NEGATIVE FEEDBACK!?


Dietmar from Trinity Audio writes: "Of course I use negative feedback since it is a really good thing, if you do it right. 

I like the quote from Paracelsus "All things are poison and nothing is without poison; only the dose makes a thing not a poison." 

Negative feedback, had a bad reputation decades ago as the people realized that a valve amp sounds even better than a transistor amp with better measurement results. The transistor amps at that time had better total harmonic distortions, caused by the fact that these are reduced by the higher open loop gain of the transistor amps. 

The question was, "Why did they sound so bad?" You have to consider the total harmonic distortions are measured with a sine wave signal in a steady state, but music is of course complete different, since you get fast attacks in the music and so on.

The questions was, how can we measure such events. Here an old but still useful article, which was one of the first publication in this filed. www.jockohomo.net/data/7470.pdf. The TIM distortions are since than well known and it was clear, if the negative feedback of a high open loop gain is "slowing" your design, it is better do reduce the open loop gain and get a faster design with more THDs, but a lower TIMs.

That is decades ago, nowadays we have much faster transistors and can realize much faster amp designs. 

What about my designs. As you know all my products are based on ultra wide band width designs, that means my feedback loops are extrem fast and can handle any kind of input signals. There is a relation between rise time and band width. The bandwidth is BW=0.35/rise time for a first order system. 

The worse case would be the power amp and here I have a group delay of roughly 30ns!!!! By the way group delay is the time a signal needs to path from the input to the output.

30ns are outstanding. if you consider the group delay in a 50Ohm coax cable is roughly 4.5ns/m depending on the cable.

To get these values I use special high frequency transistors in my power amp, which have a transit frequency in the range of 200MHz. That is the reason why my PA is so extreme fast and even the phono with 92dB gain has a bandwidth above 500kHz. That is the reason why it makes in my eyes no sense to use a TRINITY device in a "patch work" set-up, where other devices are used with a bandwidth of 20kHz. You may hear a difference, but you are far away what the single TRINITY device can delver.