Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Variable power supply troubleshooting

Status
Not open for further replies.

Kempachi343

New Member
Hi,

I've just built up a 5-15vdc power supply, which I want to be capable of about 8amps. This is my first attempt at an electronics project, I researched everything for this project off this forum.

I followed the design on the attached diagram, Even tho he is using a 26v input. I used the same components, I used a 0ohm - 20kohm tripot as the voltage adjustment. My input is 17vdc after a bridge rectifier and 4700uF capacitor. I couldn't find 0.1ohm 10w resistors, so I used 5w instead.

When I first tested it out, I had 8.5v no matter what I set the tripot too, I added in a test lead between - and the other leg of my tripot, I didn't solder it in at first because in the spec sheet for LM723 I saw a resistance through there was only for higher voltage circuits, and I wasn't completely sure I was reading the diagram right.

After I'd added this lead in, I had 8.5v - 14v adjustments, I was very happy, and proceeded to try a load on it. I hooked up a pair of motors that I expected would draw around 6A, they ran, although only drawing about 3A, and I could tell the motors had no grunt in them, after about 5 seconds the motors stopped, and my voltage dropped down to about 2v.

The tripot only adjusts around 0.1 of a volt now, yet when I remove my test lead between - and the other leg, I get the 8.5v again. I'm still getting 17v after my rectifier and capacitor, the tripot is still functioning, the resistors are still 0.10hms, the wiring still looks good, the heatsink which I used to mount 2N3055 on wasn't hot at all.

That leaves me with LM723, the 3 transistors, and the 100Pf capacitor. Is there anyway to test these components to find out which has failed? And also can anyone give me some insight as to why this happened?

Cheers,
Daniel.
 

Attachments

  • powersupply.gif
    powersupply.gif
    7.9 KB · Views: 1,981
  • DCP02732.JPG
    DCP02732.JPG
    368.6 KB · Views: 811
  • DCP02733.JPG
    DCP02733.JPG
    372.4 KB · Views: 604
Last edited:
You haven't really mounted the 2n3055s correctly.
This may cause you trouble. The right way is to use a real heat sink with fins. Use two screws, a mica washer and heat transfer silicon "grease". Using to3 sockets is a good idea too.
Some of your solder connections are not military grade :)
Make sure they are good enough.
The transistors can be more or less checked with a multimeter.
Google on how to test transistors. The 100pf is most likely ok.
If the 723 is wired up correctly it is probably still ok but if you continue to have trouble you might get an IC socket and another 723 to play with.

Edit: forgot to mention to use insolating washers on the 3055 screws. You use a to3 mounting kit.

and of course Duffy is right! a 2n2222 will not drive two 3055s.

There are many good power supply circuits on the net.
Look at one using the LM317 instead of the 723.

Good luck!
 
Last edited:
The problem looks like that 2N2222. A crappy old 2N3055 only had a gain of about 15, which means at 8A they are going to need over a half amp each, over an amp total, and the 2N2222 just isn't rated that high.

https://www.st.com/stonline/products/literature/ds/4079/mj2955.pdf

At 10A, the gain can be as low as FIVE. 2N2222 will go pop trying to drive that, may short and the whole thing goes to the rail. That circuit is LOUSY. Look for a better design, preferably something from the last twenty years or so, this one is circa 1970's.
 
Last edited:
Thanks both of you for your helpful replies.

This was a first attempt, and I agree it is quite crude, I think I will start over, with a whole new approach.

I've found this diagram in the LM317 datasheet and made some adjustments to better suit it to my needs, and to me it looks like a much better design then my previous one.

Thanks duffy, I've learned alot from your post. These new transistors have a gain of 20 and at 8amps they will need 400mA or 800mA in total. So I will need a driver transistor with at Least 1A 'max collector current' ? Can you suggest a suitable component?

Thanks flat, I will try a t03 socket with these new transistors, should I mount them on a heatsink away from the PCB, and run wires to the PCB from the socket?

Also on that diagram there is a resistor there that I've marked with a ? I'm not sure what it is, nor what value I'd put in there?

I have also bought a PCB Etch kit, So my next attempt should be much more user friendly.

Cheers all.
 

Attachments

  • Power supply design.JPG
    Power supply design.JPG
    32.2 KB · Views: 2,639
Last edited:
Can you suggest a suitable component?

I've had good experiences with Zetex BJT transistors. The ZXT1053AKTC can handle 5A continuous, has a gain of 300, withstands 75V, low saturation voltage. Drawback is they only sell it surface-mount, but it looks like you could cut a couple of traces and solder it to the back of that board.
 
"Thanks flat, I will try a t03 socket with these new transistors, should I mount them on a heatsink away from the PCB, and run wires to the PCB from the socket?"

If you use TO3 transistors, yes. Also use the insolating mounting kits (mica sheet and washers), and heat sink compound.

If you use SMD parts I don't know what to advise.
 
The SMD is only to replace the 2N2222.
 
Well, I have revised my design, based on the help of another electronics enthusiast and the figures and notes on the LM723 datasheet, but I've still got one problem that I'm stumped on.

I've made the following changes from my first design:

Replaced 2n222 with higher rated transistor (BD135)
Replaced 2n3055 with higher rated transistors (2N3771)
With the higher rated Q1, and the higher Gain of the two base transistors, this should eliminate the problem I had with 2n2222 going pop.

I have moved the potentiometer to where it should be (According to the LM723 datasheet) and added R1 and R2 resistors in (Those values are also what the datasheet recommends for variable power supply's in the 12v range)

Replaced the Freq comp capacitor with 470pF and added a 100uF Capacitor for improved dc output?

So with this new circuit I would be happy with a range between 4v and 12v but Instead I cant get over 7v, no matter what resistor values I try. I'm really stuck here, Anybody with any input on this would be very much appreciated.

Cheers,
Dan.

*Note* Fixed this problem, Am using a 15k resistor between Pin 4 and Vout, and 12k Resistor between Pin 4 and negative, Found I am now getting 1.5V-15V using a 1k Potentiometer. I removed these resistors as per LM723 instructions when adding a variable adjustment to there voltage regulator diagram, turned out to be a big waste of time :)
 

Attachments

  • Psu 1.0.JPG
    Psu 1.0.JPG
    62.1 KB · Views: 1,368
Last edited:
Well, I have revised my design, based on the help of another electronics enthusiast and the figures and notes on the LM723 datasheet, but I've still got one problem that I'm stumped on.

I've made the following changes from my first design:

Replaced 2n222 with higher rated transistor (BD135)
Replaced 2n3055 with higher rated transistors (2N3771)
With the higher rated Q1, and the higher Gain of the two base transistors, this should eliminate the problem I had with 2n2222 going pop.

I have moved the potentiometer to where it should be (According to the LM723 datasheet) and added R1 and R2 resistors in (Those values are also what the datasheet recommends for variable power supply's in the 12v range)

Replaced the Freq comp capacitor with 470pF and added a 100uF Capacitor for improved dc output?

So with this new circuit I would be happy with a range between 4v and 12v but Instead I cant get over 7v, no matter what resistor values I try. I'm really stuck here, Anybody with any input on this would be very much appreciated.

Cheers,
Dan.

*Note* Fixed this problem, Am using a 15k resistor between Pin 4 and Vout, and 12k Resistor between Pin 4 and negative, Found I am now getting 1.5V-15V using a 1k Potentiometer. I removed these resistors as per LM723 instructions when adding a variable adjustment to there voltage regulator diagram, turned out to be a big waste of time :)

I am with you on all the above.
The datasheet makes clear mention to remove those two when adding the adjustment circuit, but the TABLE1 values for voltage range is indeed NOT right, not to my (or your) experience anyway.

My revised board will be completed tonight, I will follow the same rules roughly for determining the resistor values (around the pot).

For those who don't know why he chose 470pf instead of the 100pf which became ever so popular since the 80's, see this article: **broken link removed** (among others)
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top