The ES9018 chip would in numbers cost less than than $10, how do you think these guys do a whole D/A converter using it for just $56 for the whole thing with free post. Try to find TDA1541 S1 Crowns cheap.
http://www.ebay.com/itm/ES9018-Hi-e...mplifier_Parts_Components&hash=item2ec7dda66f
Cheers George
http://www.ebay.com/itm/ES9018-Hi-e...mplifier_Parts_Components&hash=item2ec7dda66f
Cheers George
Last edited:
his TDA chip costs him $1000 each does it?
I don't they go back to the original TDA1540 - that was only FOURTEEN BITS.
Why don't we just reduce it to TWO BITS, if less is better.
Just imagine it, the appeal of Black & White stereo music?
(I realise that last line contains a joke that only serious photographers might understand.)
Cheers, Joe
it isnt less than $10 in numbers you are talking about, youre just waving your hands.
they use fake parts and autoplace/autoroute PCB 'design' techniques; using exactly the same design they used for some other piece of crap, thats how they make it so cheap.. then make the money on offering sets of fake opa627 at prices too good to be real, but high enough to pull a few sales etc. (real story, one buyer has reported dodgy markings on the 'upgrade' they bought)
these are the same sort of people that will sandblast or clean with acid and reprint/relabel parts that sell for a dollar or 2, logic does not apply.
besides, building with TDA unobtainium adds to the exclusivity/selling price
they use fake parts and autoplace/autoroute PCB 'design' techniques; using exactly the same design they used for some other piece of crap, thats how they make it so cheap.. then make the money on offering sets of fake opa627 at prices too good to be real, but high enough to pull a few sales etc. (real story, one buyer has reported dodgy markings on the 'upgrade' they bought)
these are the same sort of people that will sandblast or clean with acid and reprint/relabel parts that sell for a dollar or 2, logic does not apply.
besides, building with TDA unobtainium adds to the exclusivity/selling price
Last edited:
I think that you will find it is a TDA1541A.
Cut My teeth on those back on the 80's - ah... back to the future.
Actually, there was briefly a TDA1541 and then replaced by TDA1541A.
Don't tell that the non-A are considered better and bigger bucks... it is a never ending story... nostalgia that is.
BTW, George, haven't you figured it out... I am pulling your leg.
They were GREAT !!! Still ARE !!!
But... all I am saying is that there is other things/stuff out there worth a look... and a listen.
Cheers, Joe
PS: What are you doing Saturday week (22nd)? Definitely be hooking up a NOS-DAC or two. It's going to be great!
.
Are we getting derailed here on chip pricing?
Just to show everyone is correct , the Sabre chips come in a range from basic to outstanding. The cost of a Sabre DA chip depends on the model of course. The wholesale entry price is only $1.68 for the ES9023P but the top ES9018 goes for $ 65.60 from the US official distributor. 
Not that any of us would think for a moment that the new price of a PCM1704UK versus a ES9018 means anything in terms of sound would we?
Let alone that price would indicate R2R is better or worse then SD noise shaping. After all we all know tape is best anyway.
Back to the AD844 please?
Just to show everyone is correct
, the Sabre chips come in a range from basic to outstanding. The cost of a Sabre DA chip depends on the model of course. The wholesale entry price is only $1.68 for the ES9023P but the top ES9018 goes for $ 65.60 from the US official distributor. Not that any of us would think for a moment that the new price of a PCM1704UK versus a ES9018 means anything in terms of sound would we?
Let alone that price would indicate R2R is better or worse then SD noise shaping.
After all we all know tape is best anyway. Back to the AD844 please?
The DAC in that ebay link doesn't include the ES9018, if you read the item description you'll see:
"It is a kit,you need solder all parts by yourself.pay me more 25USD,i can solder them for you.
Don't include ES9018 IC,pay me more 56USD for one,and don't include 2pcs Single Op Amp and 2pcs dual opa amp(can use LME49710,AD797,E5534,OPA627) ,pay me more 22USD for 2pcs LM49710,pay me more 26USD for 2pcs LME49720.pay me more 25UD for the CM6631 USB card.Near the 9018IC is dual opa amp,the other 2pcs IC is single ic"
Are we getting derailed here on chip pricing?
Back to the AD844 please?
Yes I agree, back to the AD844, it doesn't take much to derail a thread, from new posters spruking other things and ruin a good informative thread.
Cheers George
Last edited:
Its back again, production is continuing according to the page currently up. PCM1704's death appears to have been somewhat exaggerated....
It's listed as "NRND" on Mouser. Is it really that much better than the PCM1794a?
If 'better' means 'better sounding' in this context then the PCM1794 is too coloured for my taste - noticeable particularly on massed brass. I haven't heard a PCM1704 myself but I know of no colouration mechanism in it. Perhaps those who've applied it will point out pitfalls of its reference section though.
Hi,
the 1704 is probabely the best R2R Chip available. In our own evaluations a implementation with 8pcs of 1704 in a parallel mode didn´t sound as well to our ears as a pair of 1792/1794 in Mono-Mode. But then there are so many points where such DACs must differ -starting by layout and not ending with parts- that I wouldn´t want to generalize about the findings. It seems to me though that modern DAC-Chips have reached such a quality, that sonic differences are more due to the digital filters, clock quality, analog filters and analog stages than to the actual DAC-core of a chip. Unfortunately the 1704 is NRND and terribly expensive now.
jauu
Calvin
the 1704 is probabely the best R2R Chip available. In our own evaluations a implementation with 8pcs of 1704 in a parallel mode didn´t sound as well to our ears as a pair of 1792/1794 in Mono-Mode. But then there are so many points where such DACs must differ -starting by layout and not ending with parts- that I wouldn´t want to generalize about the findings. It seems to me though that modern DAC-Chips have reached such a quality, that sonic differences are more due to the digital filters, clock quality, analog filters and analog stages than to the actual DAC-core of a chip. Unfortunately the 1704 is NRND and terribly expensive now.
jauu
Calvin
I've just ordered some buffer samples of LME49600 and LME49610. I'm going to try taking the signal once again from the TZ output and see if these buffers can beat the internal AD844 buffer.
They are buffer beasts unity gain no feedback, 0.000035 distortion, 250mA current output at 5.6ohm and run at quite high bias currents should sound good, I just hope they don't have too much dc offset when hooked up to the TZ output, it will have zero, so my fingers are crossed.
Cheers George
They are buffer beasts unity gain no feedback, 0.000035 distortion, 250mA current output at 5.6ohm and run at quite high bias currents should sound good, I just hope they don't have too much dc offset when hooked up to the TZ output, it will have zero, so my fingers are crossed.
Cheers George
just dont enact wide bandwidth mode if you are really worried about DC, especially if A. you dont need it and B. you dont have all your layout and parasitics in order. the reports of DC offset were caused initially by the badly implemented servo in the reference design AFAIK, but they can also come about from bad layout or noise pickup, since they are so wide bandwidth.
great chips though, enjoy.
no feedback? unity gain, no feedback? hmm ... seems an oxymoron. also the performance in the datasheet is measured with it inside the feedback loop of lme497xx afaik. performance is still excellent
great chips though, enjoy.
no feedback? unity gain, no feedback? hmm ... seems an oxymoron. also the performance in the datasheet is measured with it inside the feedback loop of lme497xx afaik. performance is still excellent
You also gain higher input impedance (7.5Mohm) by not having the higher bandwidth on, which has to be a plus for the AD844's TZ output being so high.
Do you think the web quoted output impedance of 5.6ohms will still be that low if not implemented into a feedback loop of something else? And also the input impedance, will that stay at 7.5Mohm?
I'm trying to see on the data sheet where it states that all measurments are only if it is within a feedback loop of something, but I can't see it anywhere, it only states that the distortion figure was done within a feedback loop, none of the other measurements.
Aslo if dirkwright is reading you have had a bit of experience with the LME49600 what are your thoughts? Esspecialy with the newer LME49610.
Cheers George
Last edited:
I had a look at the LME49600 DS - couldn't see the quoted output impedance but its going to be the open loop value. Putting it in a feedback loop will reduce the 5.6ohms by the excess loop gain. The input impedance stays the same yes, also be aware of the upto 2uA input bias current - higher in high BW mode.
The output impedance of 5.6ohms I got by Google to the TI forums.
As far as the input bias of 2uA, yes it's a pity they didn't use a fet input, I don't know how a bi-polar input will react to the TZ on it, it may give a high dc offset then on the output of the LME49600, I haven't knowledge on this sort of thing? Have you any suggestion on this?
Cheers George