GB700 / Trex 500 Flybarless (project 2)

[Borneoben]

Originally Posted by BobOD

That gives me an idea. I should start a "how it's made" video thread. I bet folks would enjoy that.

Finding the time is the hard part.

Bob

I would LOVE to see more info on how its made.

I would also be very interested in information like
Types of cutting tools
Cutter RPMs and feed rates for different materials.

I am making some RC Heli parts of my own and i am not happy with the finish.
Themachine shop i use is used to large stuff not small stuff so not too experienced. They use large 17ton HASS machines.
Check out my thread if your interested. its pretty long so just skip through it.
https://www.helifreak.com/showthread.php?t=108482

[BobOD]

Nice stuff. I'm thinking of doing a frame one day so it'll be an all I2 heli. Metal cutting is my favorite work though.

I'll be doing some vids one of these days. It's pretty cool when the chips get really flying.

For aluminum, I run Zrn coated micrograin carbide ground exclusively for aluminum. And I never let it touch steel...it'll never cut quite right again if it does so it's a whole other set of tooling for steel parts. I probably have invested $10000 in tooling alone.
Indeed, big stuff and little stuff are worlds apart. There's still a lot of art in getting the finish without a bunch of post finishing (which erases those nice precision CNC surfaces IMO).

Speeds and feeds...there's no one answer. I run a 1/2" endmill anywhere from 25 to over 100 in/min depending on what it's doing.

Bob

[BobOD]

Originally Posted by BobOD

From centerline of the feathering shaft to the grip ball is 17mm BUT, there is a reduction in the levers depending on the hole used on the lever. The "effective" distance would be:
25mm on the inner hole
22.5mm on the middle hole
22mm on the outer hole

Bob

BTW, that should have been:
25mm on the inner hole
22.5mm on the middle hole
20mm on the outer hole

It was a typo.
Bob

[BobOD]

Originally Posted by Wazzer

I would LOVE to see more info on how its made.

Here you go. Making some more swashplates. Won't be out of stock for long.

http://www.i2rc.com/videos/Swash1.AVI

It's cool to see how the table gets thrown around in this one. Probably a few thousand pounds with everything on it and it'll position to 0.0002". Hows that for some bad @$$ servos? http://www.i2rc.com/videos/Swash2.AVI

I slowed things down and shut the coolant off for this one. http://www.i2rc.com/videos/Coolant%20Off.AVI

Sure beats hand tapping. http://www.i2rc.com/videos/Taping%20in%20Indexer.AVI

And here's the finished goods as well as the stuff that doesn't get to fly.

[kyalamikid]

Aaah, nice vids Bob. I am really grateful that you get into this stuff. Amazing to se ehow precise the equipment is.

Maybe you should go into bearing manufacturing as well

[Borneoben]

Great Videos Bob thanks for sharing.

I promise i will reply to your email sometime over ths weekend. I have been crazy busy with my real job!

Cheers

Ben

[BobOD]

Originally Posted by kyalamikid

Aaah, nice vids Bob. I am really grateful that you get into this stuff. Amazing to se ehow precise the equipment is.

Maybe you should go into bearing manufacturing as well

Bearing mannufacturing is just a little to specialized for me to get into. Sometimes I wish I could though because sourcing a reliable supply takes more of my time than I'm happy with. I buy only ABEC5 rated bearings but still, I have to QC every one.

[kyalamikid]

Originally Posted by BobOD

Bearing mannufacturing is just a little to specialized for me to get into. Sometimes I wish I could though because sourcing a reliable supply takes more of my time than I'm happy with. I buy only ABEC5 rated bearings but still, I have to QC every one.

That wall is getting bloodied.....

[kyalamikid]

I put in 14 flights this weekend, and the T500 FBL with I2 head and swash just blows me away! I felt the control was good enough so I took out expo and put the endpoint at 60%. This thing does flips in around 1 second now, I am really enjoying it. It corners like a wild thing... crazy! I will push this more this weekend coming.

Thought I wouldn't be able to fly this weekend after toasting my HW70A with a Scorp 1400.. My trusty Jazz 40 with a heat sink came to the rescue whilst I wait for my new CC Ice75 to arrive. I am running 2800rpm headspeed with the stock motor on 13T governed and pushing 78 amps through the Jazz for 1720 peak watts... Jazz comes down at 38 degrees Celcius. No bad for a 40A ESC... Using Mavvrik blades.

[BobOD]

AAAAHHHH, there's no better sound than a happy customer. Thanks for the feedback.
What position are you using in the ratio lever and servo arm length. I think you're the first to really experiment with this so I'm curious.

Wow, you're really going to go for some power with that Ice75 aren't you. Gosh, I hope she holds together.
Just kidding. Anthony (Jag72) is running that much and all's good...well now that he found the weak link that is. :o Did you get the new steel turnbuckles yet?

They make the tracking adjustment a snap...um err...I mean a breeze.

[kyalamikid]

Originally Posted by BobOD

AAAAHHHH, there's no better sound than a happy customer. Thanks for the feedback.

The head is great Bob!

Originally Posted by BobOD

What position are you using in the ratio lever and servo arm length. I think you're the first to really experiment with this so I'm curious.

I am using the inner most holes at 25mm. I can run much higher gains at this position, around 300 on gain. At outer holes this drops to 200. I will try moving out again later, but work great for now.

Originally Posted by BobOD

Wow, you're really going to go for some power with that Ice75 aren't you. Gosh, I hope she holds together.

In all honesty I am holding on to the 1720 watt setup by the skin of my teeth, but its such a blast I cant stop. I reckon to get 2000watts with the Scorpion.

Originally Posted by BobOD

Just kidding. Anthony (Jag72) is running that much and all's good...well now that he found the weak link that is. :o Did you get the new steel turnbuckles yet?

Are you kidding, last I heard they were over New Jersey.... actually we had a postal strike last week, so I am hoping they come in this week.

Originally Posted by BobOD

They make the tracking adjustment a snap...um err...I mean a breeze.

Would that be USPS tracking

[BobOD]

Originally Posted by kyalamikid

Would that be USPS tracking

Wise arse.

Hey, that brick wall is getting a special spot just for USPS. I shipped some heads and tails over to the UK, like forever ago, for a show they are having. It got there today. Show was over yesterday.

[BobOD]

Originally Posted by kyalamikid

The head is great Bob!


I am using the inner most holes at 25mm. I can run much higher gains at this position, around 300 on gain. At outer holes this drops to 200. I will try moving out again later, but work great for now.

Just an FYI, the gain is relative to the input range. In other words, a gain of 200 at an increased throw doesn't really mean you are using a lower gain. I'm not sure if I'm explaining that well...but I'm a little tired. Anyway, most important is that it's locked in.

Cheers.

[Jag72]

Here's a crappy little vid...just to show you the head and tail movement for the I2 head and tail...with the Gyrobot 700.

Sorry for the abrupt ending...my Camera battery died and shut off

[kyalamikid]

Nice looking tail Anthony (dont tell my wife I said that). I see you are using the HD 3D pro blades... Looks like your ball links are in the middle hole, and you have the new adjustable links... Love these heads, they look the part.

[kyalamikid]

Originally Posted by BobOD

Just an FYI, the gain is relative to the input range. In other words, a gain of 200 at an increased throw doesn't really mean you are using a lower gain. I'm not sure if I'm explaining that well...but I'm a little tired. Anyway, most important is that it's locked in.

Cheers.

I would appreciate it if you can expand on this. I can see that there is a direct relationship between throw and gain, but I am trying to understand this better.

Yes I can well believe USPS did that... it seems about par for the course....

[BobOD]

I'll give it a try. The gain is basically the amount of response to a given measured error. The response (as far as the gyro is concerned) is how much to move the servo. But, in reality, the response that matters is how much thrust change is applied to the rotor disk (actually even this is simplified but good enough for now). The gyro does not know this relationship so you are given gain values to play with in order to inform it. There are many factors involved. Even increasing head speed can cause need for a lower gain. That's because when the HS is higher, less pitch change is required for a given thrust correction. But, added gyro forces may reverse this. One could go on and on...I guess I'm doing that already.:o

So, let's say you have a setup that causes a
5 degree pitch change with a 30 degree servo arm change
and a gain of 300,
and you change it so you now have a
10 degree pitch change with a 30 degree servo arm change
a gain of 150 should be comparable. Resolution has obviously changed but that is a different subject...even though it can affect the workable gain.

I hope that helps. Usually, a good understanding helps achieve the best results so I'll offer this for you as well. I believe P and I stand for proportional gain and integral gain. They're common terms in feedback loop systems so I assume this is their meaning. Some rainy day, you could do a search. Actually, this is some interesting reading.
http://en.wikipedia.org/wiki/PID_controller
I don't know why the derivative gain (used in time based calculations) is not made available but it could be that it could get you into really bad trouble. I suspect the gyro is programmed to determine the best value on it's own.

[Jag72]

I am actually flying the EL CHEAPO $20 3d blades right now and I tell you what...They actually fy VERY well overall...at least for me...

I got sick of smashing sets of $60 blades so I decided to fly these again for a while

I still prefer the maniacs and SWE's but these are decent blades for $20 a set...

[kyalamikid]

Yup Anthony, I have 4 sets waiting at the post office for me. Glad to hear they work well. You think you have problems, Mavvriks here are $110 a set!!

Thanks Bob your explanation makes sense. So you are saying there is a direct relation between gain and throw for a given servo arm movement.

I read up on PID's already with specific focus on P and I due to trying to understand the Gyrobots better. Seems to me the P is the proportion of how much to correct for a given deflection, and integral is modifier on how rigorously to hold a specific heading. This understanding seems to work OK for me, ad the algebra involved is a little more than I can take.....

[BobOD]

Originally Posted by kyalamikid

Thanks Bob your explanation makes sense. So you are saying there is a direct relation between gain and throw for a given servo arm movement.

Yes. To elaborate, since the feedback (that the gyro is measuring) is angular motion, anything that affects the amount of angular motion for a given servo signal change will affect the optimal gain value. Another example is, if you reduce the weight of the heli, it will respond more quickly to a given servo input. So, the optimum gain is expected to go down.

You can begin to see that the gain value is really a poor indicator of quality of setup...although many think that the higher their gain, the better their setup. This is true if comparing apples to apples, but quickly disintegrates as things get altered.
Things like slop and poor dampening and slow servos and poor mixing resolution all tend to reduce useable gain though so there is some merit in it as a measure of quality, but only when considering the many other variables.