Converting a U47 Phaedrus Spirit schematic to U48

[Potato Cakes]

Hello, Everyone!

I have had this U47 I have built some time ago using a Phaedrus Audio Spirit tube and it has served me exceptionally well on a full album of vocals, cello, and drum room mic duties. But my original intent was to create U48 with this particular tube I initially wired the figure side incorrectly, which is something that am trying to remedy now. I have added the additional parts copying values from the original U48 schematic, but I am not getting enough voltage to the back capsule as when I switch to figure 8 the noise floor comes up dramatically and the backside transmits no audio.

I've attached the modded schematic of what I am doing now. everything in black is the original schematic and in cardiod it works properly. The voltages measured are without the capsule attached. In relation to the backplate, what kind of voltage should I expect on the rear capsule? Should they be about the same or slightly less? I did not see voltages listed on the U48 schematic. On an U87 schematic the back plate and rear capsule are the same (47V) but I do not know if that would apply with the M7 capsule. I could change the values of the two 100M resistors that feed DC to the rear capsule till the voltage is about 4V.

Another thing is on a U47, the backplate gets a voltage of about 56V. On this schematic, it is 4.7V. Should I consider shorting the 100M resistor going to the back plate getting the voltage up to 36V that is after the 2M resistor?

Thoughts?

One side note, I had the 100M grid resistor in place since I built it and the low end response has been fine. A while back I had to use it as a backup for another U47 that I had built for a client using Oliver's EF800 alternative schematic, which uses a 1G in this same spot. The engineer told me that comparing the two, the Spirit version sounded "tinnier" prompting me to ask the Phaedrus guys about this and they said the Spirit tube can handle the increase in this resistor's value. I made the change and something sounds maybe a little weird with the low end. It might be because it is late and I'm just doing a quick test on a bench. I may change this to 300M or 500M and see what I think.

Thanks!

Paul

 

[Khron]

Another thing is on a U47, the backplate gets a voltage of about 56V. On this schematic, it is 4.7V. Should I consider shorting the 100M resistor going to the back plate getting the voltage up to 36V that is after the 2M resistor?

What's the input impedance of your voltmeter / multimeter? Unlikely to be higher than 10meg. Do you know how voltage dividers work?

 

[Potato Cakes]

I do not know the input impedance of my voltmeter.

I do have pretty good idea how voltage dividers work.

I was typing a response then I realized the voltages shown on the backplate of the U47/48 are 63V and 52.5V, are calculated from the divider on the original schematics as there is no load present after the 100M resistor, otherwise this would be a dropping resistor with lower voltage than at the divider.

But there is no divider on the Spirit tube schematic pertaining to the backplate voltage as what would be 3M on the original schematic is NF (not fitted) on this one. So from this I can deduce that the actual backplate voltage is 45V and not 4.6V. But then this would mean the polarization voltage for the rear capsule on the original schematic should be about 105V, which is not correct. Is the 150M and 10nF cap are some how creating a load and dropping the voltage? I did find this: https://www.inchcalculator.com/resistor-capacitor-circuit-calculator/. But I don't think is applies and this doesn't really help as I do not know the change in voltage on the other side of the first 150M resistor.

All this to say is that I don't think the rear back plate is getting proper polarization voltage. I know the capsule is good as when this is in omni it works just fine. The one thing I haven't checked to see if the 1000pF cap between the front and back capsules is still good. I did have a very bizarre issue on a 251 build where the stryoflex cap for the figure 8 kept failing, even with new ones put in, and I had to use a PP film cap instead.

Thanks!

Paul

 

[Khron]

I do not know the input impedance of my voltmeter.

Datasheet (assuming one exists) should mention that in the specs. 10meg is pretty usual though.

I do have pretty good idea how voltage dividers work.

But you're (still) surprised why you might be measuring a measly 4V, with a 10meg input impedance meter, "downstream" of a 100meg resistor?

I have had this U47 I have built some time ago using a Phaedrus Audio Spirit tube

Going by the schematic you attached, this thing is actually phantom-powered..?

Pretty sure you can't turn this "U47" schematic into a "U48", given that preeeeeeeeetty damn crucial difference...

https://www.wagner-microphones.com/images/u48.gif

Since the front diaphragm is basically at ground potential (0V) and the backplate is at mid-supply (what in the tube U47/U48 is ~52V), the only way to get fig-8 is to have the rear diaphragm at double the voltage difference between front diaphragm and backplate, ie. virtually the full B+ voltage (105V).

The best you can hope for with a passive capsule bias supply is either cardioid/omni, OR a noisier cardioid/fig-8 or cardioid/omni/fig-8 (since the biggest voltage difference you can have between adjacent capsule electrodes is half of what's-left-of-the-phantom-power, so ~20-25V at best.

 

[Potato Cakes]

But you're (still) surprised why you might be measuring a measly 4V, with a 10meg input impedance meter, "downstream" of a 100meg resistor?

The confusion was from the schematic. The result from the voltage divider was the same as the value listed on the other side of the 150M resistor. If there was a load present on the other side of this there would have to be a drop in voltage. So my thought was that the -52.5V was the result of the voltage divider plus the 150M resistor.

Going by the schematic you attached, this thing is actually phantom-powered..?

This is phantom powered.

Since the front diaphragm is basically at ground potential (0V) and the backplate is at mid-supply (what in the tube U47/U48 is ~52V), the only way to get fig-8 is to have the rear diaphragm at double the voltage difference between front diaphragm and backplate, ie. virtually the full B+ voltage (105V).

This answers the root of my question regarding the difference of voltages between the front and rear capsules for figure 8.

If this doubling of the voltage is indeed accurate, then how is it possible to have figure 8 on other phantom powered microphones (U87, 414, etc) when all of the voltages listed on the schematic for the back plates and rear capsule are all the same? Does this require a dual capsule?

I could add the 3M resistor back in where is says NF on the schematic to achieve the differences in voltages, but half of the VDC on the backplate may reduce the headroom.

Thanks!

Paul

Thanks!

Paul

 

[Potato Cakes]

Pretty sure you can't turn this "U47" schematic into a "U48", given that preeeeeeeeetty damn crucial difference...

https://www.wagner-microphones.com/images/u48.gif

This is the schematic that I have. The fact that there is no voltage listed for the rear diaphragm was the initial cause of confusion as I understood that the values give at the capsule back plate were with some type of load, which would have to be less than the source if this were the case.

Thanks!

Paul

 

[Khron]

If this doubling of the voltage is indeed accurate, then how is it possible to have figure 8 on other phantom powered microphones (U87, 414, etc) when all of the voltages listed on the schematic for the back plates and rear capsule are all the same? Does this require a dual capsule?

If we're talking U87 non-A, or C414EB, then yes, those do indeed require capsules with isolated backplates. The U87A and later 414's have an oscillator that puts out a positive and a negative voltage, for polarizing the capsule for the multiple patterns.

I could add the 3M resistor back in where is says NF on the schematic to achieve the differences in voltages, but half of the VDC on the backplate may reduce the headroom.

No need for that, just return the circuit to stock and you'll be back to a (hopefully working) "U47" (cardioid/omni) setup.

 

[Potato Cakes]

If we're talking U87 non-A, or C414EB, then yes, those do indeed require capsules with isolated backplates. The U87A and later 414's have an oscillator that puts out a positive and a negative voltage, for polarizing the capsule for the multiple patterns.

This.

 

[Khron]

This.

That, or the Schoeps-type oscillator, or a CMOS oscillator (OPIC, Rode NT2-A).