diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Power Supplies (http://www.diyaudio.com/forums/power-supplies/)
-   -   Request help with odd looking CCS fed shunt reg circuit and load current ability. (http://www.diyaudio.com/forums/power-supplies/230259-request-help-odd-looking-ccs-fed-shunt-reg-circuit-load-current-ability.html)

Ceglar 18th February 2013 01:45 AM

Request help with odd looking CCS fed shunt reg circuit and load current ability.
 
1 Attachment(s)
Hi,

I understand the BF245A CCS, and without source resistor it runs at IDSS which will be 2 to 6.5mA, depending.

The circuit supplies 5VDC at 25mA.

Obviously not all of the load current comes via the CCS and I'm used to CCS fed shunt regs for tube circuits where the CCS supplies ALL of the current for both the shunt and the load.. so I'm a bit lost.

The question is, if I change the load to 5VDC at 1mA, will it still function without changing any other values?

This isnt for some mental exercise, I'm just needing to use some existing supplies for a different application that require considerably less current.


Thanks,
Shane
Attachment 331169

DUG 18th February 2013 04:53 AM

The 1K resistor already draws 5mA.

You could plot the load regulation and decide then:

add 5K load, measure V

Repeat 24 times.

Have fun.

DF96 18th February 2013 09:41 AM

The Darlington pair needs to supply at least 2mA (or thereabouts) in order to switch on the second transistor. If not, the output voltage will rise by about 0.6-0.7V. Your 1k load is already taking 5mA, so all is fine. The circuit is overkill for supplying 1mA, but as you already have it it won't do any harm.

Ceglar 18th February 2013 11:35 PM

Thanks a lot, much appreciated.

Shane

Ceglar 19th February 2013 11:07 PM

Quote:

Originally Posted by DF96 (Post 3374650)
The Darlington pair needs to supply at least 2mA (or thereabouts) in order to switch on the second transistor. If not, the output voltage will rise by about 0.6-0.7V. Your 1k load is already taking 5mA, so all is fine. The circuit is overkill for supplying 1mA, but as you already have it it won't do any harm.

How to determin the maximum current the supply can provide, assuming the transformer winding is not the limiting factor?.

Would 50mA seem ok?


Thanks,
Shane

DF96 20th February 2013 09:12 AM

Maximum current depends on transistor ratings and heatsinking, and minimum voltage due to ripple (which depends on the size of the main reservoir capacitor).

Ceglar 20th February 2013 08:12 PM

I'll check the data sheet.. I can say that with 25ma the BD139 does get a bit warm.

Seems the idea with this scheme is to establish a voltage reference and then buffer its output with the darlington pair and avoiding the feedback inherent to a series ccs to shunt regulator circuit?

I wonder if theres much to be gained by using two inductors, one in the +ve rail after the ccs, and another in the ground cct between the zener and the 1k//cap, to help decouple it from the rest of the power supply and have better control of the current loops between the load and the final cap in the reg?

Application has 6 of these in a D/A converter cct.

Thanks,
Shane

DF96 20th February 2013 09:12 PM

Quote:

Originally Posted by Ceglar
Seems the idea with this scheme is to establish a voltage reference and then buffer its output with the darlington pair and avoiding the feedback inherent to a series ccs to shunt regulator circuit?

Yes, although there is heavy feedback in the follower.

I would not add inductors unless you have identified a problem they might solve.

Ceglar 21st February 2013 12:59 AM

Quote:

Originally Posted by DF96 (Post 3378723)
Yes, although there is heavy feedback in the follower.

I would not add inductors unless you have identified a problem they might solve.

Curious. Is the feedback associated with the follower of a different order than what would normally be associated with ccs fed shunt in the usual sense. (ie. where exactly is the benefit?).

Thanks again,
Shane

DF96 21st February 2013 11:13 AM

Feedback is feedback. However, within a single stage it may be easier to keep stable. Curiously, a follower is an exception to this: followers can oscillate at VHF/UHF frequencies if they see a small capacitive load. A big, or lossy, capacitive load such as an electrolytic is usually OK.


All times are GMT. The time now is 05:36 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2