Jump to content
IGNORED

SMPS and grounding


Recommended Posts

40 minutes ago, sandyk said:

 I measure around 99V AC  from the "earth" side of the output sockets on a TEAC HDB 850 STB to mains earth.

 However, with Coax SPDIF connected from it to my DAC, which is connected to an earthed Class A Power amplifier, or HDMI plugged from it into my TV, which also has an Oppo 103 connected to it via HDMI, there is no leakage voltage measurable.

 

 

There's the problem in a nutshell - the STB leakage current is now travelling through your DAC - in through the digital input and out again at analog outputs. Presumably from this, it doesn't have an input transformer. Once those noise currents are inside the DAC itself there's no telling what damage they're doing to your SQ.

Link to comment
2 minutes ago, sandyk said:

 

Given that this DIY DAC readily outperformed a Bricasti M1, the damage would have to be quite minor. Unfortunately there is no room to install a Coax SPDIF transformer.

The "earth reference''  right through the chain, is via 10 ohm resistors to PCB "earth" at the L and R inputs  of the Class A Power Amplifier. Neither the DAC, Preamp or P.A. have their  0 volts lines connected to mains earth via the power supply either, and the Source switch in the Preamp only connects through the "earth" side of the selected Input.

Only their  metalwork is connected to mains earth.

Which particular DAC is it? Given its DIY then the internal grounding can be verified to see whether it diverts the noise currents around the sensitive analog circuits. You needn't break the earth loop with a digital transformer, you could put transformers externally on the analog outs (for example).

Are the RCA socket outers on your poweramp wired direct to chassis (mains earth) or are they routed onto the PCB? If the latter then all your noise current runs through that 10ohm resistor.

Link to comment
16 minutes ago, sandyk said:

 

No, the sockets are not grounded to chassis. They connect via the 10 ohm "EARTH LIFT " resistors on  the PCB , which is a well  accepted, and  common method of reducing earth loop problems.

 

'Ground lift' resistors date back to ye olde days when there was a strong likelihood the source component would present a low impedance to mains earth on its 0V and SMPSUs weren't as prevalent as they are nowadays. When the noise generator has a high impedance (as here) they're not providing much benefit.

Link to comment
19 minutes ago, sandyk said:

 

Nobody in their right mind would do this to an amplifier with very close to 4 Zeroes in it's distortion measurements, and a high S/N ratio, along with a very good 3D presentation on good material.

 

It has been applied to a Modulus86 amp (replacing its IC input stage) with I forgot how many zeroes in its THD number (at least 3 I am sure of) with beneficial sonic results.

Link to comment
2 minutes ago, sandyk said:

 

Many designers still use earth lift resistors for their benefits.

Silicon Chip magazine for example, has published some of the lowest distortion designs, and they still use "earth lift" resistors . (Typically .0006% for their SC ULD3)

As far as "ye olde days" goes, that's exactly what using expensive transformers is when there are far better design alternatives these days.

Have you read any of the Power Amplifier Design Handbooks written by highly respected U.K. author Douglas Self, for example ?

 

Curious - what are the sonic benefits of having 3 zeroes in the distortion figure?

 

Long been a fan of Doug Self's books and his EWW articles. Think I might have one of the first editions of APAD - I got it when I was working as an amp designer professionally.

 

What 'alternatives' sound better than transformers for input and output isolation? I wind them myself so your 'expensive' is under $2 in materials for me.

Link to comment
10 minutes ago, sandyk said:

 It is a matter of record that some amplifiers such as some of the Halcro designs with even lower distortion figures are acclaimed for their 3D presentation.

 

 

I'm unclear if you're merely claiming correlation or causation?

 

Back to an earlier question still outstanding - which DIY DAC are you using?

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...