out 1 and AC out 2