Ccs

[Jack Thomas]

I've been searching for a site or information on Gary Pimm's CCS's outlining how to calculate circuit values for specific loads. Can anyone please help? The attachment will be the victim of the CCS.

 

[jeapel]

hi here is:

 

[jeapel]

here is the guide

 

[jeapel]

other option are:

C4S CCS from bottlehead company

and cascode with DN2540 (i don t try)

but i prefer the sound of C4S over Gary

 

[jeapel]

i forget

the good point for Gary ccs is the mu-follower output
for low zout and very good bandwidth to drive next
tube stage

 

[Ryssen]

I would use one with Leds instead of 1N4004.

 

[forr]

Hi Ryssen,
I got oscillations using the scheme with two LEDs in series. It was cured using a 1 kOhm resistor instead of Led1. I then discovered that Morgan Jones has done the same thing.

 

[Wavebourn]

Here is mine:

You may use a P-channel FET instead of the upper transistor.

 

[Ryssen]

I got oscillations using the scheme with two LEDs in series.

I didn´t

done several of them.Know meny people done this.How did you notice that?

Tried one with a zener diod but strange noise at high frekvensys,Leds sounden much cleaner..

Wavebourn:How do you calculate R2 and R3 in your schema?

 

[Irakli]

Hi Guys,

Walter Jung just published I really nice article about CCSs in AudioXpress. I recommend everybody to take a look at it

 

[Ryssen]

I recommend everybody to take a look at it

Okej,bring it here(on)..

 

[Irakli]

Here,

http://www.audioxpress.com/magsdirx/ax/addenda/media/jung2778.pdf

and http://www.audioxpress.com/magsdirx/ax/addenda/media/jung2779.pdf

Enjoy

 

[Wavebourn]

Wavebourn:How do you calculate R2 and R3 in your schema?

They define a working point for a tube, i.e. voltage drop accross the transistors. The resistor in emitter defines a current, for example you want a voltage drop accross it 2V, so for 5 mA it will be 390 Ohm approximately.
Now, we need about 2.5V on R2, right?
Suppose beta of Q1 is 500, so base current will be 10 uA. For a good thermal stability let's a current through resistors will be 10 times higher, 100 uA. It means, R2 is 51K. For 30V on transistors (enough for a mic- or input stage of a power- amp) voltage across R3 will be about 27V, it means R3 will be 510K.

Let's take 12V Zener, and 1 mA across it is enough. Q2 don't need a high beta, it must be high voltage, and dissipate the calculated power.

For a driver stage for output toobs a voltage drop across the CCS we want higher, it means R3 needs to be of a higher value, and Q2 needs to be able to dissipate more power.

Some 1000 uF cap in parallel with Zener may be used to get lower noise level on very high frequencies (really, you may not need it)

More questions?

 

[Jack Thomas]

CCS

Thank you, everyone, for the fantastic responses! A veritable plethora of options now awaits. The only question remaining is.....what about noise below ground source level??? Just kidding! Thanks again!

 

[forr]

Hi Irakli,
I felt puzzled by fig 3A.
http://www.audioxpress.com/magsdirx...ia/jung2778.pdf
I think splitting R1 in two 5 kOhm's and adding a decoupling cap to the +18V rail should provide much better rejection.

 

[Wavebourn]

Hi Irakli,
I felt puzzled by fig 3A.
http://www.audioxpress.com/magsdirx...ia/jung2778.pdf
I think splitting R1 in two 5 kOhm's and adding a decoupling cap to the +18V rail should provide much better rejection.

It depends on a dynamic resistance of that +18V source. If it is good enough lovering R1 value worsens results. I'd better use one more CCS instead of the R1 (for example, one JFET).