• This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn more.

New Build Finally Beginning!!!

mmeyer

Senior Member
#1
I've had this idea in the pipeline since the beginning of the year and tomorrow it will finally become a reality! The design is finalised, carbon fibre has arrived and it is all hopefully going to be successfully CNC cut tomorrow morning. I intend on posting a build log and decided while I had some time tonight that I would get it started with a picture of all the parts I intend on using.
IMG_2391.JPG
On the right hand side there are two sheets of 1.5mm CF for the main frame and one sheet of 3mm CF for the arms.
Next to them is all the gear:
  • F20A Simonk ESC
  • RC Timer 2212-13 1000kv Brushless Outrunner
  • Naze32 Acro
  • Hobbyking Power Distribution Board (I intend on making a wire harness in future)
  • Turnigy 9x with stock receiver
  • Boscam TS351 Video Transmitter
  • RC305 Video Receiver
  • Hobbyking Cloverleaf and Skew Planar antennas
One question I have is in relation to the RF interference from the CF plates. Is the heatshrink around the receiver enough to shield it from the CF or should i still take some more measures to ensure it is insulated?
Tomorrow i get it all cut and will hopefully (fingers crossed) post an update tomorrow night sometime.
 

Cyberdactyl

Misfit Multirotor Monkey
#2
Looks like you're pretty much ready. I would say 'yes' to the shrink being enough to isolate a naked' board 'electrically' from the CF. The glossy surface is not conductive. However, if the board has some really 'pointy' pins on the bottom where vibration could work through the shrink, you could insure the condition by cutting some heavy paper, like index card or a business card and putting that between the shrink and the back of the board.

But. . . I think more to what you're asking, is the shrink will have no effect concerning RF interference from the CF. What I suggest is to mount the receiver with as much gap as possible.
 

mmeyer

Senior Member
#4
So a quick update on the progress before I go to bed. Spent the whole day getting the CF cut (from 8am to 5.30pm) and spent the evening dry fitting it all together to see how well I had designed the quad. I am very pleased with the result but as expected it did require a bit of sanding here and there so that everything fit. In mark II of this design there will definitely be some changes based off what I have noticed so far in the build. I plan on posting a more in depth account of what I have done so far and the concept behind the quad tomorrow but for now I need sleep. In the meantime here are some photos of where I am up to.
IMG_2394.jpg IMG_2415.jpg IMG_2418.jpg
 
Last edited:

mmeyer

Senior Member
#6
Initially i thought CA glue but after putting it together i realised with a few extra holes i could maybe do something with zip ties. Also given that the back supports are angled i didn't put in bracing like I did with the front, i thought the angle would be enough. Apparently its not so i have to work out some solution to that as well as its a bit wobbly.
 

jhitesma

Some guy in the desert
Mentor
#7
Just to clarify on the isolation issue. Heat shrink would be enough to electrically insulate the CF so you don't short anything out against it. But...it won't do anything to change the tendency of CF to shield RF. Even with heat shrink the CF will tend to block RF and could affect range depending on where/how you plan to mount for RX antenna.
 

mmeyer

Senior Member
#8
The build log finally!

Sorry about the wait! I was planning on posting this some days ago but it was somewhat forgotten in my move back to Melbourne for Uni. Throughout the process so far I have been taking photos to help illustrate my process. This is my first build log so if there’s anything I missed out just let me know.

The picture below shows where I stand currently in the build. I am waiting on another ESC to replace one which died and the batteries from home so work has been halted until they arrive.
4in1.jpg

I thought I should also explain the basic concept behind this quad. As we all know vibrations are a big annoyance in multirotors, especially with CF. My idea with this design is to have a ‘clean plate’ which is completely suspended. On this clean plate will sit the flight controller, battery, camera and vTx. I have made a wooden prototype version of this quad and it did a very good job at eliminating vibrations so I have high hopes for this CF one.

This whole frame is CNC cut out of one sheet of 3mm CF and two 1.5mm sheets. It was cut by a guy I used to work for and it took a whole day as it was his first time cutting CF and we had to do a bit of work on the design to get it ready to cut. I was amazed at the accuracy and the amazing finish of the CNC router and would not hesitate to cut another frame with this method. For anyone that wants to know I think we were using a 1.4mm 4 flute diamond end mill. Below is a short film of the CNC router cutting out the CF (nothing special but vaguely interesting to me anyway :) )

The pieces were covered in dust so before I did anything to them I sanded all the rough edges and gave them a bit of a clean. The 3mm holes had not quite cut to size so I had to drill them all out with a 3mm drill bit.
I began by sandwiching the four 3mm CF arms between the two 1.5mm bottom plates. At the moment I am using M3 15mm bolts with lock nuts but I’m planning on changing to M3 20mm bolts as explained later. I was very happy with how it went together. It is a very sturdy and durable frame with very little flex. The one main issue I have with these two bottom plates is that I would like to put the power distribution in between them but evidently it is too thin. More on that later.
IMG_2402.jpg

As you see in the photo at the beginning of this post, the central plate is suspended between the bottom 2 plates and some 1.5mm CF supports. The below images show the supports and how they go together. Some sanding was needed for the tabs to fit into their slots as the cheap CF from ebay was 1.6mm and the slots only 1.5mm. Once I was happy with them I superglued (CA glue) them together. The front support is very very strong and is more than sturdy enough for the task it will be performing. In theory the separate pieces will break apart in a bad crash rather than the pieces themselves breaking but I am unsure if this will actually happen.
4in1 2.jpg

After putting both of the braces together I realized the rear support was not sturdy enough. For some reason while I was designing them I thought that the fact that the vertical supports in the back were angled would be enough but apparently this was not the case.
IMG_2414.jpg

To fix this I redesigned the rear supports to add the braces you see in the front support. This made the rear one just as sturdy as the front one.
ssdsd.jpg

The next step was to add the rubber o-rings to the bottom plates and the supports and then connect the supports. In the pictures above you can see the holes designed to hold the o-rings. They were too small in the initial design which made putting them onto the frame an awful task. You can see in the redesigned support in the above photo that these o-ring holes are larger than the original which makes them much much easier to put in. Some redesigning needs to happen to make this whole process of attaching the middle floating plate somewhat easier.
IMG_2424.jpg

There was a fair bit of fiddling around to try and find the correct size o-rings to get the floating platform to sit where I want it. I got it to a point where I like it but to make it perfect I think I will need to play with it some more. I noticed that when the supports are in the correct position that the o-rings hold them down quite well. This meant that I could avoid using glue which I was very pleased with. All I needed was some zip ties to make sure they don’t fall over and once they’re in it’s a very solid setup.
IMG_2425.jpg

At this point I had to dismantle it to bring it back to Melbourne with me and today I rebuilt it to this point. I made a few little tweaks including adding dedicated holes for the zip ties to hold the supports and making a few spacers out of the remaining CF to hopefully make the bottom two plates somewhat more useful. I made 8 spacers which means I can add an extra 6mm in between the plates which should allow for my power distribution board but I need to buy longer bolts for that to work (20mm M3). I might just bite the bullet and solder up a power loom thing. As yet I do not have any idea how I will do landing gear so if anyone has a good idea for that please share it. Also if anyone has any ideas about improving this design I would love to hear them as I would like to keep developing this design further. For now I just have to wait on the parts so I can finish this build and give it a maiden. I am beyond keen to get this thing in the air. Thanks to everyone who helped with the initial design feedback earlier in the year, keep it coming!
 

cranialrectosis

Faster than a speeding faceplant!
Mentor
#9
Wow, a clean plate suspended by O rings. It looks like a complicated build with all the little parts but I can't wait to see the video it shoots.
 

mmeyer

Senior Member
#10
Definately need to work on the design to make the build easier because as you say it is quite a complicated and fiddly build.
 

Twitchity

Senior Member
#11
Awesome build, mmeyer. I especially like your idea for the clean plate.

Thank you for listing the bit you used to cut the CF :) I'm still experimenting with different bits, so I'll be adding that one to my list. Just one question though, do you happen to know the cut speed of the machine for the two different thickness CF sheets?
 

mmeyer

Senior Member
#12
To the best of my memory with the 1.5mm plate we were using:
  • 1.4mm 4 flute diamond end mill
  • 22000-24000 rpm
  • 1.6mm cutting depth (only had to do one pass for each cut)
  • 1000 feedrate

With the 3.5mm plate we started with
  • 1.4mm 4 flute diamond end mill
  • 22000-24000 rpm
  • 3.15mm cutting depth
  • 600-700 feedrate (i think)

We did end up in the end changing to a 1.9mm bit as the 1.4mm bit broke once or twice. If you want i can get the exact values but i'm fairly sure these are correct (or at least very close to)
 

Twitchity

Senior Member
#13
Those are perfect, thank you. Since my CNC is different than his, I'm sure I'll need different settings, but this should help get me started.
 

mmeyer

Senior Member
#14
On the note of the bits, they cut very well and I have read online about other people also having a lot of luck with 4 flute carbide or diamond end mills. I would thoroughly recommend them. Out of curiosity what are you using to cut your G10 at the moment?
 

Twitchity

Senior Member
#15
I've mostly been using 2.55mm diamond cut carbide bits.



I do have two 2.4mm 4 flute spiral endmills that I need to try again. I only used them with the horrible spindle my CNC came with, and the bit would walk a little bit when it plunged into the material causing an indention in the cut. Since I've upgraded my spindle to something with a lot more power, I need to give them another try with some of the settings you listed.

Were the endmills you were using spiral or straight?
 

mmeyer

Senior Member
#17
I'm still waiting on the batteries and esc's but today i went to Bunnings to get some new bolts to accommodate the spacers I made. Put them into the quad and now I can put the power distribution board in between the two bottom plates. It doesn't fit perfectly but it does fit. I recently discovered $1.50 mousepads from Kmart and they are very very useful. Great for vibration dampening but in this build i used it to insulate the power distribution board from the CF.
IMG_2435df.jpg
I also bought some 1/8" rubber grommets to help further isolate the FC. I don't know if i will need this given the isolated middle plate but they were cheap so I decided to get them. I think however that the rubber is too solid so it wont do too much anyway.
IMG_2437.jpg
I'm hoping the batteries will arrive later today or tomorrow and the esc's later this week. I cannot wait to get this thing in the air, i'm far too impatient for my own good.
 

jhitesma

Some guy in the desert
Mentor
#18
You could try sections of silicone fuel line or thera band instead of the grommets. Might also work in place of the isolation o rings and give better dampening. Then again, could be worse. Just tossing the idea out there :)
 

mmeyer

Senior Member
#19
So the batteries arrived today so i could connect everything up and test motor rotations and such. Unfortunately HK is having a bit of an issue with a courier and my address so at the moment i have unmatched ESC's. I don't think this will be a problem but given that one is not Simonk is it in fact a problem?
Anyway below are some photos of the build which is pretty much ready for its maiden when the weather clears up here in Melbourne. Two other questions i had were what could i do for landing gear and does anyone have any clue about the best place to put the receiver?
Thanks for all the feedback so far, it has been very encouraging.
IMG_2440.jpg IMG_2441.jpg IMG_2442.jpg