HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Feb 15, 2012, 03:46 PM
I bail out, anywhere, anytime
Taurus Flyer's Avatar
The Netherlands, OV, Almelo
Joined Nov 2010
2,958 Posts
Zenoid

Don't shoot the messenger!! It is all to learn,

You write in your blog:

I've had some time to make some works on the android software, it is glider oriented and the goal is still to achieve racing almost like the real ones.

The location facilities are indeed amazing, GPS positioning, turnpoints, indication of the track but.
In the past we did use maps to recognize the location and route and for example a photo camara to prove we did visit the location. So of course the task of a navigator (part of the pilot) is remarkable easy with your system but only a substitude for (old) navigation tools related to the real ones
The system is also a helpfull tool to organize and control a contest of course!

So that's all right, related to the real ones but what is related to this thread important??

Weakest point of your system is the vario and airspeed measurement as expected.
Read the PDF file of Dan again and especially the last chapter about airspeed measurements.

The variometer vertical speed and the horixontal airspeed have to be measured with highest accuracy in the first place and I don't read anything about that accuracy?

So specifications of the transducers are important to know and to qualify

From you blog
- variometer, with TEK eventually, low noise
- airspeed baro sensors


The pitot tube, I see in your pictures, doesn't have static orifices and seems to be just as the TE tube "self plumbed", but If not calibrated, the signals of these elements are useless for any calculation or signal processing.

Because I read : almost like the real ones or The variometer play sounds like real glider.My impression is you are not familiar with real variometers as used for XC/thermal gliding.

This is not a shortcoming of your system, it is designed the way you intended but it is the shortcoming as result of your lack of know-how.
Be aware the content of this thread can be complicated but also will be informative.
My advice to you is do google for McCready calculations for variometers, it can be the next step after reading about the "polar(s)" of the glider to understand about flying the real gliders.

Cees
Taurus Flyer is offline Find More Posts by Taurus Flyer
Last edited by Taurus Flyer; Feb 15, 2012 at 03:52 PM.
Reply With Quote
Sign up now
to remove ads between posts
Old Feb 16, 2012, 01:54 AM
Registered User
Joined Jun 2005
156 Posts
Quote:
The pitot tube, I see in your pictures, doesn't have static orifices and seems to be just as the TE tube "self plumbed", but If not calibrated, the signals of these elements are useless for any calculation or signal processing.
You are right on this point. There is no static on the photo. But this prototype has been mounted with a sensor I already had (it have only one port for a tube, static is under the component). The next board will have dual ported air speed sensor. I will also have some pitot tube made the right way. The Easyglider bench test is just there to test the total integration of hardware/firmware/software. But the question of glider calibration is still open, so I need some hints about real glider calibration technics. I know my air speed is about right because I compared it with my car one and gps one in still air.
The next prototype will be properly mounted on a 8.5 meters ASH25 with static port and all.

Precision wise, the variometer noise is less than 10 cm/s. It is responsive and sensitive enought for RC gliders, from the experience I have with it. You can easily track some very weak thermals (< 1 m/s).

Quote:
The variometer play sounds like real glider.My impression is you are not familiar with real variometers as used for XC/thermal gliding.
I'am not a real glider pilot, I regret that, but I'am a champion at Condor simulator , so I'am kind of familiar with XCsoaring and the tools that come with it.

Could you tell us how the variometer shown here is made ? Is it all mechanical ?

Regards.
zenoid is offline Find More Posts by zenoid
Reply With Quote
Old Feb 16, 2012, 02:34 AM
I bail out, anywhere, anytime
Taurus Flyer's Avatar
The Netherlands, OV, Almelo
Joined Nov 2010
2,958 Posts
Zenoid,
While I was downloading this post I see your input.
I will make a post/answer for you!


Gents,

There isn't too much input anymore but I want to know it all, so let me give it an extra pulse.

In the descriptions and the photo's about XC flying I observe, the teams are using cars.
Does the pilot controls the plane while the driver drives the car? So on the same moment? Principial question for which I need an answer!

Why this question?

In my study about variometers I found a document in which one of the modeling variometer manufactureres did write, his variometer was to decide to stay or go (in a thermal) , or with other words his instrument was useless in the period "go"!! Which for a real vario is "nonsense" of course.

This manufacturer didn't mention any profit of the vario during " travel of the team ", read, driving the car!!


So is the pilot flying the plane while the driver is driving the car?

If so, Who is the captain, who decides the speed of the car?

Or with other words:

Does any team uses a variometer with Mc Cready facility
Does any team uses the netto facility or relative netto

In both cases these facilities as primair part of the vario instrument "real time on line" , not as result of the input of the navigator in commands as result of his "calcuations", only input of the navigator can be in this case the Mc Cready value to adjust by the pilot.

May I have your votes please?

If it isn't all clear, give me a sign and I will formulate the questions another way!


Cees
Taurus Flyer is offline Find More Posts by Taurus Flyer
Reply With Quote
Old Feb 16, 2012, 03:13 AM
launch low, fly high
New Zealand, Hawke's Bay, Havelock North
Joined Dec 2004
1,856 Posts
Cees,

I calculated the optimal speed to fly for my XC ships a looong time ago (back in the '80s), and try to fly something approximating this. In operation, the pilot usually determines the vehicle speed so as to best position the aircraft with respect to the vehicle. One also needs to take into account the road variations ahead in doing this...

When I flew XC, the usual concept for me when departing from a thermal was to yell out a target speed for the vehicle. Then, I would use delta calls (plus 5, minus 10, etc.) to modulate the vehicle speed.

A humorous anecdote on this. Turned out that my normal driver (and wife) is a competitive sort, and I found that I used about twice as many minus XX calls as plus XX calls. I'm pretty sure that if you summed up the delta calls between thermals, we shoulda been going about minus 80 mph... She knew that I flew just a little bit conservative in my speed, so she continually nudged me via attempting to make me keep up with the vehicle! I finally figured this out by watching the speedo through the cab window when I made a few speed change calls.

One item that is really important in the speed to fly assessment, is to determine the amount of lift/sink you are currently flying through. The best flight speed is dependent on both the expected thermal strength as well as the local lift/sink.
Joe W is offline Find More Posts by Joe W
Reply With Quote
Old Feb 16, 2012, 04:13 AM
I bail out, anywhere, anytime
Taurus Flyer's Avatar
The Netherlands, OV, Almelo
Joined Nov 2010
2,958 Posts
Hello Joe,

11543 miles away, amazing, 18577km!!
Nice to read the way XC is done and, from the first hand!! (or must Isay the second and your wife the first? LOL)

I read your story and will show my impression later.
One of the questions of course will be:
What kind of vario, yes or no Mc Cready facility in the instrument? (so not by separate calculation of speed).
I also can ask, did the plane have some kind of realtime airspeed measurement and feedback to the pilot?

More tt follow for your post, preparing a post for Zenoid.

Cees
Taurus Flyer is offline Find More Posts by Taurus Flyer
Reply With Quote
Old Feb 16, 2012, 09:38 AM
I bail out, anywhere, anytime
Taurus Flyer's Avatar
The Netherlands, OV, Almelo
Joined Nov 2010
2,958 Posts
Zenoid,

First back to the basic, because I read in your post 17:
But the question of glider calibration is still open, so I need some hints about real glider calibration technics.

Main goal for me starting this thread was collecting some information about used variometers systems and gain discussion
Information about used systems can be showed.

Fact is, the models and systems I write about are "as old as the way to Rome" shown by examples as the Mc Cready vario / final glide computer.
In this instrument logarithmic and antilog amplifiers and all kind of other analogue and binairy type of circuits are used to realise needed calcuations and signaling. (BTW highest possible resolution and signal processing speed for the pilots primary information!!)

The principle of TE tubes is nearly 50 years old (!!) and other TE compensating systems/principles even older.Theoretical back ground 100 year!

It is probably a lack of know-how of physics I discover (nearly) no advanced variometer systrems are used as far as I read. But my way of thinking still can change!!

Zenoid, In your post 13 you wrote:

Quote:
Originally Posted by zenoid View Post
.

So Cees if I understand correctly, what is needed to achieve those points is to be able to somehow calibrate the sailplane and its sink polar curve in still air. Then you have the material to give real time netto or speed to fly or final glide.
I've found some stuff on the www and wrote some code this afternoon that extract the famous polynomial equation from pairs of air / vertical speed. I get LD, speed to fly and all the constants of the equation. Got the experimental polar equation.
But what is of any use when you are a rc pilot ?
I'am not sure all those netto, SPDF, final glide are of any use in Europe as racing is very local. It's different in the US. Other valuable stuff :
Being able to qualify the glider after you have build it ?
Do some aerodynamic research ? May be I'll add a kind of "lab" tab in my application. Interesting thread.
And in post 17 you write:

But the question of glider calibration is still open, so I need some hints about real glider calibration technics.


Impossible I am afraid, there is more to do!

You have to analyse the plane and calibrate the system, implement all the data and theoretical needed calculations and pilot input requirements in the sotware, and for example calculate the airspeed and other data to fly.
To inform the pilot there is needed a MMC in a ergonomic way.

But, that still is a long way to go, starting for example with some kind of basic design requirements.

I give you one example.
When you generate a "speed to fly" and the airspeed measurement part of the system (pitot tube or calibration!) has a inaccuracy, the result is, you are forced to fly a speed exactly not to fly!
So people are impressed how you not win the contest!!

It looks like modeling is 50 years behind related to the real world when speaking about variometers and lack of investments and development can be the main reason.

More to come if people are interested and Joe, preparing my response.



Cees
Taurus Flyer is offline Find More Posts by Taurus Flyer
Last edited by Taurus Flyer; Feb 16, 2012 at 09:48 AM.
Reply With Quote
Old Feb 16, 2012, 10:25 AM
I bail out, anywhere, anytime
Taurus Flyer's Avatar
The Netherlands, OV, Almelo
Joined Nov 2010
2,958 Posts
Joe
Let we give your wife for a while the name Mary in the humorous anecdote!

MMC

Man Mary Communication!!


Indeed the determination of speed is the most important so my question is.
Did you use wind speed and direction, direction of flight and trafel, so related to wind and Mc Cready calculation, to calculate the speed of travel? (hope it is clear??)

Did you have any information of (climatic) changes. experienced by, and as feedback of, the plane? Downwind, airspeed? So yes, were these data useable for calculation in form of data, values or only an acoustic signal?

BTW!
Value of heigth of course important for calculations but not for me to ask!! Not primary part of the variometer!

Joe, you wrote (back in the '80s) but, do you know what is changed compared these days? Is there any pilot nowadays informed, "on line" and "real time", what airspeed to fly with feedback of the glider's real airspeed?

Will be interesting to know answers for these questions, I am curieus.
If anybody want to add information, don't hesitate it is for global humanity!

Thanks!

Cees
Taurus Flyer is offline Find More Posts by Taurus Flyer
Reply With Quote
Old Feb 16, 2012, 10:26 AM
Registered User
Beaverton, OR
Joined Apr 2002
164 Posts
Quote:
Originally Posted by zenoid View Post
... But the question of glider calibration is still open, so I need some hints about real glider calibration technics.
Zenoid, are you referring to calibrating the glider itself, meaning finding best L/D, sink rates, etc. -or- calibraiting the equipment i.e. airspeed sensor, pressure (altitude) sensor, etc. ?

- Bob -
HLGNut is offline Find More Posts by HLGNut
Reply With Quote
Old Feb 16, 2012, 12:05 PM
Registered User
Joined Jun 2005
156 Posts
From what I understand, flight computer need glider polar curves to compute speed to fly or final glide. Computers also use glider's properly certified and well calibrated sensors. Gliders polar curve are obtain from theory and flight tests corrections along research sailplanes that are constantly calibrated for that purpose. So for me to be able to compute those speed to fly and final glide, I need to somehow do what glider manufacturers do, which is impossible.
Concerning my board, sensors are already compensated at plant and my analog converter has a huge precision and powerful calibration schemes. I managed to get a very low electronic noise so sensors precision is good enough. Then for the polar acquisition, doing some statistic works on flight traces will be enough to catch a good approximation (I think it is done that way in XCsoar xcsoar/).
Regards.
zenoid is offline Find More Posts by zenoid
Reply With Quote
Old Feb 16, 2012, 12:16 PM
I bail out, anywhere, anytime
Taurus Flyer's Avatar
The Netherlands, OV, Almelo
Joined Nov 2010
2,958 Posts
Zenoid

You understand my way of thinking is topdown, so we could have a view in the specifications of the first step of the airspeed transmitter, this is the differential pressure to voltage or current transducer. You write "my analog converter has a huge precision and powerful calibration schemes". Can you show these information!.

Cees
Taurus Flyer is offline Find More Posts by Taurus Flyer
Reply With Quote
Old Feb 16, 2012, 01:09 PM
Registered User
Joined Jun 2005
156 Posts
24 bits ADC with MP3V5004
zenoid is offline Find More Posts by zenoid
Reply With Quote
Old Feb 16, 2012, 01:47 PM
Registered User
Beaverton, OR
Joined Apr 2002
164 Posts
Quote:
Originally Posted by zenoid View Post
From what I understand, flight computer need glider polar curves to compute speed to fly or final glide. Computers also use glider's properly certified and well calibrated sensors. Gliders polar curve are obtain from theory and flight tests corrections along research sailplanes that are constantly calibrated for that purpose. So for me to be able to compute those speed to fly and final glide, I need to somehow do what glider manufacturers do, which is impossible.
If you are saying that it is hard to generate the polar curves with well calibrated equipment for our models, I would disagree.

To generate a sink polar, it is fairly easy to do.
Do this in the early morning before the wind comes up or thermals are present.
Measure out a course of say 100M with people at the start and finish gates with flags.
Have your sensors outputting altitude measurements.
Launch and fly through the course, having the start gate person drop a flag when coming through the start gate. Start a stop watch. Note altitude at start.
Have the stop gate person drop another flag when going through the stop gate. Stop the stop watch. Note the altitude.
From this you can calculate the speed and sink rate.
Repeat this for several speed (elevator) settings.
This will give you the sink polar graph.

Dan's PDF on the SBXC Performance testing has a lot of the calculations explained.

The sensor that would most like need to be calibrated is the airspeed sensor.
One way to calibrate the airspeed sensor, which is normally a differential pressure sensor, outputting a voltage value; which is then just read by the A-to-D converter of the processor; is to build a "Y" tubing connection that goes between the pressure sensor, a IAS gauge, and a pressure source, which could be a simple syringe, w/o the needle. Have the processor repeatedly outputting the A-to-D values, press the syringe to get to the airspeed you want as shown on the guage, read the A-to-D value and write down the value. Repeat for all speed values. Put the values in a lookdown table that the processor can read. I used a 20 - 140 MPH gauge that I bought from Aircraft Spruce some time back.

You probably already know this....

- Bob -
HLGNut is offline Find More Posts by HLGNut
Reply With Quote
Old Feb 16, 2012, 03:02 PM
launch low, fly high
New Zealand, Hawke's Bay, Havelock North
Joined Dec 2004
1,856 Posts
Cees,

The sensor information was limited to a variometer with a TE probe. All other information is inferred. The airspeed was defined via a high static margin and somewhat calibrated trim, with additional information from the ground speed with some assumptions about the wind speed/direction. The atmospheric data (lift/sink) is inferred via the TE variometer signal and the airspeed.

As to my plane in the '80s... I built Wylie for XC flying in the late '80s, and it hasn't been beaten at an XC event yet. Of course, it hasn't been campaigned in about a half dozen years, and needs some work to return to flight status. It is a bit outdated (eppler 374 airfoil, no wing camber adjustment), but is likely to still do okay. XC is still very much a pilots event rather than an equipment event.

I do worry about the escalation of electronic technology for XC flying. I'd prefer it was the pilot that did the data integration rather than a computer. If a computer gets into the flight execution, the eventual result is likely to be that XC competitions will be won or lost via the skills of the autopilot programmer rather than pilot skills. I turned down the offer to use a sensor suite in XC a long time ago, as I felt that using it would not be beneficial to the health of XC flying.
Joe W is offline Find More Posts by Joe W
Reply With Quote
Old Feb 16, 2012, 03:05 PM
I bail out, anywhere, anytime
Taurus Flyer's Avatar
The Netherlands, OV, Almelo
Joined Nov 2010
2,958 Posts
Quote:
Originally Posted by zenoid View Post
24 bits ADC with MP3V5004
Zenoid,
"my analog converter has a huge precision and powerful calibration schemes".

Are you trying to sell me one?? I only see a incomplete number!
Keep in mind I am never impressed only want to be convinced so, 24 bits will do, it isn't the resolution of analogue but oke(LOL), but no chain is stronger than the weakest link, the MP3V5004

My information facility is reduced to basic info so please first check because you are the seller

The MP3V5004 has a accuracy of 1.5 % + or - over the lowest part (0 - 100 mmwk) of span?
Span is fixed and is 306 mmwk ( 3kPa) over delta out V FSS of 1,8 V = 1800 mV ?

Is this correct?

Cees
Taurus Flyer is offline Find More Posts by Taurus Flyer
Reply With Quote
Old Feb 16, 2012, 03:14 PM
I bail out, anywhere, anytime
Taurus Flyer's Avatar
The Netherlands, OV, Almelo
Joined Nov 2010
2,958 Posts
Joe,

Amazing, all correct and very interesting, because better no measurement than wrong measurement and than the static marging is a safe method.

I understand the disadvatage was no "netto" or "relative netto", but, you probably already know we will see, problem is measurement of speed!!

I will prepare a response, based on? You guess, Eppler 193.

Cees
Taurus Flyer is offline Find More Posts by Taurus Flyer
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Question What kind of micro servos are these? turbodriven Radios 3 Nov 30, 2011 09:37 PM
Discussion What kind of servos are best? papabearflying Electric Plane Talk 11 Oct 16, 2011 02:18 PM
Discussion Lipo Pack Failure - What Kind Of Warning Signs Are There? itsme2 Batteries and Chargers 5 Aug 15, 2011 11:12 AM
Discussion What power setups are being used for Rake models? highace1 Scale Kit/Scratch Built 31 Jan 14, 2007 04:14 AM
What Kind of Foam are used in FLying Styro? aplphabandaid Parkflyers 2 Oct 19, 2004 09:18 AM