home made desktop power supply - diyAudio
Go Back   Home > Forums > Design & Build > Equipment & Tools

Equipment & Tools From test equipment to hand tools

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 9th October 2011, 01:01 AM   #1
akis is offline akis  United Kingdom
diyAudio Member
 
Join Date: Jul 2008
Default home made desktop power supply

In the past few weeks I have built the power supply below. I have tested it to 25V and 5.5A. With an appopriate transformer it can do up to 35V.

There are two channels one positive and one negative. Voltage regulation is done with the LM317BT and LM337BT. To achieve currents in excess of 1.5A I used a bypass transistor (MJL3281A and MJL1302A).

Current limiting is performed with a .25R sense resistor and an op-amp (MC33072).

R1 slowly discharges the 10mF capacitor.
R6 presents a load on the LM317/337 otherwise the output may rise. This can be improved with a constant current source so that the current remains constant througout the range but I had already run out of board space.
R8 and R9 limit the very high voltages (both can be omitted if we are confident of no mistakes during assembly).
R14 sets a minimum level for current limiting to avoid a "dead-zone" on R13 due to the first diff op-amp having a minimum working level (eg 100-700mV). On the negative channel, there are some differences. There the op-amp "idles" close to 0V which, as far as the op-amp is concerned, is equivalent to max supply rail. The best we can get there is 500mV-1500mV off the top rail, so I used D17 to raise the idle voltage past this level. Actually in my circuit I had to add 4 * 1N4148 in series. Ideally R14 should be a trim-pot around 10K.

C16 is needed to avoid high frequency oscillations due to R26. I used 68nF randomly (it seemed to stop the oscillations), other values will also work.

The transistors are MJL3281A and MJL1302A. I mounted them on a large-ish heatsink, have not tried in anger yet to see how hot it will get. Adding a fan would be a great idea if we plan to pull lots of amps on a regular basis.

The rectifier diodes 1N5402 are only 3A - a 6A version is needed (I am running mine on the 3A ones but do not draw 5A regularly. I will replace with 6A diodes as soon as possible).

I have also added a 6.3A fuse just before the rectifier diodes and there is a 3A-5A fuse on the mains socket to protect the transformer (in my case 2x25V, 300VA).

I also added LED panel meters for voltage and current measurement. These most likely need to be powered from independent power sources, independent of the two main channels but also independent between the panel meters. In my first attempt I did not realise that a common power supply for the panel meters results in shorting the two channels so that they cannot be used in series to provide +/- voltage output with a single common. To provide power for the LED panel meters (they need 5VDC at 120mA) I wound 15 turns of 4 wires around the toroidal to get 5VAC (times 4) and built a simple rectifier for each.
Attached Images
File Type: jpg bench-psu-oct-2011-pos.jpg (133.9 KB, 389 views)
File Type: jpg bench-psu-oct-2011-neg.jpg (142.6 KB, 376 views)
  Reply With Quote
Old 9th October 2011, 01:07 AM   #2
akis is offline akis  United Kingdom
diyAudio Member
 
Join Date: Jul 2008
C11 is not 470pF, probably tens of nanos, I have to open the box and look it up. The reason is long wires running to the pots on the front which may introduce instabilities. The diagram also shows the COMMONs earthed - this is just to get the simulation running, there is no earth connection whatsoever.
  Reply With Quote
Old 11th October 2011, 05:37 PM   #3
akis is offline akis  United Kingdom
diyAudio Member
 
Join Date: Jul 2008
Some pictures of the finished product.

There is a simple wooden box made of plywood. The heavy transformer (300VA) and boards are mounted on another piece of plywood and the whole tray slides in and out of
the box. The front and back panels are etched fibreglass PCBs, the front is sprayed painted with Christmas gold decoration paint.

Only the voltmeters show, the am meters have not arrived yet.

A detail on the transformer I wound 4 extra secondaries to obtain 5VDC to power the 4 panel meters with independent power supplies.
Attached Images
File Type: jpg 20111011_170742.jpg (285.9 KB, 334 views)
File Type: jpg 20111011_171444.jpg (315.2 KB, 314 views)
File Type: jpg 20111011_171629.jpg (398.7 KB, 308 views)
File Type: jpg 20111011_171550.jpg (343.4 KB, 144 views)
  Reply With Quote
Old 11th October 2011, 06:22 PM   #4
sbrads is offline sbrads  United Kingdom
diyAudio Member
 
Join Date: Jun 2005
Location: Kent, UK
The only thing I would change (because I've blown up a lot of circuits) is the 100uF across the output. I would use something that stores a lot less energy, like 10uF or even 4u7 if it's still stable.
  Reply With Quote
Old 16th October 2011, 01:08 PM   #5
akis is offline akis  United Kingdom
diyAudio Member
 
Join Date: Jul 2008
Yes, but it is not just the output caps that are problematic. Behind the regulator sits a 10,000uF capacitor. A short at the output is catastrophic.

One of the output transistors blew yesterday and I had to replace it. This is a 200W device capable of withstanding 16A. I do not know how it died, but I suspect I shorted the output accidentally (I was just switching the PSU on - the transistor did not die of increasing heat over time). The 6.3A fast fuse on the transformer's secondary blew too. But it did not protect the output transistor.

So we need something to protect from shorts at the output, both the PSU and the equipment that might be causing the short. A 6A fuse might be helpful here?

In addition I need something to disconnect the output (open circuit) when I switch the PSU off - currently it takes a long time until the voltage drops. I am thinking of a relay. The same relay might also be employeed instead of a fuse to protect from shorts.

Is that a good idea ?
  Reply With Quote
Old 16th October 2011, 01:48 PM   #6
sbrads is offline sbrads  United Kingdom
diyAudio Member
 
Join Date: Jun 2005
Location: Kent, UK
Your current limit circuit can't be working good enough. It pulls down the LM317 o/p to effectively a shorted 1.25v in the event of a s/c o/p and it doesn't do much to limit Q2 current which looks like it's turned hard on by the LM317 conducting from Vin to Vout. Perhaps you need a transistor bypass circuit more like the one shown on page 8 of National Semi's LM317 datasheet as that version monitors the transistor current. The LM317 doesn't need protecting, it has its own internal s/c protection. A relay or fuse for longer term protection is a good idea.
  Reply With Quote
Old 20th October 2011, 07:54 AM   #7
akis is offline akis  United Kingdom
diyAudio Member
 
Join Date: Jul 2008
Yes. I am certain the 10mF capacitor provided enough current to blow a 16A transistor. I did not even see a spark, no idea how it happened. As a temporary fix, I have added a 6.3A fuse at the output, less than half of the 16A of the transistors. However if I blow it, then I will have to open the box to replace (a palavar). So I will design a relay circuit that will cut off over a certain current. Maybe also cutoff when you flick the power (as an added bonus).
  Reply With Quote
Old 20th October 2011, 05:45 PM   #8
sbrads is offline sbrads  United Kingdom
diyAudio Member
 
Join Date: Jun 2005
Location: Kent, UK
I've had another think and if it were mine I'd just try a 0R1 or thereabouts in Q2 emitter and put 2 forward biased diodes in series from base to the far end of this new emitter resistor. It should limit the s/c current to 6A or so.
  Reply With Quote
Old 21st October 2011, 04:56 PM   #9
Elvee is offline Elvee  Belgium
diyAudio Member
 
Elvee's Avatar
 
Join Date: Sep 2006
Quote:
Originally Posted by akis View Post
Yes, but it is not just the output caps that are problematic. Behind the regulator sits a 10,000uF capacitor. A short at the output is catastrophic.

One of the output transistors blew yesterday and I had to replace it. This is a 200W device capable of withstanding 16A. I do not know how it died, but I suspect I shorted the output accidentally (I was just switching the PSU on - the transistor did not die of increasing heat over time). The 6.3A fast fuse on the transformer's secondary blew too. But it did not protect the output transistor.

So we need something to protect from shorts at the output, both the PSU and the equipment that might be causing the short. A 6A fuse might be helpful here?
A fuse is a poor solution, it is too slow, and bipolars do not have I˛t ratings.

The root of the problem is the immense current regulation loop: you have 3 IC's, of which two are integrators, plus two transistors.
The time between the appearance of a stimulus at one end and the turn off of the transistor at the other is sufficient for the transistor to die 10 times.
In addition, the turn-off is not complete because of the 317.

The only good solution is to redesign the whole thing from scratch, but if you just want to avoid blowing your transistors at the drop of a hat, you could add a fast and crude limitation directly at the transistor level.

You can use the 0.25 ohm sense and connect a PNP to short the base when its Vbe is exceeded (you can use a resistive divider to adjust the current.

The transistor has to be hefty, as it will have to withstand the 317 short-circuit current.

Or you could simply use a number of diodes, as suggested by sbrads.
__________________
. .Circlophone your life !!!! . .
♫♪ My little cheap Circlophone© ♫♪
  Reply With Quote
Old 21st October 2011, 05:22 PM   #10
akis is offline akis  United Kingdom
diyAudio Member
 
Join Date: Jul 2008
Very good idea. Will do it next time I blow a fuse
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
home-made power supply AKHeathen Equipment & Tools 2 27th September 2011 11:33 PM
50 Hz in home made power amp zoubour Solid State 13 7th January 2011 07:06 AM
Home Made Tubes?? has anyone made one, is it possible?? John Biles Tubes / Valves 7 11th December 2005 01:29 AM
Home power supply for car amp? Ignite Solid State 9 14th June 2002 11:04 PM


New To Site? Need Help?

All times are GMT. The time now is 08:48 PM.


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