Sorry to bother you again on this old dac (TDA1543), but I'm going to build an active opamp IV filter at the output.
I need to know which order should be and, overall the cut off frequency.
What I don't understand is why many have cut at 10KHz (and not 20Khz or more) and they say are completely satisfied about the sound.
https://www.diyaudio.com/forums/digital-source/25893-tda-1543-stage-education-2.html
Can you please give me a quick guidance about which should be a good sounding IV bending curve at the output of a NOS DAC like TDA1543?
I need to know which order should be and, overall the cut off frequency.
What I don't understand is why many have cut at 10KHz (and not 20Khz or more) and they say are completely satisfied about the sound.
https://www.diyaudio.com/forums/digital-source/25893-tda-1543-stage-education-2.html
Can you please give me a quick guidance about which should be a good sounding IV bending curve at the output of a NOS DAC like TDA1543?
Hi I'm referring to post #22 and successive: 10Khz cutoff (I ve simulated too)
https://www.diyaudio.com/forums/digital-source/25893-tda-1543-stage-education-3.html#post367237
https://www.diyaudio.com/forums/digital-source/25893-tda-1543-stage-education-3.html#post367237
I've recently put up opamp-based schematics for I/V and filter stages for a TDA1387 DAC here : AliExpress AD1865 R2R DAC
You might find something useful there as there are similarities between the chips.
You might find something useful there as there are similarities between the chips.
But your scheme's curve is flat for hundreds of KHz?!!
This confuses me even more...
That's just to illustrate the effect of hanging a 2n2 capacitor on the inverting input of the I/V opamp - not really recommended. I posted up a post-filter which has a 3rd order roll-off beyond 20kHz and I'm going to suggest a CLC filter prior to the I/V which does NOS droop correction as well as attenuation beyond 22kHz or so.
Last edited:
Ok, assuming I'm building a separate opamp filter of 3rd (or 4th) order to cut off after 20-22KHz and provide the necessary voltage gain - thus I can run a bank (1..2..4...8...) of TDA1543 at 5V without heatsink- (I need +6.3 db gain at least because I'd like to use AD828), I need now to design the IV.
I have in mind two solutions:
1) do the IV conversion with a simple opamp with the right feedback resistor followed by the AD828 opamp filter (like in the datatsheet). In this case I need one opamp stage more.
2) do the IV conversion with just a resistor across the TDA outputs before the AD828 opamp filter. Much more simple.
Which do you suggest in such situation?
I have in mind two solutions:
1) do the IV conversion with a simple opamp with the right feedback resistor followed by the AD828 opamp filter (like in the datatsheet). In this case I need one opamp stage more.
2) do the IV conversion with just a resistor across the TDA outputs before the AD828 opamp filter. Much more simple.
Which do you suggest in such situation?
If you are using an opamp to do the IV then this stage needs to implement some filtering because an opamp output cannot follow the transitions at the output of the DAC. It need not be the main filter, though.
TDA1543 with 5V supply cannot use simple 'resistor to ground' passive IV - see my website for details. You need 6V minimum, and 8V would be better.
TDA1543 with 5V supply cannot use simple 'resistor to ground' passive IV - see my website for details. You need 6V minimum, and 8V would be better.
I don't much like either 1) or 2) as they stand as each has weakness. But whether those weaknesses matter depends what you're going to be feeding the output of your DAC into. For example, if its a datasheet application chipamp then either will be sufficient, go with whichever is easier to build.
Ok, will use a fast opamp unity gain stable IV converter supplied with +/-15 volt. As usual in all my project I will use AD826 (not AD828).
I need an advise: feedback resistor have to stay 1100 ohm (1.1kohm) and the cap across it 4.7 nF. This sounds perfect with another DAC. I won't change.
Can you please suggest me:
1) optimum supply voltage for TDA1543. I calculated 5.53 V or more (6V?)
With an assumed output current of 2.3mA, this results in a maximum voltage swing of 2.53 volt (1.1K * 2.3mA). The AC voltage swing has to stay within 1.8V and Vsupply-1.2V.
So 1.8V+2.53V+1.2V=5.53V min
2) optimum Rbias value connected as:
Ground -> Rbias ->TDA1543_vref -> AD826_non_inverting_input
I found 2.69K independently of voltage supply.
I need a biascurrent of 1.64 mA (1.8V/1100), so Iref needs to be 0.818mA, resulting in a Rref of 2.69Kohm = [2.2V/(1.64mA/2)].
Correct?
You mention using an opamp for IV, then a resistor. Which? If an opamp virtual ground then you don't have to worry about AC voltage swing on the DAC chip output because this will be approximately zero.
My calculation (done a few years ago) is that if you want 1.1k output resistor then you need a minimum supply rail of 7V.
My calculation (done a few years ago) is that if you want 1.1k output resistor then you need a minimum supply rail of 7V.
I will use exacly the solution in the datasheet, thus an opamp with 1.1k of rfeedback (instead of 1.2k) and a cap of 4.7nF (instead of 3.3nF).
So which is the optimum supply and Rbias from ground to pin 7?
Ok, thanks.
Can you please give me a simply rule of thumb to select Rbias from pin 7 to ground by using an opamp as IV converter with 1.1K feedback? Datasheet says nothing. Someone don't put at all. Someone suggested even 22k....
I think that it should increase Midpoint some way..
Is this correct?
Midpoint=2.2V+1.1K*1.15mA+1.1K*2.2V/Rbias*2 (??)
So if I want 4V midpoint should I need Rbias=7.5k?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.