Author Topic: Long distance analog signal over copper  (Read 3695 times)

0 Members and 1 Guest are viewing this topic.

Offline ScuttleButtTopic starter

  • Newbie
  • Posts: 2
  • Country: us
Long distance analog signal over copper
« on: October 16, 2017, 10:32:06 pm »
Hello all! I have lurked here for some time and figured I would finally post something. I am working on a project that requires an analog signal to be transmitted over copper for a very long distance ~500m. I have been referencing literature for telephone applications but really come up short on some good material and was hoping that someone here could point me to a good reference. To be specific, the signal is 28VPP  20mA max and the gauge wire used has to be minimal and so does shielding since space is at a premium for this application. Usually at these distances I would go straight to digital, fiber etc... but this must be an analog only operation. Sound fun right?! :-+

Edit: I was expecting a reference, but you guys wanna get engaged.... should have known! Here are some more details..
Voltage signal
28VPP 20mA maximum
Signal of interest 1Hz-50kHz
Sampling 100kHz
Open field environment
Signal is from piezoelectric transducer, amplified by 40dB no filter
Specifics for signal loss are very unclear to me at the moment

Edit2:
Thank you for all of the information you all have contributed, very helpful! After some careful consideration I was able to work out the requirement for these to be all analog due to cost savings in the end. A balanced line configuration would have surely worked but with all of the custom connectors cable etc... it would have been quite costly. I was going to post this in a separate thread since it is a different approach but it seems to make sense to keep it all here for the meantime. I feel I should share more information about this specific application.All of these components will be in a tube that is about 70mmID and would need to house all of the transducers wiring etc.. spaced along the transducer section, not a lot of space to run a lot of analog wires. I can expect somewhere from 100-200 transducers. With this in mind I was going to use a small form factor DAQ to A/D the signals from the transducers in groups and use ethernet over coax to send down the line for processing. I was thinking something similar to this https://labjack.com/products/t7 however all of these DAQs will need to share the same clock and it seems this product uses an external clock as a division from a local clock so it would not work in the end with the sampling delays, but just to show the idea. I would like to do a small demonstration to do a proof of concept for a larger one so for the small one I would do about 8 transducers and demonstrate both analog and digital. Any recommendation on small low cost daqs?

« Last Edit: October 18, 2017, 03:52:48 pm by ScuttleButt »
 

Offline cdev

  • Super Contributor
  • ***
  • !
  • Posts: 7350
  • Country: 00
Re: Long distance analog signal over copper
« Reply #1 on: October 16, 2017, 10:42:21 pm »
I'm not the person to answer your question but the information you provided is likely insufficient for others to answer it. What kind of signal, what bandwidth..

"What the large print giveth, the small print taketh away."
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 19345
  • Country: gb
  • 0999
Re: Long distance analog signal over copper
« Reply #2 on: October 16, 2017, 10:51:03 pm »
What's the maximum frequency of interest? 500m a quarter of a wavelength of 100kHz, so if the frequency is approaching a 20kHz, you need to treat it as a transmission line, otherwise it can be treated as DC.
 

Offline ataradov

  • Super Contributor
  • ***
  • Posts: 11228
  • Country: us
    • Personal site
Re: Long distance analog signal over copper
« Reply #3 on: October 16, 2017, 11:00:07 pm »
And what is your environment looks like? 500 m in an open field, and 500 m in an industrial building are two very different things.

And also, what is the nature of the signal? What sort of distortion would considered unacceptable?
Alex
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 5840
  • Country: de
Re: Long distance analog signal over copper
« Reply #4 on: October 16, 2017, 11:20:40 pm »
This can be done. Shielded balanced cable with defined impedance, and impedance matched transmitter and receiver. Your 28 VPP, 20 mA specification doesn't make a lot of sense, though.
 

Online tggzzz

  • Super Contributor
  • ***
  • Posts: 19281
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: Long distance analog signal over copper
« Reply #5 on: October 16, 2017, 11:30:45 pm »
I am working on a project that requires an analog signal to be transmitted over copper for a very long distance ~500m.

Ha! In my first job, the "very long distance" was 3000 miles; TAT-7, no less. I still regret not having hacksawed off 1" off the 3m sample in the lab before I left.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 8605
  • Country: gb
Re: Long distance analog signal over copper
« Reply #6 on: October 16, 2017, 11:45:05 pm »
I am working on a project that requires an analog signal to be transmitted over copper for a very long distance ~500m.

Ha! In my first job, the "very long distance" was 3000 miles; TAT-7, no less. I still regret not having hacksawed off 1" off the 3m sample in the lab before I left.
If you want a piece of TAT, there's plenty of it around  :)

TAT-7 was 3000 miles in total, but there were a few repeaters along the way.
 

Offline ScuttleButtTopic starter

  • Newbie
  • Posts: 2
  • Country: us
Re: Long distance analog signal over copper
« Reply #7 on: October 16, 2017, 11:56:42 pm »
This can be done. Shielded balanced cable with defined impedance, and impedance matched transmitter and receiver. Your 28 VPP, 20 mA specification doesn't make a lot of sense, though.
It is simply the max specs of the preamplifier at the transducer.
 

Online tggzzz

  • Super Contributor
  • ***
  • Posts: 19281
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: Long distance analog signal over copper
« Reply #8 on: October 17, 2017, 08:23:47 am »
I am working on a project that requires an analog signal to be transmitted over copper for a very long distance ~500m.

Ha! In my first job, the "very long distance" was 3000 miles; TAT-7, no less. I still regret not having hacksawed off 1" off the 3m sample in the lab before I left.
If you want a piece of TAT, there's plenty of it around  :)

TAT-7 was 3000 miles in total, but there were a few repeaters along the way.

Yup; quite a few.

Since there were no separate power supply wires, all the repeaters were powered in series. ISTR that implied +10kV at one end of the TAT, and -10kV at the other, but I wouldn't stake my life on it.

Each repeater could be bypassed if/when one became faulty.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 19345
  • Country: gb
  • 0999
Re: Long distance analog signal over copper
« Reply #9 on: October 17, 2017, 09:40:34 am »
This can be done. Shielded balanced cable with defined impedance, and impedance matched transmitter and receiver. Your 28 VPP, 20 mA specification doesn't make a lot of sense, though.
It is simply the max specs of the preamplifier at the transducer.
It's not the maximum output voltage and current which is important, but the smallest signal of interest.

You've also not said anything about the maximum frequency component, which is also very important.
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 8110
  • Country: fi
Re: Long distance analog signal over copper
« Reply #10 on: October 17, 2017, 09:50:06 am »
Just try it and measure if it matches your requirements. This is one of those cases that's hard to analyze without a lot of carefully paramerized information; it would take longer to come up with the right data than to just give it a go.
 

Offline Marco

  • Super Contributor
  • ***
  • Posts: 6694
  • Country: nl
Re: Long distance analog signal over copper
« Reply #11 on: October 17, 2017, 03:30:50 pm »
28VPP 20mA maximum

Just to point out the obvious, can't impedance match that (not directly on the cable any way, nothing a 1k5 series resistor can't fix).
« Last Edit: October 17, 2017, 05:37:11 pm by Marco »
 

Offline f5r5e5d

  • Frequent Contributor
  • **
  • Posts: 349
Re: Long distance analog signal over copper
« Reply #12 on: October 17, 2017, 05:08:54 pm »
4-20 ma current loop transducer? If powered by the loop there is min V at the transmitter - typ ~ 12 V
 

Offline fcb

  • Super Contributor
  • ***
  • Posts: 2117
  • Country: gb
  • Test instrument designer/G1YWC
    • Electron Plus
Re: Long distance analog signal over copper
« Reply #13 on: October 17, 2017, 05:35:44 pm »
Back in the 1990's we used to send colour video over RG58 like coax well over 5 miles. The key was the use an optoisolated video recovery unit (Addlestone Electronics) and a device I designed called the 'blatter' (basically a x10 video amplifier, and a variable pad at the other end).  The images were quite watchable.

500M over unshielded twisted pair should be straightforward, i'd treat it like audio and start out using a balanced driver and receiver. Probably float the receiver and do any ADC conversion on the floating side.

Have a look at the ti DRV134/135 and matching receivers.
https://electron.plus Power Analysers, VI Signature Testers, Voltage References, Picoammeters, Curve Tracers.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf