cpsmusic said:
Hi Folks,
I've been having a look at Douglass Self's "Small Signal Audio Design". In it the case is made for using discrete transistor circuitry. Other than this one <http://sound.whsites.net/project66.htm>, what would be a good discrete mic pre build for someone with beginner-intermediate level electronics skills and knowledge?
It depends on what you want to accomplish.
I'm particularly interested in a mic pre with high headroom - am I right in thinking that because a discrete pre can run at higher voltages than those using ICs, that there's greater headroom?
Generally no... while headroom is not well understood. In theory you could make a discrete preamp (your example is mixed or hybrid) with +/- 100V rails. A simplistic analysis would think that this delivers a lot of headroom, but what do you do with a +/-100V signal? You don't send it into a A/D convertor IC running from a 5V supply. Ultimately the important criteria is signal to noise ratio or dynamic range (both comparisons of clean signal to dirty noise floor).
Note: while a secondary design characteristic a mic preamp's gain structure or gain range can affect apparent head room. If the minimum gain cannot be adjusted low enough it can look like inadequate headroom to handle hot input signals, while this is just an arbitrary design decision and not intrinsic to the technology used..
As I mentioned above, I'm still at the beginner-intermediate level wrt electronics knowledge (I know enough to be dangerous!). If I'm totally wrong about the higher voltage-higher headroom relationship then I'm happy to be put right
Cheers,
Chris
Decades ago, the noise floor in widely available circuitry was noisier than today, so a strategy to provide better signal to noise ratio, was to make the signal larger. You quickly run into diminishing returns and ultimately need to reduce the signal level to accommodate lesser dynamic range paths in following gear.
Back before digital media was so pervasive, magnetic recording tape was more forgiving of brief or transient overloads. Tape saturation was even used as a crude form of compression. So more output could be better when carefully applied.
If we accept the premise that digital media should not be saturated (not all agree), there is little practical benefit from hotter signal levels.
I mentioned in passing that modern ICs are quieter. Not only that but several IC companies have developed dedicated mic preamps into fully integrated solutions. The schematic you linked to is one example of what "was" generally the state of the art for several decades ago. Low noise discrete transistors forming a circuitry front end, with an op amp wrapped around it for low distortion and good output performance. Now you can purchase a single IC chip set that does it all even better and far easier.
To answer your specific question it depends on what you want or need? You will learn more with an old school*** discrete front end design, you will get better performance, for less effort with a dedicated IC mic preamp.
JR
*** it is a little odd to call that hybrid or mixed design old school, since for a long time it was the new fangled approach, eclipsing even older step-up transformers in front of weaker (noisier) gain stages. Some still favor that even older, old way.
PS: Years ago Samuel Groner shared an overview of multiple different mic preamp topologies that might be worth searching out for educational purposes.