Turnigy Mega 400W V2

[stavs]

The Mega 400W V2 is also an obvious derivative of the B6 design.  The board is marked "B641C rev A3"  and "DL-B641C V3.3".

Mine was poorly calibrated at the factory so it balances reliably to 4.17-4.18V on each cell.  Service menu 1 has never been available to me.

I have very limited knowledge of electronics, but enough experience with various programming languages (including assembly) to know that flashing the original B6 firmware would probably release magic smoke from something,  if only because this is a 400W rather than 50W charger.  I guess a few bytes of code will differ to account for this,  in 5-10 different places.  There seem to be other, more profound, differences in the circuit too.

I became enthused when learning of cheali-charger, hoping that this model might one day be supported.  I got hold of a USBasp mkII, added a pin header to the charger board, traced which header-pin corresponded to which MCU-pin, and made up an adaptor lead out of modified servo leads. (my local electronics store carries *very* little stock!)

The ISP lines are routed to a 1x6 header.  1:SCK,  2:MISO,  3:MOSI,  4:RESET,  5:GND,  6:VCC

I'm guessing it might be similar in some ways to the Quattro mentioned in another recent thread.  The fans are under software control:  MCU pin 9 (RXD - PD0) runs to a D822 transistor (via a '102' resistor - 1k?) and then to one pole of each of the fans.

First I read the fuses which seem to have rational values:

Low Fuse: 0x3F

High Fuse: 0xC9

Lock fuse: 0xF8

Calibration: 0xBDBDBDBF

Unfortunately, the firmware seems to be locked for reading.  :-/

This is born out by actually reading it:  "00 00 01 01 ... FE FE FF FF" repeated 64 times.

Next, I tried reading the eeprom, hoping to unlock service menu 1 by modifying byte 0x390.  The prom data returned "00 01 02 ... FD FE FF" repeated 4 times.  I wasn't aware the prom could be protected from reading.  Is this true?

Due to laziness and eagerness, I was using eXtreme Burner - AVR v1.41 on Windows.  I continued fiddling around with it for a moment, being careful to only use the "read" buttons,  not the "write" or "erase".

I clicked the Configure AVR Clock Source button,  hoping to see the current chip configuration.  Instead, the 4 drop-down menus showed nothing selected.   So I clicked the big red cancel button.  Which appeared to do nothing.  It didn't close the window.

I tried reading the eeprom again, and got a different result: a quick glance showed almost everything was FF.  Excited, I quickly pressed "Read -> All" to try to get a dump of the firmware again.  This took a long time, then failed.  I've forgotten the exact error message. Comms failure, or something similar.

Now nothing worked.  Restarting the software and power-cycling the charger didn't make any difference. eXtreme just reported a problem at the first stage: "powering up".

With my distrust of the software rapidly mounting, I took a moment to study the avrdude flags and switch to Linux, where I'm more comfortable anyway.  I couldn't make any headway there either.

Then it struck me:  there used to be a "click" from the beeper when powering it up via the USBasp or the regular power input. That doesn't happen now.

I didn't have the screen+backlight or buttons attached when powering via the USBasp,  but after 5s what I guess was the low input voltage alarm would sound. That doesn't happen now.

I don't think I've done anything dumb here.  Before doing anything, I read until my eyes watered.  I was wearing an anti-static strap.  My adaptor lead makes solid reliable contact, and my Fluke 87 says there's effectively zero resistance or capacitance between the USBasp and the ATMega pins on the charger.  I took care that no shorts could arise, especially the charger's regular power input connectors.  All my interactions with the chip were reads, not writes.  

It worked just fine for 5 minutes, then something broke.  With my limited experience,  I can think of only two possibilities:

  1)  powering it via the USBasp rather than the regular power input slowly burned something out

  2)  eXtreme burner wrote bogus values to the atmega

Can any of you esteemed folks here suggest which is more likely, any other possibilities, or perhaps even if it's salvageable?

Without a good dump of the firmware, my gut feeling tells me this board is destined for either the recycling bin, or maybe Pawel's doorstep for experimentation purposes after the exams. Good luck with them, btw. ;-)

Cheers,

  stavs

[stavs]

On Friday, 9 August 2013 15:51:25 UTC+1, I wrote:

Without a good dump of the firmware, my gut feeling tells me this board is destined for either the recycling bin, or maybe Pawel's doorstep for experimentation purposes after the exams. Good luck with them, btw. ;-)

Of course that should be Stawel's doorstep.  Sorry about that.   I blame my stupid memory.

[yt6...@gmail.com]

Stav ,

 

switch the usb asp to 5 volt , and then try connecting to the charger.

1 = IF your firmware is gone -> nothing happens

BUT you must be able to get the identity of the atmega32 via your usbasp ( 0x1e9502 )

 

2 = IF you locked yourself out of the atmega32 -> nothing happens

In this case you will get a message from your programming software telling you that the target is not responding.

 

Most likely the lockbits of you atmega32 are messed up .

 

In case number 1 , you must be able to get some action back into the atmega.

In case number 2 ............................ you bricked it , and it will continue as a doorstopper.

 

If you are skilled enough , out on ebay are many sellers offering the 44 pin atmega32 you need to replace and for and it must be possible to get one for under 5 dollar.

But , only if you can desolder the original one without damaging the tracks .

 

 

 

 

techy 

[stavs]

On Friday, 9 August 2013 16:31:45 UTC+1, yt6...@gmail.com wrote: 

switch the usb asp to 5 volt , and then try connecting to the charger.

Techy, first of all, thanks a lot for your response.  Although it does seem to confirm my fears. :-)

This is the USBasp mkII. It doesn't have selectable voltage that I can see.  With the jumper enabled,  it seems to pass through the USB power;  I can measure 4.77V across GND and VCC on the pins of the charger's Atmega.

I'm re-iterating for the sake of clarity: things worked fine for 2-3 minutes.  I successfully saved firmware and eeprom dumps,  but they appeared to be locked against reading.

Then, the eXtreme burner software exhibited an obvious GUI bug.  It may be coincidental, but this was the last moment I was able to correctly handshake with the Atmega.  To be even more clear, there were no circuit changes or physical disturbances between when it worked, and when it ceased to work. I would be very interested to hear the opinion of anyone that has used eXtreme burner for more than a few moments.

 

2 = IF you locked yourself out of the atmega32 -> nothing happens

In this case you will get a message from your programming software telling you that the target is not responding.

This seems to be the case.  Even after passing the -F flag to avrdude to move past "target doesn't answer",  the device signature is reported as 0x000000.  :-/

In case number 2 ............................ you bricked it , and it will continue as a doorstopper.

Yes, it does seem that it's now a strategic deployment of mass.  I wish I'd started with avrdude from the outset.  I'm almost certain I'd still have a working charger.  Failing that, my .bash_history would show me exactly how I goosed it,  as a warning to other noobs.

 

If you are skilled enough , out on ebay are many sellers offering the 44 pin atmega32 you need to replace and for and it must be possible to get one for under 5 dollar.

But , only if you can desolder the original one without damaging the tracks .

I've watched in awe as folks on youtube deftly swap SMDs.  I'm unsure if my soldering skills are good enough to succeed at that. I have a temperature controlled iron, but I have been cursed with an unsteady hand.  That notwithstanding, without the original Mega 400W firmware I still have a paper-weight;  I feel certain the circuit is different enough to cause damage using the stock B6 firmware.

It seems to me that this charger's only hope of resurrection lies with cheali-charger eventually supporting this model (and replacing the mcu).

Stawel, my offer stands:  if this bricked charger would be useful to your project, I will happily donate it. Just let me know where to ship it.

Cheers,

  stavs

PS between cheali-charger, BLheli, and the exemplary deviationtx.com, I am very close to flying on entirely open-source software.

Free Walkera rx/gyro or Zyx-S firmware would be the icing on what is already a very tasty cake.  :-D

[Paweł Stawicki]

2013/8/9 stavs <stavr...@gmail.com>

On Friday, 9 August 2013 15:51:25 UTC+1, I wrote:

Without a good dump of the firmware, my gut feeling tells me this board is destined for either the recycling bin, or maybe Pawel's doorstep for experimentation purposes after the exams. Good luck with them, btw. ;-)

Of course that should be Stawel's doorstep.  Sorry about that.   I blame my stupid memory.

Pawel is fine :)

I highly encourage You to do the support yourself 

with our help of course,

It seems to me that Your charger is a bigger version of GTPowerA6-10,

so it should be not a big problem, maybe even You need only to flash it with 

cheali-charger-GTPowerA6-10-0.20.hex.

But as techy said, you probably need to replace the atmega32.

Sending it to me is probably not a a good idea, I'm currently a little overworked.

Best regards,

Paweł Stawicki sometimes stawel ;) 

[stavs]

On Saturday, 10 August 2013 01:02:05 UTC+1, cheali-charger wrote:

I highly encourage You to do the support yourself 

with our help of course,

OK, you persuaded me. :-)

There is a replacement MCU in the mail,  so I shall try my hand at this later in the week.

It seems to me that Your charger is a bigger version of GTPowerA6-10,

so it should be not a big problem, maybe even You need only to flash it with 

cheali-charger-GTPowerA6-10-0.20.hex.

I think you're correct.  Externally it seems nearly identical to the GTPower A620.

Here are some photos of the Mega 400W V2 board:

http://picpaste.com/Tgy_Mega_400W_V2_bottom.JPG

http://picpaste.com/Tgy_Mega_400W_V2_top.JPG

Cheers,

  stavs

[Paweł Stawicki]

OK, you persuaded me. :-)

There is a replacement MCU in the mail,  so I shall try my hand at this later in the week.

That's great news! 

can you please make some pictures without the MCU?

I can't see all the pcb-tracks.

 

It seems to me that Your charger is a bigger version of GTPowerA6-10,

so it should be not a big problem, maybe even You need only to flash it with 

cheali-charger-GTPowerA6-10-0.20.hex.

I think you're correct.  Externally it seems nearly identical to the GTPower A620.

Here are some photos of the Mega 400W V2 board:

http://picpaste.com/Tgy_Mega_400W_V2_bottom.JPG

http://picpaste.com/Tgy_Mega_400W_V2_top.JPG

Cheers,

  stavs

--
You received this message because you are subscribed to the Google Groups "cheali-charger" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cheali-charge...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[stavs]

On Saturday, 10 August 2013 23:02:45 UTC+1, cheali-charger wrote:

can you please make some pictures without the MCU?

I can't see all the pcb-tracks

Thanks for looking in such detail.

Here are further shots:

http://picpaste.com/Tgy_Mega_400W_V2_top_A.jpg 

http://picpaste.com/Tgy_Mega_400W_V2_top_B.jpg

Let me know if there are any other areas where you need clarification.

I used Hugin to merge those shots and remove the wires, but I've yet to figure out how to stop it vignetting the exposure:

http://picpaste.com/Tgy_Mega_400W_V2_top_merged_badly.jpg

Cheers,

  stavs

[stavs]

I finally sussed it. After importing and aligning the images,  goto "Camera and Lens",  click "Reset" and select "Vignetting".

http://picpaste.com/Tgy_Mega_400W_V2_top_merged.jpg

[Paweł Stawicki]

I've checked the atmega32 pins on your charger and

they are almost identically with GTPowerA6-10.

The "Tgy Mega 400W" has something additionally connected

to pins 25,26.

I can't see what it is, it's the black "soic-8" chip

nearby the 16MHz oscillator. 

But I'm pretty sure you can run the GTPowerA6-10 firmware on it,

You will probably need to calibrate the internal temperature probe

(and maybe other things) inside the source code.

 

2013/8/11 stavs <stavr...@gmail.com>

I finally sussed it. After importing and aligning the images,  goto "Camera and Lens",  click "Reset" and select "Vignetting".

http://picpaste.com/Tgy_Mega_400W_V2_top_merged.jpg

[stavs]

The "Tgy Mega 400W" has something additionally connected

to pins 25,26.

I can't see what it is, it's the black "soic-8" chip

nearby the 16MHz oscillator. 

That package seems to be completely unmarked. :-/

I'll take some photos to see if I can coax something out using filters,  but I really don't think there's anything there to enhance.

 

But I'm pretty sure you can run the GTPowerA6-10 firmware on it,

You will probably need to calibrate the internal temperature probe

(and maybe other things) inside the source code.

Thanks for the tips.  I look forward to trying it later this week. 

[Paweł Stawicki]

2013/8/11 stavs <stavr...@gmail.com>

Thanks for the tips.  I look forward to trying it later this week. 

 

How is your charger doing?

[Nagy Jozsef]

Hi Paweł

I installed virtual linux and compiled your source.

I create some files and modify maximum Current and power parameters for Turnigy MEGA 400W x2 charger.

http://www.hobbyking.com/hobbyking/store/__25872__Turnigy_MEGA_400Wx2_Battery_Charger_Discharger_800W_AUS_Warehouse_.html

Today testing the high current my hexfile. GTPOWER hexfile working, but only 10A charge rate.

ftp://filex.tv2.hu/7678f6cdde98d59b24cdde621dc6e6bf/ please check new source files, that there is anything wrong.
(hexfiles not tested please carefully. Dont use.)

just a thought.

stayed connected batteries for a long time after charging end. after 12 hours charger temperature of 35 ° C and battery cells only  4.12V.

It is possible that there is a bug in the balance-off after charge end or mega400W not equivalent GTPOWER?

Thank you.

Jozsef

[Paweł Si]

2013/10/17 Nagy Jozsef <nagy.jo...@gmail.com>

Hi Paweł

I installed virtual linux and compiled your source.

I create some files and modify maximum Current and power parameters for Turnigy MEGA 400W x2 charger.

http://www.hobbyking.com/hobbyking/store/__25872__Turnigy_MEGA_400Wx2_Battery_Charger_Discharger_800W_AUS_Warehouse_.html

Today testing the high current my hexfile. GTPOWER hexfile working, but only 10A charge rate.

ftp://filex.tv2.hu/7678f6cdde98d59b24cdde621dc6e6bf/ please check new source files, that there is anything wrong.
(hexfiles not tested please carefully. Dont use.)

You also need to change this values (Turnigy-MEGA400.h):

#define SMPS_UPPERBOUND_VALUE               4096

#define MAX_CALIBRATION_SMPS_VALUE          (SMPS_UPPERBOUND_VALUE/8)

probably to:

#define SMPS_UPPERBOUND_VALUE               8192

#define MAX_CALIBRATION_SMPS_VALUE          (SMPS_UPPERBOUND_VALUE/16)

the SMPS_UPPERBOUND_VALUE should be 20x higher (maybe 22x) then the calibration value used to calibrate 1A charge current.

(20x1A = 20A)

the MAX_CALIBRATION_SMPS_VALUE is a safety limit used during calibration.

 

 

just a thought.

stayed connected batteries for a long time after charging end. after 12 hours charger temperature of 35 ° C and battery cells only  4.12V.

It is possible that there is a bug in the balance-off after charge end or mega400W not equivalent GTPOWER?

Thank you.

Jozsef

I don't know, It could be a bug in cheali-charger, If I find time I check it this weekend.

Regards

Paweł

ps.

The version you are using is out of date,

please update it to the 0.31 version.

 

It's much easier to use the github repository when changing

the source code, it's also easier for me to read the changes. 

[Paweł Si]

2013/10/17 Paweł Si <sta...@gmail.com>

ps.

The version you are using is out of date,

please update it to the 0.31 version.

I'm sorry, my mistake, 

I had an older version on my computer

 

 

[Nagy Jozsef]

Ok, thank you Pavel.

Today tested the new software (20A limit)

but blowed one FET.  :-)

Between calibration process contact problem measuring wire. I stupid.

I changed one FET and one high power resistor.

working again, but the bad air in the house :)

My wife is very angry.

tomorrow changing your calibration parameters.

Please put the repositories of the Turnigy-mega400.

I will try later in the UART transmission.

Thank you again

Jozsef

[Nagy Jozsef]

Mixed mode:

Left side: absolute accurated charger.

Right side (wrong side :-) : inaccurate charger.

[Nagy Jozsef]

Hi Pawel.

This Mega400 (x2) firmware good working.

-internal Thermo sensor absolutely good working but why are there two calibration? (int/ext?)

I small problem:

this charger high current. 20A.

if charging started, then  delta current too high (0 to 20A 0sec). It would be better to smooth.

Many powers supply not good tolerate this high current change. (or gasoline generator with inverter)

decrease current of the same problem.

original(stock) software approx 3-5A/sec delta current.

I would love if possible

thank you

Jozsef

www.octomovie.com

[Paweł Si]

This Mega400 (x2) firmware good working.

-internal Thermo sensor absolutely good working but why are there two calibration? (int/ext?)

"temp extern" - is the external temperature probe (sensor).

"temp intern" - is the build-in temperature probe (if present)

 

 

I small problem:

this charger high current. 20A.

if charging started, then  delta current too high (0 to 20A 0sec). It would be better to smooth.

Many powers supply not good tolerate this high current change. (or gasoline generator with inverter)

decrease current of the same problem.

original(stock) software approx 3-5A/sec delta current.

I would love if possible

Great suggestion, I will try to implement this tomorrow, if I find the time.  

 

thank you

Jozsef

www.octomovie.com

Regards,

Paweł

ps.

congratulations, great site.

[Joukko Sool]

Hello,

I recently bought a Turnigy MEGA 400Wx2 Battery Charger/Discharger (800W). I quickly discovered that I can´t calibrate it and also the smart fan is not working. I see that you have a firmware where you can calibrate it. Does this firmware also fix the smart fan? 

I have flashed transmitter and esc atmels before but not this specific one. Is there anything different with this one that I should be aware of? And of course where can I get this firmware? Could you please share this firmware with me?

Best regards,

Joukko

[Paweł Si]

2013/12/12 Joukko Sool <joukk...@gmail.com>

Hello,

I recently bought a Turnigy MEGA 400Wx2 Battery Charger/Discharger (800W). I quickly discovered that I can´t calibrate it and also the smart fan is not working. I see that you have a firmware where you can calibrate it. Does this firmware also fix the smart fan? 

I'm not sure, don't have this model, maybe Jozsef knows,

but if the fan is connected to pin 9 of the atmega32, then it 

should be quieter. 

You can find the firmware (press the "Raw" button):

https://github.com/stawel/cheali-charger/blob/master/hex/beta/cheali-charger-Turnigy-MEGA-400Wx2-0.32.hex

or with jozsef improvements:

https://github.com/njozsef/cheali-charger-test1/blob/master/hex/modded_by_njozsef_version/cheali-charger-Turnigy-MEGA-400Wx2-0.32m.hex

[Nagy Jozsef]

Hi all

i think: big power charger is bad hardware design. but:

I increase the p0 points and increase the p1 points for better fit the big power:

discharger: p0 50mA to 300mA  p1: 300mA to 1A

SMPS: 50mA to 300mA    1A to 5A

my modded fw, I increased max_calibration_discharger_value up to 255.

Pawel, do you think that's okay?

smart fan is problem? Actually hysteresis 5 celsius. What wrong?

Jozsef

[Paweł Si]

2013/12/13 Nagy Jozsef <nagy.jo...@gmail.com>

Hi all

i think: big power charger is bad hardware design. but:

I increase the p0 points and increase the p1 points for better fit the big power:

discharger: p0 50mA to 300mA  p1: 300mA to 1A

SMPS: 50mA to 300mA    1A to 5A

my modded fw, I increased max_calibration_discharger_value up to 255.

Pawel, do you think that's okay?

Yes, any calibration points are ok.

[Nagy Jozsef]

Joukko wrote:

Hi Jozsef
Now we could calibrate it with no problems. Thank you for your help. I really appreciate your efforts in this project. Have you tested all charging and discharging programs? My main coal is to charge and discharge lipos from 2s to 6s.
best regards
Joukko

[Sergio]

On Friday, August 9, 2013 4:51:25 PM UTC+2, stavs wrote:

The Mega 400W V2 is also an obvious derivative of the B6 design.  The board is marked "B641C rev A3"  and "DL-B641C V3.3".

I have the same charger, what version is safe to use with it? I don't see specific version for Turnigy Mega 400W V2 in repos

Should I use version for  GTPowerA6-10 or Turnigy Mega 400W X2 (doubled)? Do I need to make any modification to source?

[Nagy Jozsef]

I dont know.

Please do not flash the cheali charger your v2 charger so long examine hardware.

[Paweł Si]

hm... there are definitely too many version...

Sergio can You make some pictures of you chargers PCB,

the front and back side?

I'll look into it.

Best Regars,

Paweł

 

[Sergio]

Hello Paweł,

Thanks for your response. Here is front side of PCB. I will make more detailed pics from both sides when return from vacation, in 2 weeks. This is the same model that topic starter was talking about.

[Paweł Si]

Yes, I remember that I once looked at this charger,

please see:

https://groups.google.com/d/msg/cheali-charger/nyC673SRXF4/N1nnHFqqbD8J

conclusion: if you're not afraid to take a risk,

you can try the "G.T. Power A6" Firmware or the "Turnigy Mega 400W X2"

but during the current calibration, increase the output current very slow.

Best Regards,

Paweł

[adolfo...@gmail.com]

Hello !

Just to thank you guys for your work on the firmware and to let you know that the 400w v2 works ok with the 400w x2 firmware after calibration procedure.

Thank you once again.

[Kai S]

Hey guys,

can you help me, where i can find the hex file for the turnigy mega 400 v2?

are there any known malfunctions with the firmware?

regards Kai

[Nagy Jozsef]

Hi

I can not issue unless the Shottky diode.

Need change two shottky-diode to Si-diode (example: 1N4148)

This picture represent the two shottky. This is 2x200W v2 (sorry)

https://drive.google.com/?tab=To&authuser=0#folders/0B1tSgWAfLbkBb0xDSlhQTVJncDQ

find the D7 diode

https://drive.google.com/?tab=To&authuser=0#folders/0B1tSgWAfLbkBb0xDSlhQTVJncDQ

BEst regards

Jozsef

--

You received this message because you are subscribed to the Google Groups "cheali-charger" group.

To unsubscribe from this group and stop receiving emails from it, send an email to cheali-charge...@googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Simone Fabris]

Hi Jozsef,

Your drive files are "private"... I can't access them. Can you please change the sharing settings?

Thanks

Simone

[Nagy Jozsef]

Oh Jesus.

Sorry.

I corrected

.

new link:   https://drive.google.com/?tab=To&authuser=0#folders/0B1tSgWAfLbkBUkIzZW9Rb1N3ZEE

jozsef

[Simone Fabris]

Hi Jozsef,

could you briefly explain the mod you have done on the V400? 

Unfortunately I can't understand Hungarian.

I can see the resistor upgrade. 

Why did you need to change the two scotchy diode? 

Maybe I have to try to google translate your pdfs :-)

Thanks for your help,

Simone

[Nagy Jozsef]

https://docs.google.com/file/d/0B1tSgWAfLbkBSVlsbkpKQURRck0/edit

only change any Si diode  the two red rectangle point.

Si diode need for current linearity for full range.

Original software and Shottky diode lying to the actual highest current

it gives the impression, all current good.

But original version  current lowest than displayed value in higher-range.

It is therefore not very warm.  :-)

[Kris]

Got my V2's from the HK a few days ago. They have atmega32a inside. So sould be good to go with the latest Mega 400W X2 version.
I am still waiting for my atmel USB programmer. So it takes a while until i am able to flash them and report back. So the pins should be 1:SCK, 2:MISO, 3:MOSI, 4:RESET, 5:GND, 6:VCC on this one.