(20-May-2023, 10:37 PM)e Pathfinder07 Wrote: Yeah blackbox for that sort of money.  It uses BMI270 gyro as well.  I could see me buying a couple of the 35A for a start.  That would square away two future projects for around the price of a single Speedybee stack.

 
I did buy a couple of AIO as soon as I found they are supported in Inav 6.1. Seems a perfect match to my Explorer LR 4inch to finally upgrade it to Walksnail.

It would be nice to see this help bring the price of AIOs down.. I see this hobby going sub 250 till things “cool off”… and 30x30 stacks will be stuck in 5”+ rigs..
Knowing a stack / AIO wouldn’t cost me $100+ If I crash will allow me to fly harder and maybe not scare new pilots as much..

(28-May-2023, 03:13 PM)PeladoMat Wrote:  
I did buy a couple of AIO as soon as I found they are supported in Inav 6.1. Seems a perfect match to my Explorer LR 4inch to finally upgrade it to Walksnail.

Please let us know your thoughts on them once they arrive!

What I would LOVE to see.

1-2S AM32 esc based AIO AT32 MCU.

My adventures in 1S toothpicks have shown me that bl32 blows bluejay out of the water.  Only the tmotor ($100AUD+) AIO is available.

Good to see betafpv have entered the foray too even though the esc is still 126bit based.  They have a much higher profile than nutronRC and will hopefully have an effect on their competitor's decision making.

(29-May-2023, 02:59 AM)ph2t Wrote: My adventures in 1S toothpicks have shown me that bl32 blows bluejay out of the water.  Only the tmotor ($100AUD+) AIO is available.

What difference have you noticed and this is only for 1S?

(29-May-2023, 02:20 PM)mstc Wrote: What difference have you noticed and this is only for 1S?

The overall real power to the motors is just more.  

I've been trying to build a few 1S toothpicks to compare bluejay versus bl32.  Currently rebuilding the bl32 tmotor aio toothpick on a stiffer frame and trying out another one with the 12A AIO betafpv board running RC8 of the PWM version of bluejay.  

Getting motor desyncs and some stalling although this can be because of a multitude of influences like "notchy" motors (mamba 1103 tokas) that don't spin up well through to 48 and 96Khz PWM modes having varying levels of support with the betafpv/jhemcu AIOs.   I'm still testing out different things before I post more in my main toothpick thread.  I will do this soon hopefully.

All this is focused on 1S setup, yes.   It is like the "other" extreme, getting the most out of the limited power that is available.   Digital toothpicks are always going to be in the 45 - 60 gram range and this needs a different powertrain I reakon.

I’m looking forward to reading that post once you’ve finished experimenting.

I’m well aware that different esc settings can drastically change things, but I haven’t given any thought to how different firmware might change things beyond what differences in settings they allow. I also haven’t experimented with different esc mcu’s to see if there’s a tangible difference.

Sometimes packages arrive quickly from China. Comes with some thick wires...



In order to connect to BF, it requires downloading/installing an Arterytek driver and the SHOW ALL SERIAL DEVICES option must be enabled in BF Configurator. The NeutronRC page download link is to arterychip.com, the driver is also available on arterytek.com.

And for those with itchy upgrade fingers, NeutronRC explicitly tells you that you don't need to flash different firmware.

The default FC Flash 4.3.2 branch firmware, AM32 ESC flash 1.97 firmware, usually, you don't need flash FC and ESC firmware, just install and set related options according to your settings.

About Betaflight's support for artery MCU:
The current BF 4.4.X version does not support the AT32F435 chip. Our default firmware is based on the 4.3.2 branch. At present, it is seen in the BETAFLIGHT code base that BF4.5 will soon support the AT32F435 chip, and the specific progress is uncertain.

And seems there is a bug on the AM32 web configurator?

2. About The AM32 ESC, There is a known bug in the current website version AM32, which leads to failure after writing or modifying the ESC. We have submitted a PR to fix it. For the time being, you can use our bug-modified Demo ESC Configner settings or upgrade the ESC;


Also note that per NeutronRC, the 35A version of the AIO is only suitable for 21xx or smaller motors.

That’s exciting!

I don’t really understand their note that it is only suitable for 21xx and smaller motors. It should be suitable for anything that pulls under 35A.

How does the build quality look?

Maybe it has to do with the AM32 firmware? Even for 21xx motors they go on to specify only for driving 3.5-4" props. But seems they have actually done their own testing before passing on these recommendations/cautions.

I don’t see how the firmware would limit motor size or prop size. Both of those will have a significant effect on amp draw, so maybe they’re just playing safe for people who just pick a motor and prop and throw it on a build without working out how much current it’ll actually pull?

Obviously chocolate ESC. Apply too much heat and they melt.

My guess is the VRM's are rated for 55A or 35A, but when they tested them, they died long before that, so maybe it's a bit of pseudo-honesty and a potential way to get out of any claims if people use them on bigger motors and they die. So for me I would probably employ a 10A safety net. On the 35A, go no higher than 25A and on a bigger build, I might consider sticking some micro ceramic heatsinks to the mosfets and anything else that might get hot.

Use some basic testing criteria with these. So if you have a motor with no data sheet, you could set motor limit to 30% and apply full throttle to get the max amp draw. Then increase by 5% until you find a spot that gives you much thrust but also keeps the amp draw a good 5-10A under the rating.

Even if they overrated the esc’s, I still consider it one hell of a deal. I was chatting with them on the Facebook messenger a week or two ago, I think I’ll reach back out and see if I can get ahold of their test data.

Alright, I reached out to NeutronRC to get some more info.

They actually had a tech answer me, rather than blowing me off like a number of other companies have done. For that alone, they get points from me, and I am much more likely to buy something from them.

My conversation with NeutronRC is below. The main take aways for the 35A AIO are:
The esc's are rated for 35A constant.
Peak/Burst rating is 45A for 5 seconds.
They did do testing, and their 21xx and smaller motor and prop size are just suggestions to keep you from destroying a mosfet.
They told me what mosfet they use, it is:
Toshiba TPN2R203NC
https://www.mouser.hk/datasheet/2/408/TP...rCMYjL1_qQ

Me:
Someone I know bought the 35A aio to try out. It states that it’s only suitable for 21xx and smaller motors, and only 3.5” to 4” props.
As long as the total amp draw per motor is under 35A, I don’t understand why there are size restrictions?
Do you have any test data on your ESCs that you can share with me?

NeutronRC:
Yes, we have comprehensive data tests for hardware design, MOS performance, parameters, PCB design thermal power consumption, and the reverse voltage generated by the motor when it is working and braking, and the IDm of the MOS.  , safe and reliable configuration recommendations.
How to install the 2207.2306 motor in the 35A version, it may have no problem when it starts to work, normal level flight shooting is no problem, when you do a quick U-turn, or 100% throttle steering, the MOS may burn out, causing the plane to fall
In terms of hardware design, we will take 1/3 of the mos data as the benchmark, stable and reliable working current.

Me:
Thank you. 
Can you please send me those data tests?
Can you also please tell me what the peak/burst current rating is on the 35A aio, and how long it can handle that for?
I have a number of people reach out to me for recommendations on the intofpv.com multirotor forum, and I can’t recommend your electronics without knowing exactly when and how they will fail.

NeutronRC:
https://www.mouser.hk/datasheet/2/408/TP...915945.pdf
ID 100A, according to the industrial application, the value is 35A, which is the reason for using this MOS to mark the parameter 35A. At the same time, its ID is 45A at 25 degrees, so the peak value is 45A. It is customary to follow our standard flight action, and the sudden current is full throttle. Generally, the calculation does not exceed 5 seconds, so his ID MAX burst current definition is 45A for 5 seconds. Idp anti-shock current 200A, in fact, it is very small in FPV use, relatively conservative, it is easy to reach this peak value when the motor brakes, this is the fundamental reason why I limit the 2205 motor, if you use 2207 or 2306 and a larger motor, Fortunately, light braking will not exceed this current, but considering extreme conditions, such as rapid braking to avoid obstacles, at this time, the braking reverse voltage and sink current of the motor will be far greater than 200A, which will cause MOS damage. Based on the above points and the characteristics of the actual motor, a reasonable suggestion is given. The detailed flight data test is more based on a large number of empirical tests that we actually installed on the frame. I think the actual experience is more important than the pure theoretical data. Although the MOS parameters retain a certain amount of redundancy, we still need to retain a reasonable amount of redundancy to achieve reliability and stability while ensuring the reliability of the user framework. So there needs to be a certain limit.


Me:
Thank you, I appreciate the answer and information.
I’ll be ordering one and posting my experience with it.
I really appreciate you actually giving me a proper answer.

——

For those of you that don't regularly reach out to manufacturers to try to get information, this level of detail is rare. The only other manufacturers who have been willing to interact with me on this level are Foxeer and DarwinFPV. I've reached out to just about every other major manufacturer while trying to repair boards, and I typically just get blown off or stuck in a loop of misinterpreted language translations.

Most manufacturers won't even tell you what mosfet they used, much less how they calculated the esc rating.

Never seen such a complete answer !?