FU-Dyson-BMS - An (Unofficial) Firmware Upgrade for Dyson V6/V7 Vacuum BMS

[duzycinek]

Great news, waiting for flashing method since got no Atmel-Ice

[tinfever]

Hi!

Having been inspired by tinfever's work on this series of battery pack, I've started a project to do a similar replacement firmware for the v10 battery pack series.

I shall create a separate thread, because this one is to celebrate their work, not mine.   But assuming nobody minds, I'll post a link to my github project here:

https://github.com/davidmpye/V10_Dyson_BMS/

TL;DR - mostly done now, does work. Main downside - need to figure out a better way of flashing it that doesnt need the Atmel-ICE programmer.  If anyone fancies coming up with a way to get a Pi to do it with OpenOCD then we'd be on to a winner....

David

That's awesome!

It looks like the PICkit 4 supports the ATSAMD20E15 so that would probably work too and maybe be a little cheaper?

[rulof]

Hi!

Having been inspired by tinfever's work on this series of battery pack, I've started a project to do a similar replacement firmware for the v10 battery pack series.

I shall create a separate thread, because this one is to celebrate their work, not mine.   But assuming nobody minds, I'll post a link to my github project here:

https://github.com/davidmpye/V10_Dyson_BMS/

TL;DR - mostly done now, does work. Main downside - need to figure out a better way of flashing it that doesnt need the Atmel-ICE programmer.  If anyone fancies coming up with a way to get a Pi to do it with OpenOCD then we'd be on to a winner....

David

That's awesome!

It looks like the PICkit 4 supports the ATSAMD20E15 so that would probably work too and maybe be a little cheaper?

Great news!!! I have a J-link.  he's definitely sewing.
I am ready to help in any way I can. I have serviceable and faulty batteries.

[davidmpye]

Hi tinfever,

Quite possibly - anything that supports SWD should be able to do it (though I'm not sure exactly what devices Microchip studio will talk to).

Good news though - I've proved that it *CAN* be done with a Raspberry Pi and its' GPIO pins and OpenOCD, so hopefully can produce a turnkey solution (eg SD card image) that should allow a Pi to unlock and reflash the pack without an more specialised equipment :-)

I'll create a separate EEVblog thread for it once I've finished the documentation - probably about a month til I'm ready 

David

[davidmpye]

I hope nobody minds, but to keep threads separate, I've posted a V10 thread for discussion/work on my V10 firmware :-)

It is here: https://www.eevblog.com/forum/projects/replacement-firmware-for-dyson-v10-batteries/

Best wishes,

David

[NeilP]

Following and will try as soon as my PICKit arrives

[mrjoda]

Hello, glad that I found this thread - Im trying to save my Dyson.

I have V8. It stopped working - dead battery after 5y of use.

I took it apart and one cell vas 3,8V instead of 4,2V. Ok then lets change the cell. MF! Dyson use some 20700 cells 3000mAh 10C (as far I could find info),

I had some new Sony VCT6 18650 cell so I soldered it as a replacement of the 20700 with little bit of 3M tape help. Dirty job but whatever.

However... It does not work suprisingly. The same problem. After 1 minute runtime indicator shows a flat battery. It seems to charge it somehow.

Im not sure how to continue.

I will try to replace the firmware in BMS. Its unusable so I can not brick it more    

[NeilP]

One of my boards here is a 188002, I think from a V6.  SV03

Picket 3 arrived and done what appears to be successful flash of Firmware of the V1 hex from may 2022..from GitHub. 

OK, so i did not follow instructions and balance the pack..I was too keen to see if i could make the thing flash, plus my balance kit is down the other workshop no in my living room

Do no believe this pack has a reed switch for a magnet..but maybe a Solid State device?  If so where on the board is it?  I simply ask because, the only reference to a solid blue I can find ..WHEN THE PACK IS NOT ON CHARGE is on the second page of this thread.

So i got the three colour lights to show it has your firmware.

Plug in to charger and get what appears to be a white flash followed by 8 yellow...the Balance indicator  for 50mV per flash.
So yet need to balance..


BUT ...once disconnected from charger and from the machine ...the Blue LED stays on ..constantly ..have left it 5 minutes now and it is still on


Cannot see any test telling me what SOLID BLUE when not connected means.  I have waved a few different old HDD magnets around the board with sheet of paper over the top to preventshoerign and no change in solid blue.



Yes, I'll balance the pack later and try charging again ..jsut thought i'd post this question now whilst I have moment.


Thank you

[NeilP]

AH ..the switch ..needs to be pressed!  I had it loose from the housing, so it was acting as if the grey plunger was permanently pushed.

Did basic balance with a light bulb  only 6 yellow flashes  now then blue charge light...seems to be all good now..jsut get it balanced

When pressed and release..4 green flashes, then constant green for a few seconds then it goes off. 

Got the Battery Medic balancing the pack now.

[tinfever]

However... It does not work suprisingly. The same problem. After 1 minute runtime indicator shows a flat battery. It seems to charge it somehow.   

After it shows the low voltage (flat battery) warning, I would see which cell is low voltage. One of them has to have hit the cutout, either due to lower capacity or higher ESR x vacuum current draw. Also, Since the cells in series in the battery pack, replacing one cell with the smaller 18650 might cause balance issues.

AH ..the switch ..needs to be pressed!  I had it loose from the housing, so it was acting as if the grey plunger was permanently pushed.

I'm glad you got it working. There is a magnetic reed switch on the SV11 batteries for V7 vacuums, but not for that PCB you have. If you look at the git repo, you can see it as a long black rectangle part in the photo attached.

[gold oxide]

Just put your firmware on a friend's V6 SV3 battery along with new cells. Works great.

[Lubiezny_kran]

First i want to say Hi as this is my first post.
I want ot thank u all for helping me to resurecte my battery pac.
 I curious if u herd of som one reverse engenering motor controlers for this vacs specificli for V8, as i got a heng of one and want to make fume extractor of of it.
And main cuestion is speed controll as this bastards ar plenty laud for that aplication, and i want ot taim them down.
 Sory for the gramar tho

[tinfever]

First i want to say Hi as this is my first post.
I want ot thank u all for helping me to resurecte my battery pac.
 I curious if u herd of som one reverse engenering motor controlers for this vacs specificli for V8, as i got a heng of one and want to make fume extractor of of it.
And main cuestion is speed controll as this bastards ar plenty laud for that aplication, and i want ot taim them down.
 Sory for the gramar tho

Sorry, I haven't heard of anyone reverse engineering the motor controller.

[zapta]

Looks like Dyson batteries are a common pain point. Another approach are adapters for standard tools batteries such as Dewalt and others.

[Lubiezny_kran]

The draw back i the fact id dos not fully discharge them.  Power tools bateries are 5 cell comonly and dyson one is 6 cels in series.  MOtor operating voltage is 25.2- 16.2v. And comon discharge cut off is 15v

[robca]

I have a Kobalt 24V tool set, and the Kobalt batteries are 6 elements. I got a cheap adapter off Amazon for Kobalt, and it works really well. Pretty sure Kobalt is available only in North America, though (Lowe's brand)

The 4000 mAh Kobalt battery is $70 and high quality. roughly half the price of the Dyson (which is 2800 mAh only). Kobalt also offers 2000 and 6000 mAh batteries. The best part is that if you have two Kobalt batteries, you can chare one while you use the other one.

Instead of a cheap Chinese Dyson replacement battery with unknown cells, the Kobalt is a much better option (esp if you can find the Kobalt charger used for cheap)

Right now Lowes has a deal with 2 4000mAh batteries and the charger for slightly more than a Dyson battery https://www.lowes.com/pd/Kobalt-KOB-24V-2-4-0AH-BATT-110W-CHRGR/1003139096

[sukiyakh]

你好 在这里学习了很多知识 也看见了很多文章 但是我是没看明白 我有pickit3.5 他是需要hex文件的固件 没看见这个固件在哪 我的是dyson v8 sv10

[tinfever]

你好 在这里学习了很多知识 也看见了很多文章 但是我是没看明白 我有pickit3.5 他是需要hex文件的固件 没看见这个固件在哪 我的是dyson v8 sv10

Translated via Google Translate:
Hello, I learned a lot of knowledge here and saw a lot of articles, but I didn’t understand it. I have pickit3.5, which requires hex file firmware. I didn’t see where this firmware is. Mine is dyson v8 sv10.

The hex file is located in the Releases section of the GitHub page, here: https://github.com/tinfever/FU-Dyson-BMS/releases

[PaWill68]

Dear tinfever.
dear sukiyaki asked for firmware for v8 sv10, your project for V6/7
I'm sorry for the stupid question. And how to compile HEX from the provided sources?

[tinfever]

Dear tinfever.
dear sukiyaki asked for firmware for v8 sv10, your project for V6/7
I'm sorry for the stupid question. And how to compile HEX from the provided sources?

Good point. Others have reported that the released hex file for V6/V7 will work on a V8 SV10 but the LED codes will be different or nonsensical.

If you want to compile the hex from source, you will need to download Microchip MPLABx and open the project. I think you can then just compile without extra configuration needed.

[PaWill68]

Thanks for the reply. I have already compiled in Ubuntu. But the file size was 40 KB, everything works, but the diodes are a complete mess.
I also tried to create an open-dyson-battery project. It is compiled without errors, but the size of the HEX file is 4 kbytes. After the firmware, the battery shows no signs of life.

[tinfever]

Thanks for the reply. I have already compiled in Ubuntu. But the file size was 40 KB, everything works, but the diodes are a complete mess.
I also tried to create an open-dyson-battery project. It is compiled without errors, but the size of the HEX file is 4 kbytes. After the firmware, the battery shows no signs of life.

40.5KB is correct. That is the size of my published version. This is due to the encoding. Intel .hex file is not a raw binary. You can open it in a normal text editor so I think it is ASCII encoded inside. I'm not sure what you mean about the diodes.

By the way, you shouldn't need to compile it yourself unless you want to make changes or are doing it out of curiosity.

[PaWill68]

the LED codes will be different or nonsensical.

Sorry, when I said diodes, I meant LED.
Can you explain to me why it is not possible to compile a working HEX from the project https://github.com/tinfever/open-dyson-battery

[tinfever]

the LED codes will be different or nonsensical.

Sorry, when I said diodes, I meant LED.
Can you explain to me why it is not possible to compile a working HEX from the project https://github.com/tinfever/open-dyson-battery

Ah, I understand now. That open-dyson-battery project is just an unchanged git fork I made of someone else's project. I can't remember why I forked it but I never did anything with it. You should ignore that.

[sukiyakh]

Thank you. Thank you very much for your help. I saw the file you sent. Thank you very much.