Home |
Search |
Today's Posts |
#1
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
i was reading this info from Mr. Rivers in a magazine recently:
(horribly paraphrased): "tube circuits make great DIs and amp inputs for passive electric guitars. this is because electric guitars have a high impedence output, and tubes are good at taking this stuff in and getting it where it needs to go..." then i was reading this in a book on microphones: (again horribly paraphrased): "the output of a large diaphragm condenser mic capsule has ridiculously high impedence. the tube that follows the capsule steps down the impedance into something more useful and then sends the signal down the line. in solid state mics, its usually some form of silicon blah blah blah that performs this function..." since tubes are great at processing high-impedance signals... does this mean that a tube condenser mic is inherently a better design because the component stepping down the impedance inside the mic is a tube? |
#2
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
"xy" wrote in message
since tubes are great at processing high-impedance signals... does this mean that a tube condenser mic is inherently a better design because the component stepping down the impedance inside the mic is a tube? Field-Effect Transistors ("FET"s) have equivalently high- impedance "inputs" ("gate" in FET, and "grid" in tube/valve). OTOH, the argument between the "sound" of solid-state vs. tube/valve is a whole 'nuther topic. |
#3
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
xy wrote:
since tubes are great at processing high-impedance signals... does this mean that a tube condenser mic is inherently a better design because the component stepping down the impedance inside the mic is a tube? Better than what? Better than things that AREN'T great at processing high-impedance signals? Yes. Better than FETs, which are about the only other thing used for the job? No, not really. They both work very well because they're both well suited to the task at hand. They each have advantages and disadvantages, but they both share one great big advantage over any other kind of device: Very high input impedance and substantially lower output impedance. That's pretty much the deal. ulysses |
#4
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
It's worth noting that FETs often have a nonlinear input capacitance that
causes them to generate distortion, particularly at high frequencies, when driven by high-impedance sources -- and by high-impedance I mean anything greater than about 5k. Tubes don't do that. The "crispier" sound of FET microphones as opposed to tube microphones with similar capsules *may* be in part attributable to this phenomenon. I say *may* because more research is needed to see whether this is actually going on in FET-couple condenser mics. Peace, Paul |
#5
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
JP Gerard wrote:
Crispier sound of FET mics? As opposed to tube mics with similar capsules??? I don't know... I had a similar reaction. There are plenty of well-known, highly-regarded microphone capsules for which both solid-state and tube amplifiers are available as alternatives--including the Neumann KM 84 series that we've been talking about in another thread (originally it was the KM 60 series), the AKG C 60 series whose capsules were later used for the C 451 series, and the entire present-day Schoeps line of "Colette" capsules, any of which can be used on Schoeps/Elvo M 222 amplifiers. There's even a way to fit Colette capsules onto original M 221 B amplifiers so a comparison can also be made directly to that classic model. (The result: They're noisier.) Thus a lot of folks actually can and do make these comparisons, and I just don't think it's fair to say that the "tizzy" or "cripsy" sonic character which some people attribute to FET microphones is perceived by most expert listeners most of the time. Even if it is in some cases, for the claim that it's a tube vs. FET thing to be valid, there could be no exceptions. (Kind of like, if there's even one digital recording in the whole world that sounds good, then all theories of the absolute inferiority of digital audio go out the window.) Also, it's important to check such things as frequency response in the different amplifiers, since for example when Neumann made the change from the U 67 (tube) to the U 77 and 87 (FET), they kept the capsule but they changed the equalization network in the circuit so that the newer models were brighter sounding by 2 dB and 4 dB respectively (as measured at 16 kHz). If you didn't take that fact into account and you simply used your ears to compare the two mikes, you'd draw the right conclusion about their sound, but the wrong conclusion about the nature and source of the difference. With smaller microphones that have interchangeable capsules it's generally simpler, since as a rule their amplifiers have flat frequency response with only some differences at the extremes of the spectrum. There were YEARS of overlap during which tube equipment was still being made and sold right alongside the new solid-state gear. For example, Neumann was still making and selling tube microphones as late as 1973-74, after introducing their FET microphones in 1965-68. It was similar with Schoeps--when I bought my first CMT-series microphones from them in 1974, they still had the M 221 B and CM 060 series in their catalog. So the manufacturers didn't pull all their tube microphones off the market overnight. The tubes were mostly still being made (the ones for the U 47 being one major exception, of course) so there was no absolute necessity to stop making tube gear at all, if there was a market for it. That would have been easy and would have given them a better return on investment, because the research and development cost was already long since amortized. Yet there was no big rush to hoard tube gear, even though everyone with a brain knew it would be the end of the line after a while. Why not? Doesn't it strike anyone as odd that no one did that? I bet a lot of us today would know what to do with a time machine that could let us go back and scoop up a couple dozen new microphones that have since been declared as "vintage." But evidently the comparison was viewed differently at that time--just when everyone not only could but did hear tube and solid-state gear side by side. There have always been dissidents among audio engineers and there always will be, I hope--but the fact is that the entire legend of tube audio superiority is something that emerged from the audiophile community, and only after tube equipment had become generally unobtainable. It was as much a nostalgia thing as a snob thing, in other words. And now a phony history is being conjured up in which there was always a substantial core of working engineers who thought that tubes sounded better. Uh-uh. (The part of the "tube revival" that didn't come from the audiophiles came from folks who use tubes as groovy distortion devices--an application at which tubes plainly _are_ superior to solid-state. But that's pretty much irrelevant to studio condenser microphones, unless you frequently record right at the hairy overload point of the microphone circuitry and your musicians don't vary their dynamics by more than a little bit. Which is some people, I guess.) |
#6
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
Paul Stamler wrote:
It's worth noting that FETs often have a nonlinear input capacitance that causes them to generate distortion, particularly at high frequencies, when driven by high-impedance sources -- and by high-impedance I mean anything greater than about 5k. What do you mean by "nonlinear input capacitance"? You seem to be saying that FET gate capacitance varies with signal frequency; is that it? What is the magnitude of this effect? Is it small or large relative to overall gate capacitance--and is it significant in terms of the load on the capsule? I don't claim to understand microphone amplifier circuitry 100%, but a capacitive capsule's impedance increases as frequency goes _lower_, not higher. So any adverse loading effects at the gate of the FET should tend to occur at the lowest audible frequencies, no? Not at the high end, where the capsule acts as a low-impedance voltage source. I could be wrong, though; as I said, this area isn't my specialty. There are all kinds of odd little behaviors in semiconductors, but the degree to which they affect the electrical parameters of the circuit is what's essential to the bigger picture, no? Tubes don't do that. The "crispier" sound of FET microphones as opposed to tube microphones with similar capsules *may* be in part attributable to this phenomenon. I say *may* because more research is needed to see whether this is actually going on in FET-couple condenser mics. The "tubes don't do that" part frankly sounds more as if someone's being just a LITTLE more imaginative in thinking up what COULD go wrong with FETs than with tubes, but OK; if I'm working the formula correctly, the driving impedance of a typical small-diaphragm condenser's capsule would be about 225 Ohms at 20 kHz, reaching your 5 kOhm figure somewhere around 900 Hz. That's hardly in the "crispy" region. Actually, it's right in the range where THD and IMD measurements are audibly relevant. And they're not hard to obtain, either, through the use of a "test head" (i.e. by sending test signals via a shielded capacitance equal to that of the capsule). I spent some time not long ago measuring a bunch of Neumann KM 80-series microphone bodies with a test head. A producer friend of mine was unhappy with the sound of her KM 84, and thought it might be a problem with the amplifier--and I decided to test all of my own at the same time. Plus I had just had a pair of my KM 84s modified by Klaus Heyne, and I wanted to see how those measured post-surgery. So I had myself a little test bench party one evening. Upshot is that the midrange THD was easily in the 0.1%-and-under range. These items are not crude toys. I can't see how any consistent, audible effect such as the one you've described could be lurking there. Plus, I have to say, I simply don't hear any such thing when I record music--which is this odd thing I do with microphones sometimes, when I'm not testing them or translating catalog text about them or posting here. |
#7
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
P Stamler wrote:
It's worth noting that FETs often have a nonlinear input capacitance that causes them to generate distortion, particularly at high frequencies, when driven by high-impedance sources -- and by high-impedance I mean anything greater than about 5k. Tubes don't do that. The "crispier" sound of FET microphones as opposed to tube microphones with similar capsules *may* be in part attributable to this phenomenon. I say *may* because more research is needed to see whether this is actually going on in FET-couple condenser mics. While it is true that FETs have a voltage dependent input capacitance (JFETs more than MOSFETs) and that in some condenser mics that input capacitance can be a significant fraction of the capsule capacitance, it is important to note that voltage swings in a microphone are for the most part small and therefore, such distortion does not seem to be a major factor in any "crispiness". At least not in comparison to the effects of other design parameters. Benj -- SPAM-Guard! Remove .users (if present) to email me! |
#8
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
David Satz wrote:
Paul Stamler wrote: It's worth noting that FETs often have a nonlinear input capacitance that causes them to generate distortion, particularly at high frequencies, when driven by high-impedance sources -- and by high-impedance I mean anything greater than about 5k. What do you mean by "nonlinear input capacitance"? You seem to be saying that FET gate capacitance varies with signal frequency; is that it? What is the magnitude of this effect? Is it small or large relative to overall gate capacitance--and is it significant in terms of the load on the capsule? Gate capacitance varies with level! Think of the input of the FET as being a varactor diode. All semiconductors do this. Tubes just have straight Miller capacitance, with the input capacitance pretty much only due to the device capacitance and the circuit gain. I don't claim to understand microphone amplifier circuitry 100%, but a capacitive capsule's impedance increases as frequency goes _lower_, not higher. So any adverse loading effects at the gate of the FET should tend to occur at the lowest audible frequencies, no? Not at the high end, where the capsule acts as a low-impedance voltage source. I could be wrong, though; as I said, this area isn't my specialty. The smart fix for all of these problems is to operate the input stage as a voltage follower. B&K does this effectively. But since you don't get any voltage gain out of it, you have another stage adding noise in your front end. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#9
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
xy wrote:
U77? first time i've heard of one... you learn something new everyday! OK, prepare for a blast from the past. Neumann's first solid-state microphone was the KTM in 1965, followed by the "FET 70" series in 1966. These models weren't widely sold in the United States--Gotham Audio didn't start pushing Neumann's solid- state microphones until the FET 80 series (48 Volt phantom powering) was introduced. But that was very shortly afterward. So there was about a three-year period during which tube microphones, the FET 70 series and the new FET 80 microphones were all being made and sold simultaneously. The KTM and FET 70 mikes were powered by 9-12 Volt "Tonaderspeisung" (parallel power, or "T-feed"). Either a DC/DC converter had to be used in the circuit to raise that voltage to polarize the capsule(s), or, in the models introduced by Sennheiser, the capsule was part of a radio frequency oscillator which was then frequency-modulated by the capsule; an FM detector in the amplifier circuit derived the baseband audio signal, and all of this could be implemented with available bipolar transistors. (Once low-noise FETs became available, of course, simpler circuits were possible.) Nagra recorders offered T-powering for condenser microphones prior to their introduction of "universal" preamps, so it became popular in the film sound industry. Sennheiser shotguns were made with parallel powering for decades, as were many Schoeps microphones. Schoeps even still offers an amplifier model (CMC 4) for parallel powering, though they don't sell a whole heck of a lot of those any more. There were FET 70 equivalents for the well-known KM 83, 84, 85 and 86 plus the U 87. The U 77 could be powered from an internal 9-Volt battery just as the U 87 could be powered from a pair of 22.5 Volt photoflash batteries. The KM 73, 74 and 75 don't look quite like KM 83, 84 and 85 because the different grounding and shielding arrangements made it necessary to isolate the capsule heads within screened enclosures. This also had a small but measureable effect on polar and frequency response. Sorry I can't post pictures or diagrams here--but there is at least some material on Neumann's Web site about the KTM in the historical section. Occasionally a KM 74 shows up on eBay, so some pictures of them are around. |
#10
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
David Satz wrote:
microphone capsules for which both solid-state and tube amplifiers are available as alternatives--including the Neumann KM 84 series that we've been talking about in another thread (originally it was the KM 60 series), the AKG C 60 series whose capsules were later used for the C 451 series, and the entire present-day Schoeps line of "Colette" capsules, any of which can be used on Schoeps/Elvo M 222 amplifiers. There's even a way to fit Colette capsules onto original M 221 B amplifiers so a comparison can also be made directly to that classic model. (The result: They're noisier.) Okay, maybe you can answer something. I have recently acquired a set of C60/ck28s. Some previous owner screwed around with them alot, I am in the process of undoing it. This includes th 20 dB pads that were hardwired into the power supplies, replacing the bic pen spring that is in one of the capsules, and now, currently tring to figure out if that 6 dB level difference between the power supplies is simply an unbalanced leg or supply voltage problem. But my question is...I am finding that the one, so far, correctly operating mic is quite noisey compared to a 451. I mean hiss and even a bit of 60 Hz hum. Is this normal? The level is around the same, it's just all of the added noise. If this is standard on these mics, I can't see using them for classical music. 451s are already a bit too noisey for this application. And oh yes I have done the comparison using a ck1 on the C60. BTW, I am getting the sense that a ck28 is very close to a ck1. Is the only difference due to the grills? Rob R. |
#11
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
David Satz wrote: ...for the claim that it's a tube vs. FET thing to be
valid, there could be no exceptions. Forgive me, but this is clearly false. It could well be, for instance, that certain circuit topologies increase, decrease or eliminate whatever effect causes the problem---if indeed the effect and the problem exist. Thus, there could easily be exceptions in both directions in circuits using both types of active device. Various companies definitely use different topologies. A colleague of mine who's an expert on circuit topology recently looked at schematics from a famous maker of condensers, and said (1) they differed from model to model, and (2) some were different from anything he'd seen before. On a (distantly) related topic, it's certainly true overall that most recordings are brighter and harsher than real live music (no point reading further if you disagree); and I wonder whether the change in the "sonic ideal" of some musical instruments isn't a reflection of a changing notion of what instrument sound *should* be, this being caused perhaps by listening to recordings. For instance, Steinway's ideal of sound for their pianos has in my opinion changed a lot over the last 40 years, say; and in the direction of being more like a typical recorded piano. Or perhaps it's a response to customer demand that pianos *should* sound like that. James Boyk |
#12
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
David Satz wrote: ...for the claim that it's a tube vs. FET thing to be
valid, there could be no exceptions. Forgive me, but this is clearly false. It could well be, for instance, that certain circuit topologies increase, decrease or eliminate whatever effect causes the problem---if indeed the effect and the problem exist. Thus, there could easily be exceptions in both directions in circuits using both types of active device. Various companies definitely use different topologies. A colleague of mine who's an expert on circuit topology recently looked at schematics from a famous maker of condensers, and said (1) they differed from model to model, and (2) some were different from anything he'd seen before. On a (distantly) related topic, it's certainly true overall that most recordings are brighter and harsher than real live music (no point reading further if you disagree); and I wonder whether the change in the "sonic ideal" of some musical instruments isn't a reflection of a changing notion of what instrument sound *should* be, this being caused perhaps by listening to recordings. For instance, Steinway's ideal of sound for their pianos has in my opinion changed a lot over the last 40 years, say; and in the direction of being more like a typical recorded piano. Or perhaps it's a response to customer demand that pianos *should* sound like that. James Boyk |
#13
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
James Boyk wrote:
are brighter and harsher than real live music (no point reading further if you disagree); and I wonder whether the change in the "sonic ideal" of some musical instruments isn't a reflection of a changing notion of what instrument sound *should* be, this being caused perhaps by listening to recordings. For instance, Steinway's ideal of sound for their pianos has in my opinion changed a lot over the last 40 years, say; and in the direction of being more like a typical recorded piano. Or perhaps it's a response to customer demand that pianos *should* sound like that. Are you talking about all Steinways? Or just American Steinways? I agree that the American ones have a habit of being a bit bright. Try recording a Hamburg Steinway. The other thing is tuning. Piano tuners can do all sorts of things when it comes to the brighness/darkness of pianos. It could be what they are doing that disturbs you. Off-topic: someone the other day was telling me about teflon Steinways. Not to be confused with a teflon president. Rob R. |
#14
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
James Boyk wrote:
are brighter and harsher than real live music (no point reading further if you disagree); and I wonder whether the change in the "sonic ideal" of some musical instruments isn't a reflection of a changing notion of what instrument sound *should* be, this being caused perhaps by listening to recordings. For instance, Steinway's ideal of sound for their pianos has in my opinion changed a lot over the last 40 years, say; and in the direction of being more like a typical recorded piano. Or perhaps it's a response to customer demand that pianos *should* sound like that. Are you talking about all Steinways? Or just American Steinways? I agree that the American ones have a habit of being a bit bright. Try recording a Hamburg Steinway. The other thing is tuning. Piano tuners can do all sorts of things when it comes to the brighness/darkness of pianos. It could be what they are doing that disturbs you. Off-topic: someone the other day was telling me about teflon Steinways. Not to be confused with a teflon president. Rob R. |
#15
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
Rob Reedijk wrote:
Are you talking about all Steinways? Or just American Steinways? I agree that the American ones have a habit of being a bit bright. Try recording a Hamburg Steinway. A BIT bright? They're all on the same curve on change, but NY is farther along than Hamburg. Remember that NY owns Hamburg. Hamburg just hasn't reached the same point on the curve yet. The other thing is tuning. Piano tuners can do all sorts of things when it comes to the brighness/darkness of pianos. It could be what they are doing that disturbs you. No, it's the pianos themselves. The way the hammers are treated has changed radically, I believe. Off-topic: someone the other day was telling me about teflon Steinways. Not to be confused with a teflon president. Steinway doesn't use Teflon bushings any more, thank goodness. It caused 20+ years of agony while Steinway stonewalled about the problems. By contrast, the agony of Teflon Presidents continues. James Boyk |
#16
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
Rob Reedijk wrote:
Are you talking about all Steinways? Or just American Steinways? I agree that the American ones have a habit of being a bit bright. Try recording a Hamburg Steinway. A BIT bright? They're all on the same curve on change, but NY is farther along than Hamburg. Remember that NY owns Hamburg. Hamburg just hasn't reached the same point on the curve yet. The other thing is tuning. Piano tuners can do all sorts of things when it comes to the brighness/darkness of pianos. It could be what they are doing that disturbs you. No, it's the pianos themselves. The way the hammers are treated has changed radically, I believe. Off-topic: someone the other day was telling me about teflon Steinways. Not to be confused with a teflon president. Steinway doesn't use Teflon bushings any more, thank goodness. It caused 20+ years of agony while Steinway stonewalled about the problems. By contrast, the agony of Teflon Presidents continues. James Boyk |
#17
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
By the way, back to the tube/FET business. I once had the opportunity to
compare the same capsules with FET and tubed circuits, and the difference was massive. The capsules in question were medium-diameter; in the FET microphones they were a dead ringer for Shure SM-81s, while in the tubed microphones there was a good deal less harshness on the top, richness (not mud, just richness) on the bottom, and air around the instruments. One anecdote doth not a rule make, for sure. And I had no opportunity to see if the frequency responses of the two amplifier bodies were radically different, as the U87 and U67 are. But it was an interesting and instructive experiment. Peace, Paul |
#18
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
By the way, back to the tube/FET business. I once had the opportunity to
compare the same capsules with FET and tubed circuits, and the difference was massive. The capsules in question were medium-diameter; in the FET microphones they were a dead ringer for Shure SM-81s, while in the tubed microphones there was a good deal less harshness on the top, richness (not mud, just richness) on the bottom, and air around the instruments. One anecdote doth not a rule make, for sure. And I had no opportunity to see if the frequency responses of the two amplifier bodies were radically different, as the U87 and U67 are. But it was an interesting and instructive experiment. Peace, Paul |
#19
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
P Stamler wrote:
By the way, back to the tube/FET business. I once had the opportunity to compare the same capsules with FET and tubed circuits, and the difference was massive. Sheffield Lab's condenser mikes use a well-known commercial condenser capsule; but with their high-resolution tube electronics, they sound light-years better than with the maker's solid-state circuits. James Boyk |
#20
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
P Stamler wrote:
By the way, back to the tube/FET business. I once had the opportunity to compare the same capsules with FET and tubed circuits, and the difference was massive. Sheffield Lab's condenser mikes use a well-known commercial condenser capsule; but with their high-resolution tube electronics, they sound light-years better than with the maker's solid-state circuits. James Boyk |
#21
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
P Stamler wrote:
By the way, back to the tube/FET business. I once had the opportunity to compare the same capsules with FET and tubed circuits, and the difference was massive. The capsules in question were medium-diameter; in the FET microphones they were a dead ringer for Shure SM-81s, while in the tubed microphones there was a good deal less harshness on the top, richness (not mud, just richness) on the bottom, and air around the instruments. One anecdote doth not a rule make, for sure. And I had no opportunity to see if the frequency responses of the two amplifier bodies were radically different, as the U87 and U67 are. But it was an interesting and instructive experiment. Unless the single active device was the only difference between the two circuits (which is unlikely), I don't see how any relevant or useful conclusion could be drawn from this experiment. Documenting any other differences and hypothesizing on their contribution would be a good place to start. The sonic differences you describe could very easily be attributed to simple frequency response differences. But you don't need me to tell you that. I'm positive that you personally could build another pair of electronics that would give the exact opposite results. Even I could. ulysses |
#22
|
|||
|
|||
condenser mic capsule impedance and tube circuitry
P Stamler wrote:
By the way, back to the tube/FET business. I once had the opportunity to compare the same capsules with FET and tubed circuits, and the difference was massive. The capsules in question were medium-diameter; in the FET microphones they were a dead ringer for Shure SM-81s, while in the tubed microphones there was a good deal less harshness on the top, richness (not mud, just richness) on the bottom, and air around the instruments. One anecdote doth not a rule make, for sure. And I had no opportunity to see if the frequency responses of the two amplifier bodies were radically different, as the U87 and U67 are. But it was an interesting and instructive experiment. Unless the single active device was the only difference between the two circuits (which is unlikely), I don't see how any relevant or useful conclusion could be drawn from this experiment. Documenting any other differences and hypothesizing on their contribution would be a good place to start. The sonic differences you describe could very easily be attributed to simple frequency response differences. But you don't need me to tell you that. I'm positive that you personally could build another pair of electronics that would give the exact opposite results. Even I could. ulysses |