Thread Tools
Jun 25, 2010, 01:26 PM

Pull a chair up, and maybe bring lunch.

Originally Posted by coaxialnewbie View Post

Hey killbucket, can you teach, explain/introduce to me how you make the CAD and use your tools to build it? My Tech teacher is refusing to teach something that is out of the course.

I will return in a few days.
Start with...cardboard. And take as long as possible planning, make actual hard parts only when you know they won't need to be changed (still cross your fingers).
I've included a link to a low-cost, learn it yourself CAD program at the end of my post. These rest THERE, is up to you.

(skip next section, it's mostly bragging again)
On my own learning of CAD:
I took a six-month, full-time (8 hours a day, five days a week) course in AutoCAD 2000 at:

Precision Technical Institute
3927 Lennane Dr Ste 260
Sacramento, CA 95834-1922

I learned directly from the school's founder, Martin Maloney hisself.

I really took to this. I'm sure they still remember me!
I got my $7K worth out of them, and they even offered me a teaching job when I graduated.
I ACED their course, pulling 100% on every single test taken, (and yes, I have every one here to prove it, see att. samples).
I graduated with 170% of coursework completed, due to the many extra-credit projects I turned in. I designed whole homes in a few days, including landscaping, topographical maps, utilities, even lawn drainage from sprinklers.
I also did up "plan and profiles" for highway interchanges, neighborhoods, store buildings,and warehouses, complete with plumbing layouts, pneumatic and hydraulic line spec sheets, BOM's and electrical diagrams.
I submitted countless drawings of airplane parts they had around (those are some of the most interesting items you could ever see. We "consumers" get Poo, the gov't gets the REALLY neat stuff)*.
Typically, when I had assignments done for the day, I dug through their junk cabinet for things to "mike" and draw 3-views of.

C130 Hercules Landing gear yoke, if Saddam had personally ordered one.
Aircraft parts were the most challenge, NOTHING seems to be parallel on them. "Geometric Dimensioning &Tolerancing" has a whole new meaning when you try to reverse-engineer it.
On this one, the 3D model was easier to do first, because a "flat" drawing was impossibly complex to comprehend (I did one tho!).

The teacher had worked for Schilling Robotics (arms you see on the ubiquitous ROV's) and had piles of down-rev parts. I drew most of them, I should do an off-topic and post those.

Then we can see if any are spotted in actual use, because for the most part, I could not determine their functions- shapes I have never seen on anything else.

OK, now know that I'd gotten TurboCAD earlier than this, and learned it myself, alone, "From Hell To Breakfast", before ever setting foot in CAD class, to "get my cert".

Getting Started:
1. Make a pencil drawing. I cannot stress the importance of this. If you just sit down and start measuring and plugging in numbers, you will find yourself with a file full of conflicting numbers. Measuring a part with hand tools is an operator-dependent process, and you can decide on different numbers for the same feature, if you happen to do them twice with slightly different techniques.
I know, all I did for twenty years is MEASURE PARTS!!! (sorry, flashback).
In simpler terms, avoid chasing your own tail.
If later you find an error, it will be MUCH easier to resolve with the rest of the drawing already there.

2. Then, put the part aside and take the drawing to the CAD sys.
"Complete" the drawing here, because once you have your "set dimensions", it's a lot easier to proceed to modify them.

3. Print a copy, and then check it against the item.
Keep in mind, everything has "Tolerances", and being Uber-Exact will only eat time. If things look good, you can then go back to CAD, and start making changes. This can take no time, or the rest of your life. Assuming you want to make an actual part to try out now...

4. Print the part at 100% size, and then use that as a pattern to make the part.

Obviously, you'll be making teeny, tiny stuff, but here's a pic set from a while back showing the process:


Layout for a switch panel. Drawn in TurboCAD, and printed at 100% size to use as a template.

Then I used some blue masking tape to cover my 6mm PVC scrap (I got this material just for begging a dumpster-dive at www.InterstatePlastics) and then sprayed the back of my printout with Super90 Spray Adhesive.

Note that I didn't spray the masking tape- because any areas not the covered by paper would then stick to my fingers (and the bottom of my saw guide, if using a scroll saw or jigsaw) and make cutting a pain instead of a joy. (It's fun making your own stuff, when it doesn't include bloodshed).

The tape layer allows EASY removal of the CAD artwork after cutting. You could also use cheap contact paper to do this, and I have had very good results with it. Do NOT use clear tape or "Scotch" tape, it will not come back off nicely, if at all.
Tips video for this step.

Definitely mark the hole locations in advance with a punch or similar item.
Drill bits walk around without a center mark, even on a paper surface. Also, start with a tiny drill bit, and 'pilot' all the holes first. Any error in drilling will be MUCH smaller with a small diameter bit to start. Then you can enlarge the holes later with a bigger bit, and it won't wander. That way, all the holes will line up.
Tips video for this step.

I'll go back and sand the edges, but the saw gets very close to begin with.
That's the second pattern on top there, waiting its turn at the knife.

Here's the panel, minus sanding. The front row of switches is test-installed at the moment.

The completed

(OK, nearly completed) project.

On Scratch Building:

I usually make a template of an original part off something, by covering it with the same blue masking tape, trimming the edges. The tape can then be carefully pulled off, and transferred to sheet cardboard (old cereal boxes are excellent for this)

Once I've done this, I can cut out the shape, and then use that for trial fits, and if all goes well, use that same item for measurements to put into TurboCAD**.

Most importantly, you will save money and time if you go 3-channel, because I don't see the number of channels, or even the machine's complexity, as being relevant to the grading process you hope to "PASS".

The helicopter is not the core issue. An Instructor should be more impressed by what you're seeing, understanding, re-thinking, planning, and most importantly, DOING.

That is what's going to get you the "MacGuffin" you should be aimed at, not a shiny, state of the art copter and a mediocre grade.

Very Important:

You want the evaluation-planning-making cycle to successfully fit into the time you have, before you have to turn it in.
"Quickly, Correctly, Cheaply"
Pick two (no kidding) and go for it.

*So happy to show any of this off...I still have it all!

You will like this proggie, cheap, compatible with AutoDesk's for the most part, and I easily go between the two.
I hope some of this is helpful, and whatever the rest is, entertaining.
as always, this is,
Last edited by killbucket; Jun 25, 2010 at 06:35 PM.
Sign up now
to remove ads between posts
Jun 25, 2010, 02:30 PM
Resident Skeptican
Barak1001's Avatar
Let's be realistic here.

Designing your own blades and getting to a production model in under 6 months would be hard considering that you've never flown any rc helicopters.

Learning CAD, designing an entire 650mm rotor span helicopter, and building it from scratch in under a year with no experience at all in the field will be basically impossible. I don't think anyone has actually brought up the warnings yet, but at the scale you plan on building your chopper to, if you have an accident (like it hitting you in the neck) there's a strong possibility that death can be involved.

I think you've bitten off more than you can chew with this project and perhaps you should set your sights on a goal you can actually achieve.
Jun 25, 2010, 02:41 PM
Good points, but find some path to continue:

One improved part for an existing item should meet your assignment criteria?

Yes, it's complex to design a razor, an item with no moving parts at all, and I've seen one take over a year ("...or take the rest of your life" -type boss, long story) before molds were cut.

A whole helicopter?

That's why my very first post here.
Originally Posted by killbucket View Post
First off, you need to design the mechanical parts needed, unless you're also going to use off-the-shelf's.
In which case, this is just another build, but by someone starting out.
V-tail mixers?!?!? No longer 'entry-level-speak, IMHO.
The mech's you use, to a certain extent dictate the electronics needed.
It's very easy to get mired in certain aspects of a project, and never complete the whole thing. Before you ever get into electronics, you need a chassis.
and a follow up later,
Originally Posted by killbucket View Post
Yes, chassis design will be the bigger project.
and then a bottom line,
Originally Posted by killbucket View Post
What would be the biggest hurdle for me would be deciding on what specific bearings, shaft materials, lengths, flybar/rotor lever ratios, and such.
Then spec'ing gears/pinions, THEN motors, electronics.
A good "Science project" would be taking a complete existing model and making improved parts for it, showing the improvements over original mass-produced parts.
THEN, for the win, show why they aren't done that way originally, i.e. cost, etc.
I know the initial idea was to "Build a copter" using available "gear", and then making a chassis that would show the designer's understanding of everything there. If it worked, it sure would, and surely does.

What was the assignment, rather than the goal?

Now Barak here, actually does what you're envisioning.
The whole enchilada.

Even in skilled hands, a deadline for one can be (is a) nightmare. Am I wrong?
Jun 25, 2010, 02:43 PM
Guitar Player
I have to agree with Barak on this one. The amount of training, not to mention access to equipment and time this seems like Boeing trying to design and build a 747 with 10 people. You will be undermanned, underskilled and I can pretty much guarantee you under budget (ed). There is a thread somewhere on here where a bunch of college students did this same sort of project. They designed their own unique heli with standard parts. I think that a serious senior project could be along similar lines, but the reality of the skill set needed cost, etc would be way out there. Look into standard parts without buying a kit! Why machine something that already exists. This site is full of very ingeneious modifications to the point that the end product is far beyond custom! My 2 cents!
Jun 25, 2010, 02:46 PM
But their part-scattering videos are unarguably fun to watch.

He said he'd(she'd?) be off a few, I'd like to hear how the actual assignment was worded, and maybe find an applicable task on an existing machine.

If the teacher says no CAD (maybe not unless you already have it, to avoid late projects), well, I can't design you an O-ring without it.
Jun 25, 2010, 02:48 PM
Resident Skeptican
Barak1001's Avatar
I'd also like to mention that although I don't do CAD, I do animation and modeling which mimics real life physics. It took me about 6 months to learn enough about 3dsmax to be able to make a believable model of a person. Took me over about a year to be able to do complex animations.

I imagine that CAD would have a similar learning curve, except that every part you make will have a physical cost. You'd be better off using something like Solidworks to simulate your parts before actually cutting them.


Just for you Killbucket

Chaingun Prototype Mount (0 min 11 sec)

This is a mech I was working on a while back in 3dsMax. Trying to come up with a realistic and suitable method of making the gun move.
Jun 25, 2010, 03:05 PM
Both are baffling at first, but they both have their workflow patterns, this, then that, click.

MAX is awesome, but is a timehog..
Addicting, does not cover it!
The blue CG minigun Youtube vid I linked above was done in StudioMax 3.1, forever ago. ."sleepware" it used to be called, as you'd sit all (sunny outside) day in a silicon valley loft, plotting wireframes and primitives, then finally click "Render" and...sleep for the first time in three days

(I did this!).
Another program I could confidently teach. It took me ten years to say that.

Solidworks is The Shizzle, whatever... The Bomb, the best, period.
I cringe at having to use ProEngineer so long. Ditto, the horribly laid-out Vectorworks (don't get me started on this).

All three cost more than a decent car, but TuboCAD can be had a Fry's for a measly $99, and does 3D.
Last edited by killbucket; Jun 25, 2010 at 06:38 PM.
Jun 25, 2010, 03:09 PM
Originally Posted by Barak1001 View Post

Just for you Killbucket

This is a mech I was working on a while back in 3dsMax. Trying to come up with a realistic and suitable method of making the gun move.
I like it!!
I used cut-down cordless drills to drive my RC M2HB.
I put 1/2" threaded rod into the chucks, and put limit switches on them.
More speed? More VOLTAGE...or swap in brushless motors. The drives will take it, the only question is, will it rip the gun apart.

Last edited by killbucket; Jun 25, 2010 at 03:13 PM. Reason: Note the Ghizzy Nightvision cam.
Jun 26, 2010, 03:50 AM
Registered User

The bottom line

Well, the purpose of the subject I was doing is to learn how to research, make journal, take notes, make bibliography and references effectively, combining with conducting interviews and research materials that let me achieve the goal. The subject is about the process of the research rather than the final product.

I have completed this subject for this year (B+ or A- ). As in my mail to Barak1001, I would like to turn this into a long-term project. This is the most interesting project I have been doing for my entire life and I truly want to do it. If this copter is built when I am 40 years old, costing $1000, then let it be and I shall make a new topic with the photos and give credits to everyone in this thread.

I appreciate every posts made by RCGroups members. You guys help me a lot in this project. One more years before university takes place. It's either mechanical/civil engineer, technician, or army defense job. Learning CAD will be extremely helpful for the 5-years in uni.

The number of channels is definitely 3-4 and it was stated in the first post. Thank you for helping out.
Last edited by coaxialnewbie; Jun 26, 2010 at 03:58 AM.
Jun 26, 2010, 05:08 AM
Resident Skeptican
Barak1001's Avatar
So basically it's vaporware/pipedreams?

Honestly I don't mind helping someone out, but as near as I can tell your only goal is to research how to build a coaxial helicopter from scratch, rather than build anything that will actually fly. You have no experience at all with RC, yet plan on building something fully capable of killing yourself or a bystander, and to top that off you have the expectation that someone will teach you CAD within the next year for free via the forums, and that you'll be able to piece together said helicopter by the time you're 40?

I'm sorry, but I'm unsubscribing to this thread. If and when you have a realistic and attainable goal let me know.
Jun 26, 2010, 05:13 AM
So a SALE can be made.

Compliments on your kind words of encouragement. I can tell you mean well, and are just coming off a bit strong..."The Authorities"-strong.
Just because you can tell someone off, doesn't necessarily mean you should. You might need a favor from that person one day (they might even own a CNC and be willing to make custom parts for the asking)!

You are however correct in essence. If the assignment is to do a report with sources, building a helicopter to achieve it is a bit out into left field.
If content is taking a back seat to document form (your real assignment), you may as well do a paper on the screw sizes spec'd for mounting the blades, and call it a day. It will still be about copters, and IMHO, the instructor won't care about that anyway.
Another turret airsoft gun, this one with almost two feet of vertical travel. Long since disassembled.
Last edited by killbucket; Jun 29, 2010 at 12:52 PM.
Jun 26, 2010, 07:05 AM
Resident Skeptican
Barak1001's Avatar
Honestly KB if I needed something made with a mill I'd put one together. You can build a CNC mill for under $100 using old stepper motors from printers and a dremel tool.

I don't like sugar coating things and this kid seems to need a dose of reality. Him asking for someone to teach him CAD via the forum was just too far out in left field for me. I get just as annoyed when my kids ask me questions about a homework assignments that they could answer themselves if they thought it through or research using google.

I don't get paid enough to think for people. It's hard enough thinking for myself.
Jun 26, 2010, 08:27 AM
Originally Posted by Barak1001 View Post
Honestly KB if I needed something made with a mill I'd put one together. You can build a CNC mill for under $100 using old stepper motors from printers and a dremel tool.-snip-
For somebody who mercilessly pounces on an absurd statement, I'm surprised to hear this from you.
Small toy or robot parts are about the extent of a Dremel CNC's abilities.
You can make your own CNC, there are many sites to show you how.
In the end, it all needs to seamlessly work together to even function as desired. Machine integration, your software, is a whole world into itself even after you have a solid base and drives. You pretty much have the choice of steep price (off the shelf software) or steep learning curve (open source "free" software) at this step.
Without CAD knowledge to begin with, the CNC is a path to disaster and or confusion, if something's off (it always seems to be).

Make a Dremel-based CNC?

CNC Homemade router, milling by Dremel graving tool PART 3 (1 min 10 sec)

Hear the shaft bearings?

To be useful, it needs to be robust. I've no idea how many Dremel #395's I've owned in my lifetime. At least 7 that I can recall purchasing new. The bearings in these have improved, but not by a lot. Unless you choose bits, cutting speeds, and RPM's carefully, a Dremel can easily be destroyed in one use.

You could make a small CNC to make small helis.
But what's the bigger project now?

This is a REAL CNC at work:
This is an Onsrud 1/4" single-flute bit ripping thru 1/4" ABS sheet in one pass.
This has a 3.5 horsepower Porter-Cable Industrial router mounted in it, and can do 24" x 36" x 8".

If you've cut much ABS plastic, you can appreciate how many inches per second it gobbles this material into bits.

It cost just over $6000, and it paid for itself in under a year, cranking out over 300 of these fake gun bodies. Now it sits in my garage, and I make toys with it.

Better to design a few key parts that showcase good decisions, and then hand the vectors off to somebody who already eats, sleeps, and breathes "Speed and feed".
Originally Posted by Barak1001 View Post

I don't get paid enough to think for people. It's hard enough thinking for myself.
The thread wasn't specifically asking you to come be the hero, I see no reason for condescension.

Guidance is more useful than castigation. Telling somebody off can be a lot of fun. You get to come back, see their reactions. But this is solitary activity, and destructive at best. Who wants to take advice that's shoved at them, even if they "deserve it"?

This person did not ask me to teach them internet CAD, the question was "...can you teach, explain/introduce to me how you make the CAD and use your tools to build it?"

One thing I did forget was: make and throw away a LOT of stuff, about the time this activity makes you want to cry, you will be "good at it".
Last edited by killbucket; Jun 26, 2010 at 08:40 AM.
Jun 26, 2010, 08:31 AM
Resident Skeptican
Barak1001's Avatar
Not absurd at all.

There is no part that goes on a 450 size helicopter that would require a cutting area larger than 3 x 7 inches to make. You don't need a huge cutting area to make small helicopter parts.
Jun 26, 2010, 08:54 AM
Size, is hardly the only limitation of a homebuilt Dremel CNC.

But you DO need to cut metals and Carbon Fiber (CF) to make custom helicopter parts?
No getting around: you need liquid cooling for metals, Uber-bits for CF.
Notice a lot of wood and Delrin being cut with these setups? That's why. They do best with soft, self-lubricating materials, and super low chip load.

Aluminum parts from these are invariably ragged, and those that aren't represent a LOT of experimenting.

Dremels just don't like side-loading, end-loading, or running more than about a 25% duty cycle. Hand pressure on a workpiece is all they're designed for (and I've had no problem cooking them doing just that).
He doesn't say how big his project was...or how long he ran it carving...where you lose track of time, and let it cook your hands. -been there!
Bits: Bits for these are not cheap*, and the Dremel can't hold a standard 1/4 shaft without a heavy adapter =less power available to do work.

*and they WILL be broken, even by the best machinists. They're not made for this usage.

Those stepper motors: I have Vectra steppers, and THEY are considered Poo.

A printer's stepper will not have a long life on much loading, and will not hold position tight enough to make say, an S107's outside sideplate.

Smaller parts take exponentially higher, not lower precision to produce.
On my own CNC, I'd run THREE of them to hopefully get two that fit perfectly.

Tolerances again.

EDIT: I think our discourse has proven enlightening to somebody who may be dreaming of designing their own assembly of any kind. There's a LOT to it.
Last edited by killbucket; Jun 26, 2010 at 09:02 AM.

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion Entry level outside fixed pitch? ghostware Beginner Training Area (Heli-Electric) 26 Mar 26, 2010 08:43 PM
Discussion Looking for 30 size entry level heli ApachePilot Electric Heli Talk 21 Nov 02, 2006 09:28 AM
Discussion Entry level 30 Size heli ApachePilot Mini Helis 13 Oct 31, 2006 10:04 PM
Help! Need Entry Level, limited/no build, E Glider for beginner thecirque Electric Sailplanes 8 Mar 23, 2006 05:18 PM
Wanted Entry level micro heli brianb10 Aircraft - Electric - Helis (FS/W) 1 Sep 05, 2005 02:35 AM