Cathode follower with impedence balanced drive

[Tubetec]

I want a cathode follower with unbalanced output to drive the cable back to a preamp/psu , say 15-20 feet away
I have balanced input at the far end , can I just do like in the diagram below , with the appropriate resistances ?

from this source ,
https://electronics.stackexchange.c...ngle-ended-signals-on-twisted-pair-conductors

 

[abbey road d enfer]

I want a cathode follower with unbalanced output to drive the cable back to a preamp/psu , say 15-20 feet away
I have balanced input at the far end , can I just do like in the diagram below , with the appropriate resistances ?

Ther's no reason why it wouldn't work as long as the cath-follower's output impedance is accounted for.

 

[Tubetec]

The CF bias and load resistor are located at the PSU end , the mic itself only houses the tube ,the grid resistor and the capsule , grid voltage is capsule bias , connected to the backplate , membrane grounded .

I had the idea of switching the coupling cap value to form a basic HPF , if I have the 2 core screen arrangement does the same filtering need to happen on the - input line for correct balance of the noise?

 

[Gus]

I read you are using a CF
Followers can have issues
Search for things like "emitter followers oscillations" it is something to know about when using followers

 

[abbey road d enfer]

I had the idea of switching the coupling cap value to form a basic HPF , if I have the 2 core screen arrangement does the same filtering need to happen on the - input line for correct balance of the noise?

That would be preferrable. Then you would have to double the capacitor value.

 

[abbey road d enfer]

Followers can have issues
Search for things like "emitter followers oscillations" it is something to know about when using followers

Indeed. The Rs resistors should be dimensioned as to prevent these oscillations.

 

[Tubetec]

Ive had very stable results using a variation on the B&K CF preamp circuit ,
even the unbalanced CF out makes it 15feet down the cable without degradation or noise pickup ,through the HPF and matches into 1M very well .
As I have the balanced topology in the pre the impedence balanced noise line makes even more sense .

It does make the HPF twice the trouble to do ,
I had done it simply before with a single rotary wafer switch but you end up with the extra incidentals of all the caps hanging off the line , dual wafer is better ,but if you then have to balance the the negative 'noise' input your upto a quad rotary switch assembly .
Ive done a basic 12 step HPF , its really effective for dialing in the low end on mics with proximity boost , caps dont need to be large or have very high voltage ratings as theres already a larger DC blocking cap and bleed resistor beforehand ,

A 23 step symetrical HPF using a 4 gang wafer , SMD chip caps might be an option but I used the small Wima red box caps before and they fit nice and neatly,
The Baxandall mic HPF circuit varies the coupling cap from around 0.1uF to around 800pf into 1M .
Any good coupling cap calculators that allow you overlay the various response curves of the caps ?

This one looks handy ,
https://www.ampbooks.com/mobile/amplifier-calculators/coupling-capacitor/calculator/
Handy video this one, its a very fast intro Kicad and it does just the very thing I wanted , Ill have to pluck up the courage to give it a go ,

 

[MisterCMRR]

If you don't need to vary the cutoff frequency of the HPF over too much range, you could make the coupling caps well-matched and use just one pair - moving the frequency by adding a single variable (or switched) R across points A and B (differentially). Choose the coupling cap values based on the lowest frequency - adding the single R differentially will raise frequency. Fewer caps and only one pair need be matched ...

 

[ruffrecords]

I am very impressed with the Kicad simulator. I must give that a try.

Cheers

ian

 

[Tubetec]

Thats an interesting idea , a few switched cap positions with a resistive trim might be nice way to do it
To low a load resistsance at the input might be undesirable in certain cases though .

The 12 step HPF I made previously with a single deck switch was very satisifactory in use , I have some 1% styroflex in the right values and some old ceramic rotary switches I got in Russia nearly 30 years ago , might as well double or quit .

 

[abbey road d enfer]

This one looks handy ,
https://www.ampbooks.com/mobile/amplifier-calculators/coupling-capacitor/calculator/

The "tube" calculators are nice, but for general electronics, you're better off with a Spice-based simulator, such as LTspice.

Handy video this one, its a very fast intro Kicad and it does just the very thing I wanted , Ill have to pluck up the courage to give it a go ,

This is another Spice-based simulator.
I'm not a big fan of Kicad, but I'll investigate if it has some dominant advantage over LTspice

 

[abbey road d enfer]

I am very impressed with the Kicad simulator. I must give that a try.

What is it that you see in this video that you think shows significant superiority over LTspice?
I see that the time steps are user-defined; that may be a significant advantage if it allows avoiding the dreaded "Time step too small" message of LTspice.

 

[Newmarket]

Take a look at Symmetrix

Download Free Demo - SIMetrix/SIMPLIS Elements

 

[abbey road d enfer]

Take a look at Symmetrix

I remember I had tried it. Didn't like it very much.
Didn't like at all that it's a teaser for buying the pay version ($$$).

 

[ruffrecords]

What is it that you see in this video that you think shows significant superiority over LTspice?
I see that the time steps are user-defined; that may be a significant advantage if it allows avoiding the dreaded "Time step too small" message of LTspice.

The ability to change a component value with a slider control and quickly see the difference it makes.

Cheers

Ian

 

[abbey road d enfer]

The ability to change a component value with a slider control and quickly see the difference it makes.

Cheers

Ian

OK thanks. I must admit it's quicker than using the .step command.

 

[Newmarket]

I remember I had tried it. Didn't like it very much.
Didn't like at all that it's a teaser for buying the pay version ($$$).

Fair enough. I should say that my employer has a group licence for the full version.
But I prefer its GUI / Windows interface.
I started here saying I would probably be more into LT Spice - people here use that too - but the GUI and interface has sort of won me over.

Yes - it's a lead-in for the full version. Like pretty much all free versions associated with a paid for version.

 

[ruffrecords]

OK thanks. I must admit it's quicker than using the .step command.

Absolutely. I get no end of errors with that command; matching up arguments and remembering all the commas.

 

[zamproject]

Not sure about how work netlist for usual sim, import/export, etc...
But with integrated sim in Kicad you only draw one schemo, for sim and pcb drawing/manufacturing

 

[abbey road d enfer]

Not sure about how work netlist for usual sim, import/export, etc...
But with integrated sim in Kicad you only draw one schemo, for sim and pcb drawing/manufacturing

That's an argument I hear quite often, but at my first attempts at Kicad for PCB, I was faced with a lack of electromechanical components (that are not needed in sim) such as connectors, switches, pots, brackets, and no possibility to create them. You were not supposed to create your own components, you had to request them and there was a long queue.
I don't know how it is now; I hope they came to their senses.