Single tube mic pre with continuous gain control

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Well, recording engineers and producers are a fickle bunch at best. I wouldn't pretend to try to please them and I have no interest in their approval. There is a huge amount of unscientific myth surrounding all aspects of recording technique. As a designer all I can do is try to make my designs as transparent as possible.

Cheers

Ian
 
bluebird said:
Not to discourage any one from building the best, most thought out, tube mic pre you can, But in the real word of recording, your just not going to please all of the people all of the time :mad:

If we were able to please everyone akk of the time, this forum would not exist.

Only marketing types have to pretend to please everyone. You know, like Steven Slate.  ;D

 
bluebird said:
Well that's the crazy thing, they are using the 50 ohm primary tap, so (50 ohm : 50K) is a 1:30 turns ratio, right?
In the 50 ohms position, the primary has much less inductance, which will act as a HPF with most microphones; in addition, the primary being less loaded means there will be a resonant peak at HF. That would make for a bright-sounding combination. Done on-purpose for the Japanese market.  :D
 
bluebird said:
Why they choose to use the 50ohm tap with a 47 ohm resistor had to be with a certain type of microphone in mind.

Neumann's connected for 50 ohm output? 
 
80hinhiding said:
Has anyone tried this single tube mic preamp yet?

Not yet. It is a pity nobody makes a 12V heater version of the ECC88 because that would make is easier to include as part of the EZTubeMixer project. There are NOS versions of such a tube available but they are extremely expensive.

Cheers

Ian
 
if using the 1:10 tap and a 1 M resistor  you will have a -3db HF point at close to 200 KC with a center freq of 1700 Hz.

so instead of opening up the box and soldering in a 50K resistor which would bring the HF point down to 20 KC with a 600 ohm source, they left the 1 M resistor (engineers are afraid of soldering irons) and found that by using the 50 ohm tap, they could get the HF point down to 20K and put the center freq right where it  belongs at 500 Hz.

but they found that they overshot the HF point and were down to 18 KC,  so by adding  more series resistance to the already low DCR of the 50 ohm tap, they got an extra 900 Hz out of the input circuit.

but now they are going to overload the crap out of that poor Ouncer with the 1 micro-gram core, so what do they do? grab a pad!    all this out-building for fear of soldering a resistor, wonder how long their experiment took to get the sound right compared to the soldering iron?

 
This seems like a perennial problem. I use three different modern 1:10 transformers in my mic pre designs. If fed from a 150 ohm source they are all flat as a pancake well past 20KHz. To make the primary 'look like' the recommended 1500 ohms I load the secondary with 150K. However, if you feed the same transformers with a 50 ohm source there is a pronounced peak of about 5dB at 25KHz. Since most dynamic and ribbon mics offer a 150 ohm source impedance they will see a relatively flat transformer. But most modern condensers have an output impedance around 50 ohms so they will see the peak. I suppose you could add 100 ohms in series with them to flatten the response but that would worsen the noise by about 5dB. However, condenser mics have 10 to 20dB more output than dynamics/ribbons so perhaps the sacrifice is worth it?

Cheers

Ian
 
ruffrecords said:
But most modern condensers have an output impedance around 50 ohms so they will see the peak.
It has not always been so. It can be attributed almost single-handedly to Dipl. Ing. Jörg Wuttke, who developed the CMC transformerless microphones during his tenure at Schoeps. He was so concerned with RF proptection that he neglected the effects of reducing the output impedance on a transformer input. I believe he was not so much concerned about problems that occur above 15kHz, since Schoeps microphone were used predominantly in FM broadcast.


I suppose you could add 100 ohms in series with them to flatten the response but that would worsen the noise by about 5dB. 
Ideally, the secondary should see a variable Zobel (fixed cap in series with variable resistor). It seems this is not often put into practice, in favour of the more pedestrian technique of placed a variable load on the primary, which does not achieve the same result.
In practice, though, it does not seem to be a serious issue; many people either like the added spark, or just EQ it out if they don't.
 
abbey road d enfer said:
Ideally, the secondary should see a variable Zobel (fixed cap in series with variable resistor). It seems this is not often put into practice, in favour of the more pedestrian technique of placed a variable load on the primary, which does not achieve the same result.
It is interesting to note that all the manufacturers of the three 1:10 transformers I use recommend a simple resistive secondary load.

Would a Zobel network that tames the peak with a 50 ohm source leave the 150 ohm source response unaffected?
In practice, though, it does not seem to be a serious issue; many people either like the added spark, or just EQ it out if they don't.

I have certainly had no complaints but it would be nice to be as transparent as possible at all common source impedances.

Cheers

ian
 
abbey road d enfer said:
No; I've made a number of tests and simulations, and the value of the resistor in tehe Zobel has to be tuned to the source Z.

Hmmm.... Very disappointing. I guess we'll just have to live with it.

Cheers

Ian
 
Back
Top