New Audio-gd R-7, R-7HE R-8, R-27, R-27HE, R-28 Flagship Resistor Ladder DACs and DAC/amps
May 24, 2019 at 2:29 AM Post #4,217 of 11,260
Hi all,
Kingwa told/allowed me to "Hijack/pollute" this R7/R8-thread with an OT- text. So here we go:

---
Yesterday I received a big box from China.

A few months back when Kingwa and Audio-gd announced about the ”hobby-project”, the optimized TDA1541AS2 dac, I couldn't hold myself back. I have a weakness for the 1541-sound. From memory I found the sound soft and easy flowing... Maybe too soft.. Also the regenerated powersupply in the TDA-dac attracted me. Link: http://www.audio-gd.com/TDA1541S2/TDA1541EN.htm

When Kingwa explained that the only choice for a matching circuit design was his Diamond Differential and that the capacitors had to be the RIFA PHE426 or the PHE450's, I believe him 100%. The TDA1541A can absolutely sing beautifully if they are implemented right and with Kingwa's ”body of knowledge”, I am confident that it will be at its best. Kingwa is actually living audio designs and has gathered a massive stock of this knowledge through the years. Still doing it.

In my case, I have an old Quad 66 CD-player with a reading problem that hold a TDA1541AS1 Single Crown chip. So I only need to get another one.... I let Kingwa know about my plans.. I read up on this subject with the selections criterias of the TDA15141A. I bought a cheap and ugly Rotel CD-player that hold a TDA1541A, unsoldered the chip and then I had two chips... This to start with in my continous hunt for another Quad 66 CD player to get a perfect match (S1 Single Crown)..

During my readings on the internet I found an old DIY-thread that ”died” 2007. The thread had a few members that had bought the most desired version, the TDA1541AS2 Double Crown.

I sent a PM to a member in the thread and ”Bingo!”. He answered: -Yes, I still have them.
I bought the pair, of course. The Quad 66 CD-search was called off. Kingwa got my definitive order of the DIY TDA-dac.

The Big Box came yesterday, as said and I put in the S1/A chips for a few weeks burn in and initial testing before the S2's goes in.

This HUGE dac really sings beautifully already. The ”missmatch” of the present chips inside is impossible to spot. The plain A-chip is a good pair to the S1. I think I was lucky to find the ugly Rotel player. The specific serie of the A-chip also has selected (S1 single crowns and S1 Double Crowns).

This dac is a ”Statement, a tribute to the TDA1541A. One of the best dac-chips of the 90's” The dac is HUGE, also the sound is. The sound is rich and embracing.

My new experience with this regenerated power inside the TDA-dac is definately a success. If You have the funds, go for the regenerated power supply. It brings authority, space, clarity, lush, black background...to another level

Jan
---
Edit: I started a new thread (with a picture): https://www.head-fi.org/threads/audio-gd-diy-tda1541as2-dac.907419/

ps. A comparison between the TDA1541AS2 - R2R7 - NFB27.77 will come in a while ... A bunch of XO's and the TDA-dac need some more time before it.
 
Last edited:
May 25, 2019 at 12:15 PM Post #4,219 of 11,260
@jimmychan I upgraded the R2R 7 front panel to the R7 type. Requires different DSP firmware to remap the new front panel control method to existing DAC jumper locations. The two Xilinx parts on the digital board (near the back I/O panel) do not require update. But if Xilinxs do require updating then you will need Xilinx programming software and JTAG adapter specific for Xilinx. They are fairly cheap on Ebay.

Kingwa talked me out of upgrading the back panel just to get the ability to upgrade DSP firmware via the rear panel Altera USB interace. He said it was a lot of work removing wires, unscrewing connectors/ moving to new panel, and rewiring. I am sure though he would be happy to sell you the parts. Very handy feature!
 
May 25, 2019 at 12:32 PM Post #4,220 of 11,260
The DA7 V2 modules are really starting to shine. Now 4 weeks since installation and sound still improving mainly in the highs (less grain and more lush/ silky). Sound started out OK, regressed for several weeks, and slowly blossoming. I am still not going call "done" on the V2s. They are still improving with noticeable changes in the last week.
 
Last edited:
May 25, 2019 at 1:38 PM Post #4,221 of 11,260
The DA7 V2 modules are really starting to shine. Now 4 weeks since installation and sound still improving mainly in the highs (less grain and more lush/ silky). Sound started out OK, regressed for several weeks, and slowly blossoming. I am still not going call "done" on the V2s. They are still improving with noticeable changes in the last week.
The v1s took at least 700 hours with my dac. One has to be patient. I will probably burn the r7 he, if i buy one, on a second setup driving my nfb1amp. I will give it 4 weeks non stop.
 
May 25, 2019 at 2:43 PM Post #4,222 of 11,260
@jimmychan I upgraded the R2R 7 front panel to the R7 type. Requires different DSP firmware to remap the new front panel control method to existing DAC jumper locations. The two Xilinx parts on the digital board (near the back I/O panel) do not require update. But if Xilinxs do require updating then you will need Xilinx programming software and JTAG adapter specific for Xilinx. They are fairly cheap on Ebay.

Kingwa talked me out of upgrading the back panel just to get the ability to upgrade DSP firmware via the rear panel Altera USB interace. He said it was a lot of work removing wires, unscrewing connectors/ moving to new panel, and rewiring. I am sure though he would be happy to sell you the parts. Very handy feature!

@DACLadder Thanks for your reply. In fact I have done all the upgrade including front and back panel, DAC7 V2, Accusilicon all at one time. I am able to flash the DSP (big chip), but just get stuck on the flash of the two small chips. The application gets close down every time when I flash the chip. I have tried different version of software and different PCs, and I have no clue yet.
 
May 25, 2019 at 3:33 PM Post #4,223 of 11,260
@jimmychan I had a problem with programming one Xilinx part on the DA module. I found if I slowed down the JTAG clock (Tck) from default 10Mhz to 1MHz I could program OK. I may not be using the same Xilinx programming SW as you (Impact32 14.6), On the top row find Output -> Cable Setup... and set JTAG Tck speed/ baud rate to 1Mhz or slower.

There are other JTAG tools under Debug -> (JTAG) Chain Integrity Check. This will tell you if the JTAG cable is connected correctly and if the Xilinx part ID matches what is in the programming file. I could not pass this test even until I reduced Tck speed. Other Xilinx SW versions may have these functions in different locations.

XilinxJTAGTck.jpg
 
Last edited:
May 26, 2019 at 2:15 AM Post #4,224 of 11,260
@DACLadder Thanks for you suggestion. I tried three different PC & notebook and different version of ISE. They all stuck at the "Assign New Config. File". The program just closed by itself without any error message.

But, I finally got it done by using one of my ten years old notebook with Win XP on it. The other PC that I had tried was Win 10 on it.

Now my R2R7 is 100% R7 with all the latest clock and firmware.
 
May 28, 2019 at 2:11 PM Post #4,227 of 11,260
@jimmychan I had a problem with programming one Xilinx part on the DA module. I found if I slowed down the JTAG clock (Tck) from default 10Mhz to 1MHz I could program OK. I may not be using the same Xilinx programming SW as you (Impact32 14.6), On the top row find Output -> Cable Setup... and set JTAG Tck speed/ baud rate to 1Mhz or slower.

There are other JTAG tools under Debug -> (JTAG) Chain Integrity Check. This will tell you if the JTAG cable is connected correctly and if the Xilinx part ID matches what is in the programming file. I could not pass this test even until I reduced Tck speed. Other Xilinx SW versions may have these functions in different locations.

You are a wealth of knowledge. Thank you! I could not get my Xilinx software to work properly with Win10. I will give your suggestions a try. I had just called a buddy to see if I could borrow his Win7 laptop.

On to more fun topics. I told myself I would hold off for at least a week commenting re: the DA-7V2 boards + Accusilicon clock changes in my R-8.

While "audio memory" is a funny thing, I can confidently say that the change is both noticeable and significant. To me, it was well worth it. To any potential buyers out there, feel fortunate that you get the new DA-7 boards standard and go for the clocks. Do it. I echo the words of those before me. I didn't upgrade the USB board clock b/c I don't use that input, but FWIW, it seems like an easy recommendation.

I am currently (yes, still) using V3S firmware. I'll try V3A again soon enough along with the new TDA versions.

The three things that were immediate to me and that stuck with me throughout the week are:
1) Improved sense of realism in the timbre of human voice and stringed instruments.
2) Realism in the attack, sustain, and decay of cymbals specifically.
3) Greater 'flexibility' (my word) in the soundstage offered by the modes within firmwares. I always noticed a change in dynamics and overall presentations between the modes within the firmwares, but rarely much change in the soundstage. Now, the changes between modes seem a bit more pronounced, and both the height and depth of the soundstage shifts a bit.

Overall, music just sounds better... really. The whole is greater than the sum of the parts. However those specific things made me instantly very, very pleased that I made the effort to do the updates. I can also hang my hat on those changes. Even though I "know" it sounds better, everything else is a bit more challenging to articulate.

I think the firmwares play an important role - personally. I switched from NOS2 as my favorite mode across all available V3A and V3S modes "pre-update". Now, my preferred mode is NOS3. FWIW, the OS modes were challenging for me with V3S, but better with V3A in the past. I still don't prefer the OS modes using V3S, but the OS modes are more pleasing than I remember. I preferred the OS modes in V3A to V3S and the NOS modes in V3S. That's the fun part...

Currently, things look like this...
Singxer SU-1 => HMDI R-8 / I2S (NOS 3, no PLL, no sync) => R-9 / ACSS => M-3 / ACSS => Tekton Double Impact

I've swapped amplifiers in and out a bit to see if I prefer certain modes with certain amps, but all comparisons with the R-8 before and after were done with the chain above.

As always, thanks to the kind group here for the assistance and thoughts. Mostly, thanks to Kingwa for allowing existing owners to experience the updates!

I'll forge ahead to try new firmwares on the R-8. With DACLadder's assistance, I can maybe get the FPGAs on the DA-8s in my R-1 up to the latest firmware.

-Cheers
 
Last edited:

Users who are viewing this thread

Back
Top