2

I'm having problems using HEF4051 as an analog mux. I'm using it to switch between different audio toslink signals. Right now I'm only testing it with one signal and it works for 1-2 seconds when I connect power and then the sound cuts out. I can repeat this over and over again by reconnecting power.

Vout is the "input signal" and Vin is connected to "output signal" in the schematic below. I've narrowed the problem down to the mux since it works constantly without the mux in between (Vin and Vout connected directly to eachother).

The datasheet is confusing to me since it labels y0-y7 as both inputs and outputs so I'm not sure I'm using it correctly when I'm using Y0 as an input and Z as an output. Also I'm not sure if Vee is connected properly

I connected a multimeter to see what Z is sending out and for the period of time where I get sound Z is outputting 1.6V. Then Z drops to 0.9V and I get no more sound.

enter image description here

Community
  • 1
user2665581
  • 129
  • 14
  • Is your input signal biased at half the supply voltage range of the multiplexer or is it ac coupled with an average level at gnd? – Andy aka Oct 01 '13 at 07:45

2 Answers2

2

The analogue MUX that you are using is limited to passing signals that have voltage ranges that vary between VEE and VDD. In your schematic the operational signal range is from 0 to 3.3V. If your TOS Link signals are outside that range you will need to re-think your design.

Another thing to keep in mind is that the MUX you are using may be limiting the bandwidth of the signal that you are trying to pass through it. The data sheet shows propagation delays through the part from 5 nsec to 30 nsec. Since the signal has to change state twice to make it through the part these delays correspond to a frequency range of 1.5MHz to 10MHz. If you observe in the data sheet the part gets quite a bit faster at VDD equal to a higher value. The delays shown in the data sheet start at 5V and go up to 15V. The 3.3V level you are using is not even shown and the part will be even slower at this voltage. This bandwidth thing is important to consider when selecting a proper MUX for your TOSLink signals. If the link is faster than the propagation through the part you will have troubles getting the '4051 to work as designed.

This MUX part is a bi-directional part. You can use either the Y or the Z pins as the input.

These MUX parts also have a fairly high ON resistace which you can see in the data sheet. This resistance varies according to the applied VDD voltage on the part. You need to look carefully at the circuit that drives and receives the TOSlink signals on either side of the MUX. The ON switch resistance ends up in series with the signal line and could be too high of resistance to allow your circuit to work. Look in your circuit for low value pullup/pulldown resistors at the TOSLink receivers, series resistance designed into the circuit at the TOSlink driver or higher current levels passing between the TOSLink driver and the receiver. The switch resistance could be causing signal level integrity problems taking these factors into account.

The best way to try to evaluate the real functional behavior of your circuit will come when you hookup an oscillocope and look at the waveforms of your circuit.

Michael Karas
  • 56,889
  • 3
  • 70
  • 138
  • Just to narrow it down--I also tried using higher voltage for Vdd (12V) and it seemed to work for a little longer before the sound cut out. Not sure if this helps but this is what the toslink device is outputting : S/PDIF Digital Output: 44.1K/48K/96KHz @ 16/24bit Dolby Digital, DTS, WMA-Pro. As for the resistance, I'm not sure what needs to be adjusted... http://electronics.stackexchange.com/questions/83800/toslink-fiber-optic-connectors-not-working this is the circuit I'm using with the mux in between Vin and Vout – user2665581 Oct 01 '13 at 04:47
  • 1
    @user2665581 - Did you try using an oscilloscope to look at your waveforms? After looking at the Everlight part datasheets in your other question posting I think it is going to be necessary for you to do that. It is not possible to ascertain from the information that you have provided just what frequency the link in your setup is using but if the Everlight parts are being used up to their maximum rate of 16 Mb/sec then it is almost for sure that the HEF4051B is limiting the bandwidth of the signalling and spoiling your link. You may well need to use a faster part such as Wouter has suggested. – Michael Karas Oct 01 '13 at 13:37
2

The TOSLINK signal you are trying to switch is digital in nature and high-frequency, which a general-purpose analog switch does not handle very well. The datasheet of the transmitter does not seem to specify its input impedance, so it might be rather low, which would add to the problem.

I strongly suggest that you use a digital multiplexer instead of your analog one. For 3V3 I would suggest an 74AC151.

Wouter van Ooijen
  • 48,407
  • 1
  • 63
  • 136
  • So I tried 3 different variations of the 74--151 IC and all failed to work. 74AC151, 74HC151, and the most promising 74F151 (looked to be the fastest one with tpd of 4ns). Double checked everything, I didn't even get a small period of time where the sound was working--it didn't work at all. Not sure where to go from here... – user2665581 Oct 02 '13 at 05:27
  • 1
    74F requires a supply voltage of 4.5 ... 5.5 Volt, so I would call it unsuited for your 3V3 application. 74HC has a wide supply range, but is slower at lower voltages. Does it work with a direct connection? With just oe or two 74AC gates (noninverting)? – Wouter van Ooijen Oct 02 '13 at 06:22
  • Increasing Vcc to 5V worked, thank you! – user2665581 Oct 02 '13 at 06:30