Jump to content
ThomasScherrer

Audio track as modem data NMEA

Recommended Posts

Changing R3 in the TX to a pot (to boost gain) worked to produce resonable looking digital on the RX out. It is inverted (and changing phase on TX does not change phase on RX) so I will invert with HW to a "proper" signal. Not seeing the source code, I'm not sure if the signal can be inverted in SW.

Anybody still with me on this one?...

Jeff

Share this post


Link to post
Share on other sites

I works over a wire... some of the time. Over a RF link?... I don't think so.

I've hit the time/money wall. I give up. I'll be gettin' a Maxstream.

Jeff

post-6-1174017663_thumb.jpg

Share this post


Link to post
Share on other sites

That is sad to hear. I wish I had the time to proto the modem chip set and help out. But, my schedule is hampered by the proverbial time/money wall as well.

Share this post


Link to post
Share on other sites

Still I give you two thumbs up for the effort! Never give up on any new visions and perhaps success down the road! ;)

Share this post


Link to post
Share on other sites

I usually succeed at the things I set out to do. Not always, but usually. This was a great exercise because; 1. it rekindled my electronics hobby (at least at the discrete part level), 2. I learned a thing or two about AVR microprocessors and I have a couple of leftover ATTINT2313s with which to play and 3. I found that it is easy to underestimate the difficulty of this particular task.

I talked this project over with some friends in the aerospace industry here in El Segundo. I seems generally accepted that cramming 4800 baud down a 10kHz pipe requires a more robust encoding and error correction than plain Manchester code will allow.

Thanks for everybody’s help and especially your kind words JMS!

Jeff W. Parisse

Director, kVA Effects

www.teslacoil.com

Share this post


Link to post
Share on other sites

now the page have been updated :

http://dzl.dk/projects/electronics/modem/modem.html

please also see this:

http://dzl.dk/projects/electronics/modem/s...rrer/index.html

this is just a quick mockup in both ends to prove the function,

I have added attenuators and dummy loads to the radio link so I could simulate bad/poor/dropouts and the serial downlink works fine and resync fine.

so now I am happy and will newer be loost again.

I have no idea why Jeff could not get this fantastic thing working.

1. check your audio signal input not overloaded ?

2. check rx side audio looks nice symetric ?

3. check invert/not

4. serial input from GPS must have some breaks sometimes

5. only 4800 baud input/output is accepted.

Edited by ThomasScherrer

Share this post


Link to post
Share on other sites

that's brilliant! I am really impressed with your skills.

Could you tell a little bit about your source code (if you're not going to publish it)? What kind of filtering/processing on the received signal do you use, if any? Do you measure frequencies by measuring pulse widths only? Ie was it as straightforward to do as it now seems or you've had to fight with something a bit longer?

I'd really like to learn more details on your solution, since sometime ago I was thinking about doing this myself, but got discouraged by an older colleague, more experienced in electronics who said this isn't easily doable ant pointed me towards mx614.

Do you generate a digital waveform in modulator? and what is the invert switch for?

BTW The idea with pocketpc is also brilliant. I wonder why I have not thgought of it until now!

Share this post


Link to post
Share on other sites

well as I mentioned erlier the sw is not made by me,

all I did was to encurage the right skilled person to solve the puzzle,

he is dzl.dk one of my good pals,

we often freaks out making crasy electronic projects.

the whole pointe in digital over an audio line that is AC coubled:

you need a signal stream with zero DC component,

then the rest is quite simple phase lock and invert.

the source will not be released for free !

the project is allready free to DL as hex, maybe it should not have been like that,

since it will maybe soon be sold as ready made units.

but until then I will be happy to hear more from Jeff, I also would like to understand why it did not work out for him,

if any other have the HW and programmers, please test this for me/us.

The invert could have been made in eny side of the signal,

the phase shift is made negative, so if the audio signal is inverted (quite normally)

the inverted output must be used, so the RX decoder can find the phase shift.

To resync all serial lines need a pause or delay now and then, also required on RS232 signals, else you can not find the start bit again.

on a wireless system this is so much more important, since glitches will happen

now and then on all radio links, a good system is one that can handle this with no or low notice to the user.

Share this post


Link to post
Share on other sites

the fuses:

8MHz internal clock

no div8

bor 1.8V

I suggest double checking not to drive the audio input too strong,

I have seen distortion that craped it up,

and if I run my TX module for a while with no wind and in direct sunlight,

it overheats and audio channel starts to fail, but when colder it works again,

hopefully my new tx gear will not suffer from that,

but ok normally planes fly, and are not placed for 5 mins in direct sunlight with no wind at all.

maybe its possible to add some sort of lowpass (rc) filter on the RX side also,

maybe that will improve it futher, I will see if it makes any difference after the test flights.

JEFF ?!?

are you back ? still nu luck ? I really want to know why and what happened.

Edited by ThomasScherrer

Share this post


Link to post
Share on other sites

Latest info:

I have been out flying again the whole day,

the modem works for me, if I have video I have nice action on my PDA navigator,

I dont get lost anymore and are not affraight of it.

only issue is the video tx/rx system has a lack of isolation from audio to video,

so I can see weak stripes on the picture from the data,

I dont know if it's the tx or rx side that generate the problem,

and it was so weak i did only notice it when looking at the video clips at home,

out at the field with action and sunlight in the PC, I dont see the stribes.

Share this post


Link to post
Share on other sites
so I can see weak stripes on the picture from the data

It's a problem that has been reported before when strong audio is used. Audio amplitude and bandwidth contribute to it. Try minimizing the amplitude that drives the input of the Tx. If square waves drive the audio, then try filtering them a little bit. If that does not solve it, then the modulation freq would need to be altered.

Share this post


Link to post
Share on other sites

allright I will, I am driving it close to clipping and at 5khz rounded edge square waves, so I am using my full 7khz bandwidth of the audio.

I can halve the amplityde with no problems.

Still no info from Jeff, anyone else playing with this modem ?

I was really happy about it, I used a PALM PC program called ODGPS, it can navigate to any point on earth, and make waypoints and record tracks you fly.

Share this post


Link to post
Share on other sites

I'm Back!!!!

While troubleshooting another project, I found out that the modem I was using was RS-232! So, I placed a MAX232 on it and THEN plugged it into DZL's modem and WALA!!!! The thing works GREAT over a wire but I still have to impediance match it to work with my TX/RX.

The giveaway was the TRIANGLE waves I was seeing.... they are supposed to be square.

Anyway... I'll be working on the analog stage to get it to work over my radio.

Jeff

Thomas... nice work on the OSD flying wing!!!

Share this post


Link to post
Share on other sites

allright Jeff is back !

ok you need to know the max drive level of your TX side,

check with a scope on the RX side and adjust the drive level a bit under 100%

if stribes and crud on the video try to go down to 50% or 75%.

on my TX 1vp is 100%.

the input impedance must also be known, or simply do as I did, use really low ohm value components for the filter and signal attenuator.

The TX side pulse shape must not be triangle, but square with nice rounded edges.

On the RX side I did not even add a filter to remove the hi frequency noise,

if I do that, it maybe will work a bit better on weak signals.

if your audio link has 7-10Khz bandwidth, the RX side signal will look almost the same as the tx side, square with rounded edges, the dutycycle must also look like 50/50% if all is perfect tuned, mine looks like 60/40 or even worse, but still it works.

I would like to know about what TX and RX units types you are using ?

Share this post


Link to post
Share on other sites

airwave-audio-scope07.jpg

WOW, see how the airwave 633 and 634 modules have wild audio performance !

ok I know this is a crasy test with out any filter at all on the tx side.

if you use such modules, two 4800 baud modems can be used at the SAME time, since it is in stereo. or one for sound the other for data,

and the best of all : not possible to jam the video with audio at any level and any crasy rise time.

Share this post


Link to post
Share on other sites

Airwave modules are the ones I used for my modem-less data transmission experiment I've mentioned before. Despite some doubt hinted by others, it did work on the bench. The bandwidth on the modules I tried was > 30Khz. Very decent indeed.

Yes, the stereo feature is nice for an audio channel with a companion data channel. I built up some custom Airwave Tx's and Rx's for that very reason.

Share this post


Link to post
Share on other sites

yes , but still it will make it a bit better with a modem i think.

the TD240500TXRX system I normaly use have a really poor audio compared to the airwave modules, why did I not find the good one first :-)

it is much cheaper that way, to get the right stuff !

Share this post


Link to post
Share on other sites

I heard that the RF carrier's level of the audio is (in most extreme cases) about 30dB lower on comtech transmitters to reduce harmonics in the RF spectrum.

With the Airwave modules it should be about 20dB difference between video and audio carrier level. I heard the optimum must be around 17-18dB to achieve about the same range in distance of video and audio.

From previous tests I had audio dropout much earlier than video. But I still didn't have a chance to test Thomas' modem. I will do it on the field this weekend.

BTW, -AT-Thomas: Inthe schematic diagram of the modem the resistor of the filter of the modem on the TX side is 10k. Maybe it should be 1k instead, along with the 10n cap. When I used 10k along with 10n Capacitor the output was just triangles. When I changed it to 1K I got nice square waves with round edges, as you told.

Also on your screenshot you marked the 1k values quite obviously :)

Thanks again for this modem! It's really great!

Cheers,

Hartwig

Share this post


Link to post
Share on other sites

I’m using the 2.4GHz, 600mW video transmitter / receiver from Black Widow AV. I believe that these TX/RXs are made by LawMate in Taiwan.

I’m going to get the DZL Modem to work over my radio link this weekend. I’ll report back here if I have any success. I’ll look into how I can grab screens from my scope too.

I attempted to find a link to Airwave 633 and didn’t… Got a link to those RF modules?…

Jeff

Share this post


Link to post
Share on other sites
I’m using the 2.4GHz, 600mW video transmitter / receiver from Black Widow AV. I believe that these TX/RXs are made by LawMate in Taiwan.

Hi Jeff,

I compared Lawmate and Airwave modules about a year ago for the purpose of using audio chanel for data transmission. Audio BW of Lawmate (1 watt module) was considerably worse than Airwave (612/610).

Just something to keep in mind.

Share this post


Link to post
Share on other sites

Yes Hartwig I am using much lower resistor values,

NOT the values in the schematic, that is made to a radio device with Mohm input.

P1010005_T.jpg

actually I have a 10nF sitting in parallel with the 330 Ohm also.

maybe the 330 ohm should have been 220 Ohm, but adjust to fit the audio level you desire.

I still look forward to see/hear about your modem flight test,

are you using real time displays for GPS data one the ground ?

using a PC or PDA ? show us some pictures, please.

>I heard the optimum must be around 17-18dB to achieve

>about the same range in distance of video and audio.

YES, if audio radio filters are more narrow,

a much weaker signal can be detected with same signal to noise as the

wideband video.

PS the airwave modules DONT invert on the audio !

so you must also use the non inverted data/audio output from the modulator chip.

Edited by ThomasScherrer

Share this post


Link to post
Share on other sites

OK Guys...

I'm getting ready to purchase a bunch of chips and components for the DZL Modem kit. Since the HEX is in public domain, I'm thinking of offering two versions; parts only and fully assembled. That way we (I say "we" because each sale will generate a 2% payment to DZL as per request earlier in this thread) can provide product to the hobbyist and non-hobbyist.

I have a local company willing to beta test what I produce (they download IMU data from a small UAV and this modem will help lower their on-board complexity) and I will set up a web site to act a central point of contact for the commercial version of this project.

So, with DZL's blessing, I will start a capital investment (more $$$) to produce kits.

Official Name: "DZL Modem"

Jeff

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


×
×
  • Create New...