Rectifier Huawei R4850G2 48V 42~58V 3000w

Batteries, Chargers, and Battery Management Systems.
j bjork   100 kW

100 kW
Posts: 1335
Joined: Aug 31 2018 3:01pm
Location: Linköping, Sweden

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by j bjork » May 11 2020 11:27am

rigmo wrote:
May 09 2020 3:57pm
j bjork wrote:
Mar 19 2020 8:41am
Were you (or someone else) able to find it or something simular?
I should soon have to modify 2 of these psu:s, it would be nice to be able to adjust them without the programming.
I do not know,,, only china controller.. now I have contact to get it... I have two... no used for now...
https://drive.google.com/file/d/1tu_HWx ... q9oAj/view
I was able to buy the control boards linked earlier in the thred. I haven't got them yet, but they seem to be close now. After almost 2 months..

j bjork   100 kW

100 kW
Posts: 1335
Joined: Aug 31 2018 3:01pm
Location: Linköping, Sweden

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by j bjork » May 11 2020 11:29am

eee291 wrote:
Nov 26 2018 4:53pm
There is a Chinese guy selling the programmer boards for about 12$ on Taobao
https://2.taobao.com/item.htm?id=565948 ... oSite=main

If you are having difficulty ordering it then you can use a third party service like Taobao Focus.
They will order it for you and ship it to you.
https://buy.tbfocus.com/item.php?id=565948826876
It's in English and they accept Paypal.
These ones, I didn't realize that it was still possible to buy from there still. Thanks eee291 :)

j bjork   100 kW

100 kW
Posts: 1335
Joined: Aug 31 2018 3:01pm
Location: Linköping, Sweden

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by j bjork » May 12 2020 1:46pm

I got the control boards today.
I went looking for a power supply for the board in the psu.
I found the fan had a steady 12,7v, I thought that would be perfect.

Image

Image

Image

Well, I connected the board to the 12,7v. It seemed fine.The leds lighted up, it reacted when i clicked on the knob.
I tried connecting only can bus, seemed fine (no leds or anything, but it was not powered up)
When I connected both, bang. The fuse blew in the garage and everything. The psu died. One of the big mosfets exploded.

The control board also smell a little burnt, and there are a little discoloration on the contact housing.

But I don't really understand why? Can bus should be isolated from everything else as far as I know?

User avatar
amberwolf   100 GW

100 GW
Posts: 33727
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by amberwolf » May 12 2020 2:04pm

Most likely the sections you connected with wiring to the add-on board were isolated in the PSU, and connecting them shorted things out.

You'd have to trace out and measure on a still-working PSU whether there are already connections between those sections, and if not, don't make any.
If you found this advice helpful, supporting contributions are accepted here.

Beginning of a New Life


Give a listen to my music!

(More here, including experimental)
.

j bjork   100 kW

100 kW
Posts: 1335
Joined: Aug 31 2018 3:01pm
Location: Linköping, Sweden

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by j bjork » May 13 2020 10:50am

I wonder then if it is safe to parallel the can bus from more than one psu?
I plan to run 2 psu:s in series, and control them both with one control board over can bus.
So the can bus will be connected parallel.
From my experience with can bus it should be almost indestructible.
You can short them to each other, to + or -, just repair the the fault and it works again.
And they are usually connected in big systems with several units talking to each other.
But my experience is from cars.

User avatar
amberwolf   100 GW

100 GW
Posts: 33727
Joined: Aug 17 2009 6:43am
Location: Phoenix, AZ, USA, Earth, Sol, Local Bubble, Orion Arm, Milky Way, Local Group
Contact:

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by amberwolf » May 13 2020 1:31pm

If the bus is isolated from the other connections you have between PSUs, then it shouldn't be a problem.

If it shares the ground with the power output, then you could not put them in series and still connect it, because connecting the ground would short out the more negative PSU's output.

They could still be put in parallel, however.

If they are not isolated but you can use optoisolators on the bus, then you could connect them anyway.
If you found this advice helpful, supporting contributions are accepted here.

Beginning of a New Life


Give a listen to my music!

(More here, including experimental)
.

j bjork   100 kW

100 kW
Posts: 1335
Joined: Aug 31 2018 3:01pm
Location: Linköping, Sweden

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by j bjork » May 14 2020 11:24am

There are no connection between either positive or negative output and either of the can bus wires, so I guess it should be safe..
At the moment I am a little low on psu:s, so I cant really test :roll:
Luckily I bought extras of the control board.

It seems to work fine at adjusting the voltage output, (with an external power supply for the board) but I haven't really figured out how to set the default output(when the control board is not connected) It seems to have changed, I am pretty sure I had 54,5V before. But now I have 55,7V. Maybe it gets lower under load, but I am pretty sure it has been stable at the same voltage both with and without load before.

I am also interested in those blue and red voltage and current displays that people are using. Does anyone has a link?

rigmo   1 mW

1 mW
Posts: 10
Joined: Jan 25 2019 6:44am

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by rigmo » May 22 2020 4:12pm

j bjork wrote:
May 11 2020 11:27am
I was able to buy the control boards linked earlier in the thred. I haven't got them yet, but they seem to be close now. After almost 2 months..
2. The computer serial port debugging tools, see figure set parameter writing CAN control board (USB serial port module.)
Attachments
2019021801.jpg
2019021801.jpg (107.76 KiB) Viewed 4952 times

j bjork   100 kW

100 kW
Posts: 1335
Joined: Aug 31 2018 3:01pm
Location: Linköping, Sweden

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by j bjork » May 23 2020 3:51am

Hmm,I am not sure if I understand. Do you mean that you need to connect the control board to a computer with this usb tool to change the default settings?

rigmo   1 mW

1 mW
Posts: 10
Joined: Jan 25 2019 6:44am

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by rigmo » May 26 2020 3:14pm

IMG_2637rez.png
IMG_2637rez.png (115.71 KiB) Viewed 4870 times
HI NEWS
Display monitoring of huawei and emerson PSU.

control boards
USB serial port cable
Huawei and Emerson's control board PCBA is the same, but not universal, the software communication protocol is different.

j bjork   100 kW

100 kW
Posts: 1335
Joined: Aug 31 2018 3:01pm
Location: Linköping, Sweden

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by j bjork » Jun 04 2020 10:50am

rigmo wrote:
May 26 2020 3:14pm

Display monitoring of huawei and emerson PSU.

control boards
USB serial port cable
Huawei and Emerson's control board PCBA is the same, but not universal, the software communication protocol is different.
Have you been able to set the default output of the psu?
My has changed to 55,7V after I have connected the control board.
I don't have the parts to connect it to the computer.

hyphenated   1 µW

1 µW
Posts: 4
Joined: Jan 17 2020 9:05pm

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by hyphenated » Jun 05 2020 8:57pm

BotoXbz wrote:
Apr 01 2019 1:21pm
A little update on my PSU hacking attempts:
{snip}
Work in progress still, code on GitHub https://github.com/BotoX/huawei-r48xx-esp32
Using this lib for CAN on ESP32 https://github.com/sandeepmistry/arduino-CAN
I want to use the Huawei as a jump-start power supply for a 20kW flow-battery installation using Redflow ZBM-2 zinc-bromide units. I've got them in place, but COVID-19 has sabotaged the commissioning timeline. One of the peculiarities of a flow battery is that they can be (and in fact need to be for dendrite removal) fully discharged. From that state they need a 56V power supply at high amperage to get them to the point that they can be run normally, and I wanted to set up a Huawei rectifier on the DC bus to kick in if/when required under BMS control - whilst it is possible that the Sunny Island can do this, that is not yet clear.

Because a rectifier is inherently designed to operate in a n+1 environment with high reliability and life, it seemed ideal. But of course the controls are proprietary. I decided to build a CANBus unit in the hope that I could set the voltage and current limit permanently with a view to using the rectifier under BMS control using dry contacts.

Originally I purchased an Arduino UNO and CANBus shield plus a couple of MCP2515 cards (cheap as chips) but actually implementing a CANBus tool from the tools and code out there turned out to be a lot more difficult than I thought. Plenty of mickey-mouse demos which will turn an LED on on a receiver when a transmitter is actuated, but functional transceivers either relied on obsolete and withdrawn software, or were firmly oriented to vehicles, and as this appears to be a lucrative market, pried accordingly.

As I also had a an ESPDuino clone I was seduced by BotoXbz's solution. So here are a few errors I made en route: Note that mose ESP32 boards are 3.3V, and do NOT require a CANBus chip - just a driver. I ended up cannibalising an MCP2515 (which is 5V) by stripping the CANBus chip and strapping to the driver. Messy, but cheaper - plus it saved my waiting three weeks for a part.
  • Install PlatformIO and let it pull in updates.
    Import the huawei-r48xx-esp32 GIT
    Edit main.cpp to insert your WiFi SSID and passphrase
    Edit platformio.ini to add monitor_speed=115200

    The route I took is slightly different - an ESPDuino is an Arduino card with an ESP32 and runs off 5V.
    GPIO4 is CRX, GPIO5 is CTX, although different pins can be programmed using CAN.setPins(rx, tx)
    If you do cannibalise an MCP2515 I suggest you remove the crystal first, then cut off the pins on the side nearest the crystal as close to the chip as possible to avoid tearing the tracks.then bend the chip up (trying to bend the pins without ripping the pads off) sufficiently to cut the other side.
    Now desolder each pin from the pad.
    Now strap RXD and CXD to the input pin of your choice. Lizst wire might be good, I used telephone single-strand, which gave me issues with melted insulation on the bends.
    There is a 120 Ohm resistor in the MCP2515 - you need to strap J1 - I then added a terminating 120 Ohm in the cable at the Huawei. You might get away without, but it should eliminate reflections in the cable.
When I first looked at BotoXbz's code I despaired - how did the bits fit into an Arduino code space? The answer is that they don't. I installed PlatformIO, whixch is "Open source, cross-platform IDE and Unified Debugger. Static Code Analyzer and Remote Unit Testing. Multi-platform and Multi-architecture Build System", but the big-ticket items are that it allows the use of multiple types of card, allows a GIT source to be utilised as-is, and provides all the tools you need - for free. I cannot emphasise how gob-smacking it is to use an open-source tool like this - when I was in my first job (in the seventies) I was working with UV-erasable eight-bit processors for a PABX line card, and the hardware and software tools needed were pricey, clumsy and proprietary. This is my first time playing in the software pond since that design (by the way - a lot of engineers thought it insane to have an eight port line card design with nine microprocessors 8) ) and it's a whole new ball-game.

Suffice it to say (and to my complete amazement) the GIT transferred smoothly, the code assembled and transferred, and after a hiccup with swapped CAN-L and CAN-H I was able to talk to the Huawei.
IMG_5381 - resize.JPG
IMG_5381 - resize.JPG (102.03 KiB) Viewed 4758 times
I used a connector, rather than soldering to the edge of the PCB.
IMG_5384 - resize.JPG
IMG_5384 - resize.JPG (91.18 KiB) Viewed 4758 times

hyphenated   1 µW

1 µW
Posts: 4
Joined: Jan 17 2020 9:05pm

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by hyphenated » Jun 05 2020 9:41pm

j bjork wrote:
May 14 2020 11:24am
There are no connection between either positive or negative output and either of the can bus wires, so I guess it should be safe..
There is a 5kV isolator in the Huawei, so multiple CANbus connections can be made in parallel - it's a differential transceiver, so neither leg is earthed. You should aim to have two 120 Ohm resistors in parallel with the bus to suppress reflected signal interference; one at your transceiver, the second at the furthest Rectifier. Keep stubs short; ideally just tap into the bus at each rectifier.

j bjork   100 kW

100 kW
Posts: 1335
Joined: Aug 31 2018 3:01pm
Location: Linköping, Sweden

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by j bjork » Jun 06 2020 9:50am

hyphenated wrote:
Jun 05 2020 9:41pm
j bjork wrote:
May 14 2020 11:24am
There are no connection between either positive or negative output and either of the can bus wires, so I guess it should be safe..
There is a 5kV isolator in the Huawei, so multiple CANbus connections can be made in parallel - it's a differential transceiver, so neither leg is earthed. You should aim to have two 120 Ohm resistors in parallel with the bus to suppress reflected signal interference; one at your transceiver, the second at the furthest Rectifier. Keep stubs short; ideally just tap into the bus at each rectifier.
Thanks, the control board already has 60 Ohm built in. Should I use the same on one of the psu:s, or still go with 120 Ohm at that end?

hyphenated   1 µW

1 µW
Posts: 4
Joined: Jan 17 2020 9:05pm

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by hyphenated » Jun 06 2020 5:05pm

It sounds as though he has intended the controller for use with a single rectifier, which means that the leads are very short (as per photos), which means that he could get away with consolidating the two parallel 120 ohm resistors in a single 60 ohm.

If you are using multiple rectifiers then there needs to be a variation in the code to select and communicate with a specific rectifier - I can't imagine that they will all respond to commands together (in fact if they did, they would probably jam the bus). So multiple connection will probably require either a modified unit or a different controller. I would think that the rectifiers will recognise that there are others on the bus, and change their address on startup, but what I don't know about CAN protocols would fill an encyclopedia.

Short answer - if you fix the control issue and have longer runs, you will probably need to replace the 60 ohm with a 120 ohm and add a terminating 120 ohm at the end of the cable.

rigmo   1 mW

1 mW
Posts: 10
Joined: Jan 25 2019 6:44am

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by rigmo » Jun 09 2020 6:02pm

j bjork wrote:
Jun 04 2020 10:50am
rigmo wrote:
May 26 2020 3:14pm

Display monitoring of huawei and emerson PSU.

control boards
USB serial port cable
Huawei and Emerson's control board PCBA is the same, but not universal, the software communication protocol is different.
Have you been able to set the default output of the psu?
My has changed to 55,7V after I have connected the control board.
I don't have the parts to connect it to the computer.
yes can be made default required voltage pcb control board ...

User avatar
pavlik1   10 W

10 W
Posts: 90
Joined: Aug 02 2010 2:31am
Location: Switzerland

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by pavlik1 » Aug 14 2020 5:57am

Hello rigmo

How exactly you change the default voltage? Which tool to use?

Thank you
rigmo wrote:
May 22 2020 4:12pm
j bjork wrote:
May 11 2020 11:27am
I was able to buy the control boards linked earlier in the thred. I haven't got them yet, but they seem to be close now. After almost 2 months..
2. The computer serial port debugging tools, see figure set parameter writing CAN control board (USB serial port module.)
1st radiator fan motors, 20km/h stolen after 2 days
2nd 250W-24V eZiplike motor with gear. 36V, 8Ah, peak 1000W, 40km/h
3rd GM 48V 1000W, 16S 20Ah WinForce Lipo. speed over 72km/h - LiPo FIRE
4th Recumbent HS3540 2X16s BMSbattery 48V 20Ah
5th Haibike Attack FX, e-novation M15 13S LiIon
6th 26'' with e-novation M15 10S, VESC and throttle
7th City-El 16S WBLYP100AHA TTL140B

szczur333   10 µW

10 µW
Posts: 6
Joined: Jun 15 2020 12:02pm

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by szczur333 » Sep 09 2020 9:42pm

is it possible to charge 20S battery pack and adjust current (depend of current needs?)


now i have GTK 72V (84V) 30A charger but its very heavy and big, also it doesnt have any adjustments
(this gtk 72V 30A gives only 24-25A real current, it consume 2070-2120W from wall

so if I can achieve 2500-2800W which will be about 30-33A real - this will be my best solution for my ebike, and with current adjust I will charge slower (10A on hot days or at nights, when I am not in rush)

frank_zero   1 mW

1 mW
Posts: 13
Joined: Sep 27 2020 3:40am

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by frank_zero » Sep 27 2020 3:54am

rigmo wrote:
May 22 2020 4:12pm
j bjork wrote:
May 11 2020 11:27am
I was able to buy the control boards linked earlier in the thred. I haven't got them yet, but they seem to be close now. After almost 2 months..
2. The computer serial port debugging tools, see figure set parameter writing CAN control board (USB serial port module.)
Hello Rigmo are you maybe selling any of those can bus controllers ?
Hello all sorry but i m new here i also have to get used to this kind of forum , search and answer systeem.

frank_zero   1 mW

1 mW
Posts: 13
Joined: Sep 27 2020 3:40am

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by frank_zero » Sep 27 2020 4:07am

I found out a easyier way to start the Rectifier Huawei R4850G2 48V 42~58V 3000w when you use the connector for the wiring
and you take small the start plug, the 2 wires of the start plug you just have to put both to the DC - and the Rectifier Huawei R4850G2 48V 42~58V 3000w wil start.
Does any one know what model /type for small connectors it are ?
Last edited by frank_zero on Sep 27 2020 2:00pm, edited 1 time in total.

salavati05   1 µW

1 µW
Posts: 3
Joined: Sep 27 2020 8:01am

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by salavati05 » Sep 27 2020 8:18am

rigmo wrote:
Feb 27 2019 5:28pm
I have direct contact of PCB controller builder from china.. 任星星
hi rigmo
you can help me ?
do you have cod canBUS for R4850G2.

salavati05   1 µW

1 µW
Posts: 3
Joined: Sep 27 2020 8:01am

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by salavati05 » Sep 27 2020 8:23am

rigmo wrote:
Feb 27 2019 5:28pm
I have direct contact of PCB controller builder from china.. 任星星
HI rigmo
i have controller pmu 11a
but i can't connect to controller for change voltage
do you have WhatsApp or telegram for connect
Last edited by salavati05 on Sep 27 2020 8:40am, edited 1 time in total.

john61ct   100 GW

100 GW
Posts: 8441
Joined: Dec 18 2018 2:06pm

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by john61ct » Sep 27 2020 8:36am

It is in cool weather that C-rate needs to be lowered for pack longevity

Hot weather enables fast charging with less damage

frank_zero   1 mW

1 mW
Posts: 13
Joined: Sep 27 2020 3:40am

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by frank_zero » Sep 27 2020 2:17pm

Has any of you put 2 Rectifier Huawei R4850G2 48V 42~58V 3000w in serie ?
can you also mont them standing up or down ? instaid of laying down ?

salavati05   1 µW

1 µW
Posts: 3
Joined: Sep 27 2020 8:01am

Re: Rectifier Huawei R4850G2 48V 42~58V 3000w

Post by salavati05 » Sep 28 2020 5:06am

rigmo wrote:
May 26 2020 3:14pm
IMG_2637rez.pngHI NEWS
Display monitoring of huawei and emerson PSU.

control boards
USB serial port cable
Huawei and Emerson's control board PCBA is the same, but not universal, the software communication protocol is different.
hi rigmo
do you have this control boards?
i need this
please help me

Post Reply