Tube Amps / Music Electronics
|For current discussions, please visit Music Electronics Forum.||New: view Recent Searches.
New: visit Schematic Hell!
The sunn still shines online!
|Listen to great tunes streaming live right now!|
|previous: Doug H [QUOTE]OK, guys, I give up.||View Thread|
|11/9/2000 8:29 AM|
|Bart||Re: Big Caps and Noise|
Thanks for reply. Sorry if my "exploring/endorsing urban legends" remark offended you. That wasn't my intent. I was reacting to your remark about the '308, stating,
which seemed a bit strong (perhaps just enthusiastic?)and the fact everybody ignored my attempt to get a thread going with some thoughtful discussion about op amp types and sound (especially vis-a-vis the '308). It just made me wonder if folks were just into "doing the urban legend thing" about the 308. We musicians do seem to stray there from time to time... [grin]
I meant as opposed to thinking about what the desireable characteristics of the '308 are, and discussing/exploring how to design a modern day circuit that has similar desireable sound characteristics without the excessively high noise of the LM308. I really don't like noisy op amps, and the 308 is a *real* early IC design which has an order of magnitude more (input) noise that even 15-20 year old garden variety FET op amps in typical circuits. I *really* hate noisy circuitry. One of my passions/flaws. I'm told I'm *way* too picky about playing in tune, too [sigh]
I appreciate your sharing your explorations here and the discussions that go on. This is one of the better boards/news groups for my interests and purposes.
You said in your reply,
which is confusing to me. The open loop gain of the 308 is about 105-110 db, which is fairly typical for general purpose op amps and not significantly different from the TL07x series. It rolls off from around DC at the same rate as nearly every other op amp. The 308 SMALL SIGNAL open loop frequency response curve is pretty unremarkable except that it can be adjusted with the (external)compensation cap.
The LM308 *IS* rather atypical with respect to the LARGE SIGNAL frequency response using the standard 30 pf compensation (pin1 to pin8)or the ~100 pf (pin8 to ground or V-), but this is a function of the low (read lousy, by engineering standards) slew-rate (less than 0.5 v/us) and the slew-rate limiting that occurs with large signal output swings, such as those you are using in your overdrive circuit. So the issue still seems to me to be the low slew rate (which is *very* low by BIFET standards). The LARGE SIGNAL frequency response of BIFET op amps is *much* greater than a 308 due to an order-of-magnitude higher slew rate.
I tried using the comment "(HINTS)" , an approach I must admit I copied from the posts by John Green here (my apologies if it was really someone else and my memory serves me poorly), because I hoped it would peak the interest of persons I presume lurk here who might be interested in discussing the relationship between op amp characteristics and their "sound". My post was not intended to be mysterious; it was intended to arouse interest.
And FWIW, I would (also) prefer to leave "guru" status to those here who have much more knowledge of HTML and are so selflessly willing to write and organize Web pages and share their knowledge with others over the 'net. But I won't be shy about saying I've managed to learn a thing or two in 30 years of audio design and exploration.
Glad to share ideas. I guess I presumed the "HINT" was more obvious than it may be. Well here goes...
For one, the TL070 op amp (and a number of others) can be readily externally "overcompensated" with a capacitor or capacitor-resistor in series which will allow you to set the slew-rate (and therefore the LARGE SIGNAL bandwidth) to any value you choose. Presto, low slew-rate, AND you still get low BIFET noise. There are other, even quieter, offerings from AD (pricier, though).
And if you want to make it real easy, just call a TI applications engineer and ask them what value capacitor or R-C to use for a slew rate of 0.3 to 0.5 v/microsecond on a TL070. They probably have a table somewhere and will gladly give you a value so you won't even have to guess where to start from. They may want to know why the hell you want to slow it down like that, though ...[grin]
Tweak it a bit until the circuit value sounds good to your ear... maybe even try a really low slew rate, like 0.1 v/us, to get a feel for what slew rate distortion sounds like interacting with your signal level.
Or look into circuits designed specifically to control slew-rate. Then you can select components to program whatever slew-rate limits you want. One up, and another down if you want to. (The Electronic Design 'design ideas' circuit).
Like I said in the earlier post, real loudspeakers only move so fast (i.e., they slew rate limit). An interesting sonic distortion element to experiment with.
|Doug H Bart,|