No there will not be any advantage. Just scale down your outputs to the actual resolution of the esc. Your hardware (atmega) limitations will be there before you will notice any advantage of increased throttle resolution.

If the esc firmware is feeded with a (atmega or whatever mcu) 1000-2000µSec (fast) PWM, why should we scale that down?

Originally Posted by timecop

No there will not be any advantage. Just scale down your outputs to the actual resolution of the esc. Your hardware (atmega) limitations will be there before you will notice any advantage of increased throttle resolution.

He was complaining about reduced resolution i.e. current ESC firmware going 'full throttle' for anything > 1800us thus he felt he was 'wasting' 200us of throttle range. So I recommended if he's worried about that, to recale his FC outputs to consider 1800us as max, then it won't matter.

Well here's a little further thought. If you look at this from a complete system perspective, it seems that the flight control code and the ESC should be matched. For instance, if the flight control code is written such that it can actively command up to 2000 uSec, and the ESC tops out at 1860 uSec, we're going to loose the capability to change motor commands in the upper 1040 uSec that is being ignored by the ESC.

Timecop is correct, we can accomplish this matching in 2 ways, increase the range of the ESC to match, or decrease the flight control processing to match. I think from an overall system perspective, it would be helpful to match this. If it's doable I'd rather do it by expanding the range and resolution. That's all.


I'm not fluent on the brushless motor control theory, nor the Atmega assembly code at the moment. Hence the question would expanding the range mess up the code calculations somewhere.

Hello guys.
i have a strange issue.
one of my bluesseries 20A (flashed with bs_nfet.hex) has a problem.
this happened already before the flash.
sometimes in flight one motor just stops. when it is cold outdoors and the esc gets cold. then it happens nearly every time. so i am not able to lift. if i was mounting my esc i touched a little bit the capacitor with my soldering iron. just a bit. i can see it on the shell. i ever tought this capacitors aren't that important. just for the current ripple? filterring the input to protect the mosfet. can this dropouts really have something todo with this ?
mounted is a 220uF capacitor. i only have a 170uF spare from a plush.
maybe ill give it a try and change the injured cap with this one?

useless?
still a remark. now with the flashed one. when i connect the battery the esc begins to beep in different pitch tones. when this happens i am unable to start rotating the motor.
best regards

//EDIT: problem described in post 1613

Seen little of same problem with "factory" avr firmware but does not appear to be a problem when using esc with new SIL chip. Bigger motors seem to be fine these are micro motors 1811 series, hacker 1012 and other no name cd type motor. It spins and locks onto something very slow. Throttle signal has very little effect once it locked into the slow spin and you can hear like a harmonic in the spin. The Hacker has great magnets so it is one of the most "coggy" but nothing special with the cheap 1811 and they both respond the same. They are all 2000kv - 3000kv 2s 6A motors. All high pole count 12 -14 I think. It definitely is in the starting and fixed in the SIL version . It is also definitely beyond my skills

Originally Posted by simonk

Which motors? The 6A ESCs had some problems before where I had tuned PWM-aligned comparator sampling for larger boards and this made them miss entirely. That code has all since been replaced with dynamic filtering lengths, but perhaps the smaller gate charge is still causing a difference with them.

By "unsync", do you mean it just screams and doesn't start or does it spin but aligned to a harmonic? I am seeing the screaming when starting a hard drive with platters sometimes, and if it's that same issue, it can probably be fixed. The issue is that we have to have a wide open zero crossing window while starting because we have no idea if the motor is already spinning or any idea of the mass, coggyness, KV, voltage, etc. So, sometimes we can get fooled by the demagnetization spikes and end up thinking we're running it while instead we're just commutating on a magnetization harmonic. Once we've aligned properly, the zero-cross blanking window prevents this from being a problem.

I do not wish to litter the thread.But I am excited to report that I have now successfully flashed four of the 'older type' HobbyKing Blue Series 30A ESC's. It is understood that the newer versions of this ESC only use N FETS.

From having read this topic from start to finish, I had the advantage of learning that the older versions of this ESC use both P and N FETs (Thanks, Timecop!).

Also, I had the advantage of learning of the existence of 'micro-pin-hooks', which I ordered from Digikey. (Thanks again, Timecop!)

I already had a Sparkfun AVR Pocket Programmer from some other stuff I had been doing, and I had a buspirate cable also.

I downloaded the January 4 2012 version of the modified firmware from here (Thanks, SimonK!!) and extracted all of the files.

I opened "AVR8 Burn-O-Mat v2" which I had previously downloaded for some work I was doing on my Turnigy 9X. I went into 'settings' and 'AVRDUDE Settings' and changed the programmer type to 'usbtiny', and left the 'disable fuse check' box ticked.

Then, back at the main screen on burn o mat, I chose the 'AVR Type' of Atmega 8. Under the "Flash" portion of burn o mat, I clicked on the drop down box next to 'file' and chose "Intel Hex" and then browsed to the files I had extracted, and chose "BS.hex".

I carefully connected the micro-pin hooks to the buspirate cable, which was connected to the AVR pocket programmer (which was NOT yet connected to my USB cord). Then, I carefully connected the pin hooks to the appropriate pins on the Atmel chip on the ESC. I used the pinout found on the first page of this topic, coupled with the attached AVR_ISP_Connector pinout. The trick is that MOSI and MISO are really close together, and so you have to make sure that the pin hooks do not touch each other.

Once I was comfortable with the pin hook placement, I connected the Pocket programmer to the USB cord on my computer, and just clicked on "Write" in Burn o mat. (note that I did not power the ESC separately with a lipo) After a few seconds, I saw this:
__________________________________________________ ____________

Writing | ################################################## | 100% 3.44s

avrdude.exe: 2070 bytes of flash written
avrdude.exe: verifying flash memory against C:\Users\Craig\Downloads\tgy_2012-01-04_d8a3f89 (2)\bs.hex:
avrdude.exe: load data flash data from input file C:\Users\Craig\Downloads\tgy_2012-01-04_d8a3f89 (2)\bs.hex:
avrdude.exe: input file C:\Users\Craig\Downloads\tgy_2012-01-04_d8a3f89 (2)\bs.hex contains 2070 bytes
avrdude.exe: reading on-chip flash data:

Reading | ################################################## | 100% 2.27s

avrdude.exe: verifying ...
avrdude.exe: 2070 bytes of flash verified

avrdude.exe done. Thank you.
__________________________________________________ _______________

Next I connected the ESC to my Receiver, and a test motor, and then I ran the throttle calibration by doing the following:
1) Made sure no prop was on the motor, and motor/esc was not powered up, but was connected to the Throttle channel of my RX.
2) Turned on TX with Throttle at Max
3) Powered the motor/ESC - heard the regular beeps followed by another beeep.
4) Lowered Throttle to minimum, heard the short beep beep.
5) Unpowered and re-powered both the TX and the motor/esc and tested. It worked just fine!

Now I will go try them on a multi-rotor!

Thanks!

this has nothing to do with the capacitor. this would also make no sense..
i tried something
in my room it is around 22°C. i took my quad out where it was around -2°C. waited about 5mins and then plugged in the battery. beep(normal) beep(low) beep(high) beep(high) beep(high)....beep(high) from two esc's...
when i plug them in without throttle low then it is like beep(normal) beep(low) nothing until i put my throttle to a lower (arming) position.
strange because blueseries have an external crystal?

Edit:// some tests so far:
2 of my esc's are beeping like crazy when i let them outdoors in the cold, take them back in my room, and connect a battery?
i let them warm up a bit. after 5min i did connect my battery. they did stop beeping but anyway i am not able start the motor anymore?
i noticed that they are waiting for a throttle high signal. this illustrates that my low throttle point has moved down and i can not reach it without recalibrating throttle end points.
but why do they beep when they have cold ??? and why only 2 of my 4 esc's have this behavior ?

would be great if someone has an answer for me

Hi Guys,

these are the maytech V2 harrier supreme 35 A Opto. Can one of you tell me if they are just rebranded hobbywing of different ?

Thanks

Boris

No, that is one of the ZTW units, much better n-FETs, made by International Rectifier company. This is the same ESC as the Hobby-Lobby eRC units. I was able to flash those without any problems, great ESC.

Ztw, aka blue series. Yes, obviously re labeled stuff.
Looks identical to pic of eRC stuff posted earlier.
Better fets than hobbywing.
Good news is you can reflash them

Thanks for the info ! Will try them with the given firmware just out of curiosity and that i have a comparison once i have flashed them !

Boris

The Hobby wing firmware works on the Maytech? "ZTW"

Originally Posted by BrokenRotor

No, that is one of the ZTW units, much better n-FETs, made by International Rectifier company. This is the same ESC as the Hobby-Lobby eRC units. I was able to flash those without any problems, great ESC.

This is interesting and good to hear.

hello i want to build the smallest quad/hex available. what is the smallest ESC that has pads?