Author Topic: LAB POWER SUPPLY  (Read 2984 times)

0 Members and 1 Guest are viewing this topic.

Offline glinjikTopic starter

  • Regular Contributor
  • *
  • Posts: 200
  • Country: gb
LAB POWER SUPPLY
« on: January 13, 2020, 06:20:53 pm »
Hi I'm getting the idea of creating a variable lab power supply but instead of using potentiometers I'm thinking of using a variety of fixed volt regulators with a multi position switch to switch the volt regs and thus change the output is this a viable idea or just pie in the sky?
All the truth in the world adds up to one big lie (Bob Dylan)
 

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9964
  • Country: us
Re: LAB POWER SUPPLY
« Reply #1 on: January 13, 2020, 06:34:46 pm »
Sure, this will work well and is a very reasonable approach.  +-15V +12V +9V(?) 5V and 3,3V should about cover it.  The only negative voltage you are likely to need is -15V for op amps and that dual voltage supply could be entirely separate.  I might skip the 9V.

Using regulators like the 78xx for the positive voltages and the 79xx for negative voltages should work well.

The alternative is variable regulators (that could be configured for a fixed voltage) like the LM317 for positive voltages or LM337 for negative voltages would have all the regulators identical with the output adjusted with resistors (or a trim pot).
 
The following users thanked this post: glinjik, Simon_RL

Offline rdl

  • Super Contributor
  • ***
  • Posts: 3667
  • Country: us
Re: LAB POWER SUPPLY
« Reply #2 on: January 13, 2020, 06:49:31 pm »
There's at least one commercial product that does this, but the voltages are selected in 1.5 volt increments to match batteries, so not as useful as it could be.

You could use a single LM317 and switch between various resistors electronically or with a mechanical switch.


edit: Found one

https://www.amazon.com/Philmore-Multi-Voltage-Regulated-Power-Supply/dp/B004M45MHQ
« Last Edit: January 13, 2020, 06:51:29 pm by rdl »
 
The following users thanked this post: glinjik

Offline glinjikTopic starter

  • Regular Contributor
  • *
  • Posts: 200
  • Country: gb
Re: LAB POWER SUPPLY
« Reply #3 on: January 13, 2020, 07:42:07 pm »
hmmm food fo thought I've a lot of different volt regs here like ЕН8б that's en8b in English 12volt fixed reg got lots of different ones here. i suppose the amperage heeds to be the same or at least not exceed the volt reg with the lowest amperage i have 4 identical 220/19v transformers each pushing about 1/2 amp could be used in series parallel course i need a descent schematic a friend donated I think it's a 20 pole mechanical switch Russian military origin thats, what's given me the idea the last power supply is adequate but im not entirely happy has no real oomphh i can charge a car battery but over a looong period the over current protection works well maybe im a born fiddler  :-/O but thank you for replies i think a bit of planning is needed
All the truth in the world adds up to one big lie (Bob Dylan)
 

Offline glinjikTopic starter

  • Regular Contributor
  • *
  • Posts: 200
  • Country: gb
Re: LAB POWER SUPPLY
« Reply #4 on: January 13, 2020, 07:49:12 pm »
came across this sort of idea

 
There's at least one commercial product that does this, but the voltages are selected in 1.5 volt increments to match batteries, so not as useful as it could be.

You could use a single LM317 and switch between various resistors electronically or with a mechanical switch.


edit: Found one

https://www.amazon.com/Philmore-Multi-Voltage-Regulated-Power-Supply/dp/B004M45MHQ
All the truth in the world adds up to one big lie (Bob Dylan)
 

Offline MarkF

  • Super Contributor
  • ***
  • Posts: 2764
  • Country: us
Re: LAB POWER SUPPLY
« Reply #5 on: January 13, 2020, 08:35:10 pm »
I think you will find in general use that you will want
  a +/-15V or +/-12V supply for an op-amp while simultaneously needing
  a +5V or +3.3V supply for digital logic and/or a micro-controller.

I would opt for a +/-12V dual supply for cases when you also need a 12V battery voltage.
  (For 12V lamps, relays, etc.  Not just for amplifiers.)
And then a switchable +5V or +3.3V supply for a MCU.

Think triple output supply for best flexibility.
 

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9964
  • Country: us
Re: LAB POWER SUPPLY
« Reply #6 on: January 13, 2020, 08:51:18 pm »
There's at least one commercial product that does this, but the voltages are selected in 1.5 volt increments to match batteries, so not as useful as it could be.

You could use a single LM317 and switch between various resistors electronically or with a mechanical switch.


edit: Found one

https://www.amazon.com/Philmore-Multi-Voltage-Regulated-Power-Supply/dp/B004M45MHQ


The voltage selections are not useful (other than maybe 12V):
Quote
Multi-Voltage Output: 3 , 4.5, 6 , 7.5, 9 , 12 VD
 
The following users thanked this post: glinjik

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9964
  • Country: us
Re: LAB POWER SUPPLY
« Reply #7 on: January 13, 2020, 08:52:43 pm »
In the uC world, you may find that both 3.3V and 5V are required simultaneously.
 
The following users thanked this post: glinjik

Offline glinjikTopic starter

  • Regular Contributor
  • *
  • Posts: 200
  • Country: gb
Re: LAB POWER SUPPLY
« Reply #8 on: January 14, 2020, 05:22:20 am »
i think anything is possible 4 transformers all identical so possible 4 voltages +/- i definitely need 19v maybe possible to fix the voltages without the need for a selector switch maybe 4 on/off switches to turn on or off the various transformers each with its own mini circuit independent of each other so what i think i need is a fixed voltage +/- schematic say 3.3v, 19v, 12v, and maybe 5v so it would be modular in construction no need for current regulation transformers only put out 1/2a also have a mini variable smps 4-6v i was playing with that could be used for a bit of flexibility maybe alter it to achieve 1v-6v variable below is a little module i made at 12v actually 11.89v very steady just a transformer with a smoothing cap volt reg and output cap although i think that output cap could be bigger the smps below was in a sorry state first time i tried it the output cap blew out the little transistor on top gets super hot so i put this heat sink on it
All the truth in the world adds up to one big lie (Bob Dylan)
 

Offline glinjikTopic starter

  • Regular Contributor
  • *
  • Posts: 200
  • Country: gb
Re: LAB POWER SUPPLY
« Reply #9 on: January 14, 2020, 06:13:32 am »
had a dig through my volt regs i think these 2 should be ok fo the 12 volt part ЕН8Б 12v 1.5a and L7912 12v 1.5a
All the truth in the world adds up to one big lie (Bob Dylan)
 

Offline jeroen79

  • Frequent Contributor
  • **
  • Posts: 529
Re: LAB POWER SUPPLY
« Reply #10 on: January 14, 2020, 06:47:05 am »
In the uC world, you may find that both 3.3V and 5V are required simultaneously.
So instead of a switch you could use multiple bananajacks or bindingposts. One for each voltage plus common.
 
The following users thanked this post: glinjik

Offline glinjikTopic starter

  • Regular Contributor
  • *
  • Posts: 200
  • Country: gb
Re: LAB POWER SUPPLY
« Reply #11 on: January 14, 2020, 07:15:20 am »
Y
In the uC world, you may find that both 3.3V and 5V are required simultaneously.
So instead of a switch you could use multiple bananajacks or bindingposts. One for each voltage plus common.

Hiya thanks yep that's kinda what I'm thinking now 3 jacks for each common pos neg with a simple switch on each set to turn that module on or off independent of each other but it would need a blocking diode on all terminals to prevent accidentally feeding one module into another I suspect that would create a nice meltdown :-DD
All the truth in the world adds up to one big lie (Bob Dylan)
 

Offline glinjikTopic starter

  • Regular Contributor
  • *
  • Posts: 200
  • Country: gb
Re: LAB POWER SUPPLY
« Reply #12 on: January 14, 2020, 07:23:25 am »
this is kinda what im liking although changing out the volt regs for appropriate ones
All the truth in the world adds up to one big lie (Bob Dylan)
 

Offline MarkF

  • Super Contributor
  • ***
  • Posts: 2764
  • Country: us
Re: LAB POWER SUPPLY
« Reply #13 on: January 14, 2020, 08:29:51 am »
I built mine separate. 
But, you could combine them into one case with one switch, one fuse and two transformers.
Each output rated for 1A.

You could also tap off the MC7805 regulator input with a 3.3V regulator for a fourth output.

908530-0
« Last Edit: January 14, 2020, 09:20:08 am by MarkF »
 
The following users thanked this post: glinjik

Offline glinjikTopic starter

  • Regular Contributor
  • *
  • Posts: 200
  • Country: gb
Re: LAB POWER SUPPLY
« Reply #14 on: January 14, 2020, 08:42:56 am »
I built mine separate. 
But, you could combine them into one case with one switch, one fuse and two transformers.
Each output rated for 1A.

You could also tap off the MC7805 regulator input with a 3.3V regulator for a fourth output.

(Attachment Link)

i deffo like that diagram the only problem i have is the transformers do not have a centre tap so i guess 1 transformer for pos 1 for neg unless theres a way to do it without a centre tap? but in any case that would give me 19v+, 12v+- and the smps for 0-6v leaving 1 transformer spare the only thing is getting a neg volt from the smps and getting it to under 1v a challenge and no schematic for the smps :palm:
All the truth in the world adds up to one big lie (Bob Dylan)
 

Offline MarkF

  • Super Contributor
  • ***
  • Posts: 2764
  • Country: us
Re: LAB POWER SUPPLY
« Reply #15 on: January 14, 2020, 09:19:42 am »
Multiple transformers:

908548-0


Alternately, you could use a half-wave bridge rectify.
However with the same value capacitors, the output current would be less
because of the increased ripple.

908552-1
« Last Edit: January 14, 2020, 09:41:57 am by MarkF »
 
The following users thanked this post: glinjik

Offline glinjikTopic starter

  • Regular Contributor
  • *
  • Posts: 200
  • Country: gb
Re: LAB POWER SUPPLY
« Reply #16 on: January 14, 2020, 10:55:06 am »
Thank you that looks good the transformers put 19v to for the 12v so 7v to play with the 19v I want could be on a single transformer that just leaves the smps to be modded to supply neg ? Any ideas?
All the truth in the world adds up to one big lie (Bob Dylan)
 

Offline rdl

  • Super Contributor
  • ***
  • Posts: 3667
  • Country: us
Re: LAB POWER SUPPLY
« Reply #17 on: January 14, 2020, 03:03:50 pm »
Just build two separate 12 volt supplies. It gives you the most options.

 

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9964
  • Country: us
Re: LAB POWER SUPPLY
« Reply #18 on: January 14, 2020, 03:27:43 pm »
Totally separate supplies creates an issue with switching.  Can the project accept one rail going high before the others by some relatively long period of time.  It's pretty clear the FPGAs don't tolerate this kind of thing because they invented power management chips for this very reason.

I would want the isolation that is provided by a transformer design.  In my view, transformerless is out of the question.

When you pick a preregulation DC voltage, consider the implications on the lower voltage regulators.  The preregulated voltage for a 19V supply is probably close to 24V, maybe as little as 22V.  Then think about the poor 3.3V regulator dropping 19V at 1A => 19 Watts.  Can the regulator handle it?  This is a bigger problem with 30V supplies but still worth thinking about.

That's why commercial supplies have transformers with multiple output taps.

For Op Amps, I would definitely prefer +-15V over +-12V because the older op amps just can't get within 3V of the rail and the signal swings are usually +-10V.   Many of the older analog computers use +-10V for their signal swing and +-15V for the rails.  The Comdyna GP-6 is an example:

https://oldcomputermuseum.com/comdyna_gp6.html

 
The following users thanked this post: glinjik

Offline glinjikTopic starter

  • Regular Contributor
  • *
  • Posts: 200
  • Country: gb
Re: LAB POWER SUPPLY
« Reply #19 on: January 14, 2020, 04:25:56 pm »
So the way im thinking 1 transformer 12v+-,  1 transformer 19v+ and the smps 0-6v all separated on their own boards with their own output terminals no common ground except 240v

the first step is to get this smps adjustable to 1 volt or so fist gonna upgrade the caps there only 10v so im thinking 16v caps and double the resistance of the pot
All the truth in the world adds up to one big lie (Bob Dylan)
 

Offline glinjikTopic starter

  • Regular Contributor
  • *
  • Posts: 200
  • Country: gb
Re: LAB POWER SUPPLY
« Reply #20 on: January 14, 2020, 05:31:20 pm »
So on with the smps changed out the caps from 10v to 16v I don't like the tolerance of the old caps the unit puts out 6.3v and caps 10v bit too close also the big filter cap from 200v to 400v the trimmer pot was 1k so I put in a 10k pot that works gets the voltage down to 3.2v so that's enough I think those resisters are a bit cruddy so there going to get changed out all in all I think this will do for the 3-6 v range but no neg volts but I'm not worried about that for now the pot has to be extended out of course as well as the led and the addition of a volt meter is needed also the npn transistor is getting hot that's why I put it on the heat sink but still very hot so extend it onto a bb heat sink but I'm happy now for the schematic of all this
All the truth in the world adds up to one big lie (Bob Dylan)
 

Offline rdl

  • Super Contributor
  • ***
  • Posts: 3667
  • Country: us
Re: LAB POWER SUPPLY
« Reply #21 on: January 14, 2020, 05:45:46 pm »
As long as you use transformers for the two 12 volt supplies, separate is better because it's more versatile. They could even be adjustable. Make something like this, with a 3.3 volt supply added, and you'll have almost everything covered.


 
The following users thanked this post: glinjik

Offline glinjikTopic starter

  • Regular Contributor
  • *
  • Posts: 200
  • Country: gb
Re: LAB POWER SUPPLY
« Reply #22 on: January 14, 2020, 07:05:20 pm »
thats kinda what im aiming at just trying to find a way to neg volts from the smps
All the truth in the world adds up to one big lie (Bob Dylan)
 

Offline glinjikTopic starter

  • Regular Contributor
  • *
  • Posts: 200
  • Country: gb
Re: LAB POWER SUPPLY
« Reply #23 on: January 14, 2020, 07:20:08 pm »
had a little test of the smps voltage runs 3.21v to 6.17 and 1,21a i put a 15w resister across it for amps
All the truth in the world adds up to one big lie (Bob Dylan)
 

Offline Neomys Sapiens

  • Super Contributor
  • ***
  • Posts: 3268
  • Country: de
Re: LAB POWER SUPPLY
« Reply #24 on: January 14, 2020, 10:55:40 pm »
There's at least one commercial product that does this, but the voltages are selected in 1.5 volt increments to match batteries, so not as useful as it could be.

You could use a single LM317 and switch between various resistors electronically or with a mechanical switch.


edit: Found one

https://www.amazon.com/Philmore-Multi-Voltage-Regulated-Power-Supply/dp/B004M45MHQ


The voltage selections are not useful (other than maybe 12V):
Quote
Multi-Voltage Output: 3 , 4.5, 6 , 7.5, 9 , 12 VD
Depends on the task at hand. When doing repairs, I would find those voltage selections very useful, as they correspond to widely used battery combinations.
 
The following users thanked this post: glinjik


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf