A power supply for the Squeezebox

I think we all know by now what this section is for.
User avatar
Ali Tait
Eternally single
Posts: 4374
Joined: Fri Jun 08, 2007 8:10 pm
Location: Galashiels

#31

Post by Ali Tait »

Yes, I found the same using the supply I brought to Owston. Makes a real improvement.
Neal
Shed dweller
Posts: 2299
Joined: Fri Aug 03, 2007 10:57 am
Location: From the land of the Bodgers

#32

Post by Neal »

SteveTheShadow wrote:Cheers Neal
Those LEDs will make a nice power indicator :)

I did originally have a 100uF on the output, but on the scope there was a low rate cyclical drift going on at about 0.3 Hz so in the end I used the more conventional small value cap and 2R7 resistor. The 100uF however was a low ESR Elna, so that might have been responsible for the VLF drift.
I'll try it again with a cheap cap instead.

All earth returns in the PSU are actually returned to the PSU 0V although the way I have drawn the diagram does not show it. The pre tracking option looks interesting. Is it yourself and Tony Moore who have this kind of PSU setup?

IIRC isn't it that "super-reg" thing that the Transporter had?
The transporter super regs are the ALW ones but hey use a pre-tracking regulator as well.

You can use larger caps on the output but they must not be low ESR, the series resistor you are using is a good idea.

The earth return for the sent resistor etc should be made at the load and then run back to the PSU, so three wires to the load....

Good article here about the o/p cap and set resistor earthing half way down:

http://www.tnt-audio.com/clinica/regula ... ce1_e.html
richardcooper2k
Old Hand
Posts: 780
Joined: Thu Jun 07, 2007 2:32 pm
Location: sheffield

#33

Post by richardcooper2k »

Anyone tried something like this for a laptop (used as a digital source) ?

Presumably i could make gains here ?
Cressy Snr
Amstrad Tower of Power
Posts: 10555
Joined: Wed May 30, 2007 12:25 am
Location: South Yorks.

#34

Post by Cressy Snr »

Here is the Squeezebox PSU as it is now.
No Pre-tracking reg but it is another step up in quality.

Image
Neal
Shed dweller
Posts: 2299
Joined: Fri Aug 03, 2007 10:57 am
Location: From the land of the Bodgers

#35

Post by Neal »

Excellent! Glad it worked...try the pre-tracking when you get a chance.. :wink:
User avatar
Dave the bass
Amstrad Tower of Power
Posts: 12273
Joined: Tue May 22, 2007 4:36 pm
Location: NW Kent, Darn Sarf innit.

#36

Post by Dave the bass »

Question from the back of the class Sir....

Why 5.44V? Wont the the regulator in the SBT be continually trying to scrub off 440mV and making it's life harder?

DTB
"The fat bourgeois and his doppelganger"
User avatar
andrew Ivimey
Social Sevices have been notified
Posts: 8308
Joined: Mon Jun 11, 2007 8:33 am
Location: Bedford

#37

Post by andrew Ivimey »

another question from a different part of the back - we've been split up for messin'about under the desks ... 'hands where I can see them boys!'

what's the two 1n4002s for? I'd guess the two green leds have something to do with the 5.44 volts but why not go for dftraight 5volts anyway?
Philosophers have only interpreted the world - the point, however, is to change it. No it isn't ... maybe we should leave it alone for a while.
Neal
Shed dweller
Posts: 2299
Joined: Fri Aug 03, 2007 10:57 am
Location: From the land of the Bodgers

#38

Post by Neal »

The diodes are there to protect the regulator in case the output gets shorted when using an external cap....without them the cap would discharge through the regulator hurting it so badly it may not want to play anymore.

Voltage drop across a standard green LED is around 2~2.1v so (although the ones I'm using are about 1.9v) two together could give 4.2 plus the 1.25v reference voltage of the regulator gets you to 5.45v...all depends on the type used.

You could use red LEDs which drop about 1.7v for 4.65v if it bothers you or how about a combination of the two for 5.05v... it is Xmas after all! :D

Using LED's lowers noise and output impedance.
Andrew
Eternally single
Posts: 4206
Joined: Thu May 24, 2007 2:18 pm

#39

Post by Andrew »

Hi Neal,

Daft question perhaps, but why not use a fixed 5v reg?

Andrew
Analogue, the lost world that lies between 0 and 1.
Neal
Shed dweller
Posts: 2299
Joined: Fri Aug 03, 2007 10:57 am
Location: From the land of the Bodgers

#40

Post by Neal »

....because Steve was already using an adjustable one! :D but as I understand it it's possible to improve on the cooking 7805 and LM317s etc by using the techniques above which lower the output impedance and in turn lowers the output noise plus improving the load regulation.
User avatar
Mike H
Amstrad Tower of Power
Posts: 20163
Joined: Sat Oct 04, 2008 5:38 pm
Location: The Fens
Contact:

#41

Post by Mike H »

If going to that much trouble you might design your own betterer discrete one :D

Neal wrote:The diodes are there to protect the regulator in case the output gets shorted when using an external cap....without them the cap would discharge through the regulator hurting it so badly it may not want to play anymore.
You'd think the device designer could've had them built-in? :roll:

The 2nd 1N4002 twixt adjust and out prevents adjust going more positive than out. Which would just mean the internal transistor(s) at that point (basically a Darlington combo in LM317, not sure about 338T) have their / its base emitter junction(s) possibly Zenering in reverse. Although there is also a series resistor on the adjust pin that will limit the current (e.g. 12k in the LM317, from what I've seen of a LTspice model for it).

The 1st 1N4002 prevents the whole thing being reversed if out is more positive than in, as the bulk of the circuit is between in and out.
 
"No matter how fast light travels it finds that the darkness has always got there first, and is waiting for it."
User avatar
andrew Ivimey
Social Sevices have been notified
Posts: 8308
Joined: Mon Jun 11, 2007 8:33 am
Location: Bedford

#42

Post by andrew Ivimey »

The word ( I guess some of you know all this already but) is something like this ...

Touch is programmed by Unix - it is possible to switch bits (screen) and babs(volume control) off and other 'demons' and this improves the sound.

Plugging Vortexbox straight in through ethernet was stunning!

Very polished and well designed linear PSU (cascade of voltage regulators, choke input) made a small improvement relative to the ethernet cable.

very very impressed

oh, and ... all through a mark1 RAKKDAC
Philosophers have only interpreted the world - the point, however, is to change it. No it isn't ... maybe we should leave it alone for a while.
Cressy Snr
Amstrad Tower of Power
Posts: 10555
Joined: Wed May 30, 2007 12:25 am
Location: South Yorks.

#43

Post by Cressy Snr »

Neal wrote:....because Steve was already using an adjustable one! :D
It was what I had to hand :D
Neal
Shed dweller
Posts: 2299
Joined: Fri Aug 03, 2007 10:57 am
Location: From the land of the Bodgers

#44

Post by Neal »

andrew Ivimey wrote:The word ( I guess some of you know all this already but) is something like this ...

Touch is programmed by Unix - it is possible to switch bits (screen) and babs(volume control) off and other 'demons' and this improves the sound.

....
Happy reading: http://soundcheck-audio.blogspot.com/20 ... ox-30.html

Can't say I noticed a difference though...
User avatar
andrew Ivimey
Social Sevices have been notified
Posts: 8308
Joined: Mon Jun 11, 2007 8:33 am
Location: Bedford

#45

Post by andrew Ivimey »

In the setup because of the programming involved it was easily possible to swap the Touch back to its factory default but not to quickly swap back again for a close comparison. imho, there was some loss of sound quality. The huge step was the ethernet cable straight into the back of the Touch.
Philosophers have only interpreted the world - the point, however, is to change it. No it isn't ... maybe we should leave it alone for a while.
Post Reply