Telephone handset on Telex Audiocom - getting lots of feedback.

[Lambda]

I am currently trying to build a telephone-style handset for our Telex Audiocom system.
What I've done is taken a regular Western Electric telephone handset, removed the carbon transmitter, and installed a 200 ohm dynamic microphone element in its place. I put a resistor in series with the speaker to bring the impedance up to 150 ohms. According to the technical specs for the BP-1002 beltpack, that's all OK. And when I connect it, I can hear everything fine through the earpiece. However, as soon as I hit talk there's a terrible feedback that's heard in every headset and through the telephone earpiece. I tried turning the sidetone control all the way down, but that didn't help at all, just changed the pitch of the feedback.
The best I can tell is that the telephone's speaker is vibrating the handset, and those vibrations are picked up by the microphone and sent to the speaker, thus causing a feedback loop.
Any ideas as to what I am doing wrong here?

 

[howlingwolf487]

I'm not keen on circuit bending, but it sounds like you need to make a ducking circuit of some sort. When the mic is on, the speaker should be completely muted, or at least enough that GBF is maintained. I can't remember the last time I used a Telex handset...are they full- or half-duplex? If full-duplex, then I'm not sure; if half-duplex, only one side can squawk at any one time as it is, so the "duck" idea should work.

Awaiting other thoughts...

 

[Lambda]

I'm not keen on circuit bending, but it sounds like you need to make a ducking circuit of some sort. When the mic is on, the speaker should be completely muted, or at least enough that GBF is maintained. I can't remember the last time I used a Telex handset...are they full- or half-duplex? If full-duplex, then I'm not sure; if half-duplex, only one side can squawk at any one time as it is, so the "duck" idea should work.

Telex Audiocom is full duplex. I'd rather not mod the beltpack at all, they are rather expensive. The handset is a totally passive device, it just connects to the beltpack as a headset would, with an XLR4 connector. So without a power supply and some sort of circuit to detect when the beltpack is in talk mode, I couldn't really build a ducking circuit. I know it's possible, Telex even makes telephone handsets. We just can't afford any...

 

[cpf]

I assume the mic and speaker aren't wired in parallel, like they are on some older phone handsets?

 

[Lambda]

I assume the mic and speaker aren't wired in parallel, like they are on some older phone handsets?

No, I connected the mic to two pins on the XLR connector, and the speaker to the other two, as per the pinout for the telex headsets.

 

[FMEng]

Make sure the handle of the handset is totally sealed between the ear piece and the mic. It's a hollow tube that can carry sound from speaker to mic.

 

[Lambda]

Make sure the handle of the handset is totally sealed between the ear piece and the mic. It's a hollow tube that can carry sound from speaker to mic.

Oh... OH. When I took the handset apart, the handle was stuffed with cotton. That must have been to prevent this very thing from happening. I found myself wondering why they'd ever do that, and I didn't bother to put the cotton back in.

 

[Nelson]

That's a neat project idea, something I might even try myself at some point. We also use BP-1002 packs and matching headsets.

Few random thoughts:
Ditto what FMEng said, the cotton-like stuff does help to control feedback and sidetone. You might even need to add more than it originally had.

Try removing either the mic or earpiece from the handset but leave it connected. That way you can isolate if the handset shell design is contributing to the problem.

I believe the stock headset uses an electret condenser mic. The beltpack probably provides bias voltage to the mic element. It is possible this voltage is causing the feedback since the dynamic elements doesn't need bias voltage. You could experiment adding a DC blocking capacitor in series with the mic. However, it would be much easier to grab a newer handset that has an electret element. You are using the "G" handset (my favorite, although the stock carbon transmitters always sound gritty!). You could try a "K" handset or Nortel "M" handset, for instance. Basically any handset that is not a G has an electret mic.

Let us know what you find, I'm intrigued now!

 

[Lambda]

That's a neat project idea, something I might even try myself at some point. We also use BP-1002 packs and matching headsets.

Few random thoughts:
Ditto what FMEng said, the cotton-like stuff does help to control feedback and sidetone. You might even need to add more than it originally had.

Try removing either the mic or earpiece from the handset but leave it connected. That way you can isolate if the handset shell design is contributing to the problem.

I believe the stock headset uses an electret condenser mic. The beltpack probably provides bias voltage to the mic element. It is possible this voltage is causing the feedback since the dynamic elements doesn't need bias voltage. You could experiment adding a DC blocking capacitor in series with the mic. However, it would be much easier to grab a newer handset that has an electret element. You are using the "G" handset (my favorite, although the stock carbon transmitters always sound gritty!). You could try a "K" handset or Nortel "M" handset, for instance. Basically any handset that is not a G has an electret mic.

Let us know what you find, I'm intrigued now!

I put the cotton back, it helped a bit but not much. So I stuffed it full of old gafftape and that worked pretty well.
Now I'm having a unique problem. When I turn the volume up more than 3/4, I get some sort of squeal. Not feedback though. It continues even if I take the microphone out of the handset. It changes pitch the further up I turn the volume, and begins to pulse at varying rates. It's actually a really neat sound effect and I plan to record it before I try to fix it.
But at this point, it's definitely usable at lower volumes. The speaker's far too loud but I'll install a resistor tomorrow.

The standard Telex headsets use a dynamic mic. The BP-1002s have an internal switch that you can flip in order to use an electret mic, though. I am indeed using a G-style handset from a model 500 (minus its carbon mic), simply because it was all I could dig up out of the prop closet.

 

[Nelson]

Interesting, don't know about the squeal that you are getting now. If you do record it, do you mind posting it? I might recognize it if I hear it.

Also interesting about the headsets using a dynamic mic. I could have swore ours were electret. I'll have to look closer at ours next time!

 

[venuetech]

I have something like this for a clear-com, I don't get a squeal when i actually use it with the ear piece in place next to my ear. but when i just key the mic with the set open position it will feed back. I do have a PTT button on the set itself. I think i need to change that out so that the speaker gets disconnected when the microphone is engaged. just have to find the right switch.

 

[Lambda]

I have something like this for a clear-com, I don't get a squeal when i actually use it with the ear piece in place next to my ear. but when i just key the mic with the set open position it will feed back. I do have a PTT button on the set itself. I think i need to change that out so that the speaker gets disconnected when the microphone is engaged. just have to find the right switch.

I believe this switch is something like what you are looking for. Connect the mic in series with one normally open contact, and the speaker with a normally closed contact. Of course, you'd have to find a distributor that will sell quantities of less than 1000...

I never did get the chance to record the squeal. Today when I tried it, it didn't do it. Just boring old feedback. But think of the classic ray-gun sound effect on a loop, that's what it sounded like.
I did solve the feedback, by attenuating the speaker.
Of course, now I'm having a new problem. Once in a while, the mic-kill feature seems to activate by itself unpredictably. It doesn't happen to any other headsets, just the phone. It's not that big of an issue, it doesn't happen very often. But the phone works, and I'm pleased with it.

 

[Nelson]

Glad you got it (almost) working!

@venuetech, do you have a handset like Lambda has built? If so, you can get handsets with PTT buttons in the handle. Sandman could probably build whatever you needed:
Handsets for Telephone Systems

 

[venuetech]

Nelson
I built that handset a number of years ago, It is now gathering dust on a shelf.

 

[ccm1495]

I built one of these handsets this afternoon as a fun project. I had the parts laying around including an electret mic I had. I used the existing speaker in the phone with an added resistor. I will be making a second one. The first of the two will be for use in our booth so that people without a headset can quickly pick it up and say something. The other will be used in our green room / prep area where there are multiple people that need to communicate but do not always need a headset. It is also not practical to have these people where headsets so the phone is great for a quick word.

You can see that i stuffed a cloth into the handle of the phone to prevent feed back loop or unnecessary noise. I tested the phone and it works quite well.
Here are some pictures for your enjoyment.

 

[Nelson]

Cool, thanks for sharing the pics!

I did something similar, only I just used telephones and our PBX. Its easy when you are also the PBX switchman! I set up a hot dial phone backstage so teachers/students/other untrained users can easily pick up the handset and contact the control room to tell us what we're doing wrong

 

[TimmyP1955]

Change the switch to a DPDT. Put an additional resistor in series with the speaker. Wire the switch such that when the switch is in listen mode, the speaker shorts the additional resistor. When the switch is in talk mode, the speaker will be attenuated by the additional resistor.

 

[bill pence]

I built one of these handsets this afternoon as a fun project. I had the parts laying around including an electret mic I had. I used the existing speaker in the phone with an added resistor. I will be making a second one. The first of the two will be for use in our booth so that people without a headset can quickly pick it up and say something. The other will be used in our green room / prep area where there are multiple people that need to communicate but do not always need a headset. It is also not practical to have these people where headsets so the phone is great for a quick word.

You can see that i stuffed a cloth into the handle of the phone to prevent feed back loop or unnecessary noise. I tested the phone and it works quite well.
Here are some pictures for your enjoyment.

this is great info!. did you connect the electret directly to the mic high/mic low pins of the 4-pin XLR?
I am looking to connect some light weight gaming headsets with electret mic for use in a low ambient noise production area so the headets would be more comfortable for long term wear.
Out current headsets are dynamic mics.
the dynamic/electret mic info is very sparse...

 

[ccm1495]

You can connect it directly to the mic high and low but keep in mind that the telex or other brand of communication box your using must have the ability to use both electret and dynamic mics. The ones used for these phone handsets have a switch inside that turns on the electret mic function and provides the voltage the microphones need. It is to my understanding that the new model of telex belt packs no longer have this function as telex no longer makes electret headsets compatible with there intercom gear.


Sent from my iPhone using Tapatalk

 

[bill pence]

Cam,
that is awesome info! thanks.
I'll check my belt packs. Hopefully, they have the switch.