AC/DC power supply

Thread Starter

phantomp

Joined Nov 3, 2006
8
Hi,

I'd like to build a bench power supply and feel comfortable with building a variable regulated DC supply, however it might be a bit more interesting and economical to attempt a dual output one that can provide AC or DC current, between 1.5v and 12v at 3-5amps. If someone can offer any warnings or pointers to get me heading in the right direction, i'd be most grateful!

A few of my thoughts on how to aproach the project:

Use a multi-tapped transformer - though i haven't found a suitable one.

Find a second-hand variac and build a separate transformerless DC module. (i like the sound of this, though variacs are a tad pricey!)

(this is a bit of a wild one) I've noticed that when fed a half rectified signal the LM338 chip will adjust it according to the ref voltage as with DC. There are also versions of the chip that handle negative potentials. So is it possible to adjust the voltage of each half of the AC signal separately then draw the output across both? If possible i could maybe get a variable supply with a simple 12-0-12 transformer.

It might be prudent to employ some sort current/voltage restrictors (a variable circuit breaker would be good).

Sorry if this is a bit of a noobish post but i've not got to the AC section yet but need to do some tinkering in the meantime!
 

hgmjr

Joined Jan 28, 2005
9,027
Phantomp,

I don't know what your experience level is so I will just throw this suggestion out there.

One approach you could consider would be to purchase a power supply in kit form.

I found this inexpensive power supply kit at Jameco as an example of what is available.

The advantage of working from a kit is that all of the design has been done already. You can then concentrate on understanding what goes into the design of a power supply.

Once you have successfully completed the construction and get the supply up and running, you will have a bench supply suitable for continuing your exploration of the fascinating world of electronics.

It's true that this is a DC-only power supply but it is a good place to start.

hgmjr
 

Thread Starter

phantomp

Joined Nov 3, 2006
8
Hi.
Thanks for your advice, it's a good project and i hear what your saying, but i really wanted to exersice my theory by designing a circuit as well as practice printing and etching. Though what i was aiming for was a tad optomistic!

Yes im pretty inexperienced but know enough to understand the circuit in the project you linked to and have multisim for testing out designs.

I need a 300mA 9VAC supply anyway and would also like a DC supply for testing. What are your thoughts on building a LM338 based rectifier/regulator rated at 5A as a stand-alone module and using an external transformer? This way i can keep everything simple and start with a 6VA 9-0-9 transformer that can be upgraded if needed.

In addition to a fuse it'd be good to add a variable circuit breaker to the supply but im not sure if avaliable as PCB-mounts.
 
A variac makes for a nice bench top supply for nearly any AC voltage that you require. AC devices like to use equal amounts of power in the positive and negative polarities. The amount of power you need will be the VA rating for the device because P=V*I. If you need 12 volts at 5 amps then 60 VA is required. Pick the next higher rating (i.e. 100 VA or 0.1KVA) for a well regulated voltage output. If you attach a load that uses 5 amps, it is likely that a 60 VA transformer output will drop from 12 volts to something less.

For a variable DC supply, first turn the AC into pulsed DC (with diodes), filter it (with R, L, C components of your choosing), and then dispense it using a common emitter - voltage following transistor circuit to send out only what you need from the stored DC energy.
 

Thread Starter

phantomp

Joined Nov 3, 2006
8
Again, reminding potentual corpse's that many Variacs are NOT isolated from the mains.
:eek:

Thanks, i was just thinking that using a variac or multitap before the rectifier would be a better way of attaining a range of voltages from a linear supply rather than dropping the DC voltage in a regulator.

Anyhow i've got some bit's and pieces together to make a +/- 12v supply similar to the jameco kit with a 10VA transformer from an orphaned psu.

under simulation the output seems to be identical with 1000uF smoothing caps instead of 2200uF ones, is there any reason why they must be 2200uF?

Also i've added a 30v moving coil meter to the circuit:
http://www.rapidonline.com/productinfo.aspx?tier1=Tools%2c+Fasteners+%26+Production+Equipment&tier2=Test+Equipment&tier3=Moving+Coil+Meters&tier4=Non-RoHS+Moving+coil+meters&moduleno=75577
The trouble is the coil i have found has a low impedence (57Ohm) so will effect the output. I was wondering about replacing the control pots with a set of resistors on a rotary switch so that exact voltages can be selected but a meter would be better. I guess a less conductive coil would do the trick, is there such a thing?
 

Attachments

n9352527

Joined Oct 14, 2005
1,198
You need to set a heavier load under simulation to see the difference between 1mF and 2.2mF smoothing caps. Probably more than 1A.

Your meter is probably a current meter, with deflection proportional to the current passing through it. That is why the resistance is so low. However, the fascia is calibrated with voltage marks. To use it, just place a resistor in series with the meter, choosing an appropriate value and tolerance. E.g. If the meter FSD is at 100uA, then for 30V FSD the resistance is R=30V/100uA=300K
 

Thread Starter

phantomp

Joined Nov 3, 2006
8
You need to set a heavier load under simulation to see the difference between 1mF and 2.2mF smoothing caps. Probably more than 1A.

Your meter is probably a current meter, with deflection proportional to the current passing through it. That is why the resistance is so low. However, the fascia is calibrated with voltage marks. To use it, just place a resistor in series with the meter, choosing an appropriate value and tolerance. E.g. If the meter FSD is at 100uA, then for 30V FSD the resistance is R=30V/100uA=300K
Ah thanks 100uA sounds much better than 500mA! Strange it's not in the datasheet though, only resitance and voltage is stated.
 
Top