From the diagram it read that 2x 0.00047 caps are required, is this still the case ??
Correct value for a 50khz low pass filter with 1K resistors is .0016 uF
Recently, I tried a Behringer dac (SRC2496) with a AKM AK4393 chip and i'm still amazed by the performance in upsampling mode.
Transformer mod is scheduled soon with this dac. I think that the CS4398 chinese dac will be something in the past soon.
Comparison CS 4398 and PCM 1793
I started listening to this chinese DAC a week ago.
With the standard Opamps it sounded like an average 1000 Dollar cd player. Nothing special here.
With the LM 4563 opamps in place it became somewhat better, like a better sounding 1500 Dollar cd player.
A couple of days ago, I connected two Lundahls 1674 trannies directly after the DAC chip outputs with only a 1 kOhm Welwyn RC55 resistor in series with the primary windings and a 3 kOhm resistor in parallel with the secondary windings.
( I will try a roll-off later on by adding a cap in parallel with the 1 kOhm resistor).
I have been feeding the trannies some white noise for about 48 hours.
The sound is very good now. Comparable with the highest grade cd players (3000 dollar and more).
Since I bought both the CS 4398 and the BB PCM 1793 module, I did some comparison today. I must say the difference is not subtle...
The CS 4398 has a kind of analogue quality. It offers a more relaxed, warmer sound. It gives a large soundstage by bringing out more of the mid. The top end is softer and less extended. The bass is less defined. It is very forgiving with recordings of lesser quality and it seems well adapted to shrill sounding recordings of pop music. It seems a bit compressed and it does not excel at pin point reproduction of the room ambiance, nor does it reveal the more hidden information like breathing noises etc.
The BB PCM 1793 is a more technical affair. On first hearing it is much more clinical with less midband warmth. However, it does give much more ambiance information, and it retains the original dynamics far better. Weak sounds like breathing is much more audible. The bass is much tighter and less compressed. The top end is more extended and less smeared.
When listening to high quality acoustical recordings (classcal or jazz), it is a treat, but when fed with low-fi pop music it is much less forgiving.
For my purpose, I will follow the BB route.
I would love to see more DACs available as modules. I am curious to see how the Wolfson or AD dacs would perform. Maybe one day?
I still have to do the PLL values mod.
BTW, For the PCM 1793 maybe I should alter the resistor values around the Lundahls. Kevin of K&K audio has a different proposal: http://www.raleighaudio.com/passive_output.htm
Would this be a better solution for the 1793? Any advice is welcome!
Best wishes,
Lucas
I started listening to this chinese DAC a week ago.
With the standard Opamps it sounded like an average 1000 Dollar cd player. Nothing special here.
With the LM 4563 opamps in place it became somewhat better, like a better sounding 1500 Dollar cd player.
A couple of days ago, I connected two Lundahls 1674 trannies directly after the DAC chip outputs with only a 1 kOhm Welwyn RC55 resistor in series with the primary windings and a 3 kOhm resistor in parallel with the secondary windings.
( I will try a roll-off later on by adding a cap in parallel with the 1 kOhm resistor).
I have been feeding the trannies some white noise for about 48 hours.
The sound is very good now. Comparable with the highest grade cd players (3000 dollar and more).
Since I bought both the CS 4398 and the BB PCM 1793 module, I did some comparison today. I must say the difference is not subtle...
The CS 4398 has a kind of analogue quality. It offers a more relaxed, warmer sound. It gives a large soundstage by bringing out more of the mid. The top end is softer and less extended. The bass is less defined. It is very forgiving with recordings of lesser quality and it seems well adapted to shrill sounding recordings of pop music. It seems a bit compressed and it does not excel at pin point reproduction of the room ambiance, nor does it reveal the more hidden information like breathing noises etc.
The BB PCM 1793 is a more technical affair. On first hearing it is much more clinical with less midband warmth. However, it does give much more ambiance information, and it retains the original dynamics far better. Weak sounds like breathing is much more audible. The bass is much tighter and less compressed. The top end is more extended and less smeared.
When listening to high quality acoustical recordings (classcal or jazz), it is a treat, but when fed with low-fi pop music it is much less forgiving.
For my purpose, I will follow the BB route.
I would love to see more DACs available as modules. I am curious to see how the Wolfson or AD dacs would perform. Maybe one day?
I still have to do the PLL values mod.
BTW, For the PCM 1793 maybe I should alter the resistor values around the Lundahls. Kevin of K&K audio has a different proposal: http://www.raleighaudio.com/passive_output.htm
Would this be a better solution for the 1793? Any advice is welcome!
Best wishes,
Lucas
Lucas, you are crushing the output of the Lundahls with that 3k resistor! Dynamics and extension are most assuredly compromised.
Get rid of it and parallel the secondary windings, you don't need a 4/1 step up with voltage out dacs. The Rakkdac is a current out dac, PCM1794, totally different animal. 1793s and 4398s are voltage output and have plenty of signal. Check out the new thread on this same forum about trafos, some very knowledgable people and one trafo designer have made some informative posts.
Hi Bill,
Thanks for your reaction.
I see that I wrote 3 kOhm (that is the value used by David on the Rackdac), but I used 49kOhm instead (I will try to edit that post).
I saw 40 kOhm mentioned, but I only had a 49k at hand.
I guess that is the right value?
I will read the other thread.
Best wishes,
Lucas
Many people are under the mistaken belief that a line trafo needs to be heavily loaded, but the trafo designers themselves loudly disagree.
Hehe Legarem... stop feeding the beast ;P
You have to remember to stop an enjoy sometimes, but then again... not much of a technical challenge which seems to drive many of you (understandably)
Hi folks.
Could someone please clarify for me the following:
1) Is the 'CT' (mentioned on the side of the A20 transformers) the tag marked with the earth symbol?
2) when it says (for 500-ohm connection) connect 3&4,CT and 9&10,CT does this mean then that both sets of connected pins are also joined to the earth tag on the A20, or have I got this all a bit confused? Some of the other mentioned connections on the side of the A20s do not mention CT at all, so why for 500-ohm?
Many thanks.
- John
P.S. Bill you mention that loading the secondary isn't necessary? So I can just snip out the 680-ohm resistor I have there then? I prefer the less is more philosophy whenever possible
Could someone please clarify for me the following:
1) Is the 'CT' (mentioned on the side of the A20 transformers) the tag marked with the earth symbol?
2) when it says (for 500-ohm connection) connect 3&4,CT and 9&10,CT does this mean then that both sets of connected pins are also joined to the earth tag on the A20, or have I got this all a bit confused? Some of the other mentioned connections on the side of the A20s do not mention CT at all, so why for 500-ohm?
Many thanks.
- John
P.S. Bill you mention that loading the secondary isn't necessary? So I can just snip out the 680-ohm resistor I have there then? I prefer the less is more philosophy whenever possible
Attachments
Last edited:
Hi folks.
Could someone please clarify for me the following:
1) Is the 'CT' (mentioned on the side of the A20 transformers) the tag marked with the earth symbol?
2) when it says (for 500-ohm connection) connect 3&4,CT and 9&10,CT does this mean then that both sets of connected pins are also joined to the earth tag on the A20, or have I got this all a bit confused? Some of the other mentioned connections on the side of the A20s do not mention CT at all, so why for 500-ohm?
Many thanks.
- John
P.S. Bill you mention that loading the secondary isn't necessary? So I can just snip out the 680-ohm resistor I have there then? I prefer the less is more philosophy whenever possible
The CT on the can is just to tell you that when you connect those pins that this point can be used for a center tap if needed in your circuit. If you want to keep it simple, and have no hum issues, don't connect it to anything else, just connect 3 to 4 and 9 to 10.
Yes, you can remove the 680R but you should probably add an RC filter, 1K in series with a .001uf cap in it's place, as Jensen does. Cutoff is around 150Khz, not audible but effective to control high frequency ringing.