diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Digital Source (http://www.diyaudio.com/forums/digital-source/)
-   -   Active Current Source for TDA1543 (http://www.diyaudio.com/forums/digital-source/22612-active-current-source-tda1543.html)

stefanobilliani 11th November 2003 09:07 PM

Active Current Source for TDA1543
 
2 Attachment(s)
I invite everybody to try this simple JFET current source-if you have not already made it-instead of the 1K resistor on the TDA1543 voltage reference at the pin 7.
In my set up the output resistors are 1K too. This application produce a better soundstage , better dinamics and more precision in details in the entire audio band .

... at least in my set up:)

:drink:


Cheers

Dr.H 12th November 2003 03:00 AM

What happens to the circuit values when 8 TDA1543 are used?

stefanobilliani 12th November 2003 06:31 AM

Quote:

Originally posted by Dr.H
What happens to the circuit values when 8 TDA1543 are used?
It need a different resistor value.
**********
With TDA1543 at 5volt power supply I choose the above 2.7mA current that reflects at the output - with I/V resistor=1K- 3.3volts.
It can be done also with a current like ... say 2mA.

So 2mA *8=16mA.
The resistor in the circuit above is tipically:0.3/I

For 8 TDA1543 in parallel R= 0.3/16m =0.3/0.016=18,7ohm

**********

If you want a slightly higher current decrease that resistor value.

:)

Andypairo 12th November 2003 07:43 AM

Hello Stefano,
I think I'll try your solution (among the half thousand I'd like to try it is simple enough and can be switched on/off for camparison purpose)

Correct me if I'm wrong, giving a quick look at the datasheet the current that exits from AOR and AOL is the difference between Ibias (that is the current from pin 7) and Idac, so Ibias must be > Idac(max), that is 2.65 mA.

Thus the Rload must be chosen to have Vcc/2 on AOR and AOL in order to have the maximum voltage swing..->>Rload = (Vcc/2)/Ibias

Is this consideration correct? Are there other things to take in account?

Cheers

Andrea

stefanobilliani 12th November 2003 08:31 AM

Quote:

Originally posted by Andypairo
Hello Stefano,
I think I'll try your solution (among the half thousand I'd like to try it is simple enough and can be switched on/off for camparison purpose)

Correct me if I'm wrong, giving a quick look at the datasheet the current that exits from AOR and AOL is the difference between Ibias (that is the current from pin 7) and Idac, so Ibias must be > Idac(max), that is 2.65 mA.

Thus the Rload must be chosen to have Vcc/2 on AOR and AOL in order to have the maximum voltage swing..->>Rload = (Vcc/2)/Ibias

Is this consideration correct? Are there other things to take in account?

Cheers

Andrea

Hallo Andrea,

Your reasoning looks to me pretty correct.
Probably , according to the datasheet , it is better to calculate with more precision the I/V resistor value for a low harmonic distorsion.I am however a bit confused where the Ifs (full scale 2.65mA )comes from , since the I bias is adjustable and the current trough the I/V resistor is in my case > of the Ibias

:confused:

machinow 12th November 2003 10:34 AM

Hi stefanobilliani,

As I am a certain FAN of the great LINEARITY , I should gently inform you that this current source will not be as linear as you might expect if the voltage across the fet is varying! I.e. if there is any voltage signal on it's output. I would rather prefer a more complicated current source or a BJT one with Voltage accross the Iset resistor at least 5 Volts.
Keep this in mind if you are about to use passive /resistor/ I-V convertor.
On my oppinion, the current source should sink 1/2 of the max output current of the DAC, so it outputs +/- 1/2 Ioutmax, instead of 0-(+Ioutmax) -> I.e. this is current shifting.
Regards

stefanobilliani 12th November 2003 12:20 PM

Quote:

Originally posted by machinow
Hi stefanobilliani,

As I am a certain FAN of the great LINEARITY , I should gently inform you that this current source will not be as linear as you might expect if the voltage across the fet is varying! I.e. if there is any voltage signal on it's output. I would rather prefer a more complicated current source or a BJT one with Voltage accross the Iset resistor at least 5 Volts.

machinow, thanks for your reply, well:

the Jfet is used as a current source for the pin 7 or voltage refernce for the TDA1543 and just to be precise it is not used in the output I/V stage.
From the TDA1543 datasheet Vref variations are from a min of 2.10V to a max of 2.30V.Constant current sources offer a fixed constant current at large variations of V applied. Jfets normally have a negative gate voltage and so they conveniently forms a self-biased current source as shown.The datasheet of the 2SK170 shows clearly the linearity zone at 2volts and 2.5mA of operation, so here what is the problem?

Quote:

Originally posted by machinow

On my oppinion, the current source should sink 1/2 of the max output current of the DAC, so it outputs +/- 1/2 Ioutmax, instead of 0-(+Ioutmax) -> I.e. this is current shifting.
Regards

I made experiments using the ohm's law from the original passive circuit that is 1k for the pin 7 and both 1k for the I/V stage outputs.
With that values the voltage at the output is 3volts (before the capacitor ofcourse) and the vref pin 7 is nearly the same 2.2 volts or so.As you can see I altered the current for the pin 7 from a 2.2mA to a 2.7mA that shifts the output to 3.3V.Is that too much?

Regards:)

rbroer 12th November 2003 12:42 PM

2 Attachment(s)
This should have a better output impedance,

BTW,
Have a look at Vref when music is playing :cool:

Andypairo 12th November 2003 12:53 PM

Quote:

Originally posted by stefanobilliani


I made experiments using the ohm's law from the original passive circuit that is 1k for the pin 7 and both 1k for the I/V stage outputs.
With that values the voltage at the output is 3volts (before the capacitor ofcourse) and the vref pin 7 is nearly the same 2.2 volts or so.As you can see I altered the current for the pin 7 from a 2.2mA to a 2.7mA that shifts the output to 3.3V.Is that too much?

Regards:)

I'd say it depends on which voltage you are operating it from.
If you use 5V it is probably too high, if you use 7V you can go even higher.
The output current is equal to Ibias (=2*Iref)- IDac , which is sunk by a controlled current sink.
Therefore the "zero" signal current for IDac should be 2.65 mA /2 (if we take the maximum value).

In your case with 2.2 mA Ibias = 4.4 mA and AOR should source about 4.4 - 1.325 = 3.075 mA * 1k = 3.075V

With 2.7 mA you probably saturated the bias current source at 5 mA (its rated maximum), so in this case we should have a theoretical 5- 1.325 = 3.675 V... not corrisponding to the actual value.

Can anyone explain why?

Cheers

Andrea

stefanobilliani 12th November 2003 12:58 PM

Quote:

Originally posted by rbroer
Have a look at Vref when music is playing :cool:

Hallo rbroer ,

I do it just now... it is stable at 2.120V with-or-without music.

:cool:


All times are GMT. The time now is 12:24 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2