Posted on 18 Comments

The Glide design and general frame design concepts

An overview of some concepts and considerations made in the Glide design.
This is all based on MY philosophy. That does not make it correct or incorrect…

Multi-unit vs. mono-body/single sheet base plate design:
This was a pretty major debate about three years ago.  After some design testing back then, I came to a pretty simple and straightforward conclusion.  Everything breaks.  Literally everything breaks.  It does not matter what it is.  You can chase that bullet proof carbon dream but it will still break.  Furthermore, in order to strengthen your single sheet frame body, you’ll need to fatten up all the lines because you don’t want to deal with transfering all your parts to a new frame too soon.  A multi-unit design can be made lighter with even more durability because you have a third dimension to work with.  You can sandwich plates and take advantage of various thicknesses to do as you please.  The multi-unit design is also way WAY cheaper to manufacture.  About 75% less depending on a bunch of factors.

So multi-unit is often lighter, way cheaper, maybe stronger and repairable.  Additionally, I can do some real testing on multi-unit designs because I can keep breaking the same part over and over to manage the fracture zone.  That would be really annoying and time consuming to do on a mono-body design.  Of course it’s up to you to make up your own mind but to me, it’s not worth giving all the benefits of multi-unit just to make the initial build maybe 10 min faster.

Battery on top:
This is a highly opinionated topic however, if you fly an acro frame with the battery on top after flying a bottom mount racing frame, you instantly notice a difference.  I consider optimal performance to be when the overall CG/CM of the build is right in between the props in all the X, Y and Z planes.  If the CM is not in this point, the FC will need to compensate by managing a swinging weight.  This causes control feel loss since the quad must expend energy on just stabilizing the swinging weight wherever it may be.

Disk Unloading:
The concept I present here is not anything real.  It’s merely something I’ve come up with from a few observations.  If you check thrust tests on the bench, you’ll notice the amps are quite excessive.  This is because when flying in the sky, the props are not pushing against a flat wall/table.  They are free to unload their thrust and rev up and down as needed.  This results in dramatically lower amp draw.  From this we can deduce that any solid, immobile object (like the frame, battery, HD cam…) around the props thrust path will result in higher amp draw due to a more static pressure inducing environment that the motor now must work against.

There are some frame designs that place the GoPro and battery right between the props on a low deck under the prop line.  Thrust coming off a prop is not a perfect cylinder.  It sort of spins out in a bit of a cone shape.  The battery/GoPro or anything nearby the prop disk in the thrust path will result in a partial static environment because the thrust is bumping into it and cannot flow freely away.  This results in the motor needing to work harder to overcome such issues.  This impacts flight feel.  The first thing you might notice is a rather dull yaw sensation or more forceful stick input required to execute pitch/roll movements.  BUT!  This issue is highly dynamic for every design.  Some frames that move the props farther from the body don’t cause enough of this effect to make a flight feel difference.

This is in part why the Glide has a build height taller than 20mm with mostly empty space under the prop line. Hopefully the top plate will be right at or just above the prop disk level. You can easily test this theory with any of your quads that have the same general structure. Take some wide tape and tape up the sides. Fly it and you’ll probably notice something is very different.

Of course skinny arms help.  Also, props pointing down are more effective however they’re just not practical.

A prototype with 20mm build height. Too tight for the Micro Eagle imo.

Build height:
I’ve tested this design and other similar ones with 15mm standoffs up to 35mm.  The difference between 15mm and 35mm is not huge but it is significant. It can be felt more if you have a heavy battery/GoPro on top but the difference between 20mm and 23mm cannot be felt and is purely aesthetic. We don’t really want to go below 20 due to the disk unloading issue.  Furthermore, 20mm standoffs struggle to fit a Micro Eagle camera.  Since I consider that the very best FPV camera, I want that camera to fit nicely.  It will fit inside 20mm but will be touching carbon with no buffer room.  This can cause parts of the camera to crack or get knocked off in a crash.  23mm for buffer and ease of use.

Multiple stack mounting platforms:
Making compromises to cram a stack into a 2Xmm height and still leave room for a battery strap on top is a bit ridiculous and makes maintenance challenging.  I hate maintenance more than anything else.  A laid out build with boards scattered and maybe just one board covering the 4in1 makes it super easy to reach most things without even taking the top off. 

The addition of the 20×20 platform in front was specifically for noise isolation.  We cram a whole lotta tech into a tiny space on these machines.  They don’t always play nice.  I’ve found the camera and FC to be the most sensitive.  I had been building my quads with the FC in the back for a while but I became frustrated with the power lines and moved everything up front.  Video noise is far better managed this way regardless of what components you’re using and how you set things up.

Arm interface complex:
It’s very logical and convenient to integrate the screws used for your electronic stack into the arm structure of the frame.  It’s also a great way to save weight on the frame design.  There are two glaring problems however.  First, vibrations from the motors are more directly transferred to the flight controller.  This isn’t really a big deal since vibration filtering has gotten so good over the years.  The second and much bigger issue is the aggravating annoyance of having your stack somewhat fall apart when you need to remove a screw to replace an arm.  If the arm screw is integrated into the stack, there are a multitude of ways to manage it however nothing is as simple and easy to maintain as just not having any arm screws anywhere near any electronic stack. You also really want to use just one long bolt for your main stack instead of the plastic standoffs that regularly break.  These are the reasons why the Glide has independant arm screws.

Taking it one step further, sometimes you’ll have a locking mechanism in the middle of a frame design that requires a nut or press nut or something.  If you’ve ever tried to reach the middle of a frame with an instrument to hold or fasten something, you know that this is a ridiculous stress inducing task.  The Glide does not even have a nut in the middle.  The arm screws goes directly into the standoff.  There’s nothing there to fall out, hold or manage.  Furthermore, the one screw that you do need to hold the nut to undo is totally out in the open and extremely easy to access.  I don’t use press nuts because they don’t always stay put and they’re also not standardized.  If you lose one, your day might be over but most people have a bunch of M3 locknuts lying around.

Lastly, it’s common for frames to key the arms together for added rigidity and structural support.  The glide does this and the fit is very tight.  Carbon compresses and keeps compressing over time.  That’s partly why some frames you’ve been flying a while get a little squeaky.  The first time you thread the Glide together, it might require a little pressure to get things in place.  Replacing an arm after won’t require much force.

Arm design:
This is one aspect I’ve paid very close attention to.  It may seem simple enough but in practice, you realize it’s actually quite complicated. I’m only using conventional arm designs but move to vertical or flexed plate designs like the Quad Star ARX and the challenges are far more complex. (https://quadstardrones.com/product/arx-r/)  Let’s just focus on conventional designs for now.  First off, the decision to go with 5mm over 4mm may not be the best option for weight efficiency.  This is because if the arm is 1mm thicker, the motor mount and body mounting area will also be 1mm thicker and that’s not really where you want the strength.  You can’t really narrow those areas excessively because they need to have appropriate surface area to function.  Additionally, the motor screws will need to be 1mm longer which further adds weight. 

Another point to consider is the delamination of the carbon layers in the arm which typically happens from a twisting force.  Wider arms resist delamination (in the middle, not the ends) far better than narrow arms but thicker arms do a better job at resisting delamination at the end of the arm from bumps and scrapes.

There’s also the matter of the arm shape.  There are two primary fracture points on arms.  One is at the motor base where the screw holes flow into the rest of the arm.  This is maybe not the most obvious weak zone but that motor base makes a 90deg joint with the arm so if the end of the motor gets knocked, all the forces are focused right at that joint.  The other area is where the arm flows into the body.  This is the big one.  There are numerous ways to manage this and I’ve applied the two ways I have found to be most effective on the Glide.  The arm tapers to a wider base where it attaches to the body and there are curved sandwich body plates that obscure the forces such that they’re not focused on one line.  My goal is to have the arm break somewhere in the middle.  Not at the motor mounting pad or the body interface.  That tells me I’ve done an appropriate job on the design but despite my best efforts, it still sometimes breaks in ways I don’t want it to.

There’s a lot more to this but most of my optimization has come from looking closely at literally thousands of breaks over several frame designs.  The community is fantastic at sending me pictures of my frames when they break and I highly appreciate it.  So I’ve gone with 5mm because I want the skinnier arm for reduced thrust resistance and I get the bonus of better arm end delamination management.  The body structure is wider and the arm is kept short to minimize the weight of 5mm thick carbon.  This also adversely makes the arms cheaper to produce so I can keep costs down.

Square holes:
I try to use rather large square holes in the body plates because they’re the most efficient at keeping strength and reducing weight.  Carbon sheets are commonly made of alternating layers of straight carbon fibers that are 90deg apart from each other.  The diagonal direction on the carbon is not strong.  Cutting designs or letters out of the middle is nice but wasteful.  Furthermore, I make sure to maintain a safety margin around all screw holes to avoid any of those breaking.

The lower brace has triangular holes because the plate is cut at a 45deg angle to optimize the strength in that direction. You can see the carbon fibers are running in a different direction than the main body plates.  There are access holes for the stack screws but these holes are not large enough to fit the entire screw head through.  If I widened this hole, the brace would lose a lot of strength because the fibers would be severed at that point.  Widening that area of the plate would not be an efficient way to add strength for a few reasons…

Cost:
This is a rather touchy topic because people that buy $120+ frames are truly in love with them.  That’s not a bad thing.  In fact that’s a great thing.  I love works of art as much as the next guy but my definition of ‘art’ may be different.  While I definitely love me some shiny gold motors and pretty colored plastics and metals, I find far more beauty in something that works, keeps working and doesn’t cost more than it needs to.  I don’t like non-standard hardware, custom aluminum or titanium parts, expensive carbon layups, excessive packaging, pointless frills…I don’t even like 3D printing because it’s time consuming to manufacture and often costs more than desired when made in bulk.  It’s fine if that titanium or aluminum part is actually the best way to manufacture the item but I generally have no interest in the frivolous use of materials just for the sake of being different.  I like things that are highly efficient at executing their function and every time I’ve wanted to use another material, I find a way to do it in cheap flat carbon that’s at least the same or better.  Furthermore, I always have the manufacturing method in mind in order to better optimize my parts for a higher yield per dollar.

Moving forward, I would hope that we develop improved ways to manufacture the things we want but for now into the foreseeable future, it’s really really REALLY hard to beat flat carbon.  It’s just an impressively efficient way to construct the things we fly.  My goal is to deliver the most optimized product in every way including cost.  The increase in cost of this frame compared to some of my previous designs is purely a reflection of the work and effort I put into all this stuff.  I love this stuff but am always bordering quitting because it does eat a massive chunk of my time.  I’ve got to make it at least somewhat worthwhile for myself and I hugely appreciate everyone’s support.



Topics not discussed and may be added later:
– Weight distribution
– Prop spacing
– All up weight and prop size
– Standoff placement
– Frame size and dimensions
– Dead cat vs. traditional arms
– Lots of other things…

18 thoughts on “The Glide design and general frame design concepts

  1. Knowing your mindflow is the biggest selling point for me when buying products you design or recommend. Though, I already have a Glide, thank you for this article 🙂

    I’ve got a question that is not directly related to Glide: will a battery, which is 56mm wide, hanging out of frame, but located above the prop line, affect flight performance? (the batteries i’m talking about are 6s Cnhl 1000mah – they are of an anusual shape)

    1. That should be fine. As long as it’s safely out of the way of the props if they flex up it’s all good.

  2. which fc do you recommend, at the time;)

    1. I like the crazybee boards. There’s now a 3S F4 version in the new mobula HD. The FC board should be out soon and the 5v should be more than the 2S boards.

  3. which flight controller do you recommend, at the time;)

    1. I like the crazybee boards. There’s now a 3S F4 version in the new mobula HD. The FC board should be out soon and the 5v should be more than the 2S boards.

  4. for a 5″ 😉

    1. Ah 😀 I just saw this and realized I was responding to the wrong post. The message respond thing in this web app manager doesn’t say which post these comments are from. I like 20×20 FC’s because I don’t see a need for the larger 30×30 platform. My favorites right now are the Talon and the new Radix. Both are 20×20. I haven’t used the Radix but the layout looks good and I know their things are quality. Sorry for the delay

  5. Is it possible to “toilet tank” mount a battery on this frame? Would that be likely to have any positive or negative effect? Just ordered a Glide for a 3D “cheater kwad” build with the IQ Motion Control motors that are supposed to be shipping this month and Caddx Turtles at either end, one inverted. This frame is the only one I’ve found that looks like it has enough space for the electronics and cameras, plus notches in the carbon that should allow decent tilt and FOV in both directions and orientations. Thinking the sideways battery mount might help with CoG depending how and where things fit. Really looking forward to trying this out!

    1. As is, there aren’t holes to run the batt strap the other way. The center holes aren’t quite wide enough however you could probably still do it anyway. The frame is quite spacious and ‘should’ work out for a double sided quad. The rear standoffs are the same spacing as the front so a camera should fit on the back no problem albeit a shallower angle however. I’m not sure rotating the battery sideways will make any meaningful impact.

  6. I just got a Matek F722 and it is performing fantastically. Just thought I’d throw that recommendation out there.

  7. Thanks for the reply! I see what you mean about the center holes not being wide enough and will probably stick to straight battery mounting then. Without needing to mount a GoPro onotop, it should be easy enough to adjust battery placement to optimize cog.

    Just realized there are only two camera mounting brackets for the standoffs and I’ll need another set. Is there an STL file available for them? If not I’m sure I can find something similar on Thingiverse.

    Motors shipped today. Very excited for this build and very annoyed it’s still snowing here! 🥶

    1. Sorry this messaging software on this web software is really crap. I don’t know what the last message was about :-(. Better off discussing things on the FB page. KababFPV

  8. Glad to hear the Matek 722 is doing well. Ordered a 722 mini for mine for the inbuilt camera switching.

  9. Hi kabab, thanks for sharing. I’m agree with all your design statements regarding multi unit except one. I think monoplate, with adéquate design, is lighter for given strenght, or more strenght for à given weight because you have multi layer with too much material on some areas (3 layers) and thé added Weight of additional fasteners. For shure with monoplate you also use thé 3rd dimension by choosing thé plate thickness and after choosing thé Just necessary section according each area.

    1. It would be extremely difficult for a monoplate to match the durability of multi unit frame with the same weight. Of course not all designs are good but if both are optimized, the monobody just can’t match durability. The only actual weight handicap the multiunit has are the screws. Otherwise the arms can be thicker and body can be thinner. If you make the monoplate thicker, you gain tons of weight from the body being thicker too.

  10. Hey dude ! how are you ? 😉 Just receive your frame from drone drop and I wanted to know if you have a thingiverse link or else for a gopro session mount ?
    Thanks in advance 😉

    1. Check out Brain3D.com. He does all printing for allllll things. Love the guy

Leave a Reply