Help Modify SMPS

[jkeny]

I have brought this topic over to here as it's the most logical section for it - I started the post here: http://www.diyaudio.com/forums/showt...60#post1526060

What I'm trying to do is disconnect the bottom of the schematic that supplies the main VDD (10 to 40V) & some other supplies +- 15V +-5V.

I can supply the main VDD from 24V battery & the improvement in sound is huge compared to SMPS VDD.

So I wondered would there be an improvement if I also used the battery to supply the PS to the digital ICs? I can run it for about 2-3mins off battery & it seems to be much cleaner, clearer sound but not enough time for a real evaluation.

Here's my dilemma:

When I disconnect this part of the circuit & plug in a 24V DC from SLA Battery into the red line that gives VDD & thse other supplies, I get an error message of DCDect being 0V instead of 5V.

When I leave it connected in but break the red line connection to VDD & the +-15V & +-5V and instead feed the 24V DC into the red line, I get the switching IC (IC701) beginning to click after a short while at which point I turn off the amp.

I know I could fool the circuit by putting 5V on the DCdet line but I would feel safer if I knew why the DCDET is not being set to 5V correctly with my battery connection.

The schematic comes in 3 parts, the first one I've attached (Part 1) - it is the main switcher which supplies VDD & goes on to supply +-15V and +-5V on next schematic. This schematic (Part 1) also has a standby circuit, at the top and the main VDD switcher is complicated by the fact that is is a variable voltage volume control - controlled by a reference voltage which comes from the volume control. The +-Sense signal & the volcnt signal are part of this control loop.

[jkeny]

Second part of SMPS schematic - Part 2 contains the +-15V & +-5V which is derived from the VDD feed. It also contains another switcher which generates BK+3V, BK+3.3V, BK+5.7V among others and is generated from the standby voltage.

[jkeny]

The final part of the SMPS schematic - Part 3 which contains the DCDET generation? circuit

[jkeny]

Sorry for the long post - any/all help would be appreciated!

[Eva]

Any change in sound is likely to be produced by the much higher AC output impedance of batteries and long external wirings in comparison with a well regulated SMPS. That kind of digital amplifiers exhibit 0dB PSRR.

Add some impedance in series with the speakers, maybe with non-linear elements to simulate the exponential/logarithmic I/V characteristic of batteries and you will get a similar result...

[jkeny]

Eva,
I am saying that the sound with batteries supplying the main VDD i.e. the output stage is far superior to using the "well regulated SMPS" - it has more detail & nuances in the sonic reproduction & is quieter - more of the emotion of the piece is conveyed.

From what you're saying, I gather you're not a fan of batteries and seem to have a rigid stance about this? I can tell you, in this instance, the batteries are superior by a long ways. Maybe it's the quality of the SMPS - it improved when I put a large cap between it & the output stage

Now can we not get into an argument over battery Vs SMPS - I just wanted to find out how to bypass this complicated SMPS.

I discovered that the DCDET signal is actually generated by a microprocessor (which is powered on by the standby PS), so it is based on the timing of the supplies coming up. But this should be exactly the same as if I was powering it from the SMPS?