HELP: Audio Switching

[hamopp]

Hi all
I'am trying to switch the outputs from a audio filter to a common input (audio amp) using analogue switch's (TS5A3157DBVR)
Here is the circuit:

For some strange reason the +5v from the reg (ic3) drops to about 1.7v, but when i remove sw1 to sw4 the voltage is +5v
so it looks like there is a problem with the analogue switch's, but i have no idea as to what it might be.
Here is a link to the data sheet
**broken link removed**
Hope that someone can help me with this
Thanks
hamopp

 

[audioguru]

I can't see anything shorting your 5V power supply IC.
Didn't you know that a lousy old LM358 is half of a lousy old LM324 that produce horrible crossover distortion with audio and cut sounds above only 2kHz?

 

[hamopp]

Hi

Didn't you know that a lousy old LM358 is half of a lousy old LM324 that produce horrible crossover distortion with audio and cut sounds above only 2kHz?

No i did not know that.
Maybe you could suggest a good op amp, it is for ssb audio in a hf transceiver.
Thanks
hamopp

 

[audioguru]

SSB is very muffled and very distorted so an audio opamp probably will make no difference.

 

[kubeek]

My guess is you have a short somewhere. Have you checked the resistance between +5 and gnd?

 

[crutschow]

Either that or it's wired incorrectly or you have a faulty analog switch.

 

[ChrisP58]

You might be killing your analog switches with your opamps.

You have the nominal DC level of the opamps set at 6V by resistors R2 and R3. That nominal DC level, plus the peak of your AC signal, is feed to pin 4 of your switches, which have an absolute max voltage of 6.5, so if you have 0.5V peak AC you are exceeding their spec.

Ideally, you set the DC bias to one half of the operating voltage of the analog switches, and make sure that the peak to peak AC signal does not drive that signal past it's Vcc or ground. Ideally giving it a good margin.

The simplest way to change the DC bias voltage would be to run your divider from 5V instead of 12V. Alao, the 1.2 Meg values you are using to set the DC bias seams rather high to me.

Also, as shown the Dc bias point of the net to the left of C19 is floating when none of the switches are closed. That will result in a pop when the switches turn on and off. That net should also be biased at the same DC level as the opamps are. And, asuming that your Audio_out is ground referenced, the polarity of C19 should be reversed, since the left side of that cap should be at +2.5V

 

[JimB]

SSB is very muffled and very distorted so an audio opamp probably will make no difference.

Good old AG, ranting against SSB as usual, but you forgot your usual comment that it sounds like quacking ducks.

For a better opamp, maybe a TL072.

JimB

 

[audioguru]

I designed and made a voice scrambler for a wireless boardroom conference microphone. It used SSB with a supressed carrier and reversed high and low audio frequencies so voices were completely unintelligible without my decoder. The receiver reversed everything and unscrambled the voices perfectly. Crystals and switched-capacitor lowpass filter ICs made the voices sound real, not like ducks.

 

[crutschow]

I designed and made a voice scrambler for a wireless boardroom conference microphone. It used SSB with a supressed carrier and reversed high and low audio frequencies so voices were completely unintelligible without my decoder. The receiver reversed everything and unscrambled the voices perfectly. Crystals and switched-capacitor lowpass filter ICs made the voices sound real, not like ducks.

Out of curiosity, what do you mean by reversing the high and low audio frequencies?

 

[audioguru]

Out of curiosity, what do you mean by reversing the high and low audio frequencies?

The frequencies were completely reversed so on an ordinary receiver, speech did not sound like speech, it was just many sounds.
I used an MC1496 balanced modulator/demodulator IC to suppress the carrier and I used switched capacitor lowpass filter ICs to select the "wrong" sideband which was transmitted.
The receiver added the carrier. Another set of switched capacitor lowpass filter ICs selected the "correct" sideband.

The received/descrambled sound was identical to the wideband low distortion original speech.

 

[hamopp]

Hi all
Thanks for all the comments, below is a bit of a circuit i found from a usb\lsb 9Mhz IF filter switch
showing DC bias on pin4 of the switch using 2 x 10k, top end going to +5v, is this what you are talking
about ?. So i should change resistors R2 and R3 to 10k and connect R2 to +5v, and should i also put
some DC bias on pin1 (outputs of the switch's) ?

Again thanks for all your help
hamopp

 

[hamopp]

Hi all

you have a faulty analog switch

Yes i do have a faulty analogue switch (sw2) removed switch and +5v came back up!

You have the nominal DC level of the opamps set at 6V by resistors R2 and R3.

at R2 and R3 i get 5.76v.

Thanks
hamopp

 

[hamopp]

Hi all
So should i change resistors R2 and R3 to 10k and connect R2 to +5v, and should i also put
some DC bias on pin4 (inputs) and pin1 (outputs ) of the switch's ?
Thanks
hamopp