30-Jan-2024, 02:19 PM (This post was last modified: 31-Jan-2024, 12:56 AM by ph2t.)
So this is silly.
I'm going to try and build a 1S hex. It's going to use 65mm props with 1202.5 11500kV motors. I'm 3d printing the frame for now as I try and figure out the layout.
It's tricky with six motors. Trying to get the right distance between props without making the frame too large. This first version mockup was a lot of fiddling around in tinkercad. Ripping off a pickleBig 5" STL and trying to make it work, lol. The 3d print is not final in the pics below, purely a layout test.
I hope to get it to run on 1S by:
Modifying the JHEMCU flight controller (GF16-BMI F405 2-4S) to run on 1S (or use a step-up Vreg).
Using small Bl_heli 5A ESCs rated at 1S
The GF16-BMI F405 flight controller has 8 motor outputs already broken out on solder pads, I'll use 6.
The frame will be walksnail digital by using the Betafpv digital canopy to hold the vtx. This way I don't need to worry about different mounting dimensions (16mm for the fc, 25.5mm for the vtx).
Here are some pics from this evening.
Even at 65mm for propellor size, it's quite large. I think I could move things around a bit to make it a bit smaller but it is hard! Originally, I was wanting to try a 3 inch hex but it was just too large to try and pull off on 1S, maybe possible but I'm not there yet in skills.
Man, 3d printing is such a magnificent time waster!
Very nice, I like that "dead squid" configuration How much do the ESCs weigh? I wonder if going with a 1S AIO and adding only 2 ESCs would save some weight, but I guess with 6 motors weight is not a big concern. Guessing you're going to make a big lipo to run it or maybe 2x750 in parallel?
31-Jan-2024, 12:54 AM (This post was last modified: 31-Jan-2024, 12:56 AM by ph2t.)
(30-Jan-2024, 02:57 PM)mstc Wrote: Very nice, I like that "dead squid" configuration How much do the ESCs weigh? I wonder if going with a 1S AIO and adding only 2 ESCs would save some weight, but I guess with 6 motors weight is not a big concern. Guessing you're going to make a big lipo to run it or maybe 2x750 in parallel?
That's a great name mate! Dead Squid it is! I'll rename the thread.
The ESCs are made by agfrc.com, got them of aliexpress. The model is XR-4A. They weigh 0.5g each. Rated at 1S, 4A nominal, 6A peak. I figured a net current draw at around 24A (x6 escs) is something this setup could handle. We shall see!
I checked out some AIOs and how to remap pins with DMA timers on them to actual motor pads. Was a bit tricky and just above my pay grade in knowledge so I decided on this design of six discrete escs instead.
Quote:nice! i like to get impressions, never thougt about 1s on 6 motors befor :-)
maybe a thin additional carbon circle over all motor would give stiffness? it would look like thors hammer :-)
haha, yeah the design isn't final yet. Just piecing together parts to make it work. Sizing the quad layout was the first big thing.
So i think I've got the 1S conversion sorted. I've haven't done a full test but it is promising.
This FC is rated for 2-4S, that is a 8-16V range so two voltage regulators are being used. This is pretty typical for flight controllers that have large input voltage ranges.
The first Vreg drops it down to 5V and the second Vreg drops it down to 3.3V. There is also a diode inline, between the first and second Vregs.
The first Vreg is in green, the diode in yellow, second Vreg in red.
I've soldered power to the cathode of the diode, this was the best place to connect as the pins on the second Vreg are too small and close to adjacent components. Testing the full operating range of a 1S lipo (3.5V to 4.35V) showed the FC to boot up and show activity on the status LEDs. I got it down to 2.9V before it shut off.
Now to test usb data and a single 1s esc connection to see if I can flash Bluejay.
Interesting, I was going to say what about the ESC MCUs, but they are not on the FC. Seems you can make do without a step up. But without a 5v rail, you will need to find a ELRS RX that runs on 1S, I think most of them still need 4.5v or maybe you will bypass its vreg too.
Going to be interesting where the battery will have to be to remotely get close to balance, or the FC is going to be working hard in asymetric thrust achieving hover!
You may be heading towards the quad equiv of a few of my own design planes! A mate and I used to take FIK (Slovenian) EPP chuck gliders, add RC and a motor, and see just how silly a shape would fly, and the answer was...................... VERY silly!
One of the best flying of mine had its 800mm wing now at the rear, fuselage reversed and hacked off at the wing root, pusher drive, transparent plastic fin just in front of the motor, and a 300mm front wing where the tailplane would have been now right up front as a canard with dihedral tips. Needed a much bigger battery than planned RIGHT on the nose, but it flew really very well and was fully aerobatic.
3D frame Build Point. After a lot of experiments with 3D print quad frames inc a few NanoLR's and my design changes (dare not say improvements!) to them I found that making a groove down the underside of each arm to insert and thin CA flood a flat carbon strip removed the fragility/resonance issues from an otherwise too flexible frame.
Quote:Interesting, I was going to say what about the ESC MCUs, but they are not on the FC.
Correct they run off Vbat. There is a caveat I've found though.
To get the FC AND escs to boot whilst on USB, the FC needs to be disconnected from Vbat. To properly boot everything to configure, I need to power up escs first, then plug in USB to the FC. Prolly something to do with where I tapped into power and being on the "wrong" (cathode) side of the diode. If I want to flash the esc and get BF passthrough I need to have the FC disconnected from power and only running off USB. If I don't want the esc powered on I can connect the FC to USB and it is fine for programming.
Quote:But without a 5v rail, you will need to find a ELRS RX that runs on 1S, I think most of them still need 4.5v or maybe you will bypass its vreg too.
This gave me pause, I hadn't thought about the RX yet. Turns out to be fine, I run it off the 3.3V pad on the FC (which works on Vbat power and USB power funnily enough). Originally, I thought along the lines of what you said but in reverse. ie: connect the 3.3V rail of the EP2 rx to the regulated 3.3V rail of the MCU/FC. In the end the EP2 rx runs fine below 3V. I tested bind and stick commands in BF configurator, all good.
Quote:Going to be interesting where the battery will have to be to remotely get close to balance, or the FC is going to be working hard in asymetric thrust achieving hover!
Yeah good point, I'm relying on the modern flight controller here to do its job well. The current centre line is just behind the rear canopy post. The battery will sit centre in that area hopefully.
Quote:3D frame Build Point. After a lot of experiments with 3D print quad frames inc a few NanoLR's and my design changes (dare not say improvements!) to them I found that making a groove down the underside of each arm to insert and thin CA flood a flat carbon strip removed the fragility/resonance issues from an otherwise too flexible frame.
Now that is a good idea!
This 16mm FC from JHEMCU is feature packed. They've stuffed solder pads everywhere, including eight motor outputs, lol.
Here's the first version of the frame I'm going to attempt to fly. It is 4mm in height, printed in trusty old PLA. I've included some esc holders and canopy standoffs
Prolly overdone it with the esc holders but I like them and it gives me beast class vibes, lol...
I've quickly realised that I need a PDB! I'm going to have to hack some breadboard together otherwise the cables will get very messy very quickly.
Be aware of the PID settings beacuse of the vibrations of an 3D printed frame. BotGrinder just released a funny video of a really short maiden flight of his polyacryl selfmade frame.
Draw in some bamboo skewer sized diameter troughs along arms to insert said skewers into print for support. You can pause print when at top of skewer depth then glue them in with some E6000/foam tac or similar. Resume print to cover...
Those trays definitely change the look. For small props though, that extra surface does block airflow, maybe have the ESC slotted in sideways or at an angle especially if you add CF sticks? Sorry just spewing out unwanted thoughts and comments I will have to retest my RXs, I remember the last ELRS I tried on 3.3v would not power up.
Draw in some bamboo skewer sized diameter troughs along arms to insert said skewers into print for support. You can pause print when at top of skewer depth then glue them in with some E6000/foam tac or similar. Resume print to cover...
Quote:Those trays definitely change the look. For small props though, that extra surface does block airflow, maybe have the ESC slotted in sideways or at an angle especially if you add CF sticks? Sorry just spewing out unwanted thoughts and comments
Technically the props don't overlap the escs, mind you I didn't think of this either! I put them close to the frame body to shorten the amount of "thin arm" so to speak.
Yeah CF sticks is where its at I reakon, going to do a bit more research on that.
Keep the ideas coming fellas, I'm happy to hear from all and sundry.
I'm going to use some veroboard to make a PDU that will sit under the FC, I gotta pop into my local jaycar to get that and a small isolation switch for the FC/USB functionality.
Some more progress pics, it's a soldering nightmare.