QuadMcFly

I saw your post on Facebook! Nice work on this. This lines up pretty well with what I'm thinking. I bet if you measured actual degrees if rotation you'd find the stretch axis turned less for the same amount of time rotating than the shorter axis. Thanks for trying it out!

Awesome! Testing is exactly what we need :D My suspicion is that LOS and FPV pilots are experiencing two different aspects of this. LOS from what I've seen tends to use much more rapid, sharp movements where the faster angular acceleration will play a primary role. FPV on the hand tends to use slower more controlled movements where the angular acceleration is a smaller portion of the total time in rotation, and slower angular velocity plays a larger proportional role.

Also I chatted with Mark Spatz (UAVTech) and he confirmed my thoughts.

I just chatted with a couple racer buddies and they confirmed my thoughts. Stretch X is definitely tamer on pitch than true X.

It also doesn't make sense to me that a racer would want faster rates on pitch. Philosophically you want pitch to be more stable not less stable.

Interesting, could be I've been suffering confirmation bias for three years or my style uses sustained rotation more than small moves where the acceleration would have a larger impact on the total time spent rotating. I definitely feel higher precision on the wider axis. I guess in my head what I'm feeling is the difference between trying to rotate an object in AutoCAD with the handle close to the center of rotation vs trying to rotate it with the handle moved away from the center of rotation.

In terms of what the pilot feels on the sticks anway.

Given that a motor can reach full RPMs in approximately 100ms, it seems that the resulting rotation speed in DPS is more important than how quickly it gets to that DPS.

Essentially at a static stick deflection, the axis with the extended distance will move slower than than the same static stick deflection with a shorter lever arm.

It makes sense that rotational acceleration would be faster, but not the actual degrees per second once equilibrium has been reached.

Can you explain what your claim is? In every experience I've had, a longer lever arm results in a slower rotation rate on the axis that is being extended. I have never experienced a longer axis making it feel faster on that axis in 3 years of frame design and prototyping.

Hey I moved this to chat so we can talk without cluttering the thread. I'd like to explore this a bit more.

You're right. I wasn't taking the distributed moment of inertia into account initially. This is true of acceleration but not rotational velocity though. Velocity is still lower for a given linear velocity, but it will get there faster.

They may not be rated for it, but there is a hardware limitation for how much they can create based on the prop pitch, the motor torque, and the Kv.

It's not a question of torque, it's a question of velocity per degree of rotation. The motor is inherently capable of a certain rate of change and max velocity. As you increase the distance, the motor has to travel more distance per degree of rotation. Hence, as you increase the lever arm, the rotation rate slows. The torque increase (less force required to hold against outside influences) and the effective higher resolution result in more control. I'm writing a more complete answer below.

Actually this is exactly backwards. Because of the longer rotational axis the pitch axis on a stretch X quad actually takes longer to change than a shorter distance would. What it allows is more precision of control on the pitch axis, which is why racers like it. It also means that when you're going very fast, small changes in stick movement for control are less likely to result in large changes in altitude. Which when you're pitched forward at a 85 degree angle is a very real problem :D

I mean that's pretty much been my entire relationship with Stack Exchange untill recently! Hah

not creators

Consumers

It's also worth pointing out I think a lot of the "users" of this SE will be invisible to us. Traffic only, but not user accounts or contributors.

Based on my experience in the drone community, I would say the DIY, racing, freestyle branch outstrips the research and development side in terms of size and engagement by an exponential factor. But there's no reason we can't accommodate both. As @RoryAlsop says, there's tools that come for that specific purpose.

There are a lot of different expectations, so we just need to try to balance it with what will actually work

That was mostly the point I was trying to get across.

Well google will pick up on the right threads based on the questions being asked

and the general community expectations follow it

But I would say the reason we had the massive response of new users we saw during the initial phase of trying to reach beta is centered on what I'm talking about

I think that if we tag thing properly there's no reason we can't fit both goals

I suppose whether this is the right fit or not is the question we're trying to answer with this Beta, right?

haha

For instance, that question about "why do my motors keep spinning up when I test on the bench", I answer that probably 5 times a week

the goal in setting up SE was that it would be searchable and easily accessible

which makes it hard to find answers after they've gone by

The support is pretty much community driven via facebook groups at the moment

not supporting a massive user base

they're interested in testing and developing new features

It's not exactly orphaned, there is TON of active development on the project, but it's all high level stuff

OpenTX is better

and BLHeli

I would say one major reason this entire stack exists is because of the lack of a dedicated support method for betaflight

As well as OpenTX

I agree we should avoid being too specific, but some intersection with betaflight and blheli especially will be unavoidable

And most of the flight control software now doesn't really have any dedicated support channels

Unfortunately the roll that FC software plays in general use for drones is large enough you can't really avoid it

I see it as a major reduction in the usefulness of the this stack

Unforutnately if you eliminate FC specific questions you've eliminate a MAJOR number of questions that are going to be asked, or at least eliminated a large number of answers to questions that will be asked

but they should still probably be tagged with version information

The few questions I've answered with specific screenshots are things that have stayed constant for quite a while

It depends, some of the core features tend to stay fairly consistent, but there is a lot of edge case features that tend to come and go.

Betaflight changes at a frightening pace

I think battery questions will be fairly consistent. But anything related to Betaflight DEFINITELY needs tags as to the version