SE Electronics Z3300a

[micolas]

a 4,7uF PIO won't fit in the body ! never !
PIO will be maybe in my (future) tube mic as coupling-cap...

If something doesn't fit in something else, use a bigger hammer!
(marteau plus gros)

 

[micolas]

Did you do BIAS for the jFET in the new circuit?
Or for example by simply shunting the resistance
R10 3k it is possible to increase the gain of the circuit by a few dB.

 

[Emmathom]

Did you do BIAS for the jFET in the new circuit?
Or for example by simply shunting the resistance
R10 3k it is possible to increase the gain of the circuit by a few dB.

I don't knwo how to bias a jfet... as I told you I'm more experimented with tubes than with Tr
Must have to do with IDSS, Vgs, R drain, R source, but no idea what & how to do...

 

[micolas]

I don't knwo how to bias a jfet... as I told you I'm more experimented with tubes than with Tr
Must have to do with IDSS, Vgs, R drain, R source, but no idea what & how to do...

For now,
Just do a little test:
Short R10 3k with a wire, put a jumper over it
Does the microphone output level increase?

 

[Emmathom]

Hi
Here are some measurement between original & modified audio boards...
The SA1015 emitter's voltage is quite much different but it is not the original one : we desoldered it to replace it in case of it was "burned" but I'm gonna reput the original because I think it is ok (hFE 200 / Ic 1,9mA / Vbe 633mV)
One of the ones I bought at Mouser : hFE 137 / Ic 1,3mA / Vbe 623mV

 

[micolas]

Test the connection in version A

 

[Emmathom]

Test the connection in version A

I want to reput the original Tr to see if it works... but in case not I will wire Monacor in "A"

To resume : I think there was a pb on the board tracks' because every component I tested was ok > I didn't find any failed one...

 

[micolas]

I want to reput the original Tr to see if it works... but in case not I will wire Monacor in "A"

for the A connection you only change one wire

 

[micolas]

Hi
Here are some measurement between original & modified audio boards...

K170 and A1015, both have different parameters than stock transistors.

 

[Emmathom]

I reassembled a new audio card from scratch because the one installed yesterday was a first shot (without previous experience) and I wanted to optimize my cabling which left to be desired. I put back the SA1015 that was originaly on the board, so almost everything from the original mic (except the transformer and see below) is employed.

I made some modifications:
- remove the 3KΩ resistor (R10) on the Jfet source
- replaced the 22uF tantalum by a Nichicon VR 22uF • 35v
- replaced the 4.7uF tantalum by a Muse BP of the same value

The transformer is still the Monacor wired in "E" and I have now the same gain as the original microphone. The sound is good, comparable to the original but real audio tests will be done (the Muse BP are capacitors that need to be run in to deliver 100% of their quality)
I quickly measured the tensions but did not write them down: I will do it to compare with the original...

I'm asking myself if I try to put the original transformer back... before installing Arienne K47-flat capsules !

 

[Emmathom]

Here last measurements...

I think (?) I could get more gain with the original t.bone transformer and so better bias the Jfet by adding a R source to get 1v or so.
Maybe 10KΩ after R9 (1,8KΩ) so not to drop too much to ground ? 3KΩ was installed...

 

[micolas]

Here last measurements...

I think (?) I could get more gain with the original t.bone transformer and so better bias the Jfet by adding a R source to get 1v or so.
Maybe 10KΩ after R9 (1,8KΩ) so not to drop too much to ground ? 3KΩ was installed...

R10 3k performs a kind of self-Bias, to accept more varied jFETs.
jFETs are well known to have very large parameter dispersion.
The process of selecting jFETs to work in a specific topology, with exact, constant component values (it takes enormous amounts to select identical jFETs), or making in each microphone a custom BIAS, for each individual jFET, is not feasible, cost-effective in series production. It is an enormous financial, technological, time-consuming effort, etc.
So R10 allows more tolerance for jFETs, with larger specification deviations to work in this topology. It is true, not in ideal conditions, but average. The commercial criterion, the profitability, comes first.
Technically correct is R10=Oohm and in series with R9 1.8k put a semi-variable resistor VR of about 10...15k from which the BIAS of the jFET will be adjusted.
*
Do you still have the original, functional jFET? (even with this you will see voltage variations on the drain, source, etc.)

 

[micolas]

Transformer ratio is not
2,5:1
Stock is (1,8...2) :1
Now: Vers.E is 2:1
Vers. A is 1,5:1

 

[micolas]

Stock: L2 is connected to XLR pin 2
(unless you connected the primary in reverse, and then you corrected the phase of the microphone)

 

[Emmathom]

Stock: L2 is connected to XLR pin 2
(unless you connected the primary in reverse, and then you corrected the phase of the microphone)

I wired so : 4 from emitter / 5 to gound / 1 > XLR pin 2 (+) / 2 > XLR pin 3 (-)

 

[Emmathom]

R10 3k performs a kind of self-Bias, to accept more varied jFETs.
jFETs are well known to have very large parameter dispersion.
The process of selecting jFETs to work in a specific topology, with exact, constant component values (it takes enormous amounts to select identical jFETs), or making in each microphone a custom BIAS, for each individual jFET, is not feasible, cost-effective in series production. It is an enormous financial, technological, time-consuming effort, etc.
So R10 allows more tolerance for jFETs, with larger specification deviations to work in this topology. It is true, not in ideal conditions, but average. The commercial criterion, the profitability, comes first.
Technically correct is R10=Oohm and in series with R9 1.8k put a semi-variable resistor VR of about 10...15k from which the BIAS of the jFET will be adjusted.
*
Do you still have the original, functional jFET? (even with this you will see voltage variations on the drain, source, etc.)

It's the original Jfet installed now... and from the begining ; the only tr I changed was the SA1015 but I reput the original one.
Maybe it's the lack of R10 (3KΩ) which brought back +6dB ? or the SA1015 original...
Ok for a VR at R10 (to find the right value) but which source voltage do I must look for ?

 

[micolas]

I wired so : 4 from emitter / 5 to gound / 1 > XLR pin 2 (+) / 2 > XLR pin 3 (-)

what does 4 have to do with emitter? is in the secondary.
I have drawn the correct diagram for you.

 

[Emmathom]

what does 4 have to do with emitter? is in the secondary.
I have drawn the correct diagram for you.

I mean from emitter trhu C4 (4uF) it's the signal + out

 

[micolas]

you don't have to look for a specific voltage source. They all differ. The jFETs differ from one to another.
The BIAS must be adjusted for maximum amplification, without distortion and with equal positive and negative half-sine

 

[Emmathom]

you don't have to look for a specific voltage source. They all differ. The jFETs differ from one to another.
The BIAS must be adjusted for maximum amplification, without distortion and with equal positive and negative half-sine

... and to do so I need a scope (which I don't have)
or record in Logic and watch the waveform zoomed in +++ ?