Repairing Thiele Valve Condenser Mic - G7 Mod?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

rodabod

Well-known member
Joined
May 12, 2005
Messages
2,896
Location
London
Hi guys,

Yet again, I have purchased another microphone from Ebay which is not 100% working. It is the same model, a "Thiele M4" which Cedric mentioned recently.

The problem starts with the fact that the internal power supply has been removed. This has obviously been done to reduce noise - other members here said that this was a major problem with this mic.

I was read about this mic (maybe on the old board) but can't find much info.

So, anyway, I have posted two photos of the insides of the mic below. In the lower part of the mic, there would originally have been a power transformer with a mains cable.

Now I am wondering what should I do with it?

My G7 boards look a little bit long - the board would fit from the bottom of the case all the way to the bottom edge of the capsule.

With a G7 board, the valve could be moved from behind the capsule and it would maybe be quite nice.

At the moment, there is a cable with 6 pins which should be connected to a power supply which I don't have. I'm not sure what parts of the original power supply are still in the microphone - maybe some smoothing caps and a few resistors, but I'm not sure.

I'm willing to do anything to get it going, although I'm not sure how easy it would be to trace out a schematic of the remains of the circuit.

So, do any of you have any suggestions for what I should do with it?

Cheers,

Roddy

thiele11eb.jpg


thiele28fk.jpg
 
I would highly suggest tracing out a schematic, and what tube is in use? Those would be helpful to determine the voltage. I'd build a power supply similar to what Jakob used on his G7 and hook it up to this mic. Is there an output transformer in there? If not, I'd put one in that space in the bottom.

-Dale
 
The tube is a 12AX7 / ECC83.

Apparently this is not a good choice for valve mics.

Also, the tube is very very close to the back of the capsule. It has both cardioid and omni modes (both sides are used) so this seems a bit stupid.

I will have a look at the circuit tomorrow and try to trace it.
 
Yea, I forgot about that. If I were not so adventurous, I'd then think about building a G7 circuit, but instead of a PC board, use the mechanical bits - terminals and such - that are there already. Also, move the tube where it belongs - away from the capsule.

Now, I'm more adventurous and more liberal with mods than a lot of people, I'd be really tempted to slip in a 6AH6 or 6AU6 (select for low noise), and rebuild the circuit to be very similar to a U47, in the existing point-to-point manner. The cathode resistor needs to be about 3.3 ohms if you do this instead of around 30 if you're using a 6AH6 with its 450mA filament. If you remove the tube from behind the capsule, make sure to plug the hole that the socket was in.

-Dale
 
Thanks for your advice, Dale.

I'll probably be using the G7 power supply - I have the G7 boards kit from Gustav.

Now, you made a good point about using point-to-point for the wiring. The mic is made from modular parts - each metal shaft and plastic tag board comes apart.

Is it simple to use the G7 power supply with other valve mic designs? I currently have a limited understanding of valve electronics, although a decent understanding of solid-state electronics.

I will have a look at some different circuits and see what is suitable.

What made you think of the U47? Is it a small circuit or does it just sound good?

Thanks,

Roddy
 
It's a simple circuit, it requires no electrolytic capacitors in the signal path, and it sounds good. It sounds excellent when used with the VF14 (original) tube. I think the only big differences between the G7 circuit and the U47 circuit are (besides the tube), the heater current is used to derive the bias instead of the 1.6k resistor and its associated bypass capacitors, and the tube is run at maximum gain (the 220 ohm resistor below that isn't there), relying entirely on miller effect feedback (capacitance from the plate to the grid, which is about 8pF or so). The other change is that the suppressor grid is no longer connected to the cathods as in the G7 - it is connected to the plate. Other than that they are pretty much the same circuit. If you want to nitpick, the U47 original circuit uses the main supply (105 volts) to operate the heater of the tube, rather than a separate supply.

However, if you are less familiar with tube electronics and don't want to be too risky, probably it is better to build up the G7 guts as on Jakob's page. It's a nice sounding mic, especially mated with a good sounding capsule. Some capsules would like more feedback so you can connect a 10pF mica capacitor from the plate to the grid and see if that sounds better to you. I agree with Jakob that I don't really like the output transformer feedback - you can probably skip that part safely.

-Dale
 
[quote author="rodabod"]Hi guys,
Now I am wondering what should I do with it?
[/quote]

First, I'd suggest to check if the capsule is in good working order... You know the drill...

The tube right behind the capsule sounds like a bad idea for two reasons.
1) If it is a PVC M7, the heat from the tube in close proximity will accelerate diaphragm drying out and eventual cracking.
2) The pattern and sound are completely screwed up because of such a huge obstacle right on the way to the back of the capsule. To see the effect, take something of similar size and place inside the grill behind the capsule of any mic and listen to it. Or even put your palm behind the grill and talk.

Besides ECC83 in a microphone... make a search, we already had this discussion.

I'd use the space under the capsule to build something point to point for easy tweaking. There are plenty of good designes out here.

You could probably try first a CF without output trafo, if you don't have one, as it would be the simplest and cheapest way and is a great start to become more familiar and comfortable with tubes.
 
[quote author="Marik"]
First, I'd suggest to check if the capsule is in good working order... You know the drill...

I'd use the space under the capsule to build something point to point for easy tweaking. There are plenty of good designes out here.
[/quote]

Ok, I better get the old set of 9V batteries out again.... :razz:

Cosmetically, the capsule is actually looking quite good.

Seems like a point-to-point design might be a good idea. Why did I never think of that?

I can probably invert the tube mount so that it is sitting upside down.
 
Ok,

I've gutted the mic by removing all of the components as seen in the photo below.

I am thinking of mounting a tube (eg. EF86) upside-down as shown in the photo. This would position it next to the tag-board.

Does this seem like a good idea to you? I thought I should put it there because there will eventually be a transformer in the bottom section, and obviously the capsule will be placed at the top.

I will need to drill a hole to allow the "nipple" of the valve to sit into the lower tag-board. This would only be necessary for an EF86 size valve. So, would you recommend that I go ahead with this, or do you think that a miniature valve would be a better option?

I think the G7 design appeals to me because there is so much documentation covering it.


On a side-note, I have added some pictures of the capsule. One of them is the capsule pointing at a bright light to see the holes in the back-plate.

thielechassis8gc.jpg



thielecapsule4ww.jpg



thielecapsule26kk.jpg
 
You can probably use any number of tubes in place of an EF86, and keep the rest of the G7 design the same. Mechanically, a submini tube may be easier to mount in there - you don't even need a socket, you can wire up everything directly to tube tube. A 5840 is one option. A 5744 triode also can sound good if you add a small capacitor from the plate to the grid (15 pF). Otherwise, the wiring can remain the same as a G7. Another option is the 6AU6, but I don't know if it's short enough to fit in there. Plus you'll need to get a 7-pin tube socket - likely a ceramic one so you don't have problems.

That capsule looks interesting. I would expect a bit of a coloured response from that hole pattern - maybe quite a bit of high-end? Have you heard this type of mic before? I haven't. It is not an M7, though. Some people may mistake it as one but it isn't.
 
Since no one has asked, I assume you all must know and I am the odd man out, so I must ask. What the devil does the light bulb do in this circuit? Does it just come on when you open things up, like the light unside the hood of the car? (just kidding) Or is it functioning as some sort of limiter?
 
look at the akg c60 and c61 schematics here
http://www.geocities.com/SiliconValley/Peaks/5212/start.htm

Now what tubes do you have? and what voltage power supply do you have? for CFs 100 VDC is often enought

Even the 12ax7 with both triodes in parallel might work OK in a CF a 12at7 would be good tube.

The circuit the c60 bootstraps the input R to a value of up to X8 its real value(from the big red book)

I like CF circuit like AKG used, with good parts it does not sound bad. With say a 100V supply set the junction of the three resistors at about 60V adjust the value of the two that set the bias for 1 ma to start and about 1V to 1.6V for the bias

try a 1.5K cathode to the three resistor node 60V/ 1ma 60 K try a 56K 1/2 watt. A 100meg on up should be fine for the cathode to grid

You want 2ma make it about 750ohms and about 33K etc..........................

The cool thing the circuit also charges the capsule to the voltage at the node of all three resistors and has no input cap.

You can also vary the voltage to change the operating points and capsule charge voltage for more fun.

I built a circuit with a submini I had set to about 1.5ma in a MXL V67 with the stock 2:1 transformer it did sound OK.

You might want to raise the fil above ground

Some microphones have a unity gain stage and have to be padded to sound ok into some preamps. A CF is a about a gain of 1 and a 2:1 to 4:1 transformer can sometimes work to your advantage with the reduction of output voltage but you also get more current drive.
 
I appreciate all of your advice guys.

You can probably use any number of tubes in place of an EF86, and keep the rest of the G7 design the same. Mechanically, a submini tube may be easier to mount in there - you don't even need a socket, you can wire up everything directly to tube tube. A 5840 is one option. A 5744 triode also can sound good if you add a small capacitor from the plate to the grid (15 pF). Otherwise, the wiring can remain the same as a G7

Dale, are you saying that I could put in some miniature tubes into the G7 circuit without changing any component values? If so, which equivalent would be best? Are there any advantages / disadvantages to using an EF86 compared to a miniature tube?

What the devil does the light bulb do in this circuit?

It is just an on / off bulb. A lamp limiter - you must think a lot! :razz: Nice idea though.


Now what tubes do you have? and what voltage power supply do you have?

Gus, I have a G7 board set. I plan on using the G7 power supply. The G7 mic board is too big though, so I will use point-to-point wiring.

As for tubes, I have the 12AX7 that came with it, an RFT EF86, a Zaerix ECC81/82 (12AT7/AU7) and a box of around another 100 or so radio valves (sadly I sold most of the good ones).

I like the simplicity of the C60 design (great website that one). However, I would like to use a standard G7 supply and have polar-pattern switching. My experience with valve electronics is limited.

Once I know what is my best option, I can get going with making it. I don't want to build it and then realise that I would have been better doing the X design rather than the Y design. I suppose I can always change it though.

Does it look like a G7 circuit would fit in there ok to you guys?

Thanks,

Roddy
 
The only component you might need to change is the bias resistor between the cathode of the tube and ground. Once you start up the mic for the first time, a quick check of one voltage will tell you whether you need to change that value. Make sure you can get to that resistor after the rest of the circuitry is together, and make sure you can measure the plate voltage of the tube.

I'd pick the 5840 if you wanted a subminiature tube. The famous Royer mod is based on that tube, and I've built mics around it too. But you also still do have the option of using the EF86 since you have one and it would take some small mechanical changes to fit it in there. I dunno, since I have about 20 or 30 5840's and all sorts of other submini tubes, I'd probably use one of them.

There are advantages of the C60/C61 design as Gus pointed out, so it's up to you as to which you want to use. But you can build the C60 or C61 circuit easily with the 5840 submini tube, and it will just work. And you can make it multipattern as well. Here's how:

Front diaphragm of the capsule - goes to ground
Backplate of the capsule - goes to the grid of the tube (directly - does not need a capacitor).
Back diaphragm of the capsule - goes to a 0.01uF film capacitor to ground, and a 10 megohm resistor between the pattern select signal from the power supply and that capacitor. Does that make sense? If you do that, you can test out the mic before you buy a transformer by feeding the output (capacitively coupled from the cathode of the tube) into an instrument input on your board (or preamp), and you can have the multipatterns, etc exactly as you would with the G7 circuit. Does that make sense?

Electrically, the backplate will be charged to 60 volts via the tube circuit, and the polarizing voltage typically goes from 0 (omni), to 60 (cardioid), to 120 (figure-8). You may need to tweak the resistor value a bit going from the cathode to the bias resistor a bit to get the cathode voltage to 60. The values given should be close, though.
 
I just saw this Thiele M5 on Ebay. Interestingly, the seller claims the capsule is a N*eumann M7:

8e01_3.JPG


I see they have followed the tube layout which I planned to go for with this model.
 
I don´t think those thiele capsules are M7´s (or, as mentioned in the text on this auction, "nearly perfect clone´s").
Look at the point where the leads are fixed to the central electrode (connecting the diaphragm). M7´s have a small screw here, right??

On the other hand, this mic looks much better than most other thieles I saw. No tube behind the capsule, no integreated psu... So perhaps it´s one of the very few thieles that can be used?!???

I decided to use mine for talkback, with a small electret capsule inside :). Looks nice...
I haven´t tested the Thiele capsule yet, but perhaps I´m going to do this soon. I have to change the capsule-mount in my G7 anyway, so why not try the Thiele capsule first?

Greets, Sascha
 

Latest posts

Back
Top