I am building my first lab and I think I will need a power supply. Was wondering if I can run my thinking through you guys and maybe ask a few questions, please!

1. Variable DC power supply - Budget: under £200.
I am pretty sure I'll need this. I've already looked at some circuits that require 24v.
So, I think a 0-30v supply should serve me well.

a. Do I want a linear or switching? I must say I like the idea of linear being "less noise", in case I want to do anything really sensitive
But finding a linear one without going to fully professional stuff ($$$$) seems impossible!
Is it possible that I could just put more money on a quality switching DC supply and end up with a noise level comparable or better than a cheap linear one?
b. Any product recommendations?

2. Fixed DC power supply - Budget: under £100
Most of the voltages I need are 5v, 3.3v and 12v - which an ATX power supply will provide at a pretty high standard (if it's good for a 7-Billion-Transistor CPU, it's good for my stupid 8 bit stuff).
I also like that a power supply based on an ATX box, will have a few channels. I could run something that needs 12v alongside something that needs 5v.
a. Is there a commercial product that offers a similar solution (3-channel=3.3,5&12, fixed power supply), or is this requirement not so interesting in the real world and I'll be looking to convert an ATX one?

 

As you said, an old Atx psu converted is ideal as most people on here will have one, as do i, also a switchmode psu 0 to 30V @ 10A will cover anything.

 

don't I need a linear PS, to cover low-noise work as well?

 

Build your own linear supply. It is a good way to gain design and construction experience.

Start off with a straight 5VDC 1A power supply.

 

1. Variable DC power supply - Budget: under £200.

Spend way less and save the difference for other equipment (a scope).

I'm thinking about getting one of these from Velleman. Don't need it, but the price is an incentive.

 

Build your own linear supply. It is a good way to gain design and construction experience.
Start off with a straight 5VDC 1A power supply.

+1
Buy a Toroidal with a basic secondary and it is easy to add other overwinds as needed.
I have often done this will P.S. for units in the field.
Max.

 

I reckon I'll ask while it's hot, do you have a reference to a good project (220vac->5vdc linear power supply, designed for low-noise)?
More like looking to get an idea of what a high quality project looks like. I am sure there are lots of ways to get it done and some may be more effective than others.
I'll then study the applicable theory with that project in mind. I don't know a lot about coils but willing to learn- if the result is going to be a product in the same ballpark as a commercial solution.

 

do you have a reference to a good project (220vac->5vdc linear power supply, designed for low-noise)?

Any linear regulator datasheet will have examples. You just need to provide the rectification and filtering part.

You seem to be overly concerned about noise. Why? I have a Heathkit Experimenter with a bi-polar adjustable supply and it's a crappy design with 150mV of ripple on the output. That has only caused me a problem once. I was testing a circuit that was sensitive to the supply noise and didn't work. After scratching my head awhile, I finally got around to checking the power rails. Switched to a better supply and moved on.

 

Here is your basic 5VDC PSU circuit.
Use an LM309K 5V regulator in a TO-3 package.




You will need to include the following:

metal case
power cord with 0.5A fuse
power cord strain relief
on/off power switch
power indicator (LED and 1kΩ resistor at the +5V output)
two banana posts (red and black)
third banana post (green) if you want a floating supply

 

"You seem to be overly concerned about noise. Why? "
It's a character defect. It does get occasionally out of hand.
Thanks for the ripple example. It helps put things in perspective.

 

Some details about the circuit in #9.

It looks like a 12VAC CT transformer. That will give a peak voltage of 1.414*6V-0.7V=7.8V on the filter cap.

Regulators like LM7805 can have a drop out voltage of more than 2V (datasheet says 2V typical). So the circuit could be very sensitive to input ripple.