Control via Hydraulics

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
13,035
Location
Memphis, TN
I have a hammer and every problem is a nail and I want to build a 100 story skyscraper is the analogy.
The question comes, are you going to stir and dream and go nowhere or are you going to do what you can do and make the best of it. It actually takes no imagination to come up with far fetched ideas. They don’t have to be reality. You don’t want to turn these ideas into excuses.
 

Hephaestus

Well-Known Member
Joined
Jun 25, 2014
Messages
1,752
Location
YMM
I'm not in the Lower Mainland. Never was. Was in the interior and in Alberta, and have retired to southwestern Saskatchewan where the cost of living is still sane. And the sun shines most of the time.

Don't encoders use a potentiometer signal to supply an analog voltage to the encoder?
Ah I thought you were down Burnaby ways for some reason.

Thousands of types of encoders, but no most don't rely on potentiometers, more along the lines of a hall effect sensor, still a lot of optical ones (think light flashing through a fancy wheel)
 

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
7,521
Location
North Carolina
Carbon pots? Good grief, plastic pots have been around for decades and are a cheapish drop in replacement with way, way longer life.
 

Jay Kempf

Curmudgeon in Training (CIT)
Lifetime Supporter
Joined
Apr 13, 2009
Messages
3,943
Location
Warren, VT USA
But I don't know how to make a RC autopilot work manually without the radio transmitter and receiver.
You can probably grab an RC transmitter stick and figure out how to wire it directly to a Pixhawk or something like that for cockpit level control. Or you could just go directly to the servos with the same stick and skip the autopilot if you just want to drive by trim tabs. Been looking into that as a sub system to get an pretty sophisticated autopilot out of some cheap COTS parts. They all have a good track record of quality and reliability, support, forums full of people that have a lot of knowledge and tweaks.

The UAV I am working on is using Pixhawk. Very impressive little gizmo. You can write your own software and flash it if you want and take control of anything.
 

Pops

Well-Known Member
Log Member
Joined
Jan 1, 2013
Messages
8,143
Location
USA.
Having an autopilot flying the airplane with trim tabs is really a good way to go. You can tailor the response and the control pressure to override the autopilot very easy. I start out with a little larger tabs with little tab displacement and work up to what I want.
 

Jay Kempf

Curmudgeon in Training (CIT)
Lifetime Supporter
Joined
Apr 13, 2009
Messages
3,943
Location
Warren, VT USA
Exactly. Limit the throw of the tabs mechanically so the servo can't freak out and you still have plenty of authority to fly the airplane and it's pretty safe. Trim tab flutter and maintenance to make sure they are sound as part of preflight would probably keep the system safe for a long time.

UGH! CNC router computer won't boot this AM... Another thing on the list.
 

Hephaestus

Well-Known Member
Joined
Jun 25, 2014
Messages
1,752
Location
YMM
You can probably grab an RC transmitter stick and figure out how to wire it directly to a Pixhawk or something like that for cockpit level control. Or you could just go directly to the servos with the same stick and skip the autopilot if you just want to drive by trim tabs. Been looking into that as a sub system to get an pretty sophisticated autopilot out of some cheap COTS parts. They all have a good track record of quality and reliability, support, forums full of people that have a lot of knowledge and tweaks.

The UAV I am working on is using Pixhawk. Very impressive little gizmo. You can write your own software and flash it if you want and take control of anything.
That's kind of where I started - but to avoid going massively into needing to modify the px4 code - probably easiest to make it 2 switches (altitude hold? Heading hold?) And PS2 joystick or hat style D-pad for tweaking. Then into an Arduino mini to turn it into cppm - so basically pretending to be a receiver. That way the flight controller remains unmodified and off the shelf.

Just haven't sat down to figure out the code/wiring.
 

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
7,521
Location
North Carolina
Most of us old guys have learned to think it through before we start pushing the idea on anyone else, and once all the factors are considered, things like temperature changes in a closed hydraulic system and the weight and failure points, the unworkability rules it out before we ever say anything to anyone.
This^

Many, if not most of my 'great ideas' go pretty quickly into the round file. Usually before I've even found a pencil. One of the reasons that 'new ideas' will promptly get panned here is that a lot of us have already pondered them.
When numerous seasoned blokes with years of relevant experience between them tell you that something is a terrible idea, it usually is. If you want to prove them wrong, you'll need more than a good feeling.

Now, I really should get back to designing my Coroplane...
 

Pops

Well-Known Member
Log Member
Joined
Jan 1, 2013
Messages
8,143
Location
USA.
That's kind of where I started - but to avoid going massively into needing to modify the px4 code - probably easiest to make it 2 switches (altitude hold? Heading hold?) And PS2 joystick or hat style D-pad for tweaking. Then into an Arduino mini to turn it into cppm - so basically pretending to be a receiver. That way the flight controller remains unmodified and off the shelf.

Just haven't sat down to figure out the code/wiring.
In RC you have the receiver and then goes to the decoder that separates each pulse out of the pulse train from the receiver. Each pulse output of the decoder goes to each servo.
The servo makes a reference pulse that is compared to the incoming pulse from the decoder. The motor with the gear train in the servo moves a pot in the servo to change the reference pulse width to match the incoming pulse.
Back many, many years ago I helped design the servo electronics. Also designed the first RC servo on the market that was designed to operate retractable landing gear for RC.
Gee-- Somewhere around 50 years ago.
 

Hephaestus

Well-Known Member
Joined
Jun 25, 2014
Messages
1,752
Location
YMM
In RC you have the receiver and then goes to the decoder that separates each pulse out of the pulse train from the receiver. Each pulse output of the decoder goes to each servo.
The servo makes a reference pulse that is compared to the incoming pulse from the decoder. The motor with the gear train in the servo moves a pot in the servo to change the reference pulse width to match the incoming pulse.
Back many, many years ago I helped design the servo electronics. Also designed the first RC servo on the market that was designed to operate retractable landing gear for RC.
Gee-- Somewhere around 50 years ago.
? Have I been drinking without knowing it? Because I swear BBerson asked how to make an RC autopilot operate without the rx/tx... Not sure why we're talking about how servos work...

Maybe I need to go drink a few to make sense of it 😂
 

BBerson

Light Plane Philosopher
HBA Supporter
Joined
Dec 16, 2007
Messages
13,234
Location
Port Townsend WA
I think the autopilot removed from my "Apprentice" RC trainer could work without the transmitter. I don't need to control it in flight. It could operate by itself to keep the wings level and I could manually overpower it's small tabs as needed. But if caught in clouds I could just release the controls. (if it works in clouds) Seems to require a link up first with the transmitter to get the system operating on a level table first.
 

Pops

Well-Known Member
Log Member
Joined
Jan 1, 2013
Messages
8,143
Location
USA.
? Have I been drinking without knowing it? Because I swear BBerson asked how to make an RC autopilot operate without the rx/tx... Not sure why we're talking about how servos work...

Maybe I need to go drink a few to make sense of it 😂
#147
 

thjakits

Well-Known Member
Joined
Jul 3, 2014
Messages
66
Location
Panama, Rep. of Panama
It seems this deteriorated into fly-by-wire vs. fly-by-wifi thread.....

So - back to hydraulics!

Now whatever the merits of traditional controls - I had them all go foul on me one time or the other.
Luckily never to the point of "no-control"....
My boss had a TR-control cable snap on him on short final (installed factory-fresh less than 50 hrs earlier...).
I NEVER lost hydraulics in the air - unless for practise. Usually you can "manage" a single engine helicopter without hydraulics (maybe with exception of the UH-1).

Now my background in this is "a while" fighting all the ailments of a old Bell-47G2, a few Bell-47"Frankenstein"s, dealing with R44s (never a problem...), flying AS350B3s, EC120s, Alouette III, Lama.

I am Industrial Mechanical Engineer (...long time ago), and do my own maintenace on my cars, bikes and bicycles...

First - I was quite surprised with the "...hot approach" Pilot-34 was met - ....let's say so, Mr. Zeitlin was the one I saw approaching P-34 normally in a way that would explain best why and what....

Second - to discuss this properly one needs to appraoch this from the "push/pull-tube, cable"-equivalent in hydraulics, NOT with a boosted control system or power-steering rack...

When you get down to basics, you will have the same feedback like in a cable or push-pull tube system, you will have the same forces acting on the control surfaces and in your input controls - simple physics and mechanics!

Third - besides the "traditional" ways - it is always refreshing to go over "recurring new ideas" - I am certain P-34 is not the first to get the "hydraulic idea"!!

4th - P-34, there is NO reason your idea cannot work reliably!! Looking at all the brake applications out there - I'd say some of the safest systems on moving vehicles! For actuators you might have to look at fabricating them yourself, using motorcycle and car parts where possible, but NOTHING there is rocket science. You need precise reamers, possibly wittle down a few brake cyclinders (there is no boost in there, boost is applied from outside the brake system to get the boost, engine dies you REALLY need to step on it!)
From experience - I think bicycle hydraulics are a little too small for aircraft....

5th - There is no reason you need boost at all! IF you can manage to control the aircraft with cables or push/pull tubes, you can manage with a simple hydraulic system - it's just a different medium to transfer the control forces...

6th - centering is done just the same as with other systems - control surface forces or some spring system!

7th - If unboosted and with out valves you will need TWO lines to every control surface - just like you need a two cable parts going there and back. Only push/pull will require only one item per surface...

8th - Your system would work just like a motorcycle brake with custom actuators on the other end, basically replicating your hand-grip hydraulics (if you want a 1:1 transfer of motion, otherwise adjust as required).....incl. the hydraulic reservoirs, I think these could be ganged together. Redundancy: if one line fails you have 1/2 the control available...unlike a broken cable... - if your design self-centers the control surface, you have one of the two displacement directions available..

9th - Control feedback and feel - in a lite aircraft - nice, ...or not! I wish the R22 had hydraulics to stop the shaking of the stick - ....well, I really don't care anymore, hardly ever get into one...
IF you have boosted hydraulics in an aircraft, it usually takes away ANY feedback from the rotorsystem or control-surface system!! That's how they are designed!! IF you feel feedback - you have either overpowered the design control force (too heavy, too hard maneuvering) or you have lost your hydraulics! NO control feedback in normal operation!
Your "feel" comes from your body - being attached to the airframe - you feel what the airfram does!! And that's what you want....
It is a change to having light controls all the time! .... from getting heavier controls as speed and maneuvers increase!
(Mind you - in the above simple, non-boosted system you still feel the control-surface feedback!!)

I have ..."some time" in helicopters and am used to fly without "moving the controls, but applying pressure to the controls" - by nature most helicopters are rather sensitive on the controls - compared to airplanes.
The only one I had a chance to control for a few minutes and that approached helicopter-sensitivity was a Glassair III - what a plane!! Point it where you want to go and it goes right there - a little right pressure and you roll, a little more pressure and you roll all the way around!

10th- P-34: I understand where you come from and what you know about hydraulics - I agree!
However - I assume we also agree - heavy machinery hydraulics are not going to work in a light airplane!!

11th - Pilot-34, IF you seriously want to use hydraulics in a small aircraft - I suggest you build a test rig and work it out! Maybe you want to do that anyway, just as a project!!

12th: IF it was me, I would try the utmost to go with push/pull where possible!! Once a well maintained push/pull system definitely fails - ...it doesn't matter anymore, it will likely be the last thing to fail on that airplane...
THOUGH - coming from helicopters, I sometimes shudder seeing small plane control connections!
I'd insist in double horns (forks) throughout, no sideloads possible....
Try with sealed bearings, if these have too much friction in the seals, I'd pull the seals, have me distanced covers made on a lathe, and get some superhyper-grease and check them periodically... On something like a BD-4 that would be a "quick check"....
(I was used to grease more than a 100 zerks every day - incl. seperate oil-zerks for the TR-shaft...Alouette III and Lama)

I appreaciate your "hydraulic inspiration" - always great to get to discuss a different approach...


Now - about that "adjustable canard incident angle to maximize efficiency"....


Cheers,

thjakits
 
Top