RunRyder RC
 9  Topic Subscribe
WATCH
 2 pages [ <<    <    ( 1 )     2     NEXT    >> ] 1533 views
Scorpion Power Scorpion Power
HelicopterMain Discussion › Remote recording of headspeed...
10-19-2004 08:56 AM  13 years agoPost 1
oyvindk

rrNovice

Haugesund, Norway

My Posts: All  Forum  Topic

A long time ago I created a piece of software to measure the power and torque for car engines without having a rolling road. The software used a recording of the engine RPM while it was accelerating flatout on a level piece of road. The software then needed all gear ratios, tyre dimensions, frontal area, cw value etc to do it's job.

Well, this is mostly trivial, the exciting part was how to get the engine RPM into the program.

A hint: Absolutely NO direct interfacing to the car is/was necessary...

The same method can be used to show the rotor/engine RPM of a model helicopter remotely, without anything extra installed in the helicopter. To be fair, this only works after the fact, meaning you have to analyse the data after the flight.

With this method I discovered that I actually had a hover RPM of 2200 RPM on my Raptor 50V2. I knew it was a bit high, but that came as a surprise.. During a descent my rotor RPM approached 2400 RPM...

You've probably guessed that I don't have a governor. Yet, anyway...

Intrigued? Want to know more? If there is interest, I'll consider refining the software for R/C usage, and possibly giving it away for free.

PM  EMAIL  Attn:RR  Quote
10-19-2004 12:48 PM  13 years agoPost 2
frebou

rrKey Veteran

Montreal / Canada

My Posts: All  Forum  Topic

what do you use to record your stats?

PM  EMAIL  HOMEPAGE  GALLERY  Attn:RR  Quote
10-19-2004 12:53 PM  13 years agoPost 3
oyvindk

rrNovice

Haugesund, Norway

My Posts: All  Forum  Topic

I'm using a device called MP-FUB26...

PM  EMAIL  Attn:RR  Quote
10-19-2004 12:55 PM  13 years agoPost 4
Hercdriver7777

rrApprentice

South Carolina

My Posts: All  Forum  Topic

A camera?

Good Flying!!

PM  EMAIL  HOMEPAGE  GALLERY  Attn:RR  Quote
10-19-2004 01:01 PM  13 years agoPost 5
frebou

rrKey Veteran

Montreal / Canada

My Posts: All  Forum  Topic

a mp3 player?!? Sounds interresting. How do you use it?

PM  EMAIL  HOMEPAGE  GALLERY  Attn:RR  Quote
10-19-2004 01:02 PM  13 years agoPost 6
erekose

rrNovice

France

My Posts: All  Forum  Topic

Hi,

Do you actually record the sound of the heli, and doing a FFT you get the frequency of the sound, that you use to derive the headspeed using the gear ratio ?

I've thought about that some time ago, but I don't have the knowledge to do it myself (nor did I think to use something existing as a recorder).

Am I right ?

Julien

PM  EMAIL  Attn:RR  Quote
10-19-2004 02:56 PM  13 years agoPost 7
jb_turner

rrElite Veteran

USA

My Posts: All  Forum  Topic

But why?? Just use a Tach Like MA or Model Avionics.

PM  EMAIL  GALLERY  Attn:RR  Quote
10-19-2004 03:15 PM  13 years agoPost 8
erekose

rrNovice

France

My Posts: All  Forum  Topic

Why :

- You fly alone and can't have somebody tach your heli while you're flying
- Your headspeed is so messy that it is not possible to synchronize the tach with it
-you don't want to pay 85$ and you already own an mp3 recorder
- You gave up listening to music, and you're looking for a way to recycle your mp3 recorder.
- You want to show off bringing your laptop to the field
- you don't want to do like everybody else
- you want to have fun writing a program
- you want to get your max headspeed during flight
- more generally speaking, you want to be able to calculate the mean and some percentiles, or plot a distribution graph of your headspeed during flight
- You want to be able to tell your buddies : "yes, I wrote this program Myself !!"
- You're a student in acoustics and that's part of a project

right, I'll stop there, but I'm sure that with a little bit more thinking, I could find dozens of reasons not to use a MA tach

Julien

PM  EMAIL  Attn:RR  Quote
10-19-2004 03:26 PM  13 years agoPost 9
frebou

rrKey Veteran

Montreal / Canada

My Posts: All  Forum  Topic

I hope you are not just recording the sound.

Please tell me you are recording the "data" that the sensor on your fan is sending and than "playing" it back to your program...

Sounds very interesting though. Please stop making me wait and tell me the whole story.

PM  EMAIL  HOMEPAGE  GALLERY  Attn:RR  Quote
10-19-2004 07:18 PM  13 years agoPost 10
oyvindk

rrNovice

Haugesund, Norway

My Posts: All  Forum  Topic

I've been to the field today, and I have done some more testing. I am very pleased with the results.

@erekose: Your post was hilarious! You were more or less spot on with many of your assumptions... :-)

Yes I am just using a simple and cheap mp3-player to record the sound of the helicopter. I'm sorry it isn't more exciting than that, frebou...

It would of course also be possible to generate some tones with the right pitch for different RPM's and save them as mp3 on the player. On the field, just insert one of the earpieces in one of your ears, and tune by that.

However, I am using a frequency analysis method instead. It isn't FFT, but it does more or less the same thing.

Today I recorded the sound of two different machines that had a governor, and the results are spot on for both. That proves it for me..

For testing hovering RPM, this is a great method. If recording the sound of a helicopter flying at high speed, the doppler shift of the sound will distort the readings. However, if the helicopter is doing more or less stationary manouvers, it will work just fine.

I also got another idea at the field today. I laid the recorder on the ground, and flew in a low pass directly over it. I then calculated the helicopters groundspeed from the doppler shift in the recording.

If my calculations are right, my stock Raptor 50V2 did a little over 100 km/t, equating to approximately 63 mph. Not too bad, and accounting for the very slight cosine effect from the angle between the helicopter and the recorder, the speed was actually a little higher than that.

An additional bonus is that you don't have to use a recorder like I do. Any recording device will do. You can even use existing video clips downloaded off the net. You can always get the engine RPM this way, but you will need to know the gear ratio to get the headspeed.

As long as you can get your hands on a sound recording and have some computer skills, this isn't too complicated.

PM  EMAIL  Attn:RR  Quote
10-19-2004 08:46 PM  13 years agoPost 11
JKos

rrProfessor

Redondo Beach, CA

My Posts: All  Forum  Topic

oyvindk,
It would be very kind of you to make such software available and easy to use.

> You can even use existing video clips downloaded off the net.

I did exactly that to find the gear ratio of the new MA Spectra-G gasser. MA wouldn't say, so I found it myself.

> If recording the sound of a helicopter flying at high speed, the doppler
> shift of the sound will distort the readings.

That is important to mention.

- John

PM  EMAIL  GALLERY  Attn:RR  Quote
10-19-2004 08:54 PM  13 years agoPost 12
ifixairplanes

rrApprentice

Haverhill, MA. USA

My Posts: All  Forum  Topic

i wouldnt put that software up for free if i were you. I would put a copy of it in an envelope and mail it to myself certified just to get it copyrighted. You have a potentially lucrative scheme going on there....dont give it away for a bigger company to come along and steal from you!!

sean

PM  EMAIL  Attn:RR  Quote
10-19-2004 10:31 PM  13 years agoPost 13
oyvindk

rrNovice

Haugesund, Norway

My Posts: All  Forum  Topic

I am probably going to adapt the software for this purpose, as it is now just a slightly modified version of the software I mentioned initially.

The software is not fully (and will probably not be) automatic, meaning that you can't just input a sound file and get the results. The software analyses the sound file, and shows you a grayscale image of the sound. You will typically see a lot of harmonics, but when you know what you're looking for it is easy to ignore those.

If I sold the software for, say, 10 USD, would you then buy it instead of getting a pirate copy? This is assuming of course that you are actually interested in the software.

PM  EMAIL  Attn:RR  Quote
10-19-2004 11:16 PM  13 years agoPost 14
Conrod

rrVeteran

Mount Dora , Florida

My Posts: All  Forum  Topic

Cool!
Tell us what headspeed Alan Szabo Jr is running w/ his raptor 90.
http://runryder.com/helicopter/gall...4_part1.wmv.zip
I'm guessing that he's running stock gearing.
Thanks!
conrad

PM  EMAIL  GALLERY  Attn:RR  Quote
10-20-2004 01:01 AM  13 years agoPost 15
Love2Crash

rrNovice

Spring Grove, PA

My Posts: All  Forum  Topic

oyvindk,

I assume that you are targeting the exhaust noise? We have used similar acustic techniques in the motor racing world to spy on our competitors. MathCad is great for this, it does all the work.
Have you tried to apply any bandpass filters to identify head and/or tail speed frequencies directly. If acurate enough you could use this to diagnose clutch problems too.

Thanks for sharing
Jerry

PM  EMAIL  Attn:RR  Quote
10-20-2004 02:23 AM  13 years agoPost 16
frebou

rrKey Veteran

Montreal / Canada

My Posts: All  Forum  Topic

Well, very funny.

thanks but for me I'll stay with tach and/or governor.

Your method seems to me like very aproximal

/fb

PM  EMAIL  HOMEPAGE  GALLERY  Attn:RR  Quote
10-20-2004 03:49 AM  13 years agoPost 17
JKos

rrProfessor

Redondo Beach, CA

My Posts: All  Forum  Topic

> Your method seems to me like very aproximal

With the right software, analyzing the audio can give you sub-RPM accuracy. The only two errors in the process are: 1) Motion of the helicopter which will cause Doppler shift (but that could even be averaged out with enough audio). And, 2) The deviation of the stated sampling rate of the recording device to its actual rate (and that is typically very, very small).

So, optical tacks which only give you 10 RPM increment readings are very much more "aproximal" than audio analysis can be.

- John

PM  EMAIL  GALLERY  Attn:RR  Quote
10-20-2004 03:52 AM  13 years agoPost 18
JKos

rrProfessor

Redondo Beach, CA

My Posts: All  Forum  Topic

> Have you tried to apply any bandpass filters to identify head and/or
> tail speed frequencies directly.

I tried some filters when doing my analysis, but found them unnecessary since you know pretty well where to look in the spectrum. It is possible to determine the engine RPM and the headspeed (and, thus, gear ratio) from the audio. I don't think the tail makes enough noise to be detected.

- John

PM  EMAIL  GALLERY  Attn:RR  Quote
10-20-2004 07:27 AM  13 years agoPost 19
oyvindk

rrNovice

Haugesund, Norway

My Posts: All  Forum  Topic

@Conrod:
I looked at the video you linked to, and Alan's engine is buzzing around at around 14800 RPM. If the stock gearing is 8.45:1, this equates to a headspeed of 1750 or thereabouts.

I did the measurements where he is piroflipping right in front of the camera. His engine RPM varies between 14000 and 15000 while doing this. Probably a little less, since a very slight doppler-shift will be introduced by the heli rotating.

@Love2Crash:
The sound from the rotors are very variable. The engine exhaust note is a lot easier to detect.

@frebou:
The method is actually very accurate, but it isn't as easy to use as an optical tach.

@Jkos:
I'm not using any filtering. It is pretty easy to identify the frequency you are after, since you know what to look for and where to look. I have tried filtering in an audio editor, but this is not necessary.

PM  EMAIL  Attn:RR  Quote
10-20-2004 08:43 AM  13 years agoPost 20
darwin

rrNovice

USA

My Posts: All  Forum  Topic

oyvindk,

You could refine it a bit, then release the code for all to use, learn from, and modify.

-Darwin

PM  EMAIL  Attn:RR  Quote
WATCH
 2 pages [ <<    <    ( 1 )     2     NEXT    >> ] 1533 views
Scorpion Power Scorpion Power
HelicopterMain Discussion › Remote recording of headspeed...
 Print TOPIC  Make Suggestion 

 9  Topic Subscribe

Tuesday, November 21 - 4:46 am - Copyright © 2000-2017 RunRyder   EMAILEnable Cookies

Login Here
 New Subscriptions 
 Buddies Online