If you are using digital media, you’ll use a DAC. Depending on how you store your music, this will have a bearing on how you interface this digital data out. Every format has its pros and cons, so let’s have a look here.

For digital media, to my knowledge, you have 3 choices today. SPDIF, USB and Firewire.

SPDIF is the de facto standard digital output. It’s available on CD/DVD players as “Digital Output”. Many media players like the Squeezebox has this too, as well as many PC soundcards. SPDIF even has an optical version, Toslink. A sonic upgrade can be experienced instantly if you ditch the lousy RCA socket and opt for BNC. Why oh why does everyone still use RCA then? For convenience rather than sonic reasons!

A major problem with SPDIF is that the SPDIF receiver in your DAC got to retrieve the clock signal from the incoming SPDIF. And this is a major problem. The older but still very popular obsolete CS8414 has a typical 200ps while the modern DIR9001 manages much better at 50ps. Irrespective of what comes after this, the fact that the recovered clock isn’t optimum will have a huge bearing on sonics later.

USB DAC is really popular these days. The most popular implementation is Texas Instrument’s PCM27xx chips. These are pretty good “as is” but still a lot can be improved on. Again, there is the question of how is the clock recovered?

There are better implementations of USB which do not use TI’s generic PCM27xx chips but I haven’t tried these yet.

Firewire is said to be the best but far from readily available. And too expensive at this point.

Which interface to use also depends on what is your source. If you are using CDs, SPDIF is most logical solution. If you store your music on hard disk, then USB or I2S via a SD card player is viable. The above briefly describes the pros and cons while the convenience of which format is really up to you. If you don’t mind a little trouble, go for the best sound!

My current favourite, however, isn’t any of the above. It’s via I2S! Provided you have a very good I2S source!

I started dabbling with I2S when I tapped out this signal from my old Marantz CD63 way back in 2003. The CDM12 transport isn’t that great and though the sound was already a huge improvement over SPDIF, there is still something lacking there. But hey, if you could ditch SPDIF, try this!

Later I played around with USB-I2S as well as USB-I2S-EIAJ in USB Monica. As good as all these sound, it’s still not a pure I2S solution. The interface is still via USB. You could say the weak link is how the PCM27xx chips derive the I2S out. I hope to have the resources to tackle a better way of doing USB one day…

Anyway, my current favourite digital player (The Digital Turntable) has, in my mind, the most direct route to I2S. A DSP chip reads WAV file data from a non-moving media (the SD card), then outputs I2S signal which is fed to a CS8405 chip to generate SPDIF. Both the DSP chip and CS8405 is clocked by a low jitter clock.

Since a good quality I2S is available, feeding directly to an I2S DAC chip (or through Black Crow for an EIAJ DAC chip) is all we need! Replacing the clock with a lower jitter implementation (like Kwak clock below or the plug-n-play Tent XO module) and I assure you your jaw will drop listening to music through this combo!

Okay, okay, enough about all this. So these are what we have here.

Above pics show the 3 implementation, coupled to mojo. They’ll couple nicely to mojo2 as well. mojo/mojo2 PCB has 6 PCB holes which couple nicely to the I2S-EIAJ, USB-EIAJ and SPDIF-EIAJ modules.

4 are EIAJ signals while the right 2 are your power connections to power the modules.

So instead of forcing you to buy a USB Monica or a SPDIF Monica, you now have choices and since this is in module form, you could play around slowly.

Hey, if there’s sufficient interest, how about an EIAJ selector to select from multiple inputs?

The post Let’s pay some attention to our Digital Interface appeared first on diyparadise.

Once upon a time, in a galaxy far far away, lived a Diyer by the name of CY Liew (whom by the end of this article we should give you a nickname!) who can’t decide between the TDA1545, TDA1543 and its many parallel permutations (or mutations). He could, of course, build many many DACs and choose one (what I would do) but he felt that such an approach is sooooooo “unmacho”, soooooo 20th century. So our friend here who, by the way, wears a soldering gun like a cowboy wears his gun, did the most logical thing any Diyer would like to do — build all of them in one chassis and switch to your hearts content.

And so begins the story… *cue music intro*

First of all, let me explain the features. The 4 switches at the right are for, from Left to Right, TDA1545, TDA1543 with active I/V conversion thru OPA627 opamp, then 4xTDA1543 and lastly 8xTDA1543. Cool moi?

Switch at center is power switch. At the far left, this is where it gets better… battery! Yeah our hero here can’t decide which he likes so Mr-Indecisive did the most manly thing there is. Instead of talking about the sound, he built all his doubts into this little box and listen it for himself. Now this is a MAN.

So there you have it! An out-of-this-world, whatchama-call-it, switch-as-you-please, whole-loads-of-fun DAC.

Also known as User-Selectable TDA1545, 4xTDA1543, 8xTDA1543 with passive I/V conversion and TDA1543 with active I/V conversion with both Mains AC and Battery Powered Option Non-Oversampling Digital Analog Converter(s). Phew!

Colloquailly also known as “Gua Tak Boleh Tahan” DAC! [insert foreign phrase of your language meaning “out-of-this-world cool until i’m almost stuck to the wall reading about this”.]

Anyway, one fine rainy Saturday morning, we had a listen to all the DACs in my humble system. It was so easy to select between DACs! All in all, there is a huge difference between them DACs. I prefer the 1545 in my system while CY “DAC Man” Liew prefers the single TDA1543 with OPA627 I/V conversion in his.

As you can see, we Diyers have lots of fun!

So folks, let’s give DAC-Man a round of applause for his wonderful engineering! DAC-Man welcomes praises at liew dot cy at gmail dot com.

PS: Err, DAC-Man, can we have an amp that switches between 45, 2A3 and 300B? Sorry, can’t resist!

The post A Deck of DACs appeared first on diyparadise.

Finally! The long-awaited, much-anticipated, long-overdued , oft-asked, finally see daylight… *drum roll please* Yup, in collaboration with Calv Acoustic Labs, DIY Paradise brings you… Non-Oversampling Digital Analog Converter Kits! (if you are cheap, use your business card box to house the dac) (i’m even cheaper. no box is best box!) *see below*

Half a kit, full kit, what the difference?
Yup folks. At offer here is DAC full kits. You get everything you need to build the part. Well not "full" in the strictest sense of the word, but "full" enough. You still need to get your RCA sockets, wires, casing, battery, solder, calories, sweat… The power supply side is not supplied, as it is diyparadise’s aim to promote going battery (though the PCB could be wired for separate power supplies). Remember, have battery, can travel. Besides after you built your DAC, bring your battery powered DAC and a mop to your hifi dealer, so that after listening to your DAC, they can mop up their drool! So here you go.

The TDA1543 DAC kit page.

The TDA1545 DAC kit page. DISCONTINUED

"Monica 2" – The TDA1545 DAC kit with asynch reclocking.

Update Dec 4, 2004: All you need to know about these DAC kits.

Testing, testing… Good!

Palm Top Dac, Pocket DAC, Mini DAC…

Perhaps you prefer a more conventional "box"? With power switch, LED at the front?
Shown here is 4.5Ah battery that lasts ~40 hours with this DAC. Sub for a more petite batt of your liking.

You can easily take out the guts and modify to your heart’s content!

www.diyparadise.com

The post DAC Kit appeared first on diyparadise.

Well, almost…

History of the DAC
Idea was conceived while watching some horrible B-grade movie when thoughts suddenly drifted to all those ideas/thoughts/experience “stolen” from everyone/everywhere on the Great World Wide Web. Then Head Honcho of Calv Acoustic Lab lended a GIGANTIC helping hand by laying out the PCB design. Needless to say, never bothered to finish that stupid movie.

Pre-requisite to build the DAC
As long as you can solder properly, know how to use a multimeter, know how to work logically and of course, relax with your favourite beverage (milk, coffee, tea, toddy, stout, beer, sake…), you are ready!

Work logically : Solder in all components except for CS8412 and TDA154x. Power up and measure that power supply voltages are correct. Then in goes last 2 chips. Power up again to make sure that power supply voltages are still correct. With TDA1545, check that one end of R18 is at 1/2Vdd while the other end is at some other voltage (~0.8V in mind). Then hook it up.

Don’t Worry, Be Happy
First of all, a very important point to know about these non-os DAC kits is that there is no gain stage involved. Thus the output voltage you get is just the pure voltage as generated across the current-to-voltage resistor. In other words, passive I/V conversion.

Because of this wonderful reason, these DAC kits will not put out 2V output like most conventional CDPs. If you are using single DAC chips, it’s more like 0.5-0.6V. If you parallel the chips ala 4xTDA1543 or 8xTDA1543, then the output level is higher. Remember, these are current output DACs so when you parallel the DACs, the current adds up and you get more current, hence more voltage output. 4xTDA1543 gets you to almost 2V while 8xTDA1543 gets you more than 2V.

If your system is low gain and needs all the swing from your DAC stage, do NOT be alarmed that you need to crank up your volume from 10 o’clock till 1 o’clock to get back the same listening level. There is nothing wrong going on here. It’s just that the DAC output voltage is lower, that’s all. It’s no different like connecting a lower sensitivity speakers at the output. You just need to crank up the volume, that’s all.

If you feel that it is important to listen at 10 o’clock as opposed to 1 o’clock, then feel free to add in a gain stage. This is where I’m stubborn, obstinate to the point of being a pain in the you-know-where. The decision to eschew convention is partly based on my experience of haven’t heard any gain stage at the DAC level work so brilliantly without destroying the purity of the DAC stage.

I have listened to opamp stage (even using OPA627), tube stage (using the ubiquitous 6922 variants etc) blah blah blah… but never liked its “flavourings”. Each time, my ears tell me that no gain stage is still better. Of all the gain stages I have heard (okay, I haven’t heard that many), the one most faithful to the DAC is the 12B4-based gain stage. With its mu of ~6, it’s just nice! However, due to its very microphonic nature, again, it’s hard to recommend.

Briefly pros and cons of a gain stage
Pros: more output, better dynamics, better bass.
Cons: loss of “sonic purity”, sometimes loss of micro detail as well.
So if you feel like you need a gain stage, you are on your own! *diabolical evil laughter ensues…*

Troubleshooting the DAC
This is the fun part. Okay, there is this Mr Murphy and his darn wonderful Law that makes this page necessary. After soldering everything and verifying all voltages are correct, here are some troubleshooting tips.

Problem : No sound. LED doesn’t light up when you pressed “Play” on your CDP.
The LED is just an indicator. More importantly, check pin 28 of CS8412 (the pin closest to LED). It should read ~5V when no music is playing and ~0V when you pressed “Play”. If it is ~0V when you pressed “Play”, then chances are you got the LED orientation wrong. If you still get ~5V, this means CS8412 was unable to “lock” to the incoming SPDIF signal from your CD transport. Check that you got the 75ohm, 100nF caps at the input wired correctly. Check also the PLL filter (470ohm, 220nF, 3300pF) are all placed correctly.

Also, measure Frequency at pins 1/2/3 of TDA154x pins. (Measure at the leads of the chip!) All of them should read something when music is playing. Pin 2 should read ~44.1kHz. If you could get some frequency measurement at these pins, this means all the digital signal is reaching TDA154x, then it’s highly probable that you have a dead TDA154x chip. If not, you could have a broken connection somewhere between CS8412 and TDA154x.

Problem : Distorted sound.
Check that you got the right I/V and Ref (R18) resistors in!

Problem : Got sound but I’m hearing things I’ve never heard before! Is there another backup singer on this track?
This is spooky. Call an exorcist!

Problem : My friends love the sound and wouldn’t return my DAC!
Those selfish bastards! Get new friends and new DAC.

Problem : My wife loves the sound! She said this diyparadise guy is her new hero. I’m jealous!
Lie to her that I look like a toad.

Problem : I can’t decide between mahogany wood, ABS plastic, or stainless steel chassis.
Ask your wife.

Problem : My dog/cat runs away when I play your DAC!
Are you using a super tweeter? If your system capable of up to 44kHz? Maybe your pets are irritated by this super-sonic noise. Hey, at least they can’t ruin your audio gear! They are also leaving you to listen in peace. So what’s the problem? Do you also experience less mosquito bites? I heard that mosquitoes hate this frequency too.

What else?
There’s certainly room for improvement! You could start a search for your “perfect” gain stage. You could build a super clock for asynchronous clocking between CS8412 and TDA154x. Details will be available soon.

Remember, this DAC kit is meant to give you a start in the wonderful sonic world of non-os. Now that I’ve got your attention, there are certainly many steps ahead in this wonderful journey.

Enjoy!

www.diyparadise.com

The post All you ever need to know about DAC kits appeared first on diyparadise.

Match Made in Heaven: Batteries and Non-OS DACs

2 things make my knees wobble these days. Monica Bellucci and Battery Powered DACs.

“Hi yeo, Can I come over your place to listen to your battery powered DAC…”

Battery Powered DACs… Ahh…. Like Monica, How can anyone resist?

It makes lots of sense to battery power your DAC. Think of your signal chain. First you get digital ones-zeroes-ones-zeroes from reading the CD, then this digital information is fed to various processing IC till it finally reaches your Digital Analog Converter IC. It is from here that you finally get your analog signal, which is then fed to the preamp, power amp and so on.

Now, if you think about it, analog signal first appears at the output of the DAC, and it is low level, usually in milivolt range. At this level (1 milivolt is 0.001V), your analog signal is most susceptible to pollution – power supply noise. And this is where our old friend, the plain vanilla battery comes in. Of course, parallely speaking, battery powered phono preamps should be good too, right? ; )

Batteries, by virtue of it already being a constant DC voltage, should be a perfect power source right? Err, not quite actually. Read on.

Here’s the schematic. TDA1545 (and TDA1545A) accepts only EIAJ input format (I learned this the hard way), and not I2S so CS8412 needs to be configured to output EIAJ. CS8412 datasheet says it needs to be configured for Mode 5, thus only M2 and M0 are tied to 5V. The string of diodes is to give 2/3*Vcc reference to L/R output and Reference pin. Idea given by Thorsten Loesch on diyaudio.com/forums. You could use 3 pcs of AA cells as well, if you wish.

As you can see, I got lazy in the power supply department, and forgo the usage of shunt regulators like TL431. Hey, at least I used an inductor okay? Please note that M2 and M0 are tied to VD+ and not VA+.

Err, I just remembered I didn’t mention anything about the sonics of battery power. Well, apart from the OBVIOUS lower noise floor, music flows easier, more relaxed, more natural. One of the biggest benefits of battery power is a smoooooooth mids+highs, so distinct that whenever you listen to conventional AC-DC powered DACs, you’ll have “HARSH HIGHS” written all over the place (your friends’ place). But you could easily screw up! One of the DACs I built, I forgot to wire in a capacitor before the regulators. If you look at the schematic, that’s the capacitor just closest to the battery. Without this battery, I get a slight veil, particularly noticeable in the midrange. With this cap in, the veil is gone! But why oh why? Aren’t batteries perfect DC voltage sources?

Apparently, at static conditions, batteries are perfect but music is ANYTHING but static! Under dynamic conditions, probably chemical reaction inside the battery isn’t fast enough to cope. This plot here, by DIYer ALW, is very interesting.

Oh well, I guess battery + caps + regulator will have better noise plot?

Ok! Ok! The sound! TDA1543 should be familiar to you, unless you haven’t built anything yet. (Shame! Shame!)
Aside from the common “darkness” associated with non-os DAC, here’s a (very) brief summary of general impressions. All these with batteries of course.
Single TDA1543 – Good midrange. Good highs but lumpy bass.
8x TDA1543 – Midrange loses out to Single 1543. Midrange has less prescence. But highs and bass better than single 1543.
1x TDA1545 – Best Midrange here. Best overall sonic. Makes you scream “Mama Mia”! And this is the DAC I’m going to share with Monica Bellucci.

Arrivederci!

PS: If you have two left thumbs, but still want to have a taste of… “mama mia”, contact me.

The post Match Made in Heaven: Batteries and Non-OS DACs appeared first on diyparadise.

Non-Oversampling DAC saga continues with…

8x TDA1543

First of all, my apologies for so few updates on this website. I’ve been up to my neck with my day job, and spending most of my time living out of a suitcase. I don’t have to tell you this isn’t my idea of fun at all, or do I!? That said, in the short time that I had access to some solder fumes, I’ve revisited the non-os dac as I feel it’s the current weak link in the system.

What do you do if you think your DAC is the weakest link? Build another DAC? Play around with the circuitry? Add a tube output stage? Or breakdown and plough big money for some commercial design? Fortunately or unfortunately, yours truly is also a cheapskate so the last option will always be the LAST option.

Of particular interest to me is choice number one and the TDA1541 naturally calls for attention. Also, I just heard a good implementation using the PCM63. A wonderful chip too! But the added complexity, power supply requirement and additional gain stage… make this more than a weekend affair. Yes, more stuffs to play but with so little time before I hit the road again, I’m better off with something easier…

Carsten from Germany pointed me to Doede Douma’s excellent website, where he described in good detail his 8xTDA1543 non-os dac. Now Mr Douma even claims that this DAC sounds like SACD! Wait a minute, 8 pieces of dirt cheap DAC chips (hey! it rhymes) sound like SACD?! Woo hoo! This I got to try!

The circuit is so simple you can whip up in half an hour but as with all DIY projects, it’s a good idea to factor in 3x that amount, for time wasted looking for parts, soldering/desoldring a wrong orientated part, shit-happenings, Murphy’s Law… get the idea? And sorry, I’m not going to draw the schematic. It’s so simple it’s better to describe in words.

3 I2S input lines for pins 1/2/3.

Power supply connections across pins 4/5. Pin 4 is ground.

Left output through pin 6, with 249 ohm resistor to ground.

A pot to pin 7.

Right output through pin 8, with 249 ohm resistor to ground.

Mr Douma suggests a 2kohm pot for pin 7 but I can’t find this at home so I salvaged a 10kohm pot and paralleled a 1kohm resistor. In the end, settled for about 150ohm on this pin. For I/V conversion, Mr Douma suggests 270ohm but I could only find 249ohm, so just use it. And that’s all to it!

Tower of music! The I/V resistors are all mounted underneath for shorter signal path. The power supply comes in through those ferrite beads. Those components on the right were a failed attempt with the TDA1545 DAC. Will revisit this some day… Those on the left is the single chip TDA1543 DAC.

 

 

Though I took Mr Douma’s advice to leave some air gap in between DAC chips, I didn’t use a heatsink. Waily has built this DAC too and stressed the importance of keeping this DAC cool but I wanted to live on the edge! (Actually I’m lazy.) In this arrangement, I could only play for half an hour before the DAC starts to distort. Will HAVE to incorporate some form of cooling.

 

Okay, how does it sound? Way way better than the single chip version! I remember trying 4xTDA1543 a longgggg time ago and it didn’t appeal to me, so much so that I simply prefered the single chip but this time, with so many changes to my system, 8x smokes! Forget about “one chip to rule them all”, eight is the magic number here. And the strangest thing to me is that both Waily and I don’t notice any lackings in the highs at all. This despite the sin(x)/x mumbo jumbo and abscence of the parallel resonant filter circuit so crucial for the single chip version. Huh? Maybe we need to dig out those dastard ear wax… Never mind.

 

 

But improvement in sonics is a big leap. It simply sounds more fuller and more bodied. It images better too. Now Waily has found that choice of DC blocking cap to be very very critical to the final sound. We both like the Cerafines for a more natural/musical presentation but Waily has found the non-polarized Black Gates to be even better. I guess you know what I’ll do one day… Here, read Waily’s mail to me. His enthusiasm is infectious!

> Hi Yeo,

>

> Just some updates from me:-)…

> Tried out the dddac 1543, 8 pcs tda1543 piggyback cct.

> Switched on power and fine tune the Vref to 3.85V

> according to the site. Then MUSIC…

> Headroom:- Plenty of gain and it solves my needs of

> another gain stage:-).

> Smoothness,naturality:- Lost to single chip version,

> but it’s dynamic and bit hi-fi sound:-(

> Distortion:-Wow, everything improves, the slight

> roughness of single dac gone completly, clarity and

> definition is another level that single dac can’t

> attain.

> High frequency handling: This dac wins hands down and

> all those highs is MAGIC! guess what he mentioned that

> it is close to SACD is not unfounded, nevertheless I

> haven’t really got the real chance to listen to SACD

> setup so I can’t really compare, but the high is damn

> good!

> Low frequency handling:- It wins again and the bass is

> much tighter and register better.

> Mids:- Lose out to single dac as the voice is a bit

> thinner and not as sultry as before.

> Soundstaging and seperation:- It wins again against

> the single dac, now every instrument can be heard

> individually much cleaner and won’t muddied up when

> singer voice is in even in complex passage.

> Soundstaging is wonderful, I am totally immersed in

> the music and much improves over the single dac.

>

> Summarised, it is a bit HI-FI sound but it clearly has

> more advantage over the single dac, except is does not

> feel very natural and smooth over the predessor.

>

> The dac’s last only for 10 mins as overheating

> occurs:-(!!!, Found out that my regulator

> overheats(7808), quickly whip out DIY large heat sink

> over the regulator and problem is solved..(have to

> takecare the extra current of the 8 pcs tda1543). Also

> tried out the stacked dacs without cooling, no problem

> except realised that the Vref will drift once the

> temperature increase. Whipped out another tower with

> some stupid looking antenna like heat sink over the

> dacs, and it works better as the Vref will not drift

> much..Again these dacs really can cook HOT!!!

>

> After enough listening, still found that it lacks the

> smoothness, intimacy sound of single dac…!@#@#%#*

> looks like I have to do something, agak agak my

> coupling cap, the BG PK has something to do with it(no

> doubt I prefer BG in single dac, I guess a new

> matching caps has to be in in order to match

> perfectly).

>

> And in the cerafines, $#@*&!! ha ha now this dac is

> shinning!!At last, the smoothness is back plus all the

> plus point of the 8 dacs..Now it sounds not Hi-Fi

> anymore and I really have to let the single dac to

> go:-( as this dacs is way much better(IMHO and YMMV).

> It is very relax to listen to and it brings the music

> to another level that I may put a full stop to this

> dac selection/mods:-)

>

> No doubt the single dac is good! but this 8 dacs can

> proves it’s authority and improvement over it.Guess

> what will happens with 16/32 dacs:-)..

>

> Going to try using SLA battery to drives the dac once

> I got it…Must really see why people keeps using

> battery as power supplies over the normal heavy

> regulated supply..

>

> Pics is attached..

> Yeo, try it and see how this dacs goes in your system,

> hopefully u will like it too..:-)..

>

> Regards,

> waily.

 

Waily’s DAC.

As you can see, all the credit goes to Mr. Douma and Waily for their excellent work! I’m just a “reporter” here. : )

Now Waily has moved on. He’s now using batteries to power his DAC and absolutely love it! One day, one day…

 

The post 8x TDA1543 Non-Oversampling DAC appeared first on diyparadise.

]]>

RM20 Non-Oversampling DAC project

Yeah! Believe it or not!

Heard all the hoo-haa over non-oversampling or zero-oversampling (what a mouthful) DACs? Would like to try one? For less than RM20? Read on!

If you have read the other article on this topic, you’ll know where to start. I’ll put more details here.

This the SAA7345 decoder in the CD63. Pay attention to 3 lines here, namely SCLK, WCLK and DATA (pins 21, 20 and 19 respectively).

Digital filter and dac in the CD63. Note that the 3 lines from SAA7345, each go thru a resistor before terminating at at pins 8,9, & 10 of SM58728.

Here is pinout of TDA1543. What you need to do is wire SCLK to BCLK, WCLK to WS and DATA to DATA pin. Yup, just 3 wires. Easy huh?

What else do you need? Obviously power supply and output connections.

I tapped the supply from capacitor C803. It’s about 20V here. This involves taking out the player PCB of the CD63. Hope you are comfortable with this. ; )

Know where to start? First wire the power supply first. Bring 2 wires out from +ve and -ve terminals of C803. While you have the player PCB out, desolder one end of the resistors RD11-13. Desolder the end nearer to the SAA7345. Reason for desoldering it is to lift out the leads slightly higher. Then solder it back. Without interrupting its present function, we now can solder wires on the “lifted resistor lead”.

Put the player PCB back in and solder the wires of the power supply to the 7808. Note that I drew LT1085 in the schematics. That’s because I used SwitcherCad to draw this circuit. Feel free to use any regulator of choice. With 8V, the TDA1543 is running very hot so you need heatsinks on them. Once you have wired up the regulator, you can start work on the signal lines. As the 3 I2S pins from the SAA7345 are only capable of 1mA current each, so it’s sensible to keep this wire as short as possible (mine is about 5cm from the resistors). Tin the “lifted” resistor ends of RD11-13 with some solder first. Then solder the 3 wires from TDA1543 to these ends, paying attention which goes to which. Refer schematics.

Output is straight forward. Run 2 line from Analog R/L channel and a third Ground line to output RCA jacks. I intend to drill holes on my PCB to accomodate these sockets thus I have a choice of 2 different types of sound from one CD player.

That’s it! Your RM20 DAC! Signal level was initially very low with a single TDA1543. But with some help from vt4c, I changed the I/V and bias resistors and now it’s okay. Here’s a few configurations you can play with.

1. power supply at 8V AND parallel 4 pieces of TDA1543. With this configuration, change I/V resistor to 680 ohm and bias resistor to 330 ohm. 7808 gets really hot. You need bigger heatsink on it, and the 1543s. Everyone of them.
2. add a gain stage to the output, either using opamps or tube output stage. You COULD import the whole 5687 preamp stage here.

Note that the below the 1k I/V conversion resistor, there’s a parallel resonant circuit. It’s meant to bump the 20kHz by ~3dB as this non-oversampling DAC circuit’s frequency response is 3dB down at approximately 20kHz. Simulated response is as shown below. Bottom trace is magnitude trace. It stays flat all the way till 10kHz, starts to rise and peaks ~3dB at ~23kHz. Upper trace shows phase. Stays flat all the time but off by 5 degrees at 20kHz. If this bothers you, remember your speaker has much wilder phase shifts. : )

At the moment, still letting it run in but so far, find the sound, not as smooth as the internal DAC of the CD63 but my oh my! The midrange is especially vivid! Current favourite is Willie Nelson’s Stardust. I seem to hear more echo/reverbaration in his voice. Sounds a lot more life like! At RM20, it has no reason to sound so good!

Update on April 15, 2003. Great reading here. An Interview with Kusunoki San, initiator of non-oversampling DACs theory

Update on May 5, 2003. Believe it or not, this ugly contraption makes good music. ; )

The post A Non-Oversampling DAC for RM20! appeared first on diyparadise.

Non-Oversampling DAC project

What is the easiest, cheapest way of trying out non-oversampling?

If you can trace the I2S signal in your CD player, tap it, fit in a TDA1541 (and attendent signal gain circuitry), then you have it!

You save a chassis, an interconnect cable and have the shortest signal path. What more can you ask for? Furthermore, it saves you lots of unnecessary circuitry in between. Imagine, I2S gets converted to SPDIF, then converted back to I2S… Why go through all this trouble? Why not just tap the I2S straight off? You save an expensive CS8412, you save a lot of unneccessary headache.

With the Marantz CD63 service manual, I think I have found the I2S signal.

If I’m not mistaken, signals WCLK, SCLK and DATA constitute the I2S format.

Wiring to a TDA1541 is easy and it’s only these 3 signals. However, due to the sin(x)/x filter, many builders of the non-oversamping DAC hear a marked loss of high frequency energy. Reason being, the DAC starts to roll off at 10kHz and at 20kHz, it’s already down 3dB. Hardly great stuffs huh? It’s even worse on my Hammer Dynamics as the treble is already laidback, putting down another 3dB isn’t pretty…

To get back the 3dB lost at 20kHz, a simple LC resonant filter will do. Thorsten Loesch has pointed the way in using the TDA1543. With YH’s help, I have adapted one for the 1541.

The circuit on the left is Thorsten’s for TDA1543. The right is my adaptation for TDA1541. The simulation has 2 traces. The top trace is the frequency response. Note that it gradually rises and reaches a peak of about 3dB at ~20kHz. Phase shift is about 10 degrees at 20kHz.

This is even better! (Thanks YH!) A notch filter is added to filter out the 44.1kHz CD sampling frequency. The wild swing on the phase plot is expected, courtesy of the notch filter. The 2 inductors in the circuit can be hand wound, then adjusted (add/reduce windings/ferrite) until “right”.

Do realize that the above is all just circuit simulation. When you build this circuit with real world components, all hell breaks loose. Stray inductance and stray capacitance will make their prescence felt. Imperfect components (is there any perfect component?) will make matters worse. Oh well, live with it…

Anyhow, you need to add a gain stage to amplify this signal. The output is only ~20mV at the node between the 100uF capacitor and 100kohm resistor, so a gain of 100x is needed to reach full output of 2V. Fill in your choice of gain stage, op amp or tubes. You can, of course, add another stage of RC filtering after your gain stage to reduce any output after 20kHz. The choice is yours.

That’s all to it! Again, there’s nothing new here. Stole some ideas here and there, and got the above. Thanks to buddy YH for his ideas.

Simulation was performed using Linear Technology’s excellent SwitcherCAD.

Are you ready to build one? I just did the above with a TDA1543. A non-oversampling DAC for < RM20.

The post Non-Oversampling DAC Project appeared first on diyparadise.

Computer Audio is Soooooo Much Fun!

Boy, I have been having so much fun with computer audio, or do you prefer the term “PC Audio”? Playing music through a PC is addictive. We’ll look at the hardware side first before going to the software end.

I admit I have been late to this “PC audio party” but as they say “better late than never”! And I’m glad I joined the party! The sonics has so much potential compared to conventional CD playback and most folks believe this is mainly due to much reduced jitter when playing through PC. You see, with conventional CD playback, we almost can’t escape from using digital output (SPDIF) format. The Achilles’ heel of SPDIF has to be the way the critical Bit Clock is recovered from it.

This is the work of the digital receiver (like CS8412, CS8414 etc) and we are limited by how well it recovers this clock signal. CS8412 for example has jitter specifications of Bit Clock of 200ps. Not very impressive! Monica’s asynchronous reclocking (as shown the light by Kunosoki san more than 20 years ago) ameliorates this but it’s no way a solution either.

With computer audio, let’s say this together. “To hell with SPDIF!” Ha ha!

Back to hardware, the Squeezebox is a little computer gadget that appeals to geeks like you and me. It streams (wired or wirelessly) audio files stored on your PC to your hifi system. Of course, if you do it wirelessly, you could have your noisy PC in one room and your hifi set in another. Wonderful!

To be honest with you, I know nuts about playing music through a computer but ever since I got a Squeezebox, have to start a crash course on it!

Fancy having a dig at DIYing your own Squeezebox? Well Ole Schüsseler here shows you how to do it with his Linux based Audio Appliance.

So what if you don’t want a player like the Squeezebox and prefer to play from your PC? Sure, no problem! Back then you’ll need at least, a good quality soundcard, but nowadays, there are USB DACs to consider. Yup, takes USB data and converts to analog signals.

The Wonderful World of USB DACs
You see, operating systems (at least for Windows and Macs) are smart enough to know that when they detect a USB DAC, all audio signal will be routed through there. Of course, you could control not to do this as well, but unless your soundcard has better sonics, it’s good to consider USB DACs.

USB DACs like Texas Instrument’s PCM2706 (used in My USB Convertor) not only functions as a USB DAC but also as a headphone amp, USB-SPDIF convertor as well as USB-I2S convertor. The last function is the most interesting as once converted to I2S, we could hook up many DAC chips directly. DAC chips like TDA1543, TDA1541… etc. Like this:

If you want to try this yourself with my USB Convertor,

Pinout of I2S on my USB Convertor pin 1 is GND pin 2 is Word Select leave pin 3 blank pin 4 i Bit Clock pin 5 is Data

I guess you know which pins to connect to which pins right? Told you it’s very easy! Power your DAC chip separately though. Don’t be cheap and use the USB’s 5V supply. If you do so, don’t complain if it doesn’t sound good ok?

If you are ooo-laa-laa over sonics of Monica, then this isn’t so simple. Monica’s TDA1545 chip uses EIAJ format instead of I2S. The main difference between both signal formats is the where upon the DATA signal is sent. I2S sends the DATA signal at the beginning of the Word Select signal but EIAJ sends at the end.

No, please don’t ask me why can’t those folks sit down and agree on one format. I have no idea. But look at SPDIF. The very few instances where all folks concerned agreed on a format but it turned out to be a lousy format. Oh well…

Anyway, I tested and built a pure USB Monica, taking in USB signal to I2S then converted to EIAJ then fed to TDA1545. Give us a few more weeks and you can jive with her soon! [USB Monica is ready!]

Squeezing more juice out of the Squeezebox
Back to the Squeezebox. I’m sure many will be wondering how does the Squeezebox compare to audiophile CD players? Well, stock player is nothing to shout about however never ever under-estimate the power of DIY! I’ll share with you 2 mods I tried here.

The very first upgrade every Squeezebox owner should do is to replace the stock power supply adaptor with a beefier unit. One capable of at least 1A (2A is better). Some folks prefer linear supplies while some prefer switching. My preference is switching power supplies but don’t let me influence you. Go try yourself. Just make sure it outputs 5V REGULATED and you’ll be fine. Oh yeah, pay attention to polarity too since we are dealing with DC voltages here.

For the adventurous diyers, the first mod is considered too easy, no fun at all. The second mod, you need to pull out your Torx T10 screwdriver and dismantle the Squeezebox unit.

This is a relatively easy tweak. Easy to understand and easy to implement if you are good with soldering surface mount components. If your soldering skills isn’t up to par, then forget it. Find someone to do this for you. Remember as with all mods, you can kiss your warranty goodbye so if you are not up to it, don’t try to be cheap here. It ain’t worth it brudder.

What we want to do here is modify the SPDIF circuitry. If you trace the digital output RCA socket, you’ll find it ends at pin 6 of chip U15. This is just a 7404 invertor chip. What we have here is that the invertor output goes through a coupling cap then 2 resistors to ground. The resistors form a voltage divider to scale down the output of the invertor from TTL logic level to SPDIF level. If you find all this Greek, don’t worry.

By right, there isn’t a lot of problems with this circuit but due to the very noisy environment of the Squeezebox (what more with a wireless card nearby), we don’t want all the noise to be carried all the way through the digital output (via SPDIF format) to your DAC and the rest of your audio chain. And the easiest way to accomplish this is to use a digital pulse transformer.

The digital pulse transformer accomplishes 4 things.
First it breaks the ground. The SPDIF signal ground is no longer the Squeezebox unit’s ground. This is possible due to usage of transformers.
Secondly, I believe the trans has better driving capability then just a the invertor and voltage divider circuitry.
Third, with a trans, we could benefit from its common mode rejection capability. Meaning, it rejects noise common to both signal and ground. Like in a noisy environment inside the Squeezebox.
Lastly, we could better match the impedance output impedance with the usage of suitable trans.

So first remove L8 and L9. Then desolder the R67 107ohm resistor. One end of R67 goes to ground while the other end is tied to R66. There’s a mistake in above diagram though. Primary (red dot, pin 1-4) has ratio of 1 while secondary (pin 5-8) has ratio of 2. So you should wire pins 5-8 facing the 7404 invertor chip and the other end (pin 1-4) facing output RCA socket. To anchor the trans, I soldered one leg (pin 8) to junction between 107 and 245ohm resistor. Thus pin 5 goes to ground. Pin 1 (red wire) goes to SPDIF output. Pin 4 – SPDIF ground. (If you are fussy, replace 245ohm resistor with 300ohm for even more accurate 75ohm output impedance.)

Now you know why I’m enjoying computer audio?

Moving on to the Software side

The post Computer Audio is Soooooo Much Fun! appeared first on diyparadise.