- Joined
- Mar 11, 2014
- Messages
- 199
Are the microphonics of VF14 important for the classic of U47 or is it an undesirable side effect? I decided to do some proper testing and finally this quite time consuming project is finished. I managed to measure eight VF14 and the results are very informative.
The test setup i simple. All tubes were tested in the same U47 short body (the best I've heard). It's serial number is above 4000. The transformer is BV.08, the resistors are the red Siemens carbon film, the 10nF capacitor WIMA TFM and under the transformer are the standard MP 1uF. The damper for the tube is replaced (they dry and get hard) with the same dense black foam (does anyone know the actual material?). The capsule was replaced with a 47pF NPO ceramic capacitor. I placed the microphone about 25cm from a monitor and played a loud sine sweep. The sound files can be downloaded here
Before showing the results for each tube I start with some reference measurements. This is how it looks when measured with the M7 capsule
It's a very nice picture of the rich harmonics of the microphone
From now on the amplification is 70dB louder. This is how a measurement of the tube (VF14M No1) with the least amount of microphonics look like
The microphonics of the tube can be seen as the horizontal lines and have a decay time of about 1,5 sec (i.e. a built in reverb). There's a small line around 600Hz. Besides that they show up around 1,7kHz and are strongest in the high mids. This is where we expect the resonances of the microphone body to be. Short body with headbasket is about 22x6cm ; 22cm = 1,56kHz ; 6cm = 5.7kHz
One problem is that not only the tube can be microphonic, so can the rest of the electronics in the microphone. Fortunately I have a Phaedrus solid state VF14 replacement.
With the solid state replacement the tube microphonics are gone but something shows up around 300Hz. This could be very quiet microphonics from some of the components. So quiet that it can't be seen in the measurement with the tube. The reason for this interpretation is that I noticed that the Phaedrus solid state replacement amplifies the microphonics of the other components more than the tube.
This phenomenon is very evident when I measure a very old long body (Serial No around 500) with the old GN 107 transformer, old carbon film resistors with the flat leads (various brands), 10nF paper capacitor and the flat 1uF MP capacitors. It is the only one I've seen with the original orange damper. All the voltage measurements in this microphone are correct and the capacitors are not leaky. Still I have never enjoyed the sound of it, and the same applies to another even older U47 I've heard. The reason is probably seen below.
Original damper
Measurement with VF14. A lot of microphonics without decay below 1kHz.
Measurement with the solid state replacement. The microphonics from the electronics are worse. The tube microphonics are gone.
Heres a later long body (Serial No above 2000) with BV.08, 10nF WIMA TFM, axial Bosch 1uF and the small green wirewound resistors. It sounds and performs much better. The artefacts below 1k are very subtle.
Next image is what happens if I remove the tube damper
Left is with the tube damper. Right is without.
As a reference I did the same measurement with a M269c. As expected it performs much better with a lot less resonance from the body and almost no microphonics from the AC701k.
Finally graphs from all the individual tubes. Not all VF14 without M are rejects from Neumann. For example I know that VF14 No1 and No2 were extracted from some kind of test equipment. I noticed that in most tubes the microphonics got better a while after I turned the microphone on. In one tube it got worse. In a few the noise increased. After 1 hour, probably earlier, there were no changes. Early measurement (about 5 min to the left), late measurement (about 1h) to the right.
VF14M No1
VF14M No2
VF14 No1
VF14 No2
VF14 No3
VF14 No4 - A quite large increase in noise after 1 hour
VF14 No5 - The microphonics increase after 1 hour
VF14 No6 - generally lower in microphonics except one frequency
Summary: VF14 are microphonic to a variable degree. The body of the U47 most likely enhances the microphonics. A tube damper makes a substantial difference. The tube that performs best in this test is a VF14M but the second VF14M did not perform as well. All "non M" VF14 are not Neumann rejects and some are low in microphonics. The microphonics are in most cases less after 1 hour. The loudest peaks of the microphonics in a bad tube is about 65dB lower than the signal from the capsule. The microphonics of the best tube is just above the noise floor and about 80dB lower than the signal from the capsule. Old electronics can have a clear negative effect on sound even though they measure correct. The U67/M269 body and AC701k perform much better.
If I was to choose between a VF14 with or without microphonics, I would definately choose without.
I hope you find the test as useful as I did
The test setup i simple. All tubes were tested in the same U47 short body (the best I've heard). It's serial number is above 4000. The transformer is BV.08, the resistors are the red Siemens carbon film, the 10nF capacitor WIMA TFM and under the transformer are the standard MP 1uF. The damper for the tube is replaced (they dry and get hard) with the same dense black foam (does anyone know the actual material?). The capsule was replaced with a 47pF NPO ceramic capacitor. I placed the microphone about 25cm from a monitor and played a loud sine sweep. The sound files can be downloaded here
Before showing the results for each tube I start with some reference measurements. This is how it looks when measured with the M7 capsule
It's a very nice picture of the rich harmonics of the microphone
From now on the amplification is 70dB louder. This is how a measurement of the tube (VF14M No1) with the least amount of microphonics look like
The microphonics of the tube can be seen as the horizontal lines and have a decay time of about 1,5 sec (i.e. a built in reverb). There's a small line around 600Hz. Besides that they show up around 1,7kHz and are strongest in the high mids. This is where we expect the resonances of the microphone body to be. Short body with headbasket is about 22x6cm ; 22cm = 1,56kHz ; 6cm = 5.7kHz
One problem is that not only the tube can be microphonic, so can the rest of the electronics in the microphone. Fortunately I have a Phaedrus solid state VF14 replacement.
With the solid state replacement the tube microphonics are gone but something shows up around 300Hz. This could be very quiet microphonics from some of the components. So quiet that it can't be seen in the measurement with the tube. The reason for this interpretation is that I noticed that the Phaedrus solid state replacement amplifies the microphonics of the other components more than the tube.
This phenomenon is very evident when I measure a very old long body (Serial No around 500) with the old GN 107 transformer, old carbon film resistors with the flat leads (various brands), 10nF paper capacitor and the flat 1uF MP capacitors. It is the only one I've seen with the original orange damper. All the voltage measurements in this microphone are correct and the capacitors are not leaky. Still I have never enjoyed the sound of it, and the same applies to another even older U47 I've heard. The reason is probably seen below.
Original damper
Measurement with VF14. A lot of microphonics without decay below 1kHz.
Measurement with the solid state replacement. The microphonics from the electronics are worse. The tube microphonics are gone.
Heres a later long body (Serial No above 2000) with BV.08, 10nF WIMA TFM, axial Bosch 1uF and the small green wirewound resistors. It sounds and performs much better. The artefacts below 1k are very subtle.
Next image is what happens if I remove the tube damper
Left is with the tube damper. Right is without.
As a reference I did the same measurement with a M269c. As expected it performs much better with a lot less resonance from the body and almost no microphonics from the AC701k.
Finally graphs from all the individual tubes. Not all VF14 without M are rejects from Neumann. For example I know that VF14 No1 and No2 were extracted from some kind of test equipment. I noticed that in most tubes the microphonics got better a while after I turned the microphone on. In one tube it got worse. In a few the noise increased. After 1 hour, probably earlier, there were no changes. Early measurement (about 5 min to the left), late measurement (about 1h) to the right.
VF14M No1
VF14M No2
VF14 No1
VF14 No2
VF14 No3
VF14 No4 - A quite large increase in noise after 1 hour
VF14 No5 - The microphonics increase after 1 hour
VF14 No6 - generally lower in microphonics except one frequency
Summary: VF14 are microphonic to a variable degree. The body of the U47 most likely enhances the microphonics. A tube damper makes a substantial difference. The tube that performs best in this test is a VF14M but the second VF14M did not perform as well. All "non M" VF14 are not Neumann rejects and some are low in microphonics. The microphonics are in most cases less after 1 hour. The loudest peaks of the microphonics in a bad tube is about 65dB lower than the signal from the capsule. The microphonics of the best tube is just above the noise floor and about 80dB lower than the signal from the capsule. Old electronics can have a clear negative effect on sound even though they measure correct. The U67/M269 body and AC701k perform much better.
If I was to choose between a VF14 with or without microphonics, I would definately choose without.
I hope you find the test as useful as I did
Attachments
Last edited: