Author Topic: variable power supply circuit  (Read 5200 times)

0 Members and 1 Guest are viewing this topic.

Offline aneeshTopic starter

  • Contributor
  • Posts: 19
  • Country: in
variable power supply circuit
« on: June 15, 2020, 08:09:59 pm »
hi,
i am designing a 30v5A variable supply. the circuit is similar to korad 3005 power supply, the difference is i am replacing the power transitor with pmos but it becomes very hot and temp goes on increasing. the circuit is attached.
whats wrong? why so hot ? at 20v 500ma
« Last Edit: June 15, 2020, 08:12:03 pm by aneesh »
 

Offline bin_liu

  • Contributor
  • Posts: 38
  • Country: cn
Re: variable power supply circuit
« Reply #1 on: June 16, 2020, 04:29:54 am »
You need to choose the heat sink according to the power loss of 150W.
Also choose a MOS tube that can withstand a lot of power continuously.
 

Offline aneeshTopic starter

  • Contributor
  • Posts: 19
  • Country: in
Re: variable power supply circuit
« Reply #2 on: June 16, 2020, 04:43:18 am »
mosfet that i am using has good continuously current.
my plan is to design 30v 5A supply. but i cannot go above 500mA. i am using heatsink but i think there is some other problem. As there are PS in the market like dps3005 which uses IRF5210 with the same heatsink and achieves 5A.
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 3034
  • Country: au
Re: variable power supply circuit
« Reply #3 on: June 16, 2020, 05:00:53 am »
mosfet that i am using has good continuously current.
my plan is to design 30v 5A supply. but i cannot go above 500mA. i am using heatsink but i think there is some other problem. As there are PS in the market like dps3005 which uses IRF5210 with the same heatsink and achieves 5A.
What is your source voltage? The dps3005 will likely have tap switching to reduce dissipation and a fan to reduce temperature. 
The MOSFET could also be oscillating although it need not necessarily increase dissipation.
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline aneeshTopic starter

  • Contributor
  • Posts: 19
  • Country: in
Re: variable power supply circuit
« Reply #4 on: June 16, 2020, 05:13:19 am »
input is 20v dc output is correctly regulated to 5v as per requirement  but when i start drawing current 500mA it gets heated badly... the circuit is uploaded whats the prob with the design
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 3034
  • Country: au
Re: variable power supply circuit
« Reply #5 on: June 16, 2020, 05:24:37 am »
I did a search for "dps3005". It's an efficient Switch Mode Power Supply that doesn't produce much heat.
You are experimenting  with a linear regulator. The MOSFET has to dissipate the difference between input and output power.
You will need a larger heatsink or use a SMPS IC.   
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline aneeshTopic starter

  • Contributor
  • Posts: 19
  • Country: in
Re: variable power supply circuit
« Reply #6 on: June 16, 2020, 06:16:46 am »
i am using larger heatsink and the Ic is also the same only the circuit may vary. if you can suggest me improvement in my current design to reduce heat. i will be great.
might be duty cycle, frequency, etc
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 3034
  • Country: au
Re: variable power supply circuit
« Reply #7 on: June 16, 2020, 06:38:05 am »
i am using larger heatsink and the Ic is also the same only the circuit may vary. if you can suggest me improvement in my current design to reduce heat. i will be great.
might be duty cycle, frequency, etc
The hand drawn diagram in your first post looks like you are designing a linear regulator.
Can you post pictures?
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline aneeshTopic starter

  • Contributor
  • Posts: 19
  • Country: in
Re: variable power supply circuit
« Reply #8 on: June 16, 2020, 06:45:26 am »
yes its a linear bench power supply .. i am using 20v dc from transformer rectified source.
 

Offline kallek

  • Regular Contributor
  • *
  • Posts: 59
  • Country: fi
Re: variable power supply circuit
« Reply #9 on: June 17, 2020, 06:03:01 am »
As already mentioned, with linear regulator you need much bigger heatsink than with switching regulator. Take a look at some thermal calculations and you will find out why that very small heatsink get so hot. Can you post pictures?
« Last Edit: June 17, 2020, 06:07:52 am by kallek »
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 3034
  • Country: au
Re: variable power supply circuit
« Reply #10 on: June 17, 2020, 06:09:05 am »
Start with about 1°C/W for the TO-247's  thermal pad.

EDIT: Or is it a TO-220 part? which will be much worse.
« Last Edit: June 17, 2020, 06:38:29 am by xavier60 »
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline aneeshTopic starter

  • Contributor
  • Posts: 19
  • Country: in
Re: variable power supply circuit
« Reply #11 on: June 17, 2020, 09:22:37 am »
it is t0 220 with large heatsink . plz refe the above circuit
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 3034
  • Country: au
Re: variable power supply circuit
« Reply #12 on: June 17, 2020, 09:49:41 am »
One way to calculate the power dissipated by the MOSFET, measure the voltage between Drain and Source, then multiply it by the current. 
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline aneeshTopic starter

  • Contributor
  • Posts: 19
  • Country: in
Re: variable power supply circuit
« Reply #13 on: June 17, 2020, 10:53:50 am »
drop is something nearly 15v and 0.5A
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 3034
  • Country: au
Re: variable power supply circuit
« Reply #14 on: June 17, 2020, 10:59:20 am »
drop is something nearly 15v and 0.5A


15 x 0.5 = 7.5W
What information do you have about the heatsink? 
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline rdl

  • Super Contributor
  • ***
  • Posts: 3667
  • Country: us
Re: variable power supply circuit
« Reply #15 on: June 17, 2020, 12:11:28 pm »
You probably won't be able to get by with one of those small heat sinks made from stamped and folded metal. Some kind of extrusion will be best, something like what Dave used for his dummy load (EEVBlog #102).
 

Offline aneeshTopic starter

  • Contributor
  • Posts: 19
  • Country: in
Re: variable power supply circuit
« Reply #16 on: June 17, 2020, 03:04:33 pm »
no its not about heatsink.. if u see the circuit attached i have given relevant information about the mosfet.
the internal resistance is just 3m ohm which is the best available .
 so the power dissipation should be low but i think its because of the region of operation.
 

Offline kallek

  • Regular Contributor
  • *
  • Posts: 59
  • Country: fi
Re: variable power supply circuit
« Reply #17 on: June 17, 2020, 03:21:39 pm »
Resistance is that only when mosfet is fully open. In your circuit mosfet is dissipating 15 volts.

You can reduce heat with transformer taps, but still you need bigger heatsink if your goal is 30V/5A.
« Last Edit: June 17, 2020, 03:36:39 pm by kallek »
 

Offline aneeshTopic starter

  • Contributor
  • Posts: 19
  • Country: in
Re: variable power supply circuit
« Reply #18 on: June 17, 2020, 05:37:15 pm »
look at this pic its dps3005 i.e 30v 5A with small hearsink
i am using bigger than this and better mosfet but i m not achieving this output
 

Offline kallek

  • Regular Contributor
  • *
  • Posts: 59
  • Country: fi
Re: variable power supply circuit
« Reply #19 on: June 17, 2020, 06:52:29 pm »
Just can't compare switching and linear regulator no matter what your pass transistor is.
If you want much better efficiency you need some sort of switching pre-regulator.
 

Offline TheMG

  • Frequent Contributor
  • **
  • Posts: 867
  • Country: ca
Re: variable power supply circuit
« Reply #20 on: June 17, 2020, 09:41:02 pm »
The "on" resistance of a MOSFET is a completely meaningless spec when you're using it in the linear region!!! It only comes into play when you are using it for a switching application, where the transistor is either fully off or fully on.

A linear power supply dissipates the undesired voltage as heat, plain and simple. 20V in, 5V out at 0.5A, you're sending 2.5W to the load, and the other 7.5W is dissipated in the transistor. No ifs ands or buts about it, that's how linear power supplies work.

Linear power supplies are relatively inefficient and dissipate a lot of heat. Some tricks can be used to reduce the amount of heat being dissipated, usually by switching between different input voltages (taps on transformer) depending on the desired output voltage, so you don't have as huge of a voltage drop across the transistor when outputting small voltages. Even with than, you're still not going to get anywhere near the efficiency of a purely switching power supply.

I would recommend looking up some educational electronics tutorials on linear power supply design, rather than trying to copy and modify a manufacturer's switching power supply design into a linear power supply without really knowing what you're doing. Case-in-point, the optocoupler is absolutely not necessary in your schematic! It was there in the original circuit to provide safety isolation in the feedback path between the primary (mains voltage) and secondary side of the switching supply.
 

Offline rdl

  • Super Contributor
  • ***
  • Posts: 3667
  • Country: us
Re: variable power supply circuit
« Reply #21 on: June 17, 2020, 10:12:51 pm »
I meant to put this in my previous post but forgot. Go to https://artofelectronics.net/ and download the sample chapter which is basically about power supply design. Section 9.4.1 "Power Transistors and Heatsinking" in particular applies here.

http://artofelectronics.net/wp-content/uploads/2016/02/AoE3_chapter9.pdf

 

Offline aneeshTopic starter

  • Contributor
  • Posts: 19
  • Country: in
Re: variable power supply circuit
« Reply #22 on: June 18, 2020, 03:36:03 am »
hi , plz find attached sch of korad supply which work perfectly fine and even i am using multiple tap to dissipate less heat. i completely understand the working but because i want to use single tap which is cheaper and for that reason i wanted to modify the circuit using mosfet.
i brought a dps3005 which uses mosfet and dissipate less heat. i have attached the picture as well
 

Offline xavier60

  • Super Contributor
  • ***
  • Posts: 3034
  • Country: au
Re: variable power supply circuit
« Reply #23 on: June 18, 2020, 05:54:29 am »
The dps3005 is a Switch Mode Power Supply, SMPS.
 Its MOSFET is working as a switch, turning fully on and fully off at at high frequency.
It dissipates very little power as heat. The toroidal inductor and a capacitor smooths the output voltage.

With a linear regulator, the series pass transistor functions as variable resistance. It is partly turned on, passing load current continuously.
Just like a resistor, it dissipates power as heat.
Power = Voltage x Current

The type of transistor used makes no difference. Only the amount of heat that it can safely dissipate.
HP 54645A dso, Fluke 87V dmm,  Agilent U8002A psu,  FY6600 function gen,  Brymen BM857S, HAKKO FM-204, New! HAKKO FX-971.
 

Offline aneeshTopic starter

  • Contributor
  • Posts: 19
  • Country: in
Re: variable power supply circuit
« Reply #24 on: June 18, 2020, 06:35:02 am »
yes thats what the circuit i attached above is similar, it also works in switching mode. its not on continuously . it switches the comparator checks the voltage and switches off till the output is stable.
dps 3005 also works the same way
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf