1
\$\begingroup\$

I have somewhat of a special situation. I have a telephone system at the performing arts center I work at, and I need to make some of the phones flash an LED with no audible ring. My predecessor came up with a way and did this on two of the phones, but it has a bit of a flaw that I am attempting to solve. Just as a quick background, we have an Avaya Partner phone system, and the phones that have/are to be modified are system phones with speakerphone in them. These phones ring through the speaker. On the phones my predecessor modified, he installed a SPDT switch on the positive lead for the speaker, so that when the switch was "off" an LED would flash and the ring wouldn't sound, when the switch was "on" the ring would sound but the LED would not come on. This setup is perfectly fine, except for one issue: the ring volume must be turned all the way up in order for the LED to flash. This creates a big issue in that if someone turns the switch on for either the ringer or speakerphone and forgets to turn it off, the ringing telephone creates a big disruption during the performance taking place. The ring volume on these system phones is digitally controlled in about 14 steps, and the LED will only start to flash dimly around the 10th step, and finally bright at the 14th. I bought various LEDs to see what would work best, and found two that would light bright enough, first one being FW supply 2.4v (max) FW current 20mA and the second FW supply 1.8 typ 2.4 max FW current 20mA. Both of these still require the ring volume being turned up as much as I previously stated.

What I am trying to figure out if its possible to do is to be able to increase the voltage to drive the LED with the ring volume at the lowest level. I also assume that if the voltage was increased to drive the LED, I would also somehow have to limit it to protect the LED from burning out in case someone turned the ringer up all the way. The only markings on the speaker are 24ohms 0.2w.

My questions are:

-Is there a way to make the LED work as I described?

-Is there a better way/different place to hook in the LEDs so that they will flash when the phone rings and just install a switch to turn off the speaker?

-Do I need to find a power supply on the board in the phone and tap into that to drive the LEDs instead? The phone system supplies 48vdc on pins 3 and 4 to power the system phones (was reading 45vdc with the system phone connected).

I spent hours searching around to find a way to do this but I had very little luck and most of what I found were phone flashers that were external and would need either a battery or power supply to operate, and I wasn't sure if our phone system would put out the necessary AC and Hz to make them work correctly. I also didn't want to risk burning out an extension on our phone system since some said there was a possibility of damage to the system. I would like to keep any necessary circuitry contained to the phone if possible.

I am still a bit of a beginner with this so any advice and thoughts would be greatly appreciated!

\$\endgroup\$
  • \$\begingroup\$ LEDs are better regarded as current devices. Control the current; the voltage "takes care of itself". \$\endgroup\$ – Kaz Oct 9 '15 at 17:11
  • \$\begingroup\$ LEDs can be made to appear brighter to the eye, while maintaining a low average current (and thus heat dissipation), by rapidly flashing them (many times a second) instead of delivering steady current. The persistence of vision disguises the brief periods when the LED goes dark. \$\endgroup\$ – Kaz Oct 9 '15 at 17:14
  • \$\begingroup\$ For someone to give you a solution that will surely work for you, you need to measure and report here the minimum voltage put out by the phone when you want the LED to turn on visibly. The general idea for the solution is to have some kind of comparator turn (the LED) hard on once that voltage level is exceeded. \$\endgroup\$ – Fizz Oct 9 '15 at 17:20
  • \$\begingroup\$ Also, your phone's speaker is driven by AC, so you need to use that scale when you measure. Depending what kind of wave shape your phone emits, an oscilloscope might be much more suitable to measure with than a multi-meter... but I strongly suspect you don't have one. You could improvise an oscilloscope [suitable for audio levels] with a line-for some computer and perhaps a resistor divider. \$\endgroup\$ – Fizz Oct 9 '15 at 17:33
  • 1
    \$\begingroup\$ When the ringer is low, it puts out 0.13vac and when the ringer is high, it puts out 2.07vac. \$\endgroup\$ – Dave P Oct 9 '15 at 17:34
1
\$\begingroup\$

Look at Respawned Fluff's comment. He's right on. You could use an Op Amp as a comparator and when the voltage level is exceeded, the output of the Op Amp can drive the LED. You could then control the voltage level with a pot. You should, however probably include a small buck converter to step the voltage down form 48v to like 5v or 3.3v to run your circuit off of. With that said, there is probably a power rail somewhere on the phone's pcb that has somewhere around 5v or 3.3v on it to run things like the microcontroller. You could probably tap off of that since you'll only be drawing a few mA. Finally, if you do you an op amp, it's inputs are usually high impedance meaning you will draw very, very little current from that speaker output.

Maybe something like this: (Please correct me if I'm wrong here)

schematic

simulate this circuit – Schematic created using CircuitLab

PS. That -530 ohm resistor (R4) was supposed to say 470 to 1K ohm. And the actual value of R3 will depend on your V+ and your max voltage going to the speaker.

|improve this answer|||||
\$\endgroup\$
  • 1
    \$\begingroup\$ A more workable solution (that include a feedback loop) is given in the 1st figure here. The low-level of the minimum output required to turn on the LED [0.13VAC] alas precludes (as far as I can tell) single-transistor solutions. \$\endgroup\$ – Fizz Oct 9 '15 at 17:57
  • \$\begingroup\$ I just read about the comparator and op amp and that sounds like it would work. How would the op amp know what voltage to turn on at? It would need to come on at 0.13vac, and off at a level below that. In the schematic you gave me, the input would be from one of the speaker terminals, but the V+ would be the dc in from the phone board, with both grounds running from there too? It doesn't ground back to the speaker term, right? \$\endgroup\$ – Dave P Oct 9 '15 at 18:05
  • \$\begingroup\$ Well, it really depends on what is driving the speaker. If the signal is actually AC where it goes both positive and negative in reference to ground, then the LED would only come on when the voltage when positive by a specific amount (btw, if your predecessor was only using an LED, then the same thing was happening. The LED was only turning on when the signal to the speaker was reaching a sufficient voltage). The same is true if the signal to the speaker is all in the positive voltage range. \$\endgroup\$ – Hexum064 Oct 9 '15 at 18:29
  • 1
    \$\begingroup\$ Thank you to Branden Boucher and Respawned Fluff for your explanations and help! I will try the op amp as soon as I can pick it and the rest of the components up! \$\endgroup\$ – Dave P Oct 9 '15 at 20:10
  • 1
    \$\begingroup\$ @RespawnedFluff "minimum output required to turn on the LED [0.13VAC] alas precludes (as far as I can tell) single-transistor solutions." Unless you have a vintage germanium transistor. Old Soviet MP35 will work fine. \$\endgroup\$ – ilkhd Oct 10 '15 at 5:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.