DAC ouput using Transformer

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

vaughn

Member
Joined 2008
Yeah, it is one of those things that of course makes sense once I hear it...

I had been wondering why the variable output (not digital) of my Oppo was sounding better than the digital output of my Squeezebox going through an up- sampling transformer coupled DAC. I guess in a way this is a good thing as I can make a very substantial improvement to my sound quality by simply putting a 20k pot on the input of my amp and running my digital out at maximum.
 
GerardA

GerardA

Member
Joined 2005
One way to digitally attenuate the signal without loosing the bits is done by XXHighend-player. First it adds 8 bits to a 16 bit signal and uses these resulting 24 bits to attenuate. After attenuating the first 8 bits you start loosing bits with further attenuating. But this is at a very low level so you won't hear it. Only works of course if you have a 24/32bit DAC.
 
C

ChrisA

Member
Joined 2008
Last edited:
C

ChrisA

Member
Joined 2008
GerardA said:
One way to digitally attenuate the signal without loosing the bits is done by XXHighend-player. First it adds 8 bits to a 16 bit signal and uses these resulting 24 bits to attenuate. After attenuating the first 8 bits you start loosing bits with further attenuating. But this is at a very low level so you won't hear it. Only works of course if you have a 24/32bit DAC.
Click to expand...

This is correct. If you have a 24 bit DAC and are listing to 16-bit program material (such as a CD or digital download) you can digitally attenuate all the way back to 16 bits without any loss of quality.

This is very easy to understand. The system (all of them) scale from 16 bits to 24 by simply multiplying each sample by 256. When the audio is digitally attenuated that number "256" is reduced. As long as it is not reduced below 1 you are OK. At 1 you are still getting all 16 bits of the program source material. Below 1 and you are loosing bits

But the key here is "If you have a 24 bit DAC". Almost all of the studio grade computer audio interfaces are 24 bits but lower end "gammer" and HT interfaces might be 16-bits. If you are using 16 bit DAC then the multiplier stars at 1 and goes down from there as you move away from full volume. You do want to run the 160bit DAC at full volume

The other thing is if the program source material is something you made. Most of us will record and mix at 24 bits just because we can. In that case even with a 24 bit DAC you can't digitally attenuate without loosing bits.

But in the "normal" home hifi playback case if you have a 24Bit DAC you can digitally attenuate to about 20dB without loosing bits.
 
J

jkeny

Banned
Joined 2007
brianco said:
Just an explanation from another Irishman!!

"I've been bold" (in Ireland) is a confession for having been naughty!!! :D
Click to expand...

Yes, thanks Brian, I should remember not to use Irish colloquialisms :)

BTW, guys, I don't hear any degradation in sound using my DACs digital vol control - it is a 24/192 capable DAC & I'm playing back 16/44 or 24/96 with only about 5% vol. I haven't tried any 24/192 - will do soon!
 
H

Havoc08

Member
Joined 2008
rossl said:
OK, we need a diagram. This is what I use. I hope this works, it is the first time I have tried to post a diagram with the new forum software.

The three boxes have transformers for mains and signal transfer. The only DC connection between the three is star grounding to earth through the 10 ohm resistors. The 10R is enough to prevent stray noise currents from travelling between boxes and earth. All chassis are earth grounded for safety reasons.

The signals are galvanically isolated because signal transformers break any loops between the boxes as far as the power supply noises are concerned. The 10R provides a reference for all the circuits.

Ideally, there will be no current in the 10R resistors after stray capacitance and inductance in the system are discharged through the 10Rs. Hopefully 10R is enough to dampen any interaction between those stray C and L

Only signal currents in the signal cables. Only power currents in the power connections. No mixing of currents.

The 10 ohm resistor should be high enough wattage to blow the fuse in case of a power supply fault. A tiny little SMT resistor may vaporize if the mains shorts to the circuitry.

There is a long article by Rod Elliott about a earthing scheme similar to this
http://sound.westhost.com/earthing.htm
Click to expand...


So using that layout I can use a ordinary pot on the analogue inputs to control volume instead of the digital way? I'm using it with the shigaclone -> 4398 DAC -> Lundahl LL-1517 -> T-amp
 
R

rolls

Member
Joined 2007
ChrisA said:
This is correct. If you have a 24 bit DAC and you are listening to 16-bit program material (such as a CD or digital download) you can digitally attenuate all the way back to 16 bits without any loss of quality.

In the "normal" home hifi playback case if you have a 24Bit DAC you can digitally attenuate to about 20dB without losing bits.
Click to expand...

Every bit means 6db in the binary system, means times 2 or divided by 2. If you have a 24 bit dac and a 16 bit source, you have 8 "spare" bits. So you can turn the volume down to (8 times 6 db) -48 db without losing bits. With a 32 bit dac, it would be even -96 db.
With hires 24 bit material things will be different. Then you can't use a 24 bit dac without losing bits, but you can go down to -48 db with a 32 bit dac.
 
Last edited:
rossl

rossl

Member
Joined 2004
home.roadrunner.com/~rossl/diyaudio.htm
Havoc08 said:
So using that layout I can use a ordinary pot on the analogue inputs to control volume instead of the digital way? I'm using it with the shigaclone -> 4398 DAC -> Lundahl LL-1517 -> T-amp
Click to expand...

Yes. Passive volume control sounds great. You won't have a remote control for volume any more , though. :eek:

I got rid of a preamp with a remote and went to passive between DAC and Amp. My wife was amazed. I had to get up to change the volume.
 
H

Havoc08

Member
Joined 2008
I'll take a look, thx jkeny. My LL1517 are actually on a calrec board with opamps (NE5532) so I thought that maybe they were already configured as driven by opamp as mentioned in my link. Unfortunately I can't find schematics or pinouts to the boards so I'm a bit lost atm :).
It's a calrec LN1868 (or LN811-010-5 it says on the pcb).
 
J

jkeny

Banned
Joined 2007
I was using Dave Slagle made 80% Nickel based transformers for DAC o/p duties up to now. These are not designed for this role & are earmarked for a tube preamp.
- I bought a pair of Seccom MI-97 transformers off ebay for $50 which were recommended to me.
- Just tried them tonight & they are close to the Slagles. They might be more evenly balanced than the Slagles but it's a bit early to tell. The Slagles have a shimmering & extended HF which also seems to emphasise a background hiss coming from the laptop PS. The Sescoms don't have this shimmer (it might be a HF ringing?) but also don't have the hiss. I'll solve the PS hiss first & then some more tests & tweaking needed but I'm not disappointed with these Sescoms - for the price they're bloody great!

I don't have any resistor load on the primaries yet so this may also change things :)
 
B

Bill Fuss

RIP
Joined 2005
jkeny said:
I was using Dave Slagle made 80% Nickel based transformers for DAC o/p duties up to now. These are not designed for this role & are earmarked for a tube preamp.
- I bought a pair of Seccom MI-97 transformers off ebay for $50 which were recommended to me.
- Just tried them tonight & they are close to the Slagles. They might be more evenly balanced than the Slagles but it's a bit early to tell. The Slagles have a shimmering & extended HF which also seems to emphasise a background hiss coming from the laptop PS. The Sescoms don't have this shimmer (it might be a HF ringing?) but also don't have the hiss. I'll solve the PS hiss first & then some more tests & tweaking needed but I'm not disappointed with these Sescoms - for the price they're bloody great!

I don't have any resistor load on the primaries yet so this may also change things :)
Click to expand...

Hi John,
Did you have the opportunity to look at the responses on a scope? My Sescoms have a prominent resonance up around 200khz that does show up in the top couple of octaves of the audio band. The Jensen secondary RC filter completely eliminated it.
You might try 1kohms across the primaries. That would be in parallel with the dac chip output circuitry so the trafos would see around 500ohms depending on the series resistance you are using.
Do you think the shimmer you refer to might be an artificial modulation artifact or actually be in the source material.

Best, Bill
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom