tda1541a sound signature

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I have a Marantz cd-80 with the tda1541a-s1. I was always very impressed by the deep and good bass.
In comparison with newer players f.i. a multiplayer denon the upper midrange ( bells and cymbels are sounding rather dull;
If it were an instrument you would think it is tuned down a half note or more. It also seems like the upper frequencies are overpowered by the lower.
Maybe these are three ways to describe the same problem.
Inspired by the success stories of the tda1541 goeroes i decided to do some upgrades to this machine.
So far i did:
Replaced many but not all ps-elco's as they are 20 years old
New low jitter clock for 3 chip's
resistors in i2s lines
separate regulator and ground for 7220 dig. filter
replace i/v opamp for opa627
disable spdif
disable variable output
new mains connector
schottky diodes for the dac and analog output.
poor man's dem-reclocking by changing the capacitor to match a multiple of Fs-freq

The above mentioned mod's combined did improve the sound, but the problem described has not changed. It actualy starts to annoy me as i believe some instruments sound incorrect.

As i never see negative comments about this DAC...Can anybody confirm this to be a problem of the DAC or it's implementation?
What to do to improve on this specific issue.
 
I have a TDA1541 based Satch DAC (Diy HiFi Supply) with tube output. I liked it better than Mark Levinson 360S on the type of music I listen too, vocals.

However, Levinson 360S has extreme low noise floor and everything used to come from BLACK background. It is Levinson after all and 24bit DAC. There are many areas where it beats Satch but I wanted to experiment.

I started on this route when I heard TDA1541 based AMR CD-77. Wow! Wow again :D. This is the first time ever in my life the timber of acoustic guitar was true. It was as analog as vinyl record. In this regard AMR was better than super duper dCS gear that I also heard. dCS was less organic.

However, Satch DAC proved that just having the TDA1541 chip inside is not everything. There is much more to a DAC. Satch is not CD-77. Perhaps this is what you are also experiencing :rolleyes:.

Regards
 
AMR CD-77 and Satch DAC, both are non-over sampling (NOS) and no digital filter. CD-77 actually gives many options but when I heard it was NOS with no filter. Not sure about CD-77 but I/V in Satch is passive via a resistor then the tube output.

I think CD-77 wins because of low jitter from CD drive to DAC, better digital design, better power supplies (tube rectification, cokes), better dual mono analog construction. Yes! I was impressed and sold but i am waiting for a DAC from AMR as CD player is out of my room for a long long time. It is not digital age of 80s but it is server age of 2011.
 
Last edited:
I have a Philips Cd player with TDA1541, two universal players by Denon, one by Toshiba, one by Apex... Replaced all the opamps with better ones.

I like better the sound of the newer TI DAC's (PCM series) than the sound of old Philips TDA. The TDA in "NOS mode" sounded even worse to me.
 
Hadn't much luck with changing the opamp's. The differences where to little for my shortterm soundmemory. Next time a do a swap, I try a left -right comparison.
Removing or changing the output-cap realy makes a difference (for the better)

As the sound differences are huge in all aspect's of the mentioned machines, The question is not what sounds good, but wat is a correct representation of the truth (disc-content).
Maybe a comparison with a good analog recording?
 
Hi,

I thought there was no dc servo and the cd80 uses a 220uF (c32) cap at the output anyway? Then again I don't care about caps in the signal path.

I do believe there are some circuits without a coupling cap around but you'd have to search yourself. Pedja Rogic also did some nice circuits around the 1541.

regards,
Joris
 
Try to bias the opamp into "class A ".
I use the opa627 just like you do. Put a 2k resistor on pin1 and pin4 (output and -15v).

Its will clear up the highs and mids (vocals will be more smooth and airy) bass get tighter and instruments get more dynamic and focused.

You might not like the result fisrt,I didn't, but after a couple days
I got used to it.

Try to do a proper DEM clock circuit. I use it on only pin16 connected and works well.
In NOS BCK will be 2,8224Mhz 1N5711 in series with 12 to 15K resistor to pin16
then 2K2 resistor to -15V.
In NO-NOS you can use WS signal is 176Khz.

Check ecdesigns treat about DEM clock, or I send it to you if you interested.

I have a Marantz cd-80 with the tda1541a-s1. I was always very impressed by the deep and good bass.
In comparison with newer players f.i. a multiplayer denon the upper midrange ( bells and cymbels are sounding rather dull;
If it were an instrument you would think it is tuned down a half note or more. It also seems like the upper frequencies are overpowered by the lower.
Maybe these are three ways to describe the same problem.
Inspired by the success stories of the tda1541 goeroes i decided to do some upgrades to this machine.
So far i did:
Replaced many but not all ps-elco's as they are 20 years old
New low jitter clock for 3 chip's
resistors in i2s lines
separate regulator and ground for 7220 dig. filter
replace i/v opamp for opa627
disable spdif
disable variable output
new mains connector
schottky diodes for the dac and analog output.
poor man's dem-reclocking by changing the capacitor to match a multiple of Fs-freq

The above mentioned mod's combined did improve the sound, but the problem described has not changed. It actualy starts to annoy me as i believe some instruments sound incorrect.

As i never see negative comments about this DAC...Can anybody confirm this to be a problem of the DAC or it's implementation?
What to do to improve on this specific issue.
 

Attachments

  • tda1541acir.jpg
    tda1541acir.jpg
    70.6 KB · Views: 780
Last edited:
A class-A i can test quickly.
Further i plan to move the opa627 to the buffer and use a ad811 for the I/V, as i saw a circuit from someone who appears to be more knowlegdable than m

As there is also a deemphasize circuit attached to the i/v, it is beyond my capability to cut and paste some discrete stuff.

A better dem-reclocking i save for the last mile, not for the quantumleap i need now.
 
Hi,

Yup you are right. Checked the service manual and there is no cap. sorry.

I'd still try the Pass i/v stage though, it's simple and cheap enough to try out. Maybe a good idea anyway, prototyping various i/v stages and see what you like.

regards,
Joris
 
I have CD40, with TDA1541 (not A), after i builded separated power supply for 1541 the sound improved a lot. I dont know how its in CD80, but if you have there 7220 and 7310 you can implement a separated PSU for both chips. Decopule the caps on 1541 (if its already not done) with 14 x 220 nF or 470nF.
These modes will result with deep and wide sound stage, natural highs and perfect mids.
 
Further investigation suggest that some 2mA is inserted before the i/v stage to negate a always present -2 mA that is present at the output of the dac even when there is no signal.
As long as i work with opamps, I keep that insertion.
For the discrete versions of i/v i don't see much mention of this always present offset, or it's compensation.
 
Hi,

Further i plan to move the opa627 to the buffer and use a ad811 for the I/V, as i saw a circuit from someone who appears to be more knowlegdable than m

Beware, you must make significant circuit changes for the AD811 (or LM6181 for that) to work without oscillation.

Another problem could be if you replaced the 5534 Op-Amp's without removing the Frequency compensation capacitors, they may wreak havoc with an OPA627, while pin 8 is marked "no connection", for all we know it may connect to the substrate or something.

Also, the OPA627 is quite slow for I/V. I would recommend using an OPA637 instead and using Scott Wurcers trick circuit (I came up with the same incidentally a long time back) of having a suitable value cap to ground from the inverting input, 10nF in case of the player you have and the OPA637.

As for the "dark sound", also have a look at the DEM Filter capcitors, also, I personally would consider DEM Reclocking and I2S attenuation as essential.

FWIW, the AMR CD-77 uses special PECL based reclocker/drivers for the TDA1541 which among others also perform attenuation, DEM reclocking, passive I/V and ton's of individual regulators, including some tuned to offset the temperature coefficient of the circuit they supply.

The regulators use a wide variety, from current source fed shunts to fully discrete super low noise feedback free ones, each matched precisely to the circuit they supply - there just is no such thing as a "super regulator" that is super in each and every application. They are followed by very carefully selected and designed decoupling and bypassing.

I would say the power supplies account for around 50 - 60% responsible for the final sonic result in the CD-77.

More on the principle design (block diagram) can be seen here:

Abbingdon Music Research - Product - CD - 77 Freatures

A picture of the whole DAC/digital board is found here:

Abbingdon Music Research - CD - 77 Photo Gallery

Over- and/or up-sampling are selectable as well as deflatable. Finally, for PC connections it comes with a USB input (CD Standard only) and in more recent versions also with SPDIF inputs having very low jitter on either input...

A lot of attention has also been paid to the mechanics, which avoid traditional damping and use multiple material sandwiches and application of certain geometric rules to avoid resonant reinforcement of any modes.

So in many areas it is a tour de force of what can be done.

The sound will not please everyone, nor is it meant to, it is a Player that is firmly "Digital to Analog" in sonic nature. Those who actually LIKE digital sound will be disappointed.

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