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.

Overheating power supply project

Status
Not open for further replies.

vwchico

New Member
Hi I'm new here and am a hobbiest

I'm building a variable power supply 0-30v 0-3A. I'm using the trust LM723 and 2N3055 setup. But I have an issue with the 2N3055 transistors they overheat big time if a put a small load on it say 12v 1amp. Within 1 minute it becomes untouchable. So I have added 3 attitional more transistors in parallel.

But now I have them unbalanced, one seems to over heat while others stay cooler. I have put 0.1 ohm resitors on the emmiter of all the transistors. How do I solve this and get them balanced.

What I did try is to use only 3 instead and cut off the over heating one. But still when I put a load on of 12v 1amp. They still cook, almost untouchable after 5min. This is normal how can i prevent this is this a circuit problem ? Because I've seen lots of circuits only using one transistor and i'm using 4 :shock: but it still cooks.
 
How large a heatsink are you using?, and what voltage is the incoming DC to the 2N3055's?.

Assuming 40V DC input (as you specify 30V out), 12V output and a 1A load will dissipate 28W in the transistors - this is quite a bit of heat!.
 
yup, that is the problem with linear power supplies.
You are ogin gto need some big heatsinks and paralleling (as you have tried)

but oyu have found a ballancing problem

Try having a loop for each three output transistors thus each control loop will try to maintain the correct voltage/ampage independant of each of the other 2


as Nigel stated:

30V in and trying to get 12V out will result in the trasistor dropping (30 -12) = 18V, thus 18W !!!!

if it is as Nigel says 40V, then 28W

That is alot for one to dissipate
 
Nigel Goodwin said:
How large a heatsink are you using?,

Probably none, judging from his posting. The power supply is giving out +100W heat max so he would likely need a really big heatsink where he can comfortablely mount four 2N3055s.

The emmiter current balancing resistor value is a bit small. A value of 0.5 ohm would be good in his case. But I don't think LM723 is capable of driving four 2N3055s. He would also need a buffer transistor in between.
 
Hi thanks for the info,

Ye si do have a FAT heatsink on it and this doesn't seem to help seems the longes i can run the power supply then is 5 minutes just doesnt seem normal.

There is a transistor inbetween.

Do you think the 05. ohm resistor will solve my balancing problem, i don't understand the loop theory explained.

Yes correct it is about 40volts, but becuase i have 4 transisotrs surely they shouldn't heat up as much as they currently are?

I've never owned a proper variable power supply but do these get as hot when used or not ?
 
vwchico said:
Ye si do have a FAT heatsink on it and this doesn't seem to help seems the longes i can run the power supply then is 5 minutes just doesnt seem normal.

Have you put any heatsink compound (the white stuff) between 2N3055, mica insulator and the heatsink?

vwchico said:
Do you think the 05. ohm resistor will solve my balancing problem,

It's 0.5 Ohms, not 0.05 Ohms. Yes, it probably will balance out the current.

vwchico said:
i don't understand the loop theory explained.

Give it some times. Don't worry too much at this moment.

vwchico said:
Yes correct it is about 40volts, but becuase i have 4 transisotrs surely they shouldn't heat up as much as they currently are?

The amount of heat given off is the same, just shared by four transistors instead of just one.

vwchico said:
I've never owned a proper variable power supply but do these get as hot when used or not ?

Yes, linear regulated supply do get hot when working at low voltage and high output current. A large and good heatsink will help a lot in this case.
 
On another electronics chat site is a 0V-30VDC, 3A variable supply project that melts its driver transistor and doesn't produce anywhere near its rated 30V at 3A. I have modified its circuit to allow it to work correctly and now its two 2N3055 output transistors dissipate 60W each at max. It needs a huge heatsink and a fan for such high power.

Modifying the modified supply to produce 5A made it require three 2N3055 transistors, an even bigger heatsink and more air velocity from its fan.

Adding more output transistors doesn't harm the driver transistor but actually helps it. The current gain of a 2N3055 transistor is very low at a few Amps and more. When more are added then they share the total current, reducing each one's current to 1A or less where their current gain is much higher.
 

Attachments

  • 2n3055_185.png
    2n3055_185.png
    19.2 KB · Views: 855
Hi Audio,

A CT transformer and a simple switch to select either full-wave or bridge rectifier configuration would immediately cut the heat generated by half when large current is required at low output voltage. Just wonder why so very few project designers include this simple arrangement.
 
Hi Ebic,
Yeah, I saw one power supply project that automatically switches to the lower voltage tap on its transformer when its voltage is adjusted lower at high currents. :lol:
 
BTW, my new PC revs up its fan when it does some intensive number-crunchin'. It happens each time Norton greets me: "Norton is pleased to announce that a virus has been found on your PC.". It's rated for a few hundred Watts but I never feel hot air blowing from it.

For the very 1st time, Norton greeted me like that when I turned-on my PC today. It found the Backdoor.Graybird virus and couldn't do anything about it. I updated their software (twice, a new update was released as it was installing the 1st). I did a total scan but it couldn't find it. I printed all their instructions for me to eliminate it but I couldn't find any of their listed files for it on my hard-drive. Maybe the virus jumped onto a post I made today. Maybe this post. :lol:
 
Yes, your virus has turned your "L" key into the "i" key. Twice have you called me "Ebic" instead of "Eblc". :lol:

Its time to switch to FireFox if you haven't done so yet. Highly recommended.
 
eblc1388 said:
Yes, your virus has turned your "L" key into the "i" key. Twice have you called me "Ebic" instead of "Eblc".
He, he. :lol: :lol:
Sorry Eblic, oops Eblc, my new tri-focal glasses slipped down my nose! :lol: :lol:
When positioned properly they are so perfect that I can see through clothing!

Once again, I turned on my computer and got a shocking greeting from Norton about my virus. It says where the file is located but a search doesn't show it. I think my Norton anti-virus program is sick. :(
 
Hiya Audioguru,
Eh mate I've given up on those virus scanners that seems to have an on going cost and instead opted for AVG free virus scanner. Ever since I tried out their beta program and since upgraded to the new version free I haven't had a virus since ( Touch George Bush's head)(( oopps I mean wood)). I allow the program to check for updates daily and like I say why pay for an inferior bit of software when you can have better software for free. I have suggested this software to quite a few friends and everyone has given me a pat on the back for putting them onto it.

Hope this helps

Cheers Bryan :D
 
Thanks Bryan.
I found their Professional virus scanner on Google for $20.00 (US?) per year. I didn't see and I wouldn't trust a free one.
 
why dont you try this one.....replace the fixed regulator with an LM317T and you will have an adjustable supply at 30 amps...make sure you use a large heatsink,plenty of heatsink compound and a fan would also help as well.

regards mike
 

Attachments

  • 12v30a.png
    12v30a.png
    11.6 KB · Views: 818
Last edited:
Status
Not open for further replies.

New Articles From Microcontroller Tips

Back
Top