I did the same tests as you, about 7/8 years ago, the chips I preferred were the Taiwanese double-crowned chips that were much more transparent and richer in the treble range, although sometimes I found chips without markings that sounded remarkable the important thing is to always test in pairs on a CD like the Sony ESD337, which actually used a pair........
Hi Gabby, thank you for sharing! As far as I know, those chips Looks identical are refurbished from old stocks, most salvaged from used equipments or a wholesale junk lot. This can explains why two identical S2 Chips can perform differently, because their “past life” were different 😅
Not surprised! Even two chips from the same wafer can often have differences as the coatings are not evenly spread out so the inner area will often have better coverage unlike the outer edge chips. Within the digital realm this is not that big a risk as long as the state is still set correctly. Unlike the digital space analog is very tricky! So A to D or D to A chips is an art like trying to grade fine wines!
Loved the video Gabby! Thank you for taking the time, effort and cost for doing this as it was very educational for me as someone who has an interest in the TDA 1541a dacs. Sadly it has also made some people go onto a buying frenzy as well..lol. I had a friend asking me on one that had a S7 (yeah 7 stamped crowns) chip on Ali i think, and i steered him away from that obvious fake. I do have a couple of holland chips, from the Years 87 and 88. The R1 chip is from the 87 batch and it is very analog sounding, just buttery smooth mids, bass a tad bloated and the highs are less sparkly. The standard 88 chip which is the same as your #15 is quite linear and holographic sounding and the S1 double crown from Y88 is just downright amazing but not far off the Y88 standard, but the mids smooth, yet detailed as well.
If you've seen enough of these, you can easily spot the ones that are relabeled. 1. Original chips are labeled just good enough, not perfectly bright. 2. Philips switched to BitStream DACs in the early 1990s, so production date stamps later than, say 1993, are suspicious. So are multiple chips that have the exact same date code. 3. The S1/S2 and crown on real chips were added after production to chips that were outstandingly linear (or something). They should obviously be stamps so the lines aren't as regular as the rest of the labeling; they may also be a different color and they're likely to not be perfectly aligned with the other labeling. So judging by what you're showing us: 4 and 13: I would have guessed that those are real, even if I wouldn't have known that you personally removed them from CD players. They may have the same date code but 1988 was when Philips was using these in all their CD players. The Philips logos are different between the two and the S1 and crown are obviously stamped on there. 16 also looks legit because the S1 and crown are obviously stamped on there, the date looks correct and there are just enough imperfections in the labeling. 8 has a date code that looks suspicious to me (1996) but could be legit. The color of the crowns is slightly different from the rest but it's darker so that's also suspicious. Maybe it's an older one-crown that got relabeled. 10 and 9: The ink on these ones looks a little too bright and the date code is a little on the late side (and a little bit too identical), so if I would see them on eBay, I wouldn't buy them. Nice that they turned out to be good! I think you were lucky. 6 and 11 look legit. There is just enough variation in the labeling to make me think those are the original labels, and the date codes are in the time when Philips used these for all their CD players. 12, 14, 1 and 7 all have the same date code and the date code is very late so that would make them suspicious to me. 15 and 3 look totally legit; plenty of imperfections in the labeling and the date codes are perfect. Would buy. 2 looks like a fake, the labeling is just a little too perfect and the crowns were obviously put on there at the same time as the rest of the label. The late date code is also highly suspicious. 5: I don't know. It doesn't have the Philips logo and I don't think there were any other factories that made these. Maybe it's a really early one (Philips tended to put less labeling on pre-production chips), but then the 1987 date code is a bit on the late side (they were in full production much earlier than 1987 I think). I would probably give it the benefit of the doubt if I would see it for sale somewhere. But obviously if I would be able to find a 1541A, this would drop out of my grocery cart. As for your question about whether these are still in production: Philips Semiconductors changed into NXP around 2000 and it's highly unlikely that any of these were still produced even in the late 1990s. I'm into the Philips Digital Compact Cassette and all DCC recorders have BitStream DAC's. And DCC recorders were produced between 1992 and 1996. It's possible that they kept producing the TDA1541(A) to sell them to other manufacturers but I'm fairly certain that they would have discontinued them by the mid-1990s.
Thank you for the in-depth analysis You have very good knowledge of these chips I sure learned a few things reading your comment. I really appreciate you taking the time to share your experience. Do you think that new chips are bing made in China now or maybe Taiwan I bought a few to test as I am curious I bought 6 from the same batch that have 97 year 1 is outstanding 1 very good 1 not good and 3 average. So where and when you think these were made.
@@Gabster1 It's possible that Philips actually made them long after they stopped using them in their own electronics. Maybe they sold them to OEMs in the late 1990s. But as I said, they didn't use them in any DCC recorders and those were all made between 1992 and 1996. However everything I said is just based on experience. Though I lived in Philips' home town and worked for them for a while, I don't have any insider information about chip production.
Hi, thanks for sharing this great investigation (kudos, definitely lots of time, money, precision work and mental fuel) Could you possibly by any chance consider at some point investigate/add the parameters of chip package (top side) temperature (with and without heat sink) and their current consumption to your evaluation table ? Thanks again for the always great work and sharing it
I took a thermal scan of a few but not all As they were all similar within a couple degrees They got to 40 degrees Celsius fast within a min and to around 47 degrees after 10 min But the consumption is a good idea for next time I do this
Hi Gabby, I read with interest your new found passion for tda1541a. You wanted to know if there is new production in china but I think the answer is no. Let me explain from my "research". Tda1541A ceased production in 1993/94 as Philips went on to 1 bit or bitstream dac the precusor of modern dacs ( delta sigma ) simply for cost reasons and simplicity of chip manufacture. To date Philips has the preselected S1. The reappearance of tda1541A is because Marantz which is bought over by Philips wanted to commemorate the aniversary of the introduction of the compact disk player. Philips then tasked TSMC of taiwan to make tda1541A but since the silicon technology had improved Philips decided to also make a special selection for this occasion calling it S2. According to one source 2500 pieces of S2 was shipped to Marantz so the rest was sold to chinese resellers. 😱 I believe these are those that did not pass the grading process or as well as ungraded ones. That said at best the last A version date code would be 1998. Now interestingly S2 was put to use in marantz flagship cdplayer D1. You can see the actual markings, that is the S2 and crowns ARE NOT ALIGNED. Photos are in the Dutch Audio Classics. Date code is 9713. I post links that will give more insights to tda1541A. I have bought some A version but the first batch looked genuine but underneath is marked "China" date code 9635 batch 11260. In my mind I don't think TSMC had started production in 96 or is this a trail run. I then bought a 2nd. batch but to my disappointment they were non A remarked as A version. So I am thinking would you like to test them. It could be genuine A and non A at the worst as I had made a diy dac pcb using lm317/319 regulators and cs8412, the then popular digital reciever with low jitter. Both A and non A sounded the same to me but of course you have a much better setup. If you are interested,please give your telegram handle so I can send chips to your postal address free of charge. You can keep them if they are good. A pair each to make a differential pair. www.audialonline.com/blog/tda1541a-and-model-s-usb-part-4-tda1541a-grades-and-series/ cd94blog.wordpress.com/tda1541a-tales-from-the-world-of-make-believe/ www.dutchaudioclassics.nl/marantz_project_d1_d_a_converter/
THANK you 🙏 for all this info that should keep me busy reading all these articles Very interesting information. That makes sense as it does look that the stock being sold has a mixture of Quality levels some are excellent and some ok none are bad I discovered something new since I posted my video some chips that were ok ended up being a lot better after some breaking time I discovered that as I did a second blind test one of my lame ones turned up being good after a couple days it could be moisture issues and needed some time staying hot as they usually run around 47 degrees Celsius My info can be found on my site here Feel free to reach out gabster.ca/Contacts/ I can also send you a couple boards in exchange to try Thanks again Gaby
my first chip was purchased from Speakermango from Ebay. sounded really nice and was$100 shipped. after this video, i purchased an S1 chip from a fella from Ukraine for $300. He claimed to remove it from a player (with photo). could be bias (mine, not the dac) but i think the chip sounds clearly better (vocal presence, timbre and detail). very happy. now questioning if improving the 5v ps with supercaps or using ians snooty clocks could bring more love. using ians linear pi and a 15v ifi. which is the dac ps?
The TDa chip is the best place to invest in The big caps will help but minor amount you will get a slightly firmer controlled Bass just careful switching TDa chips not to put it in reverse or off by a pin or reversed polarity Only use you S1 when things are stable these chips are hard to find
Hey I want to know what you think of the song. All the walls by erotheim, because it sounds amazing and I bet when I get done with my speakers, it'll be absolutely incredible, but I would like to see what you think of it
The Imaging Air and Musicality are far better on the TD1 Low and High frequencies detail is the same on both, Midrange Varies it depends on The TDA1541A Chip. On a very good Chip Vocals Mids are equal or better. On a average Chip very close. With a low grade TDA Chip everything suffers. 95% of the Chips selling out there are about average
Very brave and surely a labour of love! Well done and thanks! I won't suggest paralleling chips then, seeing that you've already fried a few of them! However, do you plan on ever producing a balanced dac, i.e. a truly balanced design, in the digital domain? Also, are you planning to experiment with transformers in the I/V stage?
I already did a balanced Dac with my TD1 boards but it is very dependent on the TDA1541A chip quality and matching But plan on spending more time again once I have a bit more time I have tried both Ivan’s top Transformers and they are well respected transformers. To my ears in this TDA application The OPA 861 In a single no feed-back design produced a better sound.
@@Gabster1 Excellent! Thanks! I just watched it. It's the episode uploaded 3 weeks ago right? I didn't know you also had a website. Also I wasn't aware that there were Chinese-made chips out there. Are these just unlicensed clones? Do they actually exist or it is just a false rumour? With all the fuss about the TDA1541 going on for the last 30 or so years, one would think Philips would try to release another batch on the market. One has to wonder why they haven't. Even worse, one wonders why they have not come up with an improved version. After all the TDA1541 is a 16bit version of the TDA1540.
This was Released on Patreon a week ago. You definitely have a point. I almost feel like reaching out to Philips. I sometimes dream of a TdA1541G Then I wake up :)
@@Gabster1 It does happen all the time though. If not Philips, maybe some other company could buy the rights to reproduce (not the technology rights) and simply re-release it under their own name. Of course Phlips might not want to sell the DEM technology, if the patents on it have not expired, but that is another matter.
I would have felt better about this testing if the process had been done twice, with a 3rd party randomly re sequencing the DACs at a mid point. The interesting point here is how far apart the two evaluations of the same part are - this is a quality measure of the audience. Also, it would be nice to store these parts on a grounded conductive surface (tin foil ?) and handled with some kind of ESD management. None of the latter will affect the sound, but may affect overall life.
I interpret the point of this video to be that the measurements are of low value, but listening can identify superior- and inferior-sounding devices. Yet, as another commenter said, the reliability of subjective evaluation is suspect. I propose an experiment to determine how real your evaluations are: Remove from your evaluation population those chips that measure bad. Then reevaluate some from the best and some from worst group through a new blind test with the tape and numbering redone. You should even repeat this experiment a third time, renumbering again. Can you reliably tell the best group from the worst group by listening?
you must be reading my mind I actually did something similar yesterday. I used 4 of the best ones 2 of the not so good ones and added 4 new ones. The aim was to figure out how the 4 new ones tested again covered with tape blind test I repicked the best ones easily I was off by one of the not so good ones I rated it average the 4 new ones that are Identical from the same year one was very good one good one average and one below average. The moral of the story is, the very good ones tend to stand out and are easy to rate. The other thing is that there are variations in Quality between chips even the ones that are Identical of all the S1 chips they were all very good with minor variations.
ok... but at the end of the story we are talkin about a very old chip with serious limits in frequency and resolution especially when considering the every day more and more common specs of file available from streaming... ok I know it is still plenty of 16/44 but... does it still make sense to spend that amount of money on something that born already old?
I have a PCM1794 dac thats 15 years old and the chips are much older. It sounds as good if not better than the new Gustard A26, not as hyper focused detailed but very close, more musical, better sound stage. I was thinking the same thing "after 15 years its time to upgrade", now im thinking why did I bother, So yes, the old chips still sound very good, and in some ways worse is better
May be the point is that whole technical developement is driven by sells and sells re driven by popularity and popularity needs portability and in effect miniaturization . This is the reason why new chips were developed with small energy consumption and dimensions and due to 90% earphone reproduction and low prize of smartphones I would expect no better then 1541 quality. . I used few CD players - portable rised my doubts , original japanese Onkyo with other DAC found just no problem but I wasn't fully amazed wuth any CD, and when I bought Bang Olufsen with TDA 1541 since 10 years I do not even think about any issue each CD amazes in total. If something could be 1% better it needs be already analog
None Do I even tried running the -5v only in case they did reverse the pins internally, also all need the 15v. They have a average TDA consumption and a Thermal scan showed a average of 45 degrees Celsius which is typical. they are Legit or they figured out a way to reproduce and manufacture the chip. As @mark4571 mentioned they all run in Sim mode.
Interesting that it is possible to hear individual difference between chips!👍 That begs the question if it is detectable sonically differences between between all types of different chips? We know it is sonically differences is in the "simple" OP-amps so logical their have to be POTENTIALLY as great or greater differences between more complex chips.. Or? I mean if there is individual differences between the today produced ESS/AKM DAC chips.. (Yes, some may say that we have better and more consistent production lines. But remember that manufacturers is every day pushing them self to make the transistor smaller and smaller and struggle probably likewise as they did 1981.. And they also have new issues that comes with that.) Yes, I know that those chips is far from easy to swap out, but the question still persist.😂 OK if we conclude that ESS/AKM chips individually sounds different. Then set aside that other components on the board (may/probably) also influence the sound quality. Then there should be a sonically differences between two individual complete assembled DACs that are the same (model and brand) and coming from the same production line!🎉🤔 That then this little detail opens up questions, that when you buy for example a DAC. Then order two and return the one that is performing not that good as the other one??😅😂❤🤔 Thanks for the video! As seen it started a train of thought's.😂🎉
While it is correct to notice that production variations, even among brand new genuine, chips is a very real thing. We have to appreciate that those variations will be remarkably small and will reduce in second and third production runs as the die and substrate architecture is improved. The far greater variable in listening tests is the human ear. Simply sitting in a slightly different (and I mean in millimetres) listening position can result in remarkable shifts in perceived audio quality. We must always be aware that we have two ears, listening from slightly different positions in the sound field and there are going to be cancellation and summing effects because of that. Here's a video that demonstrates the most common of these problems quite nicely ... ua-cam.com/video/T8bmASpUeh0/v-deo.htmlsi=R_RW7knxVBmGDUeI The problem is that we cannot cleanly eliminate the many many variables in listening tests so subjective impressions will always be both highly variable and suspect.
Hi Gabby, could you share the chip you bought that is decent and the price is attractive (and from where) thanks Learnt from the video that you are selling the chips, please advise price . Thanks
Sadly most sold out as I was very careful where to buy from I avoided buyers that seem to have hundred of them. That shows as most except 2 were actually bad chips.
@@Gabster1 Sorry no, i have link for white ceramic gold pinned TDA1540. There so very rare. Tbh i have never seen equivalent TDA1541 but im sure they exist.
Totally agree unless you know the person desoldering and if they have good equipment and expertise to do it With my Hako desoldering tool they come out easy
Where can I find the electrical measurements on the different chips? The SQ differences must be visual in the measurements, or a there must be trend for what sounds good or bad.
@@Pete.across.the.street That is the question, but to determine that we need measurements and listening feedback. Here we only got the subjective feedback. I would say the more measurement the better. But it looks like the harmonic response from single and multitone are where you can start to see differences in the system. How does the "forest of trees" look like in comparison etc. But the more date the better, then we hopefully can extrapolate what Gabster thinks is a good or excellent sounding DAC.
@@mk1classic you can't measure sound quality off an electrical signal. You can measure differences in the signal, whether it's good or not, you have to listen.
Generally a search for the part number plus "Data Sheet" will find the manufacturer's specifications for most chips. For example: "TDA1541 data sheet" Will find a PDF file with detailed information about the chip, including pin functions, timing, limits and internal details.
@@Pete.across.the.street The main ones are frequency response, harmonic distortion, intermodulation distortion and crosstalk. If you have a chip with 10hz to 20khz frequency response within 1db, less than 0.1% distortion and channel isolation greater than 40db... it is almost certain to sound fully transparent and detailed.
According to the Dutch Audio Classics website, all real S2s (double crown) are from the final 1997 Taiwan run and did not exist before then. None have the stripe on the left. They all have "TAI WAN" where the stripe normally appears. It seems many chips (both real and fake) have been restamped with a double crown, which is not surprising. The TDA1541A was produced in Belgium/Holland (1988-91) and the USA (1991-94). Unless one actually pulls one from an old piece of gear, it's a crapshoot. There are just too many restamps and fakes out there.
Very well said there is a lot of crown stamping Some are plain chips and can sound different from one to the other. The moral of the story is to listen 👂
My thought exactly. Looks like a fake. I have the real S2, and S1, R1 and normal. To my ear the S1, R1 and standard all sound very similar. The S2 is easy to differentiate. It is just much more smooth and analogue sounding. Not everyone likes it. I think it is perfect and feel fortunate to have one.
@@user-od9iz9cv1w I like the last batches of S2 with N1 on the description receipt they sound very similar to S2 crowns though it's pretty hard to differentiate S2's especially after DEM clocking.
@@Gabster1 There are also mystery Chinese sold and most likely Taiwanese made TDA1541A S2 with a bizarre thick printing of TDA1541A S2 and are considered to be like Crowns . They most likely kept on printing them after Philips cancelled their manufacturing there . My friend bought a few and they seemed pretty good and they were definitely not shaved and rebadged older TDA1541A or fakes . They were identical to the TDA1541A S2 in appearance except for the bizarre thick printing in a bronze gold type colour .
First time I see someone taking the time and spending a lot of money and resources to help us see the big picture. Fascinating that a lot of chips from the same batch can sound differently. Todays Chips may have better tolerances and variations but I am sure small differences exist.
They all did run in sim mode but not sure if I have any bad fakes Maybe they can reproduce good ones now no Idea if they are that is still my unanswered question maybe someone knows and can tell us
every chip was tested after manufacture, because you couldn't make chips that accurately in the early days. and they were divided into groups, because otherwise you would have to use too many different codes. It is therefore very possible that there are differences within the group.
@@jacquesw8725 Yes, remember back in the days and presumably today also. We know that some computer CPU chips is "better" and can handle over clocking better that shows us that there is individual differences between chips coming out if the same production line.. So there is never 100% the same. And it is more obvious on CPU chips when we can find and test the individual maximum clockspeed under stress test conditions. Hope this helps a little about the understanding. 👍
@@AmazonasBiotop Most people do not relize that DAC chip is digital but only in part. Of course in optimum case digital is always repeatable .But final stage must be analog and in analog 100% repeatability is never the case - we may think so but our measurements in compare to to hearing sense. are deeply not accurate
@@Mikexception Yes, then it is yet another reason to buy always two of the exact same HiFi component. (Or compare two identical individual sets somehow, to determine which to choose.) When the analog side is also not 100% the same. And return the lesser good sounding.. 🤔👍
@@AmazonasBiotop Choice is up to us- I would not categorise because bad digital is bad and bad analog is bad. And good ones are good. But in theory I expect digital cannot beat analog because is created from analog and finaly returns to analog. Reality is as is and in say 85% important is used final reproducirng system
I did the same tests as you, about 7/8 years ago, the chips I preferred were the Taiwanese double-crowned chips that were much more transparent and richer in the treble range, although sometimes I found chips without markings that sounded remarkable the important thing is to always test in pairs on a CD like the Sony ESD337, which actually used a pair........
Hi Gabby, thank you for sharing!
As far as I know, those chips Looks identical are refurbished from old stocks, most salvaged from used equipments or a wholesale junk lot.
This can explains why two identical S2 Chips can perform differently, because their “past life” were different 😅
Not surprised! Even two chips from the same wafer can often have differences as the coatings are not evenly spread out so the inner area will often have better coverage unlike the outer edge chips. Within the digital realm this is not that big a risk as long as the state is still set correctly. Unlike the digital space analog is very tricky! So A to D or D to A chips is an art like trying to grade fine wines!
I miss the days, long ago, when I ordered a TDA1541A-Crown, from Digikey, no less.
That would be nice
Loved the video Gabby! Thank you for taking the time, effort and cost for doing this as it was very educational for me as someone who has an interest in the TDA 1541a dacs. Sadly it has also made some people go onto a buying frenzy as well..lol. I had a friend asking me on one that had a S7 (yeah 7 stamped crowns) chip on Ali i think, and i steered him away from that obvious fake. I do have a couple of holland chips, from the Years 87 and 88. The R1 chip is from the 87 batch and it is very analog sounding, just buttery smooth mids, bass a tad bloated and the highs are less sparkly. The standard 88 chip which is the same as your #15 is quite linear and holographic sounding and the S1 double crown from Y88 is just downright amazing but not far off the Y88 standard, but the mids smooth, yet detailed as well.
Than you for sharing your experience
I have seen the 7 crown one it made me laugh
If you've seen enough of these, you can easily spot the ones that are relabeled.
1. Original chips are labeled just good enough, not perfectly bright.
2. Philips switched to BitStream DACs in the early 1990s, so production date stamps later than, say 1993, are suspicious. So are multiple chips that have the exact same date code.
3. The S1/S2 and crown on real chips were added after production to chips that were outstandingly linear (or something). They should obviously be stamps so the lines aren't as regular as the rest of the labeling; they may also be a different color and they're likely to not be perfectly aligned with the other labeling.
So judging by what you're showing us:
4 and 13: I would have guessed that those are real, even if I wouldn't have known that you personally removed them from CD players. They may have the same date code but 1988 was when Philips was using these in all their CD players. The Philips logos are different between the two and the S1 and crown are obviously stamped on there.
16 also looks legit because the S1 and crown are obviously stamped on there, the date looks correct and there are just enough imperfections in the labeling.
8 has a date code that looks suspicious to me (1996) but could be legit. The color of the crowns is slightly different from the rest but it's darker so that's also suspicious. Maybe it's an older one-crown that got relabeled.
10 and 9: The ink on these ones looks a little too bright and the date code is a little on the late side (and a little bit too identical), so if I would see them on eBay, I wouldn't buy them. Nice that they turned out to be good! I think you were lucky.
6 and 11 look legit. There is just enough variation in the labeling to make me think those are the original labels, and the date codes are in the time when Philips used these for all their CD players.
12, 14, 1 and 7 all have the same date code and the date code is very late so that would make them suspicious to me. 15 and 3 look totally legit; plenty of imperfections in the labeling and the date codes are perfect. Would buy.
2 looks like a fake, the labeling is just a little too perfect and the crowns were obviously put on there at the same time as the rest of the label. The late date code is also highly suspicious.
5: I don't know. It doesn't have the Philips logo and I don't think there were any other factories that made these. Maybe it's a really early one (Philips tended to put less labeling on pre-production chips), but then the 1987 date code is a bit on the late side (they were in full production much earlier than 1987 I think). I would probably give it the benefit of the doubt if I would see it for sale somewhere. But obviously if I would be able to find a 1541A, this would drop out of my grocery cart.
As for your question about whether these are still in production: Philips Semiconductors changed into NXP around 2000 and it's highly unlikely that any of these were still produced even in the late 1990s. I'm into the Philips Digital Compact Cassette and all DCC recorders have BitStream DAC's. And DCC recorders were produced between 1992 and 1996. It's possible that they kept producing the TDA1541(A) to sell them to other manufacturers but I'm fairly certain that they would have discontinued them by the mid-1990s.
Thank you for the in-depth analysis
You have very good knowledge of these chips
I sure learned a few things reading your comment. I really appreciate you taking the time to share your experience.
Do you think that new chips are bing made in China now or maybe Taiwan I bought a few to test as I am curious
I bought 6 from the same batch that have 97 year 1 is outstanding 1 very good 1 not good and 3 average. So where and when you think these were made.
@@Gabster1 It's possible that Philips actually made them long after they stopped using them in their own electronics. Maybe they sold them to OEMs in the late 1990s. But as I said, they didn't use them in any DCC recorders and those were all made between 1992 and 1996.
However everything I said is just based on experience. Though I lived in Philips' home town and worked for them for a while, I don't have any insider information about chip production.
Thanks for doing that! It's nice to see real world tests. I too have no real way to measure so I measure with my ears.
Thank You
Yup your ears are all you need for The TDA1541A Chip
Hi, thanks for sharing this great investigation (kudos, definitely lots of time, money, precision work and mental fuel)
Could you possibly by any chance consider at some point investigate/add the parameters of chip package (top side) temperature (with and without heat sink) and their current consumption to your evaluation table ?
Thanks again for the always great work and sharing it
I took a thermal scan of a few but not all
As they were all similar within a couple degrees
They got to 40 degrees Celsius fast within a min and to around 47 degrees after 10 min
But the consumption is a good idea for next time I do this
Creative video, keep it up, thank you :)
Hi Gabby, I read with interest your new found passion for tda1541a. You wanted to know if there is new production in china but I think the answer is no. Let me explain from my "research". Tda1541A ceased production in 1993/94 as Philips went on to 1 bit or bitstream dac the precusor of modern dacs ( delta sigma ) simply for cost reasons and simplicity of chip manufacture.
To date Philips has the preselected S1. The reappearance of tda1541A is because Marantz which is bought over by Philips wanted to commemorate the aniversary of the introduction of the compact disk player. Philips then tasked TSMC of taiwan to make tda1541A but since the silicon technology had improved Philips decided to also make a special selection for this occasion calling it S2. According to one source 2500 pieces of S2 was shipped to Marantz so the rest was sold to chinese resellers. 😱 I believe these are those that did not pass the grading process or as well as ungraded ones. That said at best the last A version date code would be 1998. Now interestingly S2 was put to use in marantz flagship cdplayer D1. You can see the actual markings, that is the S2 and crowns ARE NOT ALIGNED. Photos are in the Dutch Audio Classics. Date code is 9713. I post links that will give more insights to tda1541A.
I have bought some A version but the first batch looked genuine but underneath is marked "China" date code 9635 batch 11260. In my mind I don't think TSMC had started production in 96 or is this a trail run. I then bought a 2nd. batch but to my disappointment they were non A remarked as A version. So I am thinking would you like to test them. It could be genuine A and non A at the worst as I had made a diy dac pcb using lm317/319 regulators and cs8412, the then popular digital reciever with low jitter. Both A and non A sounded the same to me but of course you have a much better setup. If you are interested,please give your telegram handle so I can send chips to your postal address free of charge. You can keep them if they are good. A pair each to make a differential pair.
www.audialonline.com/blog/tda1541a-and-model-s-usb-part-4-tda1541a-grades-and-series/
cd94blog.wordpress.com/tda1541a-tales-from-the-world-of-make-believe/
www.dutchaudioclassics.nl/marantz_project_d1_d_a_converter/
THANK you 🙏 for all this info that should keep me busy reading all these articles
Very interesting information.
That makes sense as it does look that the stock being sold has a mixture of Quality levels some are excellent and some ok none are bad
I discovered something new since I posted my video some chips that were ok ended up being a lot better after some breaking time I discovered that as I did a second blind test one of my lame ones turned up being good after a couple days it could be moisture issues and needed some time staying hot as they usually run around 47 degrees Celsius
My info can be found on my site here
Feel free to reach out gabster.ca/Contacts/
I can also send you a couple boards in exchange to try
Thanks again
Gaby
Have sent you email.👌
my first chip was purchased from Speakermango from Ebay. sounded really nice and was$100 shipped. after this video, i purchased an S1 chip from a fella from Ukraine for $300. He claimed to remove it from a player (with photo). could be bias (mine, not the dac) but i think the chip sounds clearly better (vocal presence, timbre and detail). very happy. now questioning if improving the 5v ps with supercaps or using ians snooty clocks could bring more love. using ians linear pi and a 15v ifi. which is the dac ps?
The TDa chip is the best place to invest in
The big caps will help but minor amount you will get a slightly firmer controlled Bass just careful switching TDa chips not to put it in reverse or off by a pin or reversed polarity
Only use you S1 when things are stable these chips are hard to find
I've actually had good luck purchasing DAC chips on eBay.
Hey I want to know what you think of the song. All the walls by erotheim, because it sounds amazing and I bet when I get done with my speakers, it'll be absolutely incredible, but I would like to see what you think of it
Between your Terminator DAC and this TDA1541 DAC with Ian Canada stacks, which one is best to you?
The Imaging Air and Musicality are far better on the TD1 Low and High frequencies detail is the same on both, Midrange Varies it depends on The TDA1541A Chip.
On a very good Chip Vocals Mids are equal or better. On a average Chip very close.
With a low grade TDA Chip everything suffers.
95% of the Chips selling out there are about average
Very brave and surely a labour of love! Well done and thanks! I won't suggest paralleling chips then, seeing that you've already fried a few of them! However, do you plan on ever producing a balanced dac, i.e. a truly balanced design, in the digital domain? Also, are you planning to experiment with transformers in the I/V stage?
I already did a balanced Dac with my TD1 boards but it is very dependent on the TDA1541A chip quality and matching
But plan on spending more time again once I have a bit more time
I have tried both Ivan’s top Transformers and they are well respected transformers.
To my ears in this TDA application The OPA 861
In a single no feed-back design produced a better sound.
@@Gabster1 Excellent! Thanks! I just watched it. It's the episode uploaded 3 weeks ago right?
I didn't know you also had a website.
Also I wasn't aware that there were Chinese-made chips out there. Are these just unlicensed clones? Do they actually exist or it is just a false rumour? With all the fuss about the TDA1541 going on for the last 30 or so years, one would think Philips would try to release another batch on the market. One has to wonder why they haven't. Even worse, one wonders why they have not come up with an improved version. After all the TDA1541 is a 16bit version of the TDA1540.
This was Released on Patreon a week ago. You definitely have a point. I almost feel like reaching out to Philips. I sometimes dream of a TdA1541G Then I wake up :)
@@Gabster1 It does happen all the time though. If not Philips, maybe some other company could buy the rights to reproduce (not the technology rights) and simply re-release it under their own name. Of course Phlips might not want to sell the DEM technology, if the patents on it have not expired, but that is another matter.
I would have felt better about this testing if the process had been done twice, with a 3rd party randomly re sequencing the DACs at a mid point. The interesting point here is how far apart the two evaluations of the same part are - this is a quality measure of the audience. Also, it would be nice to store these parts on a grounded conductive surface (tin foil ?) and handled with some kind of ESD management. None of the latter will affect the sound, but may affect overall life.
I interpret the point of this video to be that the measurements are of low value, but listening can identify superior- and inferior-sounding devices. Yet, as another commenter said, the reliability of subjective evaluation is suspect. I propose an experiment to determine how real your evaluations are:
Remove from your evaluation population those chips that measure bad. Then reevaluate some from the best and some from worst group through a new blind test with the tape and numbering redone. You should even repeat this experiment a third time, renumbering again. Can you reliably tell the best group from the worst group by listening?
you must be reading my mind
I actually did something similar yesterday.
I used 4 of the best ones 2 of the not so good ones and added 4 new ones.
The aim was to figure out how the 4 new ones tested again covered with tape blind test
I repicked the best ones easily I was off by one of the not so good ones I rated it average
the 4 new ones that are Identical from the same year one was very good one good one average and one below average.
The moral of the story is, the very good ones tend to stand out and are easy to rate.
The other thing is that there are variations in Quality between chips even the ones that are Identical
of all the S1 chips they were all very good with minor variations.
ok... but at the end of the story we are talkin about a very old chip with serious limits in frequency and resolution especially when considering the every day more and more common specs of file available from streaming... ok I know it is still plenty of 16/44 but... does it still make sense to spend that amount of money on something that born already old?
I have a PCM1794 dac thats 15 years old and the chips are much older. It sounds as good if not better than the new Gustard A26, not as hyper focused detailed but very close, more musical, better sound stage. I was thinking the same thing "after 15 years its time to upgrade", now im thinking why did I bother, So yes, the old chips still sound very good, and in some ways worse is better
May be the point is that whole technical developement is driven by sells and sells re driven by popularity and popularity needs portability and in effect miniaturization . This is the reason why new chips were developed with small energy consumption and dimensions and due to 90% earphone reproduction and low prize of smartphones I would expect no better then 1541 quality.
. I used few CD players - portable rised my doubts , original japanese Onkyo with other DAC found just no problem but I wasn't fully amazed wuth any CD, and when I bought Bang Olufsen with TDA 1541 since 10 years I do not even think about any issue each CD amazes in total. If something could be 1% better it needs be already analog
Do any of them work on a single 5V supply (pin 28) only? Is supply current correct?
Apparently none of them, since they are all able to run in simultaneous mode. Based on this, all the chips are legit.
None Do I even tried running the -5v only in case they did reverse the pins internally, also all need the 15v.
They have a average TDA consumption and a Thermal scan showed a average of 45 degrees Celsius which is typical.
they are Legit or they figured out a way to reproduce and manufacture the chip. As @mark4571 mentioned they all run in Sim mode.
Interesting that it is possible to hear individual difference between chips!👍
That begs the question if it is detectable sonically differences between between all types of different chips?
We know it is sonically differences is in the "simple" OP-amps so logical their have to be POTENTIALLY as great or greater differences between more complex chips.. Or?
I mean if there is individual differences between the today produced ESS/AKM DAC chips..
(Yes, some may say that we have better and more consistent production lines. But remember that manufacturers is every day pushing them self to make the transistor smaller and smaller and struggle probably likewise as they did 1981.. And they also have new issues that comes with that.)
Yes, I know that those chips is far from easy to swap out, but the question still persist.😂
OK if we conclude that ESS/AKM chips individually sounds different.
Then set aside that other components on the board (may/probably) also influence the sound quality.
Then there should be a sonically differences between two individual complete assembled DACs that are the same (model and brand) and coming from the same production line!🎉🤔
That then this little detail opens up questions, that when you buy for example a DAC. Then order two and return the one that is performing not that good as the other one??😅😂❤🤔
Thanks for the video! As seen it started a train of thought's.😂🎉
I like your Thinking, you have a god point or shall we say good thinking. :)🙂
While it is correct to notice that production variations, even among brand new genuine, chips is a very real thing. We have to appreciate that those variations will be remarkably small and will reduce in second and third production runs as the die and substrate architecture is improved.
The far greater variable in listening tests is the human ear. Simply sitting in a slightly different (and I mean in millimetres) listening position can result in remarkable shifts in perceived audio quality. We must always be aware that we have two ears, listening from slightly different positions in the sound field and there are going to be cancellation and summing effects because of that.
Here's a video that demonstrates the most common of these problems quite nicely ... ua-cam.com/video/T8bmASpUeh0/v-deo.htmlsi=R_RW7knxVBmGDUeI
The problem is that we cannot cleanly eliminate the many many variables in listening tests so subjective impressions will always be both highly variable and suspect.
I have one of those chips in an old Philips CD player that doesn't work, of course. Are they still competitive with modern chips?
Then you are lucky yes they are even better
@@Gabster1 Oh, great I suppose if I can do something with it.
Hi Gabby, could you share the chip you bought that is decent and the price is attractive (and from where) thanks
Learnt from the video that you are selling the chips, please advise price . Thanks
Sadly most sold out as I was very careful where to buy from
I avoided buyers that seem to have hundred of them. That shows as most except 2 were actually bad chips.
Chip rolling is a thing. Keep an eye out for the white ceramic gold pin versions , these are the very best. Ceramic body chips are superior.
I have only seen a photo of a non A white one is there a White Ceramic A version any links you know of
@@Gabster1 Sorry no, i have link for white ceramic gold pinned TDA1540. There so very rare. Tbh i have never seen equivalent TDA1541 but im sure they exist.
If chip is excessivelly overheated during desoldering, it will be damaged. So it is better to desolder it yourself.
Totally agree unless you know the person desoldering and if they have good equipment and expertise to do it
With my Hako desoldering tool they come out easy
Where can I find the electrical measurements on the different chips? The SQ differences must be visual in the measurements, or a there must be trend for what sounds good or bad.
What measurements would you use to determine sound quality?
@@Pete.across.the.street That is the question, but to determine that we need measurements and listening feedback. Here we only got the subjective feedback. I would say the more measurement the better. But it looks like the harmonic response from single and multitone are where you can start to see differences in the system. How does the "forest of trees" look like in comparison etc. But the more date the better, then we hopefully can extrapolate what Gabster thinks is a good or excellent sounding DAC.
@@mk1classic you can't measure sound quality off an electrical signal. You can measure differences in the signal, whether it's good or not, you have to listen.
Generally a search for the part number plus "Data Sheet" will find the manufacturer's specifications for most chips.
For example: "TDA1541 data sheet" Will find a PDF file with detailed information about the chip, including pin functions, timing, limits and internal details.
@@Pete.across.the.street
The main ones are frequency response, harmonic distortion, intermodulation distortion and crosstalk.
If you have a chip with 10hz to 20khz frequency response within 1db, less than 0.1% distortion and channel isolation greater than 40db... it is almost certain to sound fully transparent and detailed.
According to the Dutch Audio Classics website, all real S2s (double crown) are from the final 1997 Taiwan run and did not exist before then. None have the stripe on the left. They all have "TAI WAN" where the stripe normally appears. It seems many chips (both real and fake) have been restamped with a double crown, which is not surprising. The TDA1541A was produced in Belgium/Holland (1988-91) and the USA (1991-94). Unless one actually pulls one from an old piece of gear, it's a crapshoot. There are just too many restamps and fakes out there.
Very well said there is a lot of crown stamping
Some are plain chips and can sound different from one to the other. The moral of the story is to listen 👂
How can your TDA1541A S2 dual crowns be S2 if they are not even made in Taiwan and not have Taiwan stamped on them .
My thought exactly. Looks like a fake. I have the real S2, and S1, R1 and normal. To my ear the S1, R1 and standard all sound very similar. The S2 is easy to differentiate. It is just much more smooth and analogue sounding. Not everyone likes it. I think it is perfect and feel fortunate to have one.
as I mentioned in the Video The crowns are likely fake but interesting that one was Perfect and one was almost bad.
@@user-od9iz9cv1w I like the last batches of S2 with N1 on the description receipt they sound very similar to S2 crowns though it's pretty hard to differentiate S2's especially after DEM clocking.
@@Gabster1 Yes I know . Though I thought you might of known that S2 crowns only come with the Taiwan print and dual crowns printed on them .
@@Gabster1 There are also mystery Chinese sold and most likely Taiwanese made TDA1541A S2 with a bizarre thick printing of TDA1541A S2 and are considered to be like Crowns . They most likely kept on printing them after Philips cancelled their manufacturing there . My friend bought a few and they seemed pretty good and they were definitely not shaved and rebadged older TDA1541A or fakes . They were identical to the TDA1541A S2 in appearance except for the bizarre thick printing in a bronze gold type colour .
First time I see someone taking the time and spending a lot of money and resources to help us see the big picture. Fascinating that a lot of chips from the same batch can sound differently. Todays Chips may have better tolerances and variations but I am sure small differences exist.
Does the possibly fakes run in simultanous Mode?
They all did run in sim mode but not sure if I have any bad fakes
Maybe they can reproduce good ones now no Idea if they are that is still my unanswered question maybe someone knows and can tell us
@@Gabster1 I believe that too, the pins looks incredebly new on this China Chips.But if they work well why not.
Gabby?
Chips 9 and 10 are identical by label, so now I am wondering how this could be? You rate #10 chip better in SQ! Hmmmm 🤔
every chip was tested after manufacture, because you couldn't make chips that accurately in the early days. and they were divided into groups, because otherwise you would have to use too many different codes. It is therefore very possible that there are differences within the group.
@@jacquesw8725 Yes, remember back in the days and presumably today also.
We know that some computer CPU chips is "better" and can handle over clocking better that shows us that there is individual differences between chips coming out if the same production line..
So there is never 100% the same. And it is more obvious on CPU chips when we can find and test the individual maximum clockspeed under stress test conditions.
Hope this helps a little about the understanding. 👍
@@AmazonasBiotop Most people do not relize that DAC chip is digital but only in part. Of course in optimum case digital is always repeatable .But final stage must be analog and in analog 100% repeatability is never the case - we may think so but our measurements in compare to to hearing sense. are deeply not accurate
@@Mikexception Yes, then it is yet another reason to buy always two of the exact same HiFi component.
(Or compare two identical individual sets somehow, to determine which to choose.)
When the analog side is also not 100% the same.
And return the lesser good sounding.. 🤔👍
@@AmazonasBiotop Choice is up to us- I would not categorise because bad digital is bad and bad analog is bad. And good ones are good.
But in theory I expect digital cannot beat analog because is created from analog and finaly returns to analog. Reality is as is and in say 85% important is used final reproducirng system
You have accent , where are you come from ? It isn't American 😮😮
And that matters because??
@@Douglas_Blake_579MAG Trump 😅