Tubes v. SS: can you mix and match?

The main parts of your question have been answered (The first rule is that there are no rules! - I am running SS preamplification with tube power amplification right now), so I'll try to answer your last question. The only real qualification that occasionally arises has to do with impedance mis-matches: Tube preamps *can* have highish output impedances, and SS power amps *can* have lowish input impedances, and the two factors combined *can* roll-off bass response. Axendo said it though - you rarely have to worry about this happening as a practical matter with most gear. One other thing to think about: The well-established existence of 'hybrid' gear, where tube and transistor circuits are combined within a single amplification chassis.

What are the parameters of highish output impedance in tube preamps and lowish input impedances in SS power amps to cause concern?

No real rules of thumb, eh?

I agree that you can do whatever you want. While I have used a tube pre with solid state power and all tube system as well, it occured to me (after many months) that mixing the two can be very nice but, I started wishing I had the best qualities of tubes or the best qualities of SS thru out the amplifier chain. In other words when you mix the two it compromises the best qualities of either, the sound was neither glowing with the sound of tubes or the real direct quality I associate with SS, it was smooth and detailed but missing a little of the magic of either.

Having said the above, I have always wondered what a SS pre and tube power would be like as you would have SS amplifying the lower signals and it would (probably) be quieter than a tube pre and the majority of the amplifying would be done with the tube power amp and you could get the benefits of the tube sound there.

Also what about the posts that mention that SS has only odd harmonics (or is it even?) and tube have only even harmonics? When mixing the two do you get neither or both? I have a hard time believing the even/odd posts, does anyone know of a link that proves this?

Anyway, enough babbling, just have fun :)

i think it also depends on the kind of music you listen to. if you listen to heavy metal only . ss would be better. i think the combo of tubes and ss is a good idea for the audiophile who is on the fence. most audiophiles will eventually prefer tubes i think. the pieces i have heard that would do double duty for the big bucks gryphon has more slam than any ss i have ever heard .it also has midrange like tubes. extention on top like the best ss. if you took my tubes away give me gryphon. tube research gives up nothing to ss in any way. for the person on a budjet the conrad ss amps are awsome as they do vey good double duty as well.

OK, I'll put my unqualified neck on the chopping block...

Webnick: It's the ratios, not the absolute values involved, that matter most. Convention says that there should be a factor of at least 10X between the output impedance of the soucre component (the lower of the two) and the input impedance of the load in order to minimize LF frequency response roll-off (keep in mind that in actuallity, these impedances will vary with frequency, which is usually not specified). Again, the vast majority of preamp/amp combo's will satisfy this criteria. (The other thing that should be watched for if one is using a tube preamp having a highish output impedance, say more than a few hundred ohms, is to avoid long runs of highish-capacitance interconnect, the combination of which can cause premature HF roll-off.)

Philojet: Simplistically speaking, the differences between tubes and transistors that people talk about as far as distortion spectra go has a lot to do with the way they behave when driven into or near clipping (tubes behave more 'nicely', meaning higher overall levels of THD can be less objectionable with tubes). Below clipping, the main differences between distortion spectra have as much to with circuit design as the type of active device(s) used, although tubes will usually display higher overall distortion and noise levels. Simpler tube designs (triodes) are usually considered to display 'nicer' distortion properties than the more powerful tetrodes and pentodes, and a similar characterization is often applied to FET's as compared with bipolar transistors.

As for the distortion spectra themselves, low-order products (mostly the 2nd and 3rd harmonics) are considered less undesirable ('nicer') than higher-order harmonics, a smoothly descending harmonic series (from low-order to high-order) is considered preferable to an unevenly descending/ascending series, and even harmonics tend to be surpressed by typical push/pull amplification circuits (yielding a harmonic series that emphasizes the odd-order products, which by definition are higher-order than the even-order products, and results in an uneven roll-off of the harmonic series). This description ignores a lot of other considerations and types of distortions, including the questions of distortion levels vs. distortion qualities vs. other considerations (such as higher power capability) that may come into play as design trade-offs. (EE's forgive me...)