Tube Mic Design Idea

[analag]

What if I configured a medium Mu tube as a microphone head amp and instead of an output transformer, I use an opamp which buffers the tube without the associated gain loss of the OPT. In other words a line level input amp setup could be used. Remember as you increase the flow of current through the tube the higher the input resistance becomes thus a better match is made between the grid and the capsule. If a mic is the first device in a recording chain should'nt it be equipped with the best electronics? Gentlemen I own such a device, Schematic?

Analag

 

[Kev]

For many people it is the combinations of Mic and Pre that make all the fun.
Just as experimenting with Guitars and Amps to create the tone they are looking for.

It is also just as valid to get from Mic diaphragm to digital asap.
get a Neumann Mic with AES output.
:wink:

 

[gyraf]

Let's see it!

Jakob E.

 

[Gus]

In one of the AES books there is a solid state line level microphone schematic IIRC it is the Microphone book. One of the the custom manley microphones is a line level tube design. I have drawn up some solid state and hybrid line level microphones.

I do think you should do some more tests with the current flow in tube vs the grid charge build up and also look into the grid cathode plate construction of tubes.

I like high current but you need to get away from the standard tube microphone tube to capsule front end.

 

[analag]

I use one of three pre amps with this mic. An all discrete high voltage solidstate, a transformerless input/output tube mic pre, and a transformer input tube mic pre. That's a wide palette of sounds. I also modified a Nady TCM1050 with wonderful results.

Analag

 

[Rossi]

There are a couple of tube + solid state hybrid designs such as the Rode K2 or the MXL V69 that are like that. Kind of. The output circuit on the MXL is covered with epoxy, so we don't know exactly what it is. The Rode K2 has a discrete transistor output circuit after the tube. Those mics are pretty high output, but not quite line level. Although that pretty much depends on what kind of source you're recording. A very loud singer close mic'd would be about line level, I guess. Actually, it may not be wise to build a mic that's much hotter than that. You would need an attenuator for loud signals.

 

[analag]

The tube stage has line level input capability. So no matter what happens the amp inside the mic cannot be overdriven. This stage is configuered for low gain, high dynamic range, and wide frequency response. Normally a tube needs high gain to drive the transformer. And while the transformer is balancing the output and stepping down the tube resistance, it is also stepping down the output level. I bypass all that with a low gain stage, which provides me with a wider choice of tubes, and circuit topology. I played around with it until I got it right.

Analag

 

[alk509]

[quote author="analag"]I also modified a Nady TCM1050 with wonderful results.[/quote]

What did you use in place of the Alctron capsule?

Peace,
Al.

 

[analag]

When I examined the mic, I found the capsule to be ok, believe it or not. Remember these things are Neumann clones, quality control in their only flaw. The resistors around the tube are of the most ridiculous values I have seen, so I changed them. Changed the electrolytic cap they had coupling the CF to the OPT with a polyprop. Switched it on and the power supply caved in, so I gutted it and rebuilt. And she because a beautiful lady with a gorgeous voice.
Analag

 

[gyraf]

Let's see the schematic..

Jakob E.

 

[analag]

Coming soon

 

[alk509]

[quote author="analag"]When I examined the mic, I found the capsule to be ok, believe it or not.[/quote]

Errr.... Is this the capsule?

Peace,
Al.

 

[analag]

Nope that's not it....whew

Analag

 

[alk509]

Not to hijack the thread, but could you possibly post a picture of the capsule, or at least tell us what Alctron part number they use on that Nady TCM1050 mic? That would be awesome!

Peace,
Al.

 

[Gus]

I own two modded 1050s the capsules suck compaired to the better china capsules like the 797 and tenlux ones even the mxl 32mms often sound better.

Tthe nady and mxl seem to be the same design but the mxls seem to have better QA or are built at a different factory.

The holes are not the same as a real k67/87 and the skin is 3 micron and not glued to the ring.

Look at the ntk thread in the meta meta the k2 is alot like the ntk I like the ntk/k2 circuit my k2 has been modded.

 

[analag]

How do you test the capsules

Analag

 

[Gus]

I test by ear, curves with LD microphones are almost a waste of time.

I have a number of good capsules and microphones MBHO capsules Neumann capsules. Some 34mm china capsule and alot of china 32mms the 32mms just are not right compared to a realy nice capsule. I do move capsules to different circuits and body grills. The 32mm are not bad but are not as real sounding as better capsules.

Ifyou are going to go to the extent of building a realy nice circuit try at least a 797 or tenlux 34mm mult pattern

 

[Kev]

I use a test speaker and a small presure driven chamber.

so I guess my tests are strictly comparisons and not LAB quality actuals
... as the components and room have not been calibrated therefore I have no compensation curves

 

[PRR]

> as you increase the flow of current through the tube the higher the input resistance becomes thus a better match is made between the grid and the capsule.

I don't believe that is true. Certainly the capacitance (which directly loads the capacitor capsule) will not change in any large way with change of tube current. "Input resistance" covers many things, but in this application we are mainly concerned with grid current and its shot noise. I believe grid current generally reduces with tube current; electrometer inputs run very low cathode current. Or this may really be a function of grid voltage: large negative grid voltage repels electrons and reduces grid current (and also cathode current).

Grid current shot noise sets the optimum grid resistor value. If there were no grid current, bigger would be better, because the fixed capacitance of the capsule "shorts out" resistor thermal noise better. With small-area FETs, "grid" resistors in the GigaOhm range give lowest noise. With most tubes, even low-grid-current types, the grid current shot noise times the grid resistor leads to large noise voltages with large resistors. There is no "optimum" because the spectral shape of the noise sources are different, but values like 100Meg-300Meg are typically "good".

We don't want to "match", we want to "unload". This criteria is not really a problem: 100Megs gives "no-load" and flat bass response with typical capsules.

The amp in a typical mike is "about" unity-gain. Sand-state AKG 414 is just a couple buffers and 1:1 iron.

There have been tube mikes made this way. But a plate-loaded triode has higher power gain than a cathode follower. Usually we work grounded-cathode, take a gain of 5-10 in the tube, a step-down of 5 or 10 in a transformer. If the tube's plate resistance is OTOO 10K, this gives a suitable impedance for mike-line.

 

[Rossi]

[quote author="analag"]The tube stage has line level input capability. So no matter what happens the amp inside the mic cannot be overdriven. [/quote]

I wasn't thinking about internal distortion. A very hot output mic may overdrive whatever device it is connected to when the source is loud.