Common Secondary Winding for Two Rectified Supplies?

[Unlikekurt]

Hi All,

I have a bipolar linear power supply derived from a center-tapped secondary winding with a bridge rectifier and reservoir capacitors.  Within the same project I also have a positive supply node that would be better of if it weren't influenced by the state of the main positive supply's reservoir capacitor.  It is a fairly low current node with a draw of abut 20mA.  I started by trying regulation and it helped, but it would still be better if it were isolated from the main reservoir cap.
The bias supply of a guitar amplifier without a bias tap on the secondary came to mind and I started wondering.  Would adding a half wave rectifier to one end of the secondary winding have any ill impact on either the transformer or the B+/B- supplies?  As it would be sharing the same common as both of the main supplies it wouldn't be a threat of shorting on either AC interval.  Is there anything else I'm missing?  Any sage advice or input would be helpful.
Thanks!

[antonis]

What is the current of +/- VDC1..??

[Unlikekurt]

Antonis,

The winding itself is very beefy with plenty of headroom.  That said, it is probably about 3A max for each supply, but not continuous by any means.
The additional supply is very low current; probably a few 10s of mA

[antonis]

Nornally, you shouldn't face any issue.. 
(not even in case of  VDC2 taken right from VDC1 'cause positive supply's reservoir cap should be able to handle voltage ripple due to extra 20mA current..)

[Unlikekurt]

Antonis,

Out of curiosity now, why would the additional current from the half wave rectifier cause an increase in ripple to the output of the bridge rectifier?  The extra current pull from that half of the secondary winding?

[antonis]

I didn't mean that..

I was talking about the additional current drawn from VDC1 in case of VDC2 was part of it..
i.e. -VDC1 for negative supply, as seen between 4 & GND and +VDC1 plus +VDC2 for possitive supply, as seen between 1 & GND..
(better 0V instead of GND, terminologically speaking..)

edit: Now I got you..
Short answer is YES.. 
Long answer is YEEEEESSSSSS... 
Detailed answer as below:



Where I is the DC load current in amps, ƒ is the frequency of the ripple or twice the input frequency in Hertz, and C is the capacitance in Farads.

You can see that I(load) of 2.02A (VDC1 +VDC2) makes no difference compared to 2A (VDC1)

[Unlikekurt]

Antonis,

I think what I'm failing to understand is that as VDC2 is not sharing a filter capacitor with VDC1, why would it have an impact on it?

[antonis]

As you've designed it, it hasn't..!! 

The whole assumption above is made for the case of getting VDC2 straight from VDC1..
(no extra diode & smoothing/reservoir cap..)

[Unlikekurt]

Haha so, is it permissible then to do as i designed?  Or is there some overlooked item aside from making sure it doesn't put too much current demand on the transformer?

[antonis]

Despite bad regulation due to half-wave rectification, you may proceed on your own risk..

[Unlikekurt]

thanks antonis.  i fixed the broken link to the image of the schematic for posterity.

[Ben N]

The bias supply of a guitar amplifier without a bias tap on the secondary came to mind and I started wondering. 

I'm guessing you meant "without a separate secondary bias winding", right?

[Unlikekurt]

Ben,

Nope.  Although lots of amps (many Marshall's come to mind) have a separate winding for the bias supply, plenty of amps have a tap coming off the HT winding at a lower voltage than the full HT. 

And of course, then there is the method that came to mind, when neither of the above are available: a half rectifier employed at one end of the HT winding with a suitable dropping resistor to bring the voltage into more manageable range before the bias circuit.  You might also see an additional full wave employed in this situation, but it is far less common.