Originally Posted by plusultra

Hi All!
Please help me. What .hex file i must use to flash "Maytech 45A-OPTO with NO BEC"?
Thank you.

I believe it is "bs.hex". Let me know if it works and I will add it to the spreadsheet.

Originally Posted by chessmaster

I have a quick question, what is bootloader exactly? I flashed 6 turnigy plushes that had atmel chips and didnt enable it.

It's a chunk of code at the end of the flash that, if the fuses are set, gets executed first on reset/power up. The one I wrote that is included supports flash and EEPROM writes and reads via the Turnigy USB linker. This allows you to update to a newer version later without having to access the ISP pins.

Originally Posted by simonk

I believe it is "bs.hex". Let me know if it works and I will add it to the spreadsheet.



It's a chunk of code at the end of the flash that, if the fuses are set, gets executed first on reset/power up. The one I wrote that is included supports flash and EEPROM writes and reads via the Turnigy USB linker. This allows you to update to a newer version later without having to access the ISP pins.

All you have to do is enable when you flash your esc? Is there anything different you have to do when flashing it? Thanks

Well in fact, default behavior of kkMulticopter Flash Tool is perfect, I didn't notice that flashing through USB Linker also erase the EEPROM, so after each flash, I'm using default values, perfect for me !
In tgy.asm I change STOP_RC_PULS to 1100us and FULL_RC_PULS to 1900us, but when I send a 1100us signal at start time, my ESC doesn't beep, it's like the signal is to high.
When I change STOP_RC_PULS to 1105us, there's a beep for a 1100us signal when my esc start, and after it can work. Is it normal ?

Thanks

Originally Posted by airmamaf

When I change STOP_RC_PULS to 1105us, there's a beep for a 1100us signal when my esc start, and after it can work. Is it normal ?

Yes, this is an "on the fence" case of what happens when there is a slight oscillator mismatch. There always has to be some margin somewhere, even if it's less than 1%, when working with pulse lengths. What flight board are you using in this case? Most boards will send something much lower, like 1000µs or 900µs to arm the ESCs, and then jump up to 1100µs or so to start the motor.

Hi!

What parameter / code do i have to change so that the calibration really starts on "low throttle" during calibration? When i test with a servo tester, there is a much higher start position.

manu

Originally Posted by simonk

Yes, this is an "on the fence" case of what happens when there is a slight oscillator mismatch. There always has to be some margin somewhere, even if it's less than 1%, when working with pulse lengths. What flight board are you using in this case? Most boards will send something much lower, like 1000µs or 900µs to arm the ESCs, and then jump up to 1100µs or so to start the motor.

Can the first post be updated with the SiLabs programming?

I have flashed a lot of Atmel chips, but never SiLabs.
Don't want to read the whole thread to find out.

Thanks.

OK...

I found this

http://www.rcgroups.com/forums/showthread.php?t=1686498

Some one a hint for my problem? -> http://www.rcgroups.com/forums/showp...postcount=6784

Fuses look like this -> http://s7.directupload.net/file/d/3130/3bnkujsn_png.htm

Also flashed tp_8khz.hex firmware with no success.

@Simonk,
This is the layout of my Mystery 40A ESC,
I guess you are right the S29B ( LM3485) probably is set to 5V not 3.3V
The internal BEC (red cable) was not connected, so I don't see any reason why this 3 components would burn down.
The datasheet of the 4421 is clearly limited to 18V, so I assumed this could be the possible cause for the problems.
Maybe this layout is the previous version of the ESC which was limited to 4S?

Originally Posted by alll

What parameter / code do i have to change so that the calibration really starts on "low throttle" during calibration? When i test with a servo tester, there is a much higher start position.

Hi Manu! PROGRAM_RC_PULS sets this. It is normally halfway between the default neutral and full throttle, so that a slightly higher throttle does not start calibration at power on. The trade-off is that the maximum throttle cannot be calibrated below that point.

Originally Posted by MasterG

Some one a hint for my problem? -> http://www.rcgroups.com/forums/showp...postcount=6784

Fuses look like this -> http://s7.directupload.net/file/d/3130/3bnkujsn_png.htm

Also flashed tp_8khz.hex firmware with no success.

Without poking at it with an oscilloscope, I cannot guess what the problem may be. I suspect something has failed.

Originally Posted by hexacop

@Simonk,
This is the layout of my Mystery 40A ESC,
I guess you are right the S29B ( LM3485) probably is set to 5V not 3.3V
The internal BEC (red cable) was not connected, so I don't see any reason why this 3 components would burn down.
The datasheet of the 4421 is clearly limited to 18V, so I assumed this could be the possible cause for the problems.
Maybe this layout is the previous version of the ESC which was limited to 4S?

Yes, you definitely have a different layout. Yours does not seem to have the 78L05 that was heating up at 6S in our case. By the way, isn't the 4421 an AO4421 P-channel FET with -60Vds rating?

Originally Posted by simonk

Yes, this is an "on the fence" case of what happens when there is a slight oscillator mismatch. There always has to be some margin somewhere, even if it's less than 1%, when working with pulse lengths. What flight board are you using in this case? Most boards will send something much lower, like 1000µs or 900µs to arm the ESCs, and then jump up to 1100µs or so to start the motor.

It's an home made tool, I used to bench some motors. To arm my ESC, I use a 1100µs signal. I was wondering if the difference between 1105µs and 1100µs was a symptom of something wrong, but it seems, that I may need to send lower signal to arm my ESC (beep) and then send a idle signal of 1100µs.

Thanks for your reply

Hi,
Can Multiwii handle motor outputs at a rate of 1KHz (1000Hz), is it a parameter that can be set? Hardware PWM?

@Simon
I had to set ultrapwm=1 to be able to use a 200...1200µs signal.

i saw this :

ldi    temp1, 32 * 17/16    ; Stop/reverse pulse (slightly above exact)

This was my question
I changed it to

		
ldi	temp1, 32 * 64/63;

and now it is more "exact" start at low throttle.

I still can't understand this asm logic, but could calibration be done this way?
Set throttle at desired max pos = rxMaxInit (enter calibration if > 1200µs), power up esc, you enter calibration as usual, read rxMin, you leave calibration if rxMin<(rxMaxInit-"10") and rx >= (rxMaxInit +- 2),finally save to eeprom.
This will let you more precisely set rxMin and you have the time to do it.

Also, why are we using "real" timings.
Assumption:any PWM length from 200-2400 is valid (only time you use the osc as a reference)
In code you only use a "steps" resolution (2x800=1600 for example or more )
After calibration you map rxMin/rxMax to 1600 ....

Thanks
manu

Originally Posted by simonk

Hi Manu! PROGRAM_RC_PULS sets this. It is normally halfway between the default neutral and full throttle, so that a slightly higher throttle does not start calibration at power on. The trade-off is that the maximum throttle cannot be calibrated below that point.

Originally Posted by simonk

I really don't know what happened here. At first it used to work, and then somehow they started flashing a version with broken calibration (even though I don't see how, it works when I flash it). I sent them the latest version, so newer ones should be OK.

Thanks for all the info, i confirm my ESCs, and i guess all the ESCs sold by RC Timer during that period, have not bootloader enabled.

Regards

Originally Posted by simonk

Yes, you definitely have a different layout. Yours does not seem to have the 78L05 that was heating up at 6S in our case. By the way, isn't the 4421 an AO4421 P-channel FET with -60Vds rating?

Hi Simon
According to some other pictures I found, it is the layout of the Blue Series ESC, which would also require a different firmware.
Btw. what is the difference then between bs40a.hex and bs.hex, could this have been the cause for the smoke in the end?

Hi Simon,

I made some tests with parameter COMP_PWM = 1 for an HobbyKing 20A BlueSeries ESC, i didn't notice any change in consumption / noise from low throttle to high throttle. Maybe at high throttle there were little less consumption for a little more thrust, not quite sure, it can be the measure error margin.
It'd like to know if this parameter (COMP_PWM = 1) could have any advantage, and for which conditions ?
Another question, for MOTOR_ADVANCE parameter, by default it's set to 18, is there any advantage to change this value or not ?

Thanks