Hi everyone,
I'm a newbie at audio amplifiers and I'm studing in Bob Cordell's book "Designing Audio Power Amplifiers", I've only made two designs but never tested it yet.
I've a question to do:
Which is the best current source, in terms of linearity, thermal stability and output resistance?
In both books I've got from Bob Cordell and Randy Slone the best current source is allways the feedback current source, in which both transistors act together to mantain their collector currents. Another form of making a current source is using a current mirror with a known voltage and a resistor (I think that this could also be made with a wilson current mirror or enhanced current mirror), this current mirror should be degenerated with 10:1 degeneration or a little more. The other type of current source which is also an acceptable design is the current source based on diode anode-cathode (Vak) voltage, you could use normal silicon diodes such as 1N4*** series, a led or a zerner diode.
Thank you very much for your atention,
Best regards,
Daniel Almeida
I'm a newbie at audio amplifiers and I'm studing in Bob Cordell's book "Designing Audio Power Amplifiers", I've only made two designs but never tested it yet.
I've a question to do:
Which is the best current source, in terms of linearity, thermal stability and output resistance?
In both books I've got from Bob Cordell and Randy Slone the best current source is allways the feedback current source, in which both transistors act together to mantain their collector currents. Another form of making a current source is using a current mirror with a known voltage and a resistor (I think that this could also be made with a wilson current mirror or enhanced current mirror), this current mirror should be degenerated with 10:1 degeneration or a little more. The other type of current source which is also an acceptable design is the current source based on diode anode-cathode (Vak) voltage, you could use normal silicon diodes such as 1N4*** series, a led or a zerner diode.
Thank you very much for your atention,
Best regards,
Daniel Almeida
Last edited:
For the LTP, a simple Zener current source is the best. It can also be arranged for low(er) noise as well. For TIS load I think the two transistor variant probably the best bet.
Hello Bonsai and thank you very much for your help,
I can use two feedback current sources one for the LTP and another one for the TIS, instead of zener current source for ltp and feedback current source for TIS?
Best regards,
Daniel Almeida
I can use two feedback current sources one for the LTP and another one for the TIS, instead of zener current source for ltp and feedback current source for TIS?
Best regards,
Daniel Almeida
Thank you very much for posting, this project with depletion-mode MOSFET seems very interesting, those CCS have very good output impedance characteristics even at higher frequencies, but I've a few questions to make:
The output current in those current sources are adjusted by the current set resistor, right?
What's the minimum current that can be set?
Whats's the minimum operational voltage?
The MOSFET draws current with the gate shorted to ground?
Thank you very much for your help,
Best regards,
Daniel Almeida
The output current in those current sources are adjusted by the current set resistor, right?
What's the minimum current that can be set?
Whats's the minimum operational voltage?
The MOSFET draws current with the gate shorted to ground?
Thank you very much for your help,
Best regards,
Daniel Almeida
Last edited:
Thanks for the review of these topologies. I had seen some of these in schematics here, wondered what they were for. My transistor books are very old. Wikipedia constant current source wasn't very useful, had one schematic, a transistor with a zener on the base leg. To keep my parts count down I ended up using JFET's with a gate tied to the source, a part which went obsolete last month, 2000 minimum.I have one J174 left. !@#$$%$^ Yes, I can buy them somewhere else, for $8-10 freight.
Last edited:
Hello Bonsai and thank you very much for your help,
I can use two feedback current sources one for the LTP and another one for the TIS, instead of zener current source for ltp and feedback current source for TIS?
Best regards,
Daniel Almeida
Of course.
The reason I like. Zener reference is tat it is extremely wide band. You do need to a. filter it and b. decouple it so that means two extra caps though.
My second favorite CCS is the two transistor active type.
I had problems with just a CCS in the tail of the LTP.
I found I had to use a current mirror too in the two legs.
Then I found I got a DC offset on the output so I made one of the current mirror resistors a preset.
I found I had to use a current mirror too in the two legs.
Then I found I got a DC offset on the output so I made one of the current mirror resistors a preset.
...
The reason I like. Zener reference is tat it is extremely wide band. You do need to a. filter it and b. decouple it so that means two extra caps though.
My second favorite CCS is the two transistor active type.
Someone did a series of simulations of different current sources to choose for an amp. That was you IIRC?
Did you ever compare any of the commercial CCSs? Seems like a suitable use for an IC.
Also did you miss my question on your transformer thread?
And best source of power supply capacitors -about 10000Uf 100V - to use with it?
Best wishes
David
Last edited:
The output current in those current sources are adjusted by the current set resistor, right?
Yes.
What's the minimum current that can be set?
1 uA to 10mA with LM334. Look at the datasheet. For higher current see LT3092 or use a trimmer.
Whats's the minimum operational voltage?
Just a few volts (1.8~2.2 DN2540, 2.9~3.5 10M45S).
The MOSFET draws current with the gate shorted to ground?
Yes, depletion mode (normally on). Make drain positive vs. gate to turn off.
Very, very good CCS. Ixys IXCP 10M90S is good for 900V.
Of course.
The reason I like. Zener reference is tat it is extremely wide band. You do need to a. filter it and b. decouple it so that means two extra caps though.
My second favorite CCS is the two transistor active type.
Zener references need lots of filtering because they are noisy. This is why I prefer a low Goss J-fet bias at Idss, cascode with a high Goss J-fet in series with a resistor||small COG cap. This is how Borbely does it. Although the output Z may be higher, this makes a very quiet voltage reference. If you like J-fets as much as I do then this is not a problem, using another J-fet as a buffer. I honestly cannot see how anyone who designs audio circuits could not like J-fets, they are so wonderful. I use them for whatever, whenever I can.😎
BTW, N-ch J-fets typically are less noisy. Keeping a constant Vds for the low Goss device helps a lot, it is worth the cascode. J177 would make a good P-ch source, cascode that with the J174.
Last edited:
This is why I prefer a low Goss J-fet bias at Idss, cascode with a high Goss J-fet in series with a resistor||small COG cap. This is how Borbely does it.
Hi CBS240,
Could you possibly post just a simple schematic of this? I can't exactly see how things are connected ...
Greetings,
Jesper
Hi everyone,
Which is the best current source, in terms of linearity, thermal stability and output resistance?
Firstly, all current sources are negative feedback current sources. What Bob refers to as the negative feedback current source is, in fact, an amplified negative feedback current source.
The amplified negative feedback current source would be my first choice for both the TAS and the TIS, as its output impedance is very high, approaching that of a cascode source.
Danny, you can read more on current sources and much else besides in the late great Hans Camenzind's book which he made available free to download here:
http://www.designinganalogchips.com/_count/countdown.pl?designinganalogchips.pdf
The book is also available to buy from Amazon. Highly recommended.
http://www.designinganalogchips.com/_count/countdown.pl?designinganalogchips.pdf
The book is also available to buy from Amazon. Highly recommended.
Last edited:
Did any of you read Walt Jung's series on current sources with measurements and such, a few years back in aX?
That has all the info one could ever want.
jan
That has all the info one could ever want.
jan
- Status
- Not open for further replies.
- Home
- Amplifiers
- Solid State
- Types of current sources for audio amplifier applications