how to design sinx/x filter for TDA1541A?

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.

Hi Patwen & Fedde,
My opinion on the sin(x)/x issue, i.e. slight rolloff of at 20kHz in a NON-OS DAC is reflected here:

It is quite clear that Thorsten and me disagree about the subject.
I made the choice for a perfect <B><i>square wave</B></I> and needle <B><I>pulse</B></I> response; Thorsten chose a perfect <B><I>amplitude</B></I> response. Given the phase response of his network has square wave must be looking horrible. I tried notch filters too and was not satisfied with these. Please also note that most of my experiments are with the AD1865, while Thorsten uses TDA1543 and TDA1541. The AD DAC is a much more bright and alive DAC than the Philips~!:yinyang:
patwen said:
I am planning to built a non-oversampling DAC based on TDA1541A.
Would someone kindly point me to some place where I can learn how to built a sinx/x filter to compensate the non-oversampling's roll off?

May I ask why you are planning to use this chip? Have it?

Haven't you thought about of using a top modern chip instead?

Make it easy for you and skip the sinx/x thing.

I don't want to spoil your fun but you may experience more fun with a new cool chip.

Crystal CS43122
Crystal CS4390
AKM AK4395
Analog Devices AD1955
Re: How?

Jocko Homo said:
When it causes the cheap op-amp they are using to generate IMD products that fall into the audio band.

IMD= Intermodulation Distortion. The occurrence of sum and difference frequencies when two frequencies are amplified in the same amplifier.:)
Though the frequencies are above the audible audioband the IM products can be in the audible region.;)
Re: Re: How?

Elso Kwak said:
Though the frequencies are above the audible audioband the IM products can be in the audible region.;)

It's a good rule to reduce HF noise even if your amps don't make it audible. Our surroundings don't like HF noise. Here in Europe it's much stronger regulations for that. For instance, american cordless phones are strictly forbibben here in Sweden. They radiate way too much. Always thrive for a clean audio signal.

A cell phone has to radiate an RF signal to work. It is a radio frequency transmitter/ receiver. Banning American products in Europe is not a technical issue but a trade protectionist issue. You buddies at Ericsson are just trying to protect thier market share. Probably a good idea since they are losing their shirts in the US Telecom market. :crackup:

Spot The Troublemaker

"Because he is Peranders, and he is am extreme designer, man, where you are usually a troublemaker.....

Sin Bin by Sundown,


Contrary to your opinion Mr Ren, Jocko and Harry have done their homework and freely give good advice derived from study, actual experiment, measurement and listening.
Perhaps the sticking point to you is that this advice is often given in a one line statement - IOW clear, concise and straight to the point and correct.
The Aussie expression for you Ren is "Pull your head in (before it gets knocked off) mate".
If you have sensible things to say by all means say so, if not then just zip it thanks.


Yes, indeed, you hear the most outrageous lies about it. Half-baked goggle box do-gooders telling everybody it is bad for you.

Pernicious nonsense! Everyone can stand to have a hundred X-rays a year. And they ought to have them, too.

Thanks to ****** companies like Ericsson, Alcatel, and Marconi, by the time the US telecom industry is set right, I'll be old enough to collect Socialist Insecurity. EC certification amounts to nothing more than a way to screw the country that bailed your butts out of the fire on more than one occasion. And how do they repay us? By leveraged buyouts of our copmpanies, and keeping the rest from selling anything over there.

Right, Arty?

End of rant.

Jocko "Sin Bin Bound" Homo

P. S. See ya' there Ren.........
I eat neither. Harry is the carnivore. I eat non-GM salad.

Sorry I forgot to put [joke] around everything.

But while I'm at it......has anyone else (excluding Harry.......I've heard his horror stories.....) gone through the frustrating experience of getting a CE sticker for some electronics gear? Seems as though they make up arbitrary rules every time you resubmit it.

Back to the subject at hand.....which was??

yes, but it showed up we could not get that sticker, because we only deliver equipment for bugger machines. So the complete system needed to be checked before you could mount the CE sticker on it.

But then our products needed to be tested to ensure that this would not be where the machines failes!

I do not think the EMC rules are that a big problem. Mostly we call 'Jydsk EMC' to find out which part of the regulations our parts have to comply with. They deliver a good service.

What i think is a problem, is the company's who have certified test labority charge 120USD/man hour. This can get costly!!


PS : Try to get CE-sticker on a spot-welder .. they normaly uses a 100KVA weld transformer.
Wow Jocko, that's a deep hole you're digging for yourself there... ;) I wonder if the internationals outnumber the Americans on this forum?

Maybe I respect your engineering opinions, but your political ones... ??? :rolleyes: I hope you aren't really naive enough to believe that the US doesn't use the very same tactics... actually worse in many cases for leveraging their disproportionate economic and political influence to strong-arm other countries. Well, enough anyway, I won't discuss it further.

So, back to the topic... sinx/x to me automatically means a filter in the digital domain - So doesn't the use of sinx/x filtering imply oversampling? Or am I missing something here... are we really trying to talk about brickwall filters in the analog world which therefore have a causal impulse response?
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.