Chinese U87/emitter follower circuits mod - MXL, SE, Audix and others

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

ln76d

Well-known member
Joined
Aug 11, 2012
Messages
2,486
Location
Gallifrey
If someone would like to mod his chinese u87 type circuit.
It's simplified U87 circuit, with full low end response,
Cardioid only version.
Transformer 6:1 - 10:1
Really simple and effective!
 

Attachments

  • U87 mod.png
    U87 mod.png
    44.6 KB · Views: 66
This won't work... (That is: the microphone will produce signal, but the HF rolloff won't...)
R8/C6 don't have any 'load' on the low side to make the HF filter work. (Neumann had the "Messeingang" here, about 600 ohms).
Also the implementation of the low-cut filter depends on the behaviour of the transformer.
It might even be possible to get  a nasty resonance on a low frequency, depending on the ratio of  the 0.22 uF and the inductance of the transformer. But in all cases the LF rolloff frequency will depend on the input impedance of the following amplifier.
 
For the low cut - did you ever had any "resonance"?
Me not - when i had to use low cut switching am making it usual this way - never had any problem with any kind of transformer i used and doesn't matter is it tube or fet microphone.
 
Yes, I have had problems with series resonance caused by an output capacitor and the inductance of an output transformer.
This is the reason why I don't like a frequency dependent component connected directly to a transformer.

The idea behind the HF rolloff is that you introduce a frequency dependent feedback.
To do so, you will need a form of a frequency dependent attenuator: low feedback at low frequencies, higher feedback at high frequencies. In your example you have 1 M / 220 pF feedback network, that is 'loaded' by a 1 G.ohm (gate) resistor in series with the capsule capacitor.  That is your 'attenuator'. In a 1 M.ohm/1 G.ohm attenuator, the influence of the 1 M can be neglected, no matter if it is bypassed by a 220 pF capacitor, or not.  Neumann did not include the 6,8K + 560 Ohm without a reason!
But if (you think) it works for you, it is fine with me...
 
Rudolph is right as usual :)

I had shorted switch during tests.
It did "click" but it didn't switched out C5 from the circuit :D
That's why it worked :D
Anyway still extended low end with updated schematic ;)
 
You can adjust the circuit keeping the stock PCB and transformer
I posted an adjustment years ago at this forum showing a low parts change and adding one resistor, more can be done
The thread seemed to upset some people so I did not post any more advanced changes
Not all have a 170 at the first stage. 
 
I am sure the corrected circuit diagram works a lot better than the previous one!  ;D
You can tailor the LF response by changing the value of C5, to adjust the LF roll-off frequency.
 
RuudNL said:
I am sure the corrected circuit diagram works a lot better than the previous one!  ;D
You can tailor the LF response by changing the value of C5, to adjust the LF roll-off frequency.

So i did that on this version of schematic ;)
Usual i used 100nF in this position, but after i screwed up switch tests i moved to 330nF.
This value should give maximum low end response in usable frequency range.

Below is simplified version, without pad and hpf, if someone would like to use it with schoeps circuit PCB or build it with perfboard, It's a less parts.

Really good results are with double bipolar 10uF in parallel for C8.
 

Attachments

  • U87 mod 2.png
    U87 mod 2.png
    38.3 KB · Views: 57
ln76d said:
Below is simplified version, without pad and hpf, if someone would like to use it with schoeps circuit PCB or build it with perfboard, It's a less parts.

Really good results are with double bipolar 10uF in parallel for C8.

"Below", where? :)
 
1
I think the 33VDC supply is better.  I would copy that instead of the 24VDC one.

 

Latest posts

Back
Top