Balanced input for JFet mic pre advice?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
hi,

I am designing a Jfet preamp similar in design to the Hamptone preamp, two Jfet stages, although I have completely redesigned the circuit to suit the 2sk117 JFet's of which I have many.

I need to add phantom power via a transformer balanced input but I don't want any additional gain, I was looking to use a 1:1 mic input transformer from here: EDCOR - MXL1

is this the best idea since I am effectively ruining the designs high input impedance of 10M ohms by adding input transformer, can anyone advise on this? can I use some other form of balancing scheme?

thanks
 

Attachments

  • fetass.jpg
    fetass.jpg
    957.9 KB · Views: 647
[QUOTE=evengravy

is this the best idea since I am effectively ruining the designs high input impedance of 10M ohms by adding input transformer

hello.
i am no mic expert but
iirc a mic only needs a few kiloohms input impedance (typ. 2.......to 5 k or so; sometimes 600 ohm).........so a transformer should be no problem.
greets
 
mjf, thanks for the response, I am FAR from an expert on transformers so I would appreciate any help with selecting the right type of transformer, am I right in thinking to stick with a 1:1 xfmr as the gain stages of the preamp give 60dB of gain as is and the additional gain stage on the xfmr shouldn't be needed. using this logic I have narrowed it down to these: (all 1:1 ratio from edcorusa.com)

MXL1 - m6 steel core, input >150ohm, THD 0.05%, Insertion Loss: 0.5 dB, Freq Response 20-20k<1dBu
MXL1cs (as above but copper shielded)

MX1 - nickel core, input: mic level(?), THD 0.01%, Insertion Loss: 0.5 dB, Freq Res. 20-20k<0.5dBu
MX1cs (as above but copper shielded)

I'm a little unsure if I have worked this out right as I have never worked with xfmr's before, the impedance thing is a little confusing to me. Is my thinking along the right tracks? any suggestions for better solutions very welcome,

thanks
 
I see a major flaw in your design, and it's not about an input transformer.

You are WAY over biasing the CCS fets under the darlington trasistors. R2 and R10 go directly to the positive rail - which will way over bias the fets, maybe even damage them.

They should be biased with a pair of resistors, one to gnd and the other to + rail, the values to bring the correct amount of cuurent you want in the CCS fets.

If you simply cap couple the input you don't need an input transformer unless you have some magical reason for wanting one. Just feed the phantom power to the mic through some 6.8k R's. Easy and far less expensive.

Regards, Allen
 
allen, thanks for your input, after looking at the circuit a little more I have made some modifications which are posted. Basically I have removed the darlington and CCS fet from between the first and second amplification stage as the extra drive is needed more on the output and this will reduce my part count. I have added two 1n4148 diodes to limit the current through the transistor pair also. this seems to have solved it, according to my simulations, and reduces current through Q5 to ~5ma,

comments allen?

I dont necessarily have a reason for having a transformer bar the simplicity of converting microphone balanced input signal to unbalanced for amplification by the circuit. can you comment on my transformer choices? I am not sure about this at all, particularly choosing a xfmr with the correct impedance.

thanks, even
 

Attachments

  • fetass_v2.jpg
    fetass_v2.jpg
    932.1 KB · Views: 377
thanks allen, I probably will turn this into a fully balanced pre at some stage but I dont have any match jfet pair at the moment so I would like to try adding a xfmr to this design for now,

Out of interest could I use a coupling cap on the input and just dump the balanced signal to ground rather than amplifying the balanced signal also? I realise it would no longer be balanced but would it be practically usefull?

also, could anyone advise on choosing a transformer for this application?

any help is appreciated,
even
 
Last edited:
[QUOTE=evengravy;
Out of interest could I use a coupling cap on the input and just dump the balanced signal to ground rather than amplifying the balanced signal also? I realise it would no longer be balanced but would it be practically usefull?

also, could anyone advise on choosing a transformer for this application?

hello.
only my opinion...........
yes.........this is an unbalanced circuit.........as allen w. stated a transformer is rather useless except you want to break groundloops ...........
transformer with this app. depends on the specification of your mic.......with a dynamic mic often used 600 ohms.........and with an electret mic 1kohm or so..........but you can do some experiments with this........the input impedance of the x former is often equal or bigger than the output impedance of the mic............
voltage gain of mic amps is often around 100x/40db or so............
greets
 
is this the best idea since I am effectively ruining the designs high input impedance of 10M ohms by adding input transformer, can anyone advise on this? can I use some other form of balancing scheme?

thanks

Mic preamps need a "ton" of gain. At least 50 dB. Transformers can help because you can get say 5X voltage gain with no noise. But you will have o sheild the transformers or keep them away from any eetric noise. Comercial mic transformers are inside cans and triple sheilded and expensice. The edcore xfoormer is open and might pick up noise.

Do you really want a 10M input impedance? I think that means the mic cable will pick up noise. That's the other thing a transformer does is allow a better input impedance. You amp's input impedance will be 10M times the transformer's impedance ratio which is the square it its voltage or turns ratio
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.