Tuesday, October 30, 2007
The USB board is DDDAC1753 mkII. See picutre(s).
USB board output I2S but I need AD1865 compatible format (18-bit, MSB, standard). I do not like using a digital filter like DF1704, since I will have to oversample to 96Mhz. So I built (PCB in construction) a circuit using shift registers and flip-flops. I learned Eagle (and Eagle 3d).
Note: I also take a loot at gates jitter (and also CPLD, FPGA) and found out that HC is good enough (having not bad PS).
Tuesday, September 25, 2007
Last week I did quite a lot of studies on my ultimate DAC + amp all-in-one design. Since I will put DIY aside for a couple of weeks (but still some minor work) I will describe my design here so that I will not forget it.
Inspiration of Retro DAC project with exception of the PSU and output stage. For PSU I will use my battery power supply Daja (maybe try some Young super-regulator later). I will bypass SLAs w/ 22000uF low ESR caps and WIMA MKPs (I can put some 5R before the caps not to have huge currents when connected to SLAs). I plan to connect Sowter 9762 transformer directly to the Iout of the DAC and then vary (volume control) voltage with the resistance on the secondary. Read this:
Yes, a transformer transforms the impedance by the square it's winding ratio. And it will transform the current or voltage by the winding ratio (voltage up current down for stepup transformers).
In other terms, if you have a DAC with an Output current of +/-1.2mA for full scale and you apply this current to the primary of a 1:10 Transformer you will give +/- 120uA on the secondary. To get a 5.65V Peak Peak signal from this current you need around 23k5 as I/V resistor. This 23k5 will be transformed into the transformers primary by the square of the 1:10 winding ratio, namely 100. So the load reflected upon the DAC's current output is around 235 Ohm.
Clearly, an output impedance of 23KOhm is impractical, so if we follow the Transformer plus I/V resistor with an active stage we can reduce the I/V resistor.
Let's use a 6922 as Valve for the analogue stage. It has a gain of around 29db with a 22k Anode load and around 3K output impedance. For 2V RMS (5.65V P-P) from the Anode we need around 200mV P-P or 100mV Peak. So we require 100mV from 120uA or 833 Ohm. Let's use 1k for arguments sake, so we require a 1:10 transformer that can be loaded with 1k and this will then reflect an impedance of 10 Ohm plus all direct current resistive losses.
While looking good on paper this is not a very practical transformer to find of the shelf, so we are back to first converting our current to Voltage with a suitable resistor (say 10 Ohm) and then stepping it up with a traditional 1:10 Stepup Transformer (for microphones and or MC Pickups) loaded with it's usual secondary load (say 10k giving a 100R Load in parallel with the 10R I/V conversion resistor) to give use 100mV Peak from the DAC's output into our Valve stage.
The idea is to connect it to the gain stage of the power amp directly, if possible. Otherwise, either Aikido 6SN7/6SL7 octal output stage or UGS V3 preamp driven via SLAs (+ huge caps)!
I will be very short here since it is very simple: UGSs V3 preamp + FirstWatt F4 output buffer. Something like this.
All is completely balanced (differential) design.
I did quite some reading on speaker protection (DC at speaker output can destroy speaker drivers). Initially I planned to use Velleman K4700, but later I found ESP P33 much more advanced.
- Pathos ClassicOne mkIII - very musical, but a little but to gentle for me. For romantics :)
- Copland CTA305 preamp + Audio Research 150.2 amp - this was not a real test of each component. I found Copland one step backwards to my Aikido preamp - lack of air and detail. I cannot judge the AR.
- MBL 7006 - a complete new league compared to the amplifiers I've tested. Very airy, a new definition of bass (more tight), noiseless, outstanding dynamics, but...
- Burmester 051 - a real gem, especially in lower frequencies is a class of its own! Noiseless, muscial, airy, almost too analytica, but with atouch of softness. I really like this one...
- NAD M3 - a very good integrated amplifier. With exception of the lower frequencies in class of Burmester 051. In terms of highs and dynamics maybe even slightly better. I prefer NAD M3 to MBL 7006.
Wednesday, September 12, 2007
Therefore, I will most likely DIY one.
One DIY project: TNT TTS
And maybe more scientific approach: VH Audio
And another: Scoobidoo
Intersesting it is that no shielding is used (to achieve lower capacitance) - also Brahma does not have it.
Friday, August 31, 2007
I have to solve one issue - one channel has 160V other 210V. Aikido balances this, but I have to solve this.
Thursday, August 30, 2007
Coupling caps: Mundorf Silver/Oil
Polypropyene caps: WIMA MKP
Low-noise (digital) caps: BG NX
HV PSU caps: EPCOS 435xxx series (did not try others)
Resistors: Vishay VTA52/55 (of course, tantalums are not bad)
Riken Ohm : Good balanced sound; improvement on cheap metal films.
Tantalum (AN and Shinkoh) : A little more detail than Riken, nice midrange, but upper midrange and treble can be a bit harsh.
Vishay VTA52 / VTA55 : Amazing. By far the best. More detail, smoother sound, very lifelike.
Friday, August 17, 2007
I plan do do a class-A power amplifier based on Aikido input stage and MOSFET output stage - called MOSKIDO ;)
Today I've finally completed Battery Powered PSU (named Daja). Its is fully automatic (no switches at all, triggered by client - 12V trigger) based on relays. When not used battery are being charged. Charging is advanced using several stated: batery revival (trickle charge), current limited charging, voaltage limited charing and finally float charge. No maintainence is needed. Connectors are RCA and all fully fused.
See pictures with comments.
Thursday, August 2, 2007
I've removed digital PSU out of DAC and mounted new PSU. Now everyhing is nice and clean. No bruuum and background noise. Great!
SLA battery chager (UC3906) prototype works. I plan to built 4 on them on a board and put SLAs in nice separate chassis. I will use RCA conectors (standard and thay have good connections).
Monday, July 30, 2007
Value is determined by C = 1/(2*pi*R*f), where f is a roll-off frequency. Coupling cap forms a high-pass filter with resistor and impedance of the next stage (power amplifier). I recommend values 470nF and larger.
Wednesday, July 25, 2007
Last days I've done and learned lots of things, experiments, playing with the osciloscope, etc.
However today I've installed my new HT PSU and the sound is to cry for: soundstage is wide, instrumets got played by musicians, everyhining is so much real and alive.
SiC diodes, EPCOS 220uF LL caps, 2k2 resistors (2W) in CRCRC configuration bypassed with terrific WIMA MKP4 4,7uF.
Got to go now... litstning to the music :)
Saturday, July 21, 2007
DIY Audio thread (very looooong).
Use low noise PSU (SLA battery) and add capacitor (good 100uF low-LSR; consider Murata 47uF ceramic) in parallel to LDR LEDs.
I recommend usage of Silonex NSL-32SR2S (sorted, matched).
Sunday, July 15, 2007
this weekend I am working on the PSU reimplementation.
After lots of thinking I end up with.
HT PSU - simple (R)CRCRC using high quality electrolyte aluminium capacitors (220uF; low ERS, high ripple current - e.g. EPCOS), bypassing them with polypropene (4,7uF; WIMA MKP4). Since ElCo-s are slower than polypropene and do not hanle high requencies well, I will use decoupled electrolytic capacitor technique - patented by Audio Research. See AR-LS22 schematic for realy good design. I like it very much.
LV PSU - simple C(R)C with LM317 regualtor (initialy low ripple configuration, later with slow start-up). This power could also be used as digital power supply (receivers, USB, XO-DAC).
I am considering buying an XO-DAC, but I have one technical question before purchasing.
I have use CS8414 and AD1865N (Audio Note config).
Using XO-DAC will get MCK (and DGND) from the CS8414 and will output clean clock. Since no filter is being used, connection is not so straight forward.
CS8414 has outputs:
- this is being replaced by XO-DAC
- SCK (=MCK/4)
- I can get this by dividing new MCK (2 flip-flops)
The question is what to do with these:
I guess simply aligning them (w/ D flip-flops) to the new MCK, is the best ting to do? Right.
Well, decoding SDATA and FSYNC to left/right data before D flip-flops is slightly better - to avoid gate's jitter.
Saturday, July 14, 2007
Tuesday, July 10, 2007
I noticed some serious humming sound when I connected DAC to a wall socket (even though not switch was not turned on). Chaning polarity (L with N) did helped, but not completely.
I found out that the output GND of the PSU board has to be grounded.
Since I do not use PSU to supply power to the digital board, grounding was not done via digital board.
Now I still hear some humming (at very unreasonablly high values), but this is result of heaters/B+, I guess.
Well, the break is over - I have to go back to work.
Monday, July 9, 2007
DAC has been playing for the whole two days now - really nice sound, but know it can be improved a lot.
I've done some measurements:
TP1 = +4.85V (R24 = 1k7)
TP2 = -5.25V (R25 = 2k0)
TP3 = +4.66V (R26 = 1k6)
I really need to "equalize" resitors.
FIL1 = 6.2V
FIL2 = 6.3V
I really need to "equalize" these also. I am thinking of using one PS for both fillaments - this will quarantee same voltage (but will require 2x higher current, which I do not like).
I_dac1 = 22.5mA
I_dac2 = 25.7mA (since TP2 > TP1, I guess)
I_receiver = 18,5mA
Currents are quite small, which is good for my batteries :)
Sunday, July 8, 2007
- V/I transformers
- USB in (TOSLINK in?)
- Black Gates on 6922 board (2x 470uF)
- Mundorf's SIlver/Oil decouplung caps (2x 0,470uF)
- other tubes (Amperex 7309?)
- equalize TP1 == TP2, higher TP3
- equalize filament voltages (redesign filament PS?)
- cap before filament VRs (100nF)
- cap before gretz (cca 10nF)
- PS (C/10 + C + R + C+ C/10)
- smart SLA charger
- caps on SLAs
Interesting AN DAC upgrade.
As you can see, I am using batteries for analog voltage section of DAC (currently I have no charger done yet) - cleaner sound.
I had to lower output level of the DAC since it was to high for my amplifier. This is what I do not like - soundstage suffers. But this only means that DAC is capable of quite high output level and can be used without preamplifier :)
Well, now I only have to made a power amplifier and connect it directly to the DAC.
Still, I plan to add USB input to the DAC and do the reclocking (Tentlabs XO-DAC).
Tuesday, July 3, 2007
Things to be done:
- make tweaks final
- check Digital board volages (they really should be +/-5V)
- "nicefly" all together
- test w/ battery power supply for DAC analogue section
- solid state (JFET) output stage
Sunday, July 1, 2007
Here are some of pictures (package and a building process) with comments.
I did not manage to built it during the weekend - firstly I had some problems with the power supply board (I switched 200V and 40 diodes - you might guess what happened) and this took me a few hours to figure it out, but I did a PhD from the PS board :)
Now I have to build a 7805/7905 regulator for the tube filaments, but I ran out of 7905s. Tomorrow...
I hope to get this DAC working ASAP.
P.S. I decided to go with the battery power supply at least for the digital PCB. I will also try solid state (JFET) output stage, of course battery powered.
I've also found out that class-D design power amplifier became quite good nowdays (for sure I do not want to have class-A owen at home). Especially, Hypex moules. I might build one of those (UCD400HG based).
Monday, June 4, 2007
The week-end is over, have to work now... and wait for the AN DAC Kit to arrive ;)
Sunday, June 3, 2007
My opinion is that #3 is the only clean way to go, but #5 is best price/value performance (how it is usualy done) - also Mr Kusunoki described this method (comment on his article: since this method is not 100%, transports sound differently).
1.) Get rid of the problem right at the source, e.g. the player. The player and the DACs would be fed from the same local low-jitter clock. PSU design and clock distribution scheme would be optimized to have the lowest jitter possible at the DAC, while the jitter at the rest of the circuit isn't that important.
For external DACs, we either have to isolate the timing of incoming signal from the timing of the DAC chip, or recover the original clock better than usual.
2.) Design a high quality PLL which recovers the original clock better than a simple CS841x receiver chip. The Pass Labs D1 (service manual available from their site) is a nice example of a good implementation of this technique.
3.) Let the input data stream fill a FIFO buffer. The DAC is fed from a local low-jitter clock and the data comes from the buffer. Due to the separate transport and DAC clocks, the buffer will empty or fill up slowly, depending on which clock is faster. Thus, we must provide some means to slightly "tune" the local clock source to keep the buffer about half filled.
4.) Use an ASRC to complete isolate the timing of input data stream and DAC chip. This approach is cheap and easy, but there are some caveats.
5.) Asynchronous reclocking right before the DAC. A nice idea, cheap to implement and it works good.
- low jitter
- clean, low noise power supply for analog part (and for clock)
Audio Note digital board (signature version) is my DAC board base. It is really well good designed, but still not perfect. Audio Note completely ignores the jitter issue. I take low jitter as important component of DAC design - reclocking must be done just before DAC. This is the thing I will have to do on my own.
Secondly, I will investigate analog voltage regulator design for the DAC chip analog voltage supply - this has to be clean as possible. AN board does not use classic LM317/TL471 regulators (they are good, but not good enough), but LF351/pass transistor/LED combination. I hope it does its work good or I will provide it myself.
I also consider using battery power supply (SLA/VRLA) as they provide low-noise DC (yet to be measured under load).
I also have to provide low-noise DC voltages of 6.3V and 230V for the 6DJ8 tubes.
So said - "I'll do it myself. A very gode one".
As many of you already know CD plater consists of 2 parts:
- transport (a mechanism that reads CD and outputs data - usually SPDIF or even I2S) and
- DAC (digital-to-analog converter) w/ output stage (2Vrms output is required, for unbalanced RCA).
Choosing a transport was an easy thing - CD-PRO2LF. Great transport, no clumsy tray, very turntable alike.
A great DIY (including controller board, remote, PSUs) can be purchased here.
Quite a straight-forward implementation. The only think-about thing is: should I inject DAC clock (and reclock SPDIF) or not. Tentlabs X03 is what I have in mind.
By-the-way, CD-PRO2LF has an I2S output (advantage over SPDIF of avoiding additional jitter), but there are many different I2S standards. CD-PRO2LF can be configured to output a specific standard, but this would require additional work (to sent DSA commant to the control unit). But I want to have transport DAC independed and I will most likely use SPDIF output.
DAC part is trickier.
I will not write an essay here. I agree with the Kusunoki's "Non-oversampling Digital filter-less DAC Concept".
This left me the following DAC chip choices: PCM1704 (new vesion of PCM 63), TDA1541 (and its economy versions: TDA1543, TDA1545), AD1865.
TDA1541 is very nice (also accepts I2C outputed by CD-PRO2LF), but requires 15V (and lots of decoupling caps).
PCM1704 is also great, but requires additional filter to convert 16-bit to 20-bit.
I found AD1865 most appropriate (all over 5V), very good specs and nice sounding (Audio Note preffered DAC chip).
Simpler is better.
By-the-way, there is quite good CD player available - fully differential (4x PCM1704-K), non-oversampling - Primare CD31. Worth listening, if you do not want to DIY.
I found these three DAC implementations to be really good one (my reference):
At the end, I ordered Audio Note 2.1 Kit w/o M2 PSU (that was to expensive for me). It shoud arrive in a few days. Weeeee :)
But I have some improvements in my mind, keep on reading...