Quickie power supply advice?

Help Support GroupDIY:

sircletus

Well-known member
Joined
Nov 19, 2008
Messages
165
I'm doing an op-amp project with a textbook ±17V supply and was wondering how you guys would recommend BEST,  most-efficiently (in terms of cost, component count and energy wasted) doing an additional 5V supply for logic without adding another transformer (I've already got a toroidal guy with two 18V secondaries lying around that I'm going to use).  I could obviously grab the post-bridge-rectified, pre-regulated positive rail and just dump it into a 7805, but I hate wasting that much voltage and generating that much heat.  Is it stupid to do a resistive divider to step it down and then regulate?  Or a zener to bump it down to 9-ish volts and then regulate?  Or how 'bout this absurd one: grab one leg of one of the 18V secondaries, half-wave rectify it with a beefy power diode and huge reservoir cap then regulate?  That's probably a really dumb idea.
 

ruffrecords

Well-known member
Joined
Nov 10, 2006
Messages
13,771
Location
Norfolk - UK
All of those options waste exactly the same amount of energy which will appear as heat. You could try a small dc/dc converter fed from the unregulated +ve supply

Cheers

Ian
 

sircletus

Well-known member
Joined
Nov 19, 2008
Messages
165
ruffrecords said:
All of those options waste exactly the same amount of energy which will appear as heat. You could try a small dc/dc converter fed from the unregulated +ve supply

Cheers

Ian

Hadn't even thought of that.  But here's the real question, Ian, what would YOU do?  :)
 

Tubetec

Well-known member
Joined
Nov 18, 2015
Messages
3,125
Is your toroid open in the middle or  epoxied ?
Maybe you could add a low voltage winding , you can easily determine the volts per turn and work out how many turns you need .
Then you'll need to make up a wooden shuttle , wind with the appropriate gauge of magnet wire ,but it has to be able to pass  through the centre  of the toroid  so you can add turns.
 

JohnRoberts

Well-known member
Staff member
Moderator
Joined
Nov 30, 2006
Messages
21,019
Location
Hickory, MS
this is a classic old problem... the old solution was a 5V winding on transformer.

Back in the 80s I did a trick power supply that pulled a unregulated supply for 5V rail from the 20v winding with a swich (transistor) that stopped charging the cap at 10V. This did not result in the conventional inefficiency of a pass element dropping it  down to 5V. But was more complex than typical PS designs.

A this century solution is single rail with DC-DC switching supplies for +/-15V and +5V...

JR 
 

sircletus

Well-known member
Joined
Nov 19, 2008
Messages
165
JohnRoberts said:
this is a classic old problem... the old solution was a 5V winding on transformer.

Back in the 80s I did a trick power supply that pulled a unregulated supply for 5V rail from the 20v winding with a swich (transistor) that stopped charging the cap at 10V. This did not result in the conventional inefficiency of a pass element dropping it  down to 5V. But was more complex than typical PS designs.

Verrrrrry interesting, JR.  I'm making a little line-level thing for a friend and I'm trying to keep costs down,  but  I  have a nasty habit of getting hung-up on "the best" way of doing things.  There are so many options and ways to design things that I assume there MUST be one way that's better than all the others...
 

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
13,036
Location
Marcelland
sircletus said:
There are so many options and ways to design things that I assume there MUST be one way that's better than all the others...
Best depends very much on how much time you're willing to spend on this.
Is energy waste a serious issue? It becomes an issue when the global consumption approaches the xfmr's power rating and/or when heat becomes hazardous. Anyway, the only way to dispense with energy waste is an smps.
 

sircletus

Well-known member
Joined
Nov 19, 2008
Messages
165
abbey road d enfer said:
Best depends very much on how much time you're willing to spend on this.
Is energy waste a serious issue? It becomes an issue when the global consumption approaches the xfmr's power rating and/or when heat becomes hazardous. Anyway, the only way to dispense with energy waste is an smps.

Fair enough!  Was looking at Meanwell supplies but find myself wishing they made either a.) one ±17V or b.) trimmable bipolar outputs.
 

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
13,036
Location
Marcelland
mjrippe said:
Why does the textbook specify 17v?  Could it be 15 or 18?
Because it's the max voltage that can be safely applied to the device and guaranteed by the mfgr. Indeed, it could probably be powered by anything from +/-5 to +/-20 with different consequences ranging from reduced headroom to excessive heat dissipation.
 

moamps

Well-known member
Joined
Sep 24, 2005
Messages
1,835
Location
Croatia, HR
@sircletus
Maybe you can use CMOS ICs  for the logic (just additional 7815 for supplying) and  relays rated to 24V (supplied directly from a smoothing capacitor). 
 

mjrippe

Well-known member
Joined
Dec 27, 2011
Messages
2,575
Location
Brooklyn, NY
abbey road d enfer said:
Because it's the max voltage that can be safely applied to the device and guaranteed by the mfgr. Indeed, it could probably be powered by anything from +/-5 to +/-20 with different consequences ranging from reduced headroom to excessive heat dissipation.

Quite right, Abbey.  My poorly worded question was meant to ask sircletus why they were fixated on having exactly 17v.
 

Whoops

Well-known member
Joined
May 9, 2008
Messages
5,720
Location
Portugal
sircletus said:
I could obviously grab the post-bridge-rectified, pre-regulated positive rail and just dump it into a 7805

That's how was done in a lot of different circuits, quite an easy solution will work great and problem solved
 
Top