I want a bigger S/PDIF !

[gmarsh]

Hi,

The circuit in my first post seems to use a HD74HC04, and has some feedback resistors. Removing them seems to do nothing to the output level, so should I remove them? Do gates like these not have a near infinite open loop gain, like an op-amp?

A HCU04 is built with a single CMOS inverter. A HC04 is "fully buffered", and is equivalent to three HCU04 gates in series.

If you tie the input and output of a HCU04 inverter together with a feedback resistor, it will bias itself and its output voltage will be stable around half of Vcc. Biased as such, it can be used as an inverting amplifier, which is why you often see HCU04's being used for things like crystal oscillators.

But a HC04 gate has way more gain (HCU04 gain ^3), and the self-biasing voltage of the three gates isn't guaranteed to be the same. Which means if you attempt to self-bias the thing with feedback, you're far more likely to end up with a ring oscillator than a stable amplifier.

With an input signal applied, your circuit probably works fine. But unplug your input, and I wouldn't be surprised if the circuit oscillates.

The only place where you really need feedback is the inverter next to R6, so keep R6. You don't need R8 or R7.

 

[Tenson]

Hi that makes a lot of sense. It does indeed oscillate with no input. I'll try pulling R8 & R7 and report back.

10min later - Okay I tried that, and it didn't stop it oscillating with no input, but it did produce a small DC offset on the output which isn't good. So I put them both back in place.

I think I'll leave it alone now. It works reliably so that's all I really care about!

 

[gmarsh]

Just put in the HCU04 chip...

 

[Tenson]

Hi,

I want to follow up on this. I recently did a bit more experimenting with the feedback and found that adjusting R8 from 0 to 300Ohm will increase the output level from nothing to 500mV. Increasing R8 to greater than 300Ohm doesn't produce any difference in output level and it just stays at 500mV. The same story is true for R6, only it can increase to 1.2K before the output tops-out, again at 500mV.

What is going on, why won't the thing amplify the signal to any greater than 500mV? Is this something to do with the circuit usibng an HC04 rather than an HCU04?

Thanks.

Circuit repeated for ease -

An externally hosted image should be here but it was not working when we last tested it.

 

[Andrew Eckhardt]

What is the supply voltage on the HC04? Changing to a U04 will not increase gain.

 

[Andrew Eckhardt]

On my second statement, consider that an HC04 does not really have a linear region. Below a certain somewhat supply determined band of voltages there is practically no gain at all, and beyond that the output will saturate.

This circuit does look like it was designed with a U04 in mind. Using one will get rid of the handy no-signal oscillator in the first stage and provide some more predictable gain for low signals.

Here's some info on what to expect from an unbuffered inverter in "linear" mode:

http://www.discovercircuits.com/PDF-FILES/invertamp1.PDF