Fantastic, Chris! Thanks a bunch for the testing and comparison! 😃 I really miss Drone Mesh's tests. I know many people contested his methodology, but at least he was trying! Anyway, stay safe there with your family! 🖖😊
I live in the low-to-moderate range. I'm eager to see what you and the BlueJay devs will come up with. It's pretty sweet as-is, truth be told, at least for my set up. Bring it on, Wizards!
Yes, awesome work providing consistent, confident data that can't be ignored, (already-purchased) products improve, which generates value to all us consumers, while closes competition to a point that will make all FW a viable tough compromise, broadening FW choices to delay the HW churn or cover corner applications, and esp on budget to standardize on a HW platform.
Very good job, my friend. Can't wait to see this be implemented as an improvement in the code as a result. Along with this random spin up to 100 percent, people have been seeing with blheli 32.9, it looks like we are due for a new update.
Does the flywheel's moment of inertia torque equal propeller resistance at 50% throttle or 100%? I'm wondering if the startup torque with the flywheel does not represent the real start up torque of a propeller, and that could be causing some of the extra hesitancy to deliver amps. Also, it would be interesting to see your take on the optimal graph line. What graph line would be realistic for an ESC firmware to achieve.
Hi Chris, thanks for the work you do. A couple questions I haven’t seen videos on from anyone: 1) what is the real world cost to using a motor output limit instead of having the exact kv that you want? How much more inefficient is it? 2) is there a correlation between kv in a given range for prop size, and how high you can push gains? For some reason I have been able to push gains higher on lower kv motors. Is this a good thing? Hope I’m not out of line posing these questions here. Your skill set would be perfect for answering these questions. Thanks again.
Now that you fixed tuning by dragging it out of that "magical mystic art"-realm and turning it into a systematic step by step approach... you're fixing the esc firmwares for us. Love it, this contributes so much! (my 5.5 tracks critically damped at around 4ms vs. setpoint
@ChrisRosser great measurements, I would be curious how the ESCape32 stacks up against the bunch. I have the impression it is very well on torque. Any insights? Thanks!
Hi Chris. Do you recommend switching to AM32? Or I have to go with "if it works don't touch it"? My 4in1 is supported by AM32, but I'm scared of burning one esc or something else that I would regret. Thanks.
Maybe I missed this during the video. But was there a reason to not compare bluejay/blheli_s with the others? Would be interesting to see that as well, since nowadays the decision is still very much blheli32 or blheli_s (with possibility to flash bluejay)
The graphs of the two are directly comparable. Sorry I didn't think to plot them all on the same chart. In general I would recommend bl32 over bluejay for the variable pwm and because the ESCs with BL32 tend to be of higher quality and rating in most cases.
Would comparing current within 8500-11000 RPM, the (lower) efficiency of BLHeli vs BlueJay - where the former is peaking current while the latter is already at max torque - infer that BLHeli is way over-thinking lower-RPM protection interventions, but not smoothing the transition once removing the interventions, ie., the main program playing catch-up might be causing that wasted current peak trying to ramp torque? Still, BlueJay seems to ramp torque efficiently at 5500 / 8000 RPMs. From a (cruise-to-far-spot-and-do-your-thing) range-efficiency or cruising 5/6-inch cinematic POV, is pursuing 32-bit / G071 gaining any over decent (old) tech/profiles available on BlueJay equivalents?
Whilst pretty graphs are nice to look at I would be interested to compare the off throttle control of a drone one blheli_32 to an am32 drone considering it seems that blheli_32 seems to generate more torque are low rpm.
Do you have the test code still? Any help in flashing this since the bl_heli32 company decided to 🤬 the community over as far up and as hard as they could?
Perhaps it's worth identifying a few ESCs with very different dead time and seeing how differently they perform on the test stand. Any suggestions for ESCs with large and small dead time?
Very interesting. I think the closed source ESC developers should be contributing to your channel or buying you lots of coffees for these videos, not penniless hobiests. Pretty sure they're all closed source apart from BLheli S.... Closed source in that they earn money from licensing anyway.
I never let facts and figures get in the way of flying! That said if we can spend some time and make a few changes to make things fly even better I'm 100% in.
Fantastic work Chris, I'm really glad you're helping so much in bringing more science to our hobby and improving it to everyone. Thank you!
I'm a simple man: if a new Chris Rosser video comes out I smash the like button
Fantastic, Chris! Thanks a bunch for the testing and comparison! 😃
I really miss Drone Mesh's tests. I know many people contested his methodology, but at least he was trying!
Anyway, stay safe there with your family! 🖖😊
Chris I'm so happy we have people like you with your knowledge and expertise to make our hobby better for the future. Thank you Sir!
I live in the low-to-moderate range. I'm eager to see what you and the BlueJay devs will come up with. It's pretty sweet as-is, truth be told, at least for my set up. Bring it on, Wizards!
Yes, awesome work providing consistent, confident data that can't be ignored, (already-purchased) products improve, which generates value to all us consumers, while closes competition to a point that will make all FW a viable tough compromise, broadening FW choices to delay the HW churn or cover corner applications, and esp on budget to standardize on a HW platform.
Fantastic work! please consider comparing JESC vs BlueJay
Great stuff Chris. I'm up to 56.7 per cent understanding now 😂
Very interesting comparison and result Chris,
many thanks for your efforts💪
Very good job, my friend. Can't wait to see this be implemented as an improvement in the code as a result. Along with this random spin up to 100 percent, people have been seeing with blheli 32.9, it looks like we are due for a new update.
Could you please overlay BLHeli_32 with BLHeli_S, since that's the choice most of us have to make when purchasing?
This!!!
Does the flywheel's moment of inertia torque equal propeller resistance at 50% throttle or 100%? I'm wondering if the startup torque with the flywheel does not represent the real start up torque of a propeller, and that could be causing some of the extra hesitancy to deliver amps.
Also, it would be interesting to see your take on the optimal graph line. What graph line would be realistic for an ESC firmware to achieve.
Hi Chris, thanks for the work you do.
A couple questions I haven’t seen videos on from anyone:
1) what is the real world cost to using a motor output limit instead of having the exact kv that you want? How much more inefficient is it?
2) is there a correlation between kv in a given range for prop size, and how high you can push gains? For some reason I have been able to push gains higher on lower kv motors. Is this a good thing?
Hope I’m not out of line posing these questions here. Your skill set would be perfect for answering these questions.
Thanks again.
Now that you fixed tuning by dragging it out of that "magical mystic art"-realm and turning it into a systematic step by step approach... you're fixing the esc firmwares for us. Love it, this contributes so much!
(my 5.5 tracks critically damped at around 4ms vs. setpoint
@ChrisRosser great measurements, I would be curious how the ESCape32 stacks up against the bunch. I have the impression it is very well on torque. Any insights? Thanks!
This is pretty neat stuff, I run the FETTEC stuff and have always wondered why this esc is so aggressive/unreliable :)
Hi Chris. Do you recommend switching to AM32? Or I have to go with "if it works don't touch it"? My 4in1 is supported by AM32, but I'm scared of burning one esc or something else that I would regret. Thanks.
Have you considered trying a VESC and seeing how that performs. Also might be worth trying the crawler mode from AM32
Been waiting for this.
Hello, I have a beta 95 and it has blheli 32, you can flash a blue Jay. I can't find information on this topic, greetings and an excellent video
Excellent video
Love your ability to geek out to the Nth degree. Feeds my soul hahaha
Maybe I missed this during the video. But was there a reason to not compare bluejay/blheli_s with the others? Would be interesting to see that as well, since nowadays the decision is still very much blheli32 or blheli_s (with possibility to flash bluejay)
The graphs of the two are directly comparable. Sorry I didn't think to plot them all on the same chart. In general I would recommend bl32 over bluejay for the variable pwm and because the ESCs with BL32 tend to be of higher quality and rating in most cases.
@@ChrisRosser ah good to know. Thanks for the clarification :)
Would comparing current within 8500-11000 RPM, the (lower) efficiency of BLHeli vs BlueJay - where the former is peaking current while the latter is already at max torque - infer that BLHeli is way over-thinking lower-RPM protection interventions, but not smoothing the transition once removing the interventions, ie., the main program playing catch-up might be causing that wasted current peak trying to ramp torque? Still, BlueJay seems to ramp torque efficiently at 5500 / 8000 RPMs. From a (cruise-to-far-spot-and-do-your-thing) range-efficiency or cruising 5/6-inch cinematic POV, is pursuing 32-bit / G071 gaining any over decent (old) tech/profiles available on BlueJay equivalents?
Very cool testing.
Beautifully illustrated Chris. But I’m just curious why no KISS ESC’s in your testing?
Really awesome stuff. Is this new BLHeli_32 testcode available to download and install yet? At my own risk of course!
If you take a look in the GitHub there is some test code for a certain ESC Mamba 55A 128K 4110 but not for all types yet.
Whilst pretty graphs are nice to look at I would be interested to compare the off throttle control of a drone one blheli_32 to an am32 drone considering it seems that blheli_32 seems to generate more torque are low rpm.
Do you have the test code still? Any help in flashing this since the bl_heli32 company decided to 🤬 the community over as far up and as hard as they could?
super interesting results!
Anyone know how BLHELI 32.8 compares to 32.9?
(That new test code looks promising!)
Well done.
Ummm.. Very interesting. I'm still using the old firmware 16.9
Thank you
I wonder how much deadtime plays into this.
Perhaps it's worth identifying a few ESCs with very different dead time and seeing how differently they perform on the test stand. Any suggestions for ESCs with large and small dead time?
@@ChrisRosser I dont, but I would also be curious how esc settings play a role and if different motors just work better with different settings etc.
Very interesting.
I think the closed source ESC developers should be contributing to your channel or buying you lots of coffees for these videos, not penniless hobiests. Pretty sure they're all closed source apart from BLheli S.... Closed source in that they earn money from licensing anyway.
AM32 is open source too
I think BLHeli_32 and FETTEC are the only ones who are closed source
👍
First
Imagine not being able to enjoy flying because you're so concerned with the numbers.
Some of us enjoy the technical side of the hobby more than actually flying
Imagine profiting from awesome flight performance for free because of people looking at the numbers.
There are people who actually enjoy numbers more than flying. And there are other people who enjoy building drones but don't care about flying them
@@FiorixF1 and then there are people really enjoying a perfectly tuned quad while flying ;P
I never let facts and figures get in the way of flying! That said if we can spend some time and make a few changes to make things fly even better I'm 100% in.