Switched Mode Power Supplies, Changed to Variable OP?

Log in or register to post comments
Go To Last Post
8 posts / 0 new
Author
Message
fever2tel
fever2tel's picture
Level: Hangaround
Joined: Thu. Jan 15, 2009
Posts: 175 View posts
Location: South Africa
#1
Posted by fever2tel: Thu. Aug 20, 2009 - 10:52 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi,
I have been reading up on switch mode power supplies, and one that has peaked my interest is the attached schematic from the datasheet of a TOP249Y.

If you look at the bottom right of the schematic you will see three zeners, along with R6 create a voltage reference of +/-48V (the desired output).
When the zeners forward bias, it turns on the opto which allows current to flow into the SMPS Chip U1, which controls the switching.

I was wondering if I just vary that combined zener voltage, do I now have a variable power supply (0-48VDC)? I am pretty sure it is not that simple because all the examples in the datasheet are fixed voltage outputs.

The only reason I can think of why it wouldnt work is because that high frequency transformer is set up for certain conditions that only "work" for 48V.

Can anyone see any limiting factors as to why one couldnt simply change that supply to be a variable one?

Thanks for any insight!

Attachment(s): 

Image icon SMPS.GIF

Just a noob in this crazy world trying to get some electrons to obey me.

Tags:

Learning and Information, General Electronics
Top
Kartman
Kartman's picture
Level: 10k+ Postman
Joined: Thu. Dec 30, 2004
Posts: 24785 View posts
Location: Melbourne,Australia
Posted by Kartman: Fri. Aug 21, 2009 - 11:38 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

You can only put X amount of energy into a given transformer and the resultant voltage is based on the turns ratio. The zener circuit turns on the opto to tell the controller to back off the pulse width and thus causing the system to regulate. If you use a circuit that can vary this voltage you will get a variable power supply but the amount of energy you get will decrease with the voltage to a point where the pulse width is so small the system won't regulate. If you REALLY want a variable supply, wind the transformer with a number of windings and use a linear regulator to do the fine regulation. Switch in the required winding for the voltage range you want to use. An AVR might come in handy to do this.

Top
ka7ehk
ka7ehk's picture
Level: 10k+ Postman
Joined: Fri. Nov 22, 2002
Posts: 21822 View posts
Location: Tangent, OR, USA
Posted by ka7ehk: Fri. Aug 21, 2009 - 03:52 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

That circuit is really NOT designed for variable output. First, it is totally isolated, and that type circuit is much harder to make variable than non-isolated ones.

It is also step-up, and if you want a variable supply that goes below the input voltage, you cannot use a step-up circuit.

Thirdly, it requires a 3 winding transformer. That will be quite expensive and adds some odd characteristics that are hard to deal with.

Jim

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

Top
dbvanhorn
dbvanhorn's picture
Level: Resident
Joined: Thu. Jan 31, 2008
Posts: 700 View posts
Posted by dbvanhorn: Fri. Aug 21, 2009 - 05:49 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If you already had this supply, then I'd say yes, you can do it.
I'd agree with Kartman also that it would be best to use a linear regulator on the output, and an AVR to set the switcher up to deliver the minimum voltage to the linear to keep dissipation down, and to program a Dpot or similar to set the linears output voltage where you like.

The magnetics will be a big problem though, if you haven't already got it in hand.
Custom transformers are a pain, and magnetics companies traditionally don't sample worth a damn.
Even worse if you need gapped magnetics.

You seem to understand the principle of operation though, when the opto turns on, the SMPS is turned off, till the voltage comes down far enough to turn off the opto and allow the SMPS to restart.

Top
fever2tel
fever2tel's picture
Level: Hangaround
Joined: Thu. Jan 15, 2009
Posts: 175 View posts
Location: South Africa
Posted by fever2tel: Fri. Aug 21, 2009 - 09:34 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

ka7ehk wrote:
That circuit is really NOT designed for variable output. First, it is totally isolated, and that type circuit is much harder to make variable than non-isolated ones.

It is also step-up, and if you want a variable supply that goes below the input voltage, you cannot use a step-up circuit.

Thirdly, it requires a 3 winding transformer. That will be quite expensive and adds some odd characteristics that are hard to deal with.

Jim

Jim

Thanks for the reply Jim, always an honor to get some feed back from the pros!

1. I was thinking of using some sort of op amp comparator on the output side that when the voltage rises above some set value then it allows a path for the opto's LED to ground.
That TOP248Y has a soft start feature, that slowly ramps up the duty cycle at start up until it gets some feedback.

2. Thanks for the reply, when you say step-up, Im guessing you dont mean step up in voltage? (since the circuit says 250-380V in and 48V out).

3. I see, those "odd characteristics that are hard to deal with" concerns me.

Just a noob in this crazy world trying to get some electrons to obey me.

Top
fever2tel
fever2tel's picture
Level: Hangaround
Joined: Thu. Jan 15, 2009
Posts: 175 View posts
Location: South Africa
Posted by fever2tel: Fri. Aug 21, 2009 - 10:26 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman wrote:
You can only put X amount of energy into a given transformer and the resultant voltage is based on the turns ratio. The zener circuit turns on the opto to tell the controller to back off the pulse width and thus causing the system to regulate. If you use a circuit that can vary this voltage you will get a variable power supply but the amount of energy you get will decrease with the voltage to a point where the pulse width is so small the system won't regulate. If you REALLY want a variable supply, wind the transformer with a number of windings and use a linear regulator to do the fine regulation. Switch in the required winding for the voltage range you want to use. An AVR might come in handy to do this.

Kartman to the rescue! Thanks for your really informative reply as always! That is very clever about changing the number of windings.

With regards to REALLY wanting a variable supply... I was thinking of making a DC motor drive, say 48V but didnt want to buy a big huge expensive transformer so was looking into SMPS. And then to control the motor speed I was going to switch the SMPS's output 48V with a varying duty cycle. But then I starting thinking I can switch the supply directly, and that would eliminate the next stage.
But upon reading your recommendation, I will just add the chopper stage.

Just a noob in this crazy world trying to get some electrons to obey me.

Top
ka7ehk
ka7ehk's picture
Level: 10k+ Postman
Joined: Fri. Nov 22, 2002
Posts: 21822 View posts
Location: Tangent, OR, USA
Posted by ka7ehk: Fri. Aug 21, 2009 - 10:35 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Pardon, I misread the schematic. It has to be step-down.

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

Top
Kartman
Kartman's picture
Level: 10k+ Postman
Joined: Thu. Dec 30, 2004
Posts: 24785 View posts
Location: Melbourne,Australia
Posted by Kartman: Fri. Aug 21, 2009 - 11:33 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Upon a closer look, I see the power supply is a lot bigger than I'd first thought - usually the topswitcher designs are like 5-10W. At 250W, its at the upper end of what you'd design using a flyback style converter as the magnetics start getting large - but that's really just a cost consideration. Being a flyback means there's a DC component and you need to gap the core correctly - not enough and things will go pop and too much it will be inefficient. For a one off, I'd look at rewinding a PC power supply - they're everywhere and most likely cheap or free. For the purposes of motor control, adding a chopper is probably the best solution. That's how my stepper drives do it - power supply to give me 42V DC and the stepper drives look after the rest.

Regarding the schematic, it shows R1 as a 2M 1/2W resistor. With over 300V across it, it won't last. I'd be putting three resistors in series. This was a common problem with consumer electronic devices until there was enough field failures. The old Macintosh was one of them - it had a 330k 1/2w resistor that would go open circuit. Something to do with electromigration due to the voltage gradient.

Top