ImpulseRC Apex Vibration analysis. Find out why the Apex is so good to fly. 

I learned that we are going to see a bunch of copies of the apex arm design.

I like in the comments how people appreciate the content but like a good student I like answering or atleast trying to answer the question he asks. I really like the way ignite curiosity by asking a related question on the topic , you must be a very good teacher.

It could be intentional, there is also a lot of pre loading on the carbon fibre parts. The rear standoffs are slightly shorter than you might expect so the top and bottom plates are flexed toward each other. The countersunk stack screw heads sit slightly proud and cause flexing of the sandwich plate, they probably generate heat through friction too. I was concerned about the screw heads when I first assembled an Apex and considered counter sinking them more but a friend who was involved in the development phase told me it is designed that way for a reason. The arms are so strong on the Apex that I’ve broken the steel bolts leaving the carbon fibre intact and I’m not the only person to notice this, so spare hardware is probably more useful than spare arms 😂 Great video, look forward to your insights on motor choices 👍

I swear I will hunt down any pilot who down votes this video. God the knowledge he provides is amazing.

What an incredible amount of analysis on a drone frame. Most just build, crash, repeat and would never even spend 2 seconds thinking about this stuff. I'm glad this hobby has people going this deep.

Thats funny, i tryed a Apex to see if there was a difference from my other quads and i was amazed how good that thing flew, i could not figure out what it was it was just more locked in, so much that i switched all my other quads out with the Apex. So now i know it was not just me. Thanks

Quality equipment goes a long way. Maintaining that equipment is pretty important as well. Awesome video brotha man!!! 🦄🦄🦄

Excellent analysis and presentation of your results!

Haven't seen your videos before and I've already learned a bunch. Those mode visualizations are fantastic. Can't wait to dive into your other videos!

Love the content Chris, especially how you break things down into understandable chunks. It would be great to see your analysis of the Glide frame 😁❤️👍

Great video Chris, it really helps to understand all the processes and actions a quad goes through in flight. The more i know the easier it is to troubleshoot and find solutions to problems. Keep up the good work, the content and the way you explain it to thickies like myself is top notch.

Incredibly informative video, hope other frame manufacturers try this out in their frames too.

Excellent video man. Clearly you have education in some type of mechanical engineering. Excellent analysis, and excellent visualization. This is probably the most educational fpv I've seen in years. Can't wait to look at your other videos. I hope you have a vibration analysis on the armattan chameleon

just logged my Ummagawd Remix for the first time, its a 2 year old build and has always taken a tune well. Now i know why, amazing resonance characteristics, thank you for the tutorials and check out that frame if you want something interesting. Id love to see an in depth analysis of it

This is very informative bro! Keep up the videos! I am very intrested in your next topic

Brilliant video! Never knew mini-quads and acoustic musical instruments had so much in common :) Although with opposite goals, trying to damp out all the vibrational modes rather than maximize and balance them.

Awesome Chris, thanks for this video 👍🏼

Awesome video Chris!

Very cool videos and analysis !! Good engineering insights!!

I think they added a press nut screw so the arm can be fixed with more than one screw but the user doesnt need to remove the stack screws as that particular screw doesnt serve that purpose.

Hi Chris, very interesting and very well described and analyzed topics. I wonder which mod would add coulomb damping. For example, adding a small layer (such as a gasket in a soft material such as tpu) between the CF layers would add Coulomb damping or introduce even more vibrations into the system. A video might be interesting in which you give some suggestions on how and where we can add more damping in any form (both in the design phase but also to improve an existing frame). I was also wondering if the cutouts in carbon plates (used primarily for weight reduction) might play a role in vibration propagation. Also it would be very interesting to analyze a deadcat style frame. Does this frame geometry create more frame resonance? Finally, I always wonder how much the vtx antenna is responsible for vibrations. In fact in your simulations the antenna is never taken in consideration, such as other "free" components like wires. Thanks, have a nice day 💪💪

Whoaaa we need ppl like u in our hobby ❤️❤️

Thanks for the amazing content. I think I have never learned that much on youtube, in such a little amount of time.

Great break down. I wonder if the design was a happy accident. Or maybe from test pilots trying different setups over the years. Great video. I learned something today.

The level of detail in these videos Is great. Found this channel on a discord server and glad I did. Learned a lot from one vid can't wait to watch the rest

After watching ur videos.. my frame must vibrate like crazy. I need to put the arm braces that came with it

quality content

I've been using tape and foam in different places between the arm and frame to absorb vibrational energy. I haven't done direct 1:1 controlled comparisons with/without, but it seems to work (on similar builds).

Another great piece of design in the Apex is their "Through bolt" as you call them are proprietary and etched-thin at the top and bottom of the arm... thus on a bad crash on the arm the bolt is more likely to shatter at those points and not rip either of the sandwich plates; after 15 months flying Apex I've broken 5 of these bolts and only one Sandwich plate and never an arm (though I very well should have)... never knew about what you're describing in their video but this "etching" could also allow for more coulomb-dampening due to the "through bolt" being less rigged because of etching perhaps? (i mean, in my thought experiment atleast).

If you're considering any other frames to analyze I'd vote for the hyperlow airshot. It seems like it may benefit in similar ways, but I've never seen any logs so it's just a guess. And also the mad kwads frame because they're the only frame company I've heard even mention vibrational analysis as being a driving force for design choices. Great job. Your coulumb damping explanation was great and the drawing top notch but I feel like a great animation and maybe a little more detail or depth on the friction/bending/compression/stretching aspect could have really driven the concept home for everyone.

Really interesting videos. Just cut my first frame design. I should've tried this

I don't know if this was by purpose but I know that my Apex's (I have 2) are my best frame ever for stability and durability. I tried a clone (shame on me) that doesn't reaches his anckle. ImpulseRC did a great job with this frame (thx to Mr Steele ?). However, once again a great video, thank you Chris :-)

I don't have one but the knowledge is my wisdom

Wow, cool video subbed! BTW, I've read that when arms and rear bottom plate are fully assembled, it does a subtle bent in this plate and create a little gap. Maybe this is some kind of material damper absorbing energy?

second for sure :) that was not with intention to lower vibrations :D Amazing content for tech savvy pilots !

Could be wrong but I'm not sure if impulse would have the same test equipment to be able to dermine that the arm friction dampening mechanism provided good result and so was a sole design reason? Could it just be a co-incidence? However, by having seperate countersunk bolts on the upper plate, they certainly did chose to decouple the flight controller somewhat by choosing not to have the flight controller mounting bolts going through the arms.. which would transmit a lot of vibrations directly to the FC. Most other manufacturers would typically try to save a few grams by incorporating the arm bolts into the FC without considering the bigger picture of piping vibrations directly to the FC. I would love to see you do a test on using soft motor mounts and maybe also have a look at the iflight Xing motors which use a rubber o-ring between the bearings and motor base to reduce vibes... its the first video I've watched of yours, so maybe you already have. Very well done, I love this sort of stuff. You just got a new Sub, :)

If you read the ImpulseRC web page they specifically mention the shoulder bolts as being unique to this frame. I guess with a shoulder bolt you get a much tighter tolerance fit between the arm and the bottom plate as apposed to a fully threaded bolt? Nobody's talking about the shoulder bolts. I thought it sounded like an interesting point to think about.

Great video. Do you balance your props before analysis? Do you think this would have much affect on results?

Chris I would (and I'm sure others would find it interesting as well) like to see a video that shows the process you use to do the vibrational mode sims. Would you please make one?

ImpulseRC use partially threaded screws intentionally to get tighter tolerances on the screw-arm interface and a tight fit to the centerwedge where other manufacturers use threaded screws, maby that factors also into the better dampening/ressonance vs other frames

i believe the skids under the motors also do their job for. dumpening vibrations from motor screws

Very interesting especially your drawing examples I wonder if that was pre design with that thought or just dumb luck

So it's unicorn magic indeed ;D I think learning from mechanical engineering it would still be better to make the frame as stiff and lightweight as possible and thus moving resonances into higher frequencies, break major resonances down and reducing their amplitudes. I think even those triangular design-arms could really need one more arm to make it a pyramid shaped motor-mount to achieve this. I would be so curious how that translated into a drone. Also so far attention is only given to the center of weight in terms of front-heavy or back heavy. But not in terms of top-heavy or bottom-heavy. Many of those modes which would normaly not turn into gyro-noise but only accelerometer noise (straight movements instead of angular movements) can still cause the quad to all of a sudden get gyro-noise if center of weight is off-axis. And those resulting movements out of a different mode are usually much harder to filter as they don't show up as smooth sinus-waves. (I work in the construction of checkweighers and there torsion modes actually turn into linear weighing-motion if the conveyor-motor sits off-center.). I really have a quad frame in mind that would incorporate these 2 effects - shifting center of weight down and allowing for a pyramid-shaped motor mount... maybe this should be a approached further...

Learning from you. I got apex cean new build, but it shows resonance at 180hz cca. Don't know where to start searching for cause

12:50 - I noticed on my apex, the standoffs and top sandwich plate, were almost 1mm short. So did not meet flush with the top of the arms. The through screws pulled this plate down, but leaves it in a arc which you can see looking in between the two sandwiched plates, there is an obvious gap above the arms. I have 4 apex frames, all are the same. This lead me to two things. One, all the complaints of burning those kiss 4 - 1 escs? As the top sandwich plate is flexing, This leads to the stack screws being bent outwards or inwards when arms are pushed up and down. Next thing i noticed, You can physically move an arm up and down and it shifts permanently. Like a short leg on a table, You can press down on the arm again to shift it back into place. You can see this by placing the frame on a table, Press one the legs up or down. Now back to flat table, you will see the frame has changed shape. Now as much as the resonance on this frame is great, these 2 factors leave me wanting and hoping the next Impulse frame sorts this out. The reverb before this was just perfect. On another note as well, the apex is the only frame, where every fly day, i have to tighten up the frame. All bolts have come loose. I can understand the absorption of the vibration would probably be aiding this. I'm gonna use better lock tight, but would be good if impulse also included pre lock tighted bolts, or a small tube of lock tight. I do not think it was intentional. I believe it was done do allow for the 20x20 and the 30x30 stack screws. You cant offer stack screws when arms bolts are in the way for 20 x 20. As well as the wolf pdb then wont fit as it sits flush on the plate.

Does that heat cause a change in it's coefficient of friction and so dampening effect? cold day or warm day?

Hi, I know it's too much to ask but could you check Chimera7 frame as it's one of the most used with AOS for long range and seems to be very low noise. Cheers. Your tutorials helped me a lot to transition knowledge from planes to quads.

Intresting conclusion, I have a diffrent theory, as the Friction force would be negligible. I do not think there would be any movement, I mean maybe a tenth or hundredth of a millimeter. I think the key is rather a combination of things. Firstly the interlocking arms, a keystone piece in the middle , two screws in the arm instead of one, nuts imbedded in the arms (this one I think is the most brilliant as carbon fiber can not be threaded well, and therefore must have wider holes than ideal), and tight tolerances for a very tight fit of every piece. These little factors is what I believe add up. I am sure the friction exists, although laminate on laminate most likely has a relatively low coefficient of friction, but the Normal force is most certainly quite grand. Simply the fact that two plates can slide next to each other when they are bolted together a few centimeters away seems unreasonable to me, especially considering they all connect together at the middle of the frame. You know I actually am so curious now I want to make an experiment, I will make both screws just go halfway and see if the black box logs are diffrent. You could also test this in your simulation I suppose. In fact I was meaning to ask you, what software do you use for these simulations? I am extremally happy that your channel is as awesome as it is! I love this content, you are a much needed "UAV Tech" of vibrations. The hobby is just that much brighter with you in it. I apologize if I sound harsh in any way, that is truly not my intention, but I know that tone is hard to judge when behind a screen.

On their website, they mention the keystone as a part of the flying performance. So I would say it was kind of intentional ? I think they were maybe looking for robustness when they found out a way to contribute to the vibrations at the same time... Now, is there a way to use that knowledge with existing frame design and convert them to magic unicorns? Is there a way to add damping to a carbon frame with easy modifications ?

Hi @Chris Rosser, with some of your other vibration analysis videos you have provided insights or recommendations into how you'd tune a particular frame (for example the video on the QAV S JB edition) do you have any recommendations for the Apex? - also was this for the regular Apex frame or the Apex HD?

Awesome video. Have a odd question - i have a cheap Chinese clone of the apex. Do you think it performs the same as the original ignoring worse tolerances and lower quality materials? Thanks

happy accident - great designs from Impulse RC since warpquad days

Hey Chris! I watched all of your videos, they are insanely informative and I really love the work you do. I like to design and manufacture my own frames, so I can tailor them to my needs. They work fine, but I'm having issues with vibrations, as I have never run vibration analysis on them. From a designing standpoint, what would be the workflow, to design a frame that handles vibrations best? The majority of vibrations (if not all) comes from the motors. So my question is, is it a valid idea, to somehow measure what kind of vibrations does the motors generate, and then try to design a frame that tries to suppress these frequencies? Or this would be pointless as each motor has unique vibration frequencies? Or even if you change the props, these frequencies might shift, so your design would be pointless? What would be your thought-process or workflow in designing "the perfect frame" purely from vibration standpoint? Thank you, if you take your time to answer my questions, and keep up the good work!

Great and informative video! Could you possibly do the same for Armattan Marmotte/Badger frames please?

Hello Chris i would like to ask you about your graphs. All graphs that you provided into the video are having x axis limited from 0-1000hz. Does that mean that higher frequencies are not that important for drone flight? Because i'm imagining this system to fly on frequencies much higher than 1000hz. #question

Great content..tnx..i have one quick question,which frame is better than these two: Ummagawd moongoat or armattan chameleon ti?Frame with separate arms or unibody frame?

Please do an analysis of the Source V!

Please test the CatalystMachineWorks BangGod and Shocker Tank frames. They are so unique let’s see how they compare to the more simple design of the Apex. Please!!

And with all this insight and ultra in depth viewing he will still say check motor temps. Reduce filtration follow a good strategy check motors. Can you reliably log this as a given cos different flight controllers produce different noise. Same frame different flight controllers different readings. The theory is one thing but it will come back to checking motor temps for feel.

How can I get resonance mode for my frame? Do you use blackbox explorer for that? Thank you Chris!

Hi Chris! What do you tjink of the Catalyst Machineworks arm lock design? It seems to me that it could probabily act similar to the apex. I have 2 Cmw frames and the vibration performance is great

Chris, while the choice to not run the bolt through both plates on the inner arm mount may seem like a happy accident I think it was deliberate. The keystone is what makes the design work, as without it the arms have too much movement and the design hurts more than it helps. The keystone also introduces sliding friction at the tip of the arm on the vertical axis, and I think acts similar to an anti-roll bar on a car in that it couples the arms together and total deflection of the arms is reduced. EDIT: also, the Moongoat uses a similar design, the inner most mounting hole on the arms doesn't use a through-bolt, and butts up against the other arm in the middle. I wonder if it has a similar effect to the Apex.

Hi, you have amazing videos, can you do this analysis for the Vega gravity frame?

Could you effectively dampen all quads by simply using a viscoelastic dampening layer between carbon fiber plates? I am thinking of the same type of material that is used unconstrained in aircraft fuselages to assist in active noise cancelling and as well used to dampen buildings in shear layer constraint for earthquake dampening?

Do the banggod next! I believe it was not intentional, but a side effect.

If you notice the fc mounting screws don't go through the arms like some other frames. This was all intentional as it has been know to be problematic for quite some time. This is why I moved away from unibody designs a while back. No secrets discover here.

So was the dampening built in by accident or design? Do you think Apex having 7” arms at 5.5mm thick will have too much vibration or resonance?

What if you didn’t have a through bolt and instead had two press nut bolts. One going top to bottom and the other going bottom to top?

Do you have any filter suggestions for the apex by chance?

Dear Chris, I was referred to you by Bob F. Here is a question that I bet you have an answer to. I am an experimenter. So I sometimes do silly things like trying to make a FC stack waterproof by submerging it in 3M dielectric FC 84 as a means to liquid cool it. Even with Heat pipes and Radiator....Why? For grins. HOWEVER, for no good reason, I am seeing upon arming all is good. Any input to yaw, pitch or roll, gives motor response as expected. If I use any throttle, the motors ramp up without any other input and other inputs are then useless. I had to do an emergency landing/crash but retrieved it. I thought maybe I did something wrong with wiring etc ( not that submerging FC under fluid is a good idea....) but the cooling system works well. That being said, would a dielectric liquid like the Fluorinert cause an aberration in the FC harmonics enough to create a maelstrom that the filters could not handle? Is there anything I can change in Betaflight to allow for the algorithm to see through the fluorinert influence if that is what is causing an unsolicited runaway type motor effect? I sent a video to Joshua Bardwell this evening on this very issue. Thank you for any insight and your expertise is welcomed! Thank you for your consideration of my query. If you are in contact with Joshua, please request from him the link to the video I sent him or he can also forward to you my email so I can send you the clip. Thanks again.! Very Best Regards, James

Or would the fluid sloshing dampen the vibration?

Maybe u can input vibration data to your analysis to find how long the period of mode shape. If periode is to long maybe the natural frequensi of the frame same as frequensi of vibration on a motor it will cause big resonance.

I know you did the QAV-S JB edition but can you do the QAV-S Special Edition Johnny FPV?

i have apex hd kit and 6inch arms... i cant get my build to function properly, no matter what i do... i get HUGEEE oscilations and stutters on throttle punches. I think i exchanged all components other than vista,nano rx and FRAME... its kiss/fettec build with f40 pro4 1950 motors (limited to 80%), tried different mounts (soft,hard,different soft)... always the same problems... Frame seems good, but i start to suspects its his fault for my almost unflyable quad... If you have time, i can show you few of my logs, so you can give me an opinion

I wanna see a analysis of a gravity loss FPV pro 2

Are you taking into account a good quality motor vs a poor quality motor in these BB readings? What happens if you put a good motor on an average frame and viceversa?

can you do one on the glide 5" frame please

Can you make a tutorial how you do the mode analysis?

Is it possible to get you to check out the source v

ImpulseRC propably used press nuts because press nuts look nice which is attractive to customers. But they can wear out and when they do it is much easier to fix if the press nut is in an arm instead of the bottom or sandwich plate. Once they did tests they found out that a design with only one screw through arm and both plates had its advantages. So I don't think, ImpulseRC specifically designed it with the intent to dissipate vibration energy but realized after the fact that it had desirable characteristics. That's my guess of how they ended up with this design. tl;dr: Somehow on purpose but through trial and error. P.S.: Could it be possible that this specific quad had lower electric noise than the other quads you tested? Gyros tend to pick up a lot of noise from the ESCs which leads to noisy spectrums and the PID controller reacting to phantom rotations.

+1👍 💪

I can’t get rid of vibes min my micro apex. Caddx peanut is full of jello and Caddx polar even has vibrations. Brand new build all new parts.

Now, theory is one thing, simulation also and experiment is the missing piece here. Thermistors on both sides of the friction damped arm while pumping the throttle with unbalanced props would help confirm your reasoning. Anyone has a frame lying around? I'd ship it back after. I'd bet they didn't think about these dynamics in the design stage. They would have made marketing around it, it's a great idea. I'll try it on soft mounted cameras with the right materials for the first harmonics.

I really don’t understand why with single bolt it sliding but with 2 screws is bending

Speaking of material dampening, I wonder if something like the hyperlite frames by pyrodrone will have different characteristics due to the alu bottom plate.

What if on top of taking advantage of the coulomb damping on the apex, we also used Nyogel 767a to further dissipate vibrations. Would this work, or would we be approaching the point of diminishing returns?

your coulomb dampening hypothesis seems to ignore the fact arms are buttressed against each other which would prevent any sliding.

apex ftw

15:00 - no, of course, it was a mistake. this is FPV.

I think it was Intentional..I have never flown an impulserc frame, but I know that they take their frame engineering very seriously, and build on what was cutting edge in a previous frame.

I’m going to say it was unintentional because the 3” and 4” versions of the Apex don’t use the same design. Cosmetically similar but functionally very different design than the 5”.

I would love to see TimmyRc's tubular frame tested. Hint hint

I hundred percent believe that's an intentional design to avoid the vibration.

Looking at the atrocious aerodynamics of the mainstream quads it can't be anything but a sheer coincidence that this form of damping has made it's way into that design. If people were paying any attention to the aero things would look more like the Moka Gravity but with tilted rotors or the DJI fpv which is a masterpiece of design in this regard.

It's great frame it's such a shame that there is NO motor protection, You just keep smoking motors in a crash, if it was not for that I'd fly it .. *( I used to fly it )

11:35 - I'm sorry, but I watched this segment multiple times, and couldn't grasp your idea of what moves, what deforms and how it's the "Apex" case where the arm slides more. I can see how the arm can slide between the two plates: all the holes the bolts go through have some thread-hole clearances. If the clearance (D_hole - OD_thread) is e.g. 0.2 mm, the arm can slide as much as 0.4 mm along its own main axis in the top drawing, but only 0.2 mm in the bottom drawing (because the press nut has no clearance). But you're talking about bending... Well, when the arm bends downwards, there's no difference, as the upper plate will be bent by the outer bolt, pivoted on the arm's inner end, so the length of the lever is virtually the same. When the arm bends upwards, the upper plate, which now has a much longer stretch that's structurally elastic, will be bent only by the outer bolt tilting and pressing it in, but... my napkin geometry fails me here, but it's either the same radius, so it doesn't slide against the arm, or a larger radius, and it detaches from the arm's surface as soon as they start to bend... OK, hand-wavy reasoning and Paint pictures just don't work with this. We need to either get a civil engineer who can apply his materials mechanics 402 knowledge and evaluate the two configurations using differential equations and shit, OR do an experiment: get an Apex frame (or, hell, a quarter-frame), ditch the press-nut, drill through the whole thing, install a (larger diameter) through-bolt and compare it against the original. Otherwise, I'm not convinced a single bit. This is quite an unusual frame. Its excellent vibration properties can come from any one of its features or all of them together.

Any one else in here looking for Steeles comments 👀 haha

okay, so quad arms should be two arms bolted together and not one laminated arm.

Just make the frame out of Jello

high resale lol