Why do people still build tube microphones?

P48 limits you.

Click to expand...

I'm designing some SS condenser mic circuits these days, and I started with these premises:

1) unit should operate with preamp input impedances as low as 600 ohms (even though 2k is typical)
2) should handle the 140+ dBspl levels I have often measured with close miced drums

with typical p48 zener regulated circuits I can't do it without an attenuator. I just can't swing enough voltage into that load.

Can do a little better with a switching supply of course.

With an external supply it's no problem at all to swing 10+ volts, so no attenuator is needed.
Seems advantageous. With the right capsule, it would be nice to have a condenser mic that could handle levels like an SM57. And still have very low noise.

Gus, as far as liability with external supplies for SS mics...doesn't the ubiquitous wall wart solve that problem?

Les

Les

I don't know if a wall wart will solve the problem.  I guess I need to find a lawyer to talk with.



I guess I expected different posts from some of the posters about why I would use external power.  I know about available power from phantom, you need to think POWER (IXE) and what to can add and do when you have more available.

I am thinking about "better" than tubes(not needing to worry as much about the tube going bad) once you start to understand what matters in a tube circuit you can transfer that to solid state.

I am working on not so clean circuits for LD vocal etc microphones.  And have one that will work with different capsule types M7,china ones, CK12 type etc.

 FWIW I lent out a P48 powered adjusted MXLV67 circuit(stock PCBs) in a MXL 1006 type body with the stock 32mm capsule and the person using it likes it better than some other good microphones they have used.

It is all about details and understanding the circuits.

leswatts said:

P48 limits you.

Click to expand...

I'm designing some SS condenser mic circuits these days, and I started with these premises:

1) unit should operate with preamp input impedances as low as 600 ohms (even though 2k is typical)
2) should handle the 140+ dBspl levels I have often measured with close miced drums ... With an external supply it's no problem at all to swing 10+ volts, so no attenuator is needed.

Click to expand...

It all depends on the actual sensitivity. If you want your mic to exhibit the typical LDC sensitivity of 20mV@1Pa (94dBspl), then for 140dBspl the output level would be 4V. And you're gonna need an attenuator somewhere. But if you reduce the sensitivity to 2mV@1Pa, the output level would be 400mV@140dBspl

with typical p48 zener regulated circuits I can't do it without an attenuator. I just can't swing enough voltage into that load.
Can do a little better with a switching supply of course.

Click to expand...

There's always a trade-off between sensitivity, noise and headroom, and I don't think P48 is the bottleneck.
A microphone manufacturer wants to control the operational parameters of his product as much as he can.
The rather high sensitivity of most LDC's concentrates the optimisation of Noise Factor on the head amp, so, unlees the operator makes a gross mistake, no one will complain that "this" mic is noisy.
But, OTOH, if the mic is used inside a kick drum, something will clip, in fact probably the whole chain will clip. But the mic manufacturer has placed his warning in the user's manual, so if someone says: "This mic distorts", the answer is already there: RTFM, dumbo!

Seems advantageous. With the right capsule, it would be nice to have a condenser mic that could handle levels like an SM57. And still have very low noise.

Click to expand...

Yes, but if I were you, I would first establish a detailed level diagram of the whole chain, and make sure that all the components clip more or less at the same time. This will give you a value for the sensitivity of the mic; then you have to work out all the parts. And probably find out you're asking too much. Getting an equivalent noise of 15dB and a max spl of 140dB computes to 125 dB S/N ratio, which is next to impossible to achieve in the current state of electronics, without some gain switching.

I don't know if a wall wart will solve the problem.  I guess I need to find a lawyer to talk with.

Click to expand...

Me too, as one entering the kit market. Casual observation suggests that practically the whole purpose of the wretched things is to avoid having to go through UL/CSA etc. It's been a while, but I have been through it with products, and small companies simply don't have the resources for that, unless something has changed.

I think the best liability protection for us is to simply make it virtually impossible for the home constructor to electrocute themselves or burn their house down, no matter what they do.
Kinda rules out tube mics, huh? Too bad.

I'll check more on this issue...you have me thinking about it.

Yes, but if I were you, I would first establish a detailed level diagram of the whole chain, and make sure that all the components clip more or less at the same time. This will give you a value for the sensitivity of the mic; then you have to work out all the parts. And probably find out you're asking too much. Getting an equivalent noise of 15dB and a max spl of 140dB computes to 125 dB S/N ratio, which is next to impossible to achieve in the current state of electronics, without some gain switching.

Click to expand...

I've certainly done that. The biggest SPLs I see are drums...specifically cymbal crashes when close miced.

That's where I measure 140+. And of course this means mics basically putting out line level signals. Certainly 125 db s/n is at the bleeding edge of what modern line level converters/analog buffers can do.
I find the biggest noise contributors are room noise and pressure noise across capsule acoustic resistance anyway.

I'm just thinking it might be nice to be able to stick a condenser mic anywhere, like a 57 (which can do 175db+ in the midrange), turn down the preamp gain to just about nothing, and go. With no mic attenuators to fiddle with.

One might wonder about capsule mechanical overload, but modern designs like concave backplates can take care of a lot of that with fewer trade-offs.

I calculate max theoretical phantom power is when the load  impedance equals source...6.81k/2.
170 milliwatts. Need switching regulators to use much of it. That would be 17 [email protected] some from efficiency, isolation resistors, etc.

Les

leswatts said:

The biggest SPLs I see are drums...specifically cymbal crashes when close miced.
That's where I measure 140+. And of course this means mics basically putting out line level signals. Certainly 125 db s/n is at the bleeding edge of what modern line level converters/analog buffers can do.

Click to expand...

Maybe not at the bleeding edge, but still requires a lot of attention in the implementation. Even harder in the head amp.

I find the biggest noise contributors are room noise and pressure noise across capsule acoustic resistance anyway.

Click to expand...

Again, it depends on the level diagram.

I'm just thinking it might be nice to be able to stick a condenser mic anywhere, like a 57 (which can do 175db+ in the midrange), turn down the preamp gain to just about nothing, and go.

Click to expand...

That's the virtue of being passive; the whole signal optimization is left to the rest of the chain.

That's the virtue of being passive; the whole signal optimization is left to the rest of the chain.

Click to expand...

Yeah, for sure. I use a zener/pass transistor regulator for phantom, so I had to have an attenuator. Was going to use a reed switch, but ended up with a conventional switch. Have to be careful with that since it's a potential moisture/dirt ingress point in the very high impedance front end.

But i'll agree with Gus (I think) that a separate supply for SS mics would offer an advantage...if you consider a condenser mic that you could treat more like a dynamic as an advantage.

Do people care? I don't know. A wall wart powered +/- 15v supply could be pretty inexpensive and open up a lot of possibilities...

Well back to mowing. Another 3 acres to do.

Les

hop.sing said:

ppa said:

I suppose there is a small market for solid state mics with external supplies, but there will never be a big market.

Click to expand...

there are sound engineers who know that's possible having different sound look and performances with external PSU, so there is a market for these mics.

Click to expand...

There is this guy Dieter Schöpf who builds non phantom powered solid state mics: http://www.sommercable.com/cardinal/2/index.html
We have one of his Tube mics at our Studio and that is certainly a world class microphone (MBHO capsule and an Andreas Grosser designed circuit with an ef12 tube).
Since he is working with A. Grosser, I have the feeling his solid state design might be the one Grosser uses in his VF14 replacement circuit which is supposed to sound very much like the tube and runs on high voltages.

Click to expand...

I actually met Dieter a couple of times at the Musikmesse. I spotted the MBHO capsules but I didn't know the circuits are designed by Andreas Grosser. Do you know that for sure? I assumed he did them himself, my impression was that Dieter knows his way around electronics. It also clashes with the fact that A. Grosser does his own  U47 style mic for Vertigo. But I know they know each other.

Yes, I saw his PSU powered FET mic, but as I said, that's a small market. For Dieter that doesn't matter as he doesn't cater to the high volume market, anyway.

leswatts said:

I'm designing some SS condenser mic circuits these days, and I started with these premises:

1) unit should operate with preamp input impedances as low as 600 ohms (even though 2k is typical)
2) should handle the 140+ dBspl levels I have often measured with close miced drums

with typical p48 zener regulated circuits I can't do it without an attenuator. I just can't swing enough voltage into that load.

Can do a little better with a switching supply of course.

With an external supply it's no problem at all to swing 10+ volts, so no attenuator is needed.
Seems advantageous. With the right capsule, it would be nice to have a condenser mic that could handle levels like an SM57. And still have very low noise.
Les

Click to expand...

Listen to Abbey (whom I met, too, at the Musikmesse  ): Gain staging is the key. First of all you should find out what kind of capsule is used with your circuit. Sensitivity varies a lot, even with similar capsule sizes.
A TLM103 has a high sensitivity capsule similar to a K67. The mic sensitivity is given at 23 mV/Pa. For all I know the circuit is unity gain and impedance balanced. So there is no voltage doubling in the output stage and there is likely to be some small gain loss. The capsule sensitivity, thus, must be around 25 mV/Pa (at the polarization voltage chosen).

A Sony-style 1-inch capsule is about 10-12 dB lower in output.  If you really want enormous 140 dB-SPL, have a look at the Josephson C715. It has a Sony-inspired low output 1-inch capsule and cascode FET input (probably unity gain) followed by a step down transformer. That's a way to do extremely high SPL, but sensitivity is only 1.77 mV/Pa. So it requires a very low noise clean mic preamp.

That said, many engineers would rather back off a little or use an attenuator switch than deal with such low sensitivity. Which is to say, you have to know your target audience and the the mic's main applications.

BTW: A colleague of mine measured a high sensitivity phantom powered measurement mic at about 40V peak to peak before clipping. I have no idea how they achieved that, though.

Gus said:

I guess I expected different posts from some of the posters about why I would use external power.  I know about available power from phantom, you need to think POWER (IXE) and what to can add and do when you have more available.

I am thinking about "better" than tubes(not needing to worry as much about the tube going bad) once you start to understand what matters in a tube circuit you can transfer that to solid state.

I am working on not so clean circuits for LD vocal etc microphones.  And have one that will work with different capsule types M7,china ones, CK12 type etc.

Click to expand...

Gus, to be honest, I still don't quite know what you are aiming at.

Obviously, one thing you could add is a preamp stage in your PSU. That's probably (?) not what you were thinking about, but it would certainly help in a commercial design as it would lessen the disadvantage of having to deal with an external box. Plus you could use it on location without a mixer or preamps.

I guess people get stuck with certain ways of building things.  There is more that can be done with tube microphones that I have not seen posts about.

> I think you can do just about anything with 5 mA.

The maximum power available happens when you match the 3.4K series resistor(s). For exact P48 this gives 24V at 7mA, 0.169 Watts.

None of the "normal" tubes will heat with 0.17W. (6C4 single med-Mu triode: 6.3V 0.15A = 0.945W, five times more than is available.)

Someone was selling a tube mike with a hearing-aid tube, low-power filament. Magic switcher electronics to convert the dozens of volts at low current into 1.4V at higher current. But hearing-aid tubes are scaled for long battery life and therefore low power output. There should be ample to feed a mike preamp input, but details may be tricky.

Hmmmm.... 1C5 (beefy octal for ancient battery radio) is only 1.4V at 0.1A or 0.14W heat, within available power (but needs conversion). If 100V and 12mA were available it could deliver 200+mW or +23dBm toward a load, far more than we need. At 24V supply and our last mA of P48 power it might do over 0dBm, ample. If you don't want a soup-can diffracting your capsule, later beach-radio miniatures have similar parameters.

Gus,

I'm going to guess that you've designed a way to get a tube-like "transconductance curve" from a solid state FET or something? I still am really curious about my whole concept of the tube space charge having an effect on the audio result. Problem is you can't isolate just the space charge and change it without changing plate voltage / heaters or making a new tube.

I second the note that Sony capsules can take higher source SPL. My C48 handles quite a bit without needing the pad. However I imagine the body is capable of mucho gain in itself as I'm just fine with a 40dB preamp for even classical work.

One thing cool about your own PSU versus phantom is that you have control over the quality of the power, instead of some unknown off the users mixer or preamp.

If it's possible that increased headroom allows or better transient response will result from a better PSU or higher voltage topology than it is worth it.

I know this is not a high power application but could you get some love from a MOSFET?

Rossi said:

hop.sing said:

ppa said:

I suppose there is a small market for solid state mics with external supplies, but there will never be a big market.

Click to expand...

there are sound engineers who know that's possible having different sound look and performances with external PSU, so there is a market for these mics.

Click to expand...

There is this guy Dieter Schöpf who builds non phantom powered solid state mics: http://www.sommercable.com/cardinal/2/index.html
We have one of his Tube mics at our Studio and that is certainly a world class microphone (MBHO capsule and an Andreas Grosser designed circuit with an ef12 tube).
Since he is working with A. Grosser, I have the feeling his solid state design might be the one Grosser uses in his VF14 replacement circuit which is supposed to sound very much like the tube and runs on high voltages.

Click to expand...

I actually met Dieter a couple of times at the Musikmesse. I spotted the MBHO capsules but I didn't know the circuits are designed by Andreas Grosser. Do you know that for sure? I assumed he did them himself, my impression was that Dieter knows his way around electronics. It also clashes with the fact that A. Grosser does his own  U47 style mic for Vertigo. But I know they know each other.

Yes, I saw his PSU powered FET mic, but as I said, that's a small market. For Dieter that doesn't matter as he doesn't cater to the high volume market, anyway.

Click to expand...

Andreas Grosser recommended us the Schoepf Microphone (2 Years ago a bargain, less than 2000 Euro), and I am pretty sure that Andreas mentioned that it is the same circuit like in the Wagner u47, which he designed.
Schoepf is very secretive about his mic, he would not even confirm that he uses a mbho capsule and it is impossible to open the mic (no screws, but rivets, we would have to send him the mic to change the tube ???), but nonetheless a nice guy who cares a lot about details and builds a great sounding microphone.

A few things you could do for fun (but possibly not with P48)

1. Polarise the capsule at >48V, directly without a DC converter.

2. Build a microphone that delivers a line level output - skip the mic preamp altogether.

3. Make a noise cancelling circuit in the mic (like the Aikido preamp).

4. Make separate head amplifiers for the two sides of your capsule (like the Josephson C720, which uses 2 x 48V).

5. Use a high voltage FET device.

Just brainstorming really.

PRR said:

> I think you can do just about anything with 5 mA.

The maximum power available happens when you match the 3.4K series resistor(s). For exact P48 this gives 24V at 7mA, 0.169 Watts.

None of the "normal" tubes will heat with 0.17W. (6C4 single med-Mu triode: 6.3V 0.15A = 0.945W, five times more than is available.)

Click to expand...

I was actually thinking about solid state, not tube. You can get more than 5 mA from P48, but most manufacturers try to stay below that figure because - as you know - lot's of preamps don't fully comply with the (revised) P48 specification. When one mic works with a preamp and another does not, people assume it's the mic's fault. Often it isn't, it's a crappy P48 supply.

There were two P48 tube mics, one from Microtech Gefell and one from Audio Technica. AFAIK the AT one is discontinued, or maybe they don't deliver it to Europe anymore. The Gefell one is still available, but not exactly cheap.

riggler said:

I second the note that Sony capsules can take higher source SPL. My C48 handles quite a bit without needing the pad. However I imagine the body is capable of mucho gain in itself as I'm just fine with a 40dB preamp for even classical work.

Click to expand...

I should have been more specific: By Sony-style capsules I meant the older Sony designs which were single diaphragm with mechanical switching between cardioid and omni as found in the C37, C38 and (non-G) C800. Later Sony mics such as the C800G use a dual diaphragm capsule with higher output, basically a K67 knockoff. AFAIK the C48 has the same or a similar capsule as the C800G.

To be honest, I don't even own a Sony, my experience is mostly based on Dale's Sony-style capsules (which I quite like!).

hop.sing said:

Andreas Grosser recommended us the Schoepf Microphone (2 Years ago a bargain, less than 2000 Euro), and I am pretty sure that Andreas mentioned that it is the same circuit like in the Wagner u47, which he designed.
Schoepf is very secretive about his mic, he would not even confirm that he uses a mbho capsule and it is impossible to open the mic (no screws, but rivets, we would have to send him the mic to change the tube ???), but nonetheless a nice guy who cares a lot about details and builds a great sounding microphone.

Click to expand...

Yeah I know, he likes to keep his secrets (just like a lot of small manufacturers). But it's not that hard to spot an MBHO capsule. There aren't that many capsule manufacturers anyway. The rivets thing is a bit weird. I think you should be allowed to open a tube mic and replace a faulty tube. I mean what are you supposed to do if the mic fails just prior to an important session or, worse, during a session? It also makes it hard to service the mic if anything should happen to its ceator. After all it's just a one-man-show. There's no one to carry on business if he becomes ill, for instance. That said, he is a nice guy.

Well, during concept phase I fiddled with external supplies, switchers to get most  of the 170 mW phantom, instrumentation amps with LT1792 FET IC front end (4.2nV/sqrt hz), etc.

But I didn't use them. I was afraid of less than ideal P48, and the hassle of a lunchbox and special cables vs the hassle of having another gain staging point at the mic.

I used something much more conventional and limited to 5ma. With attenuator.

At this point I'm done with it, and grinding away on the deep drawing dies for the grilles.

But I think the ideas have merit... both for tube and SS.  I just didn't do it this time. Might later.

Les