Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Peak clamp or snubber network - how to calculate in flyback smps

Status
Not open for further replies.

obiwan

Newbie level 4
Joined
Aug 9, 2009
Messages
7
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
Argentina
Activity points
1,373
Hello ! I consulted Power Integrations about a power supply design using one of their TOPSwitch IC's. The details of my questions are here:

**broken link removed**

I have almost everything figured out, except for the snubber network for the StackFET configuration. This is used to improve the ability of the IC to handle high voltage.
I am definitely not an expert in this kind of circuits, but to me this snubber network shown at

**broken link removed**

... is just a peak clamp because it doesn't use a capacitor in a classic RCD configuration.
ST has a component called PEAK CLAMP in a single package that is very much the circuit proposed plus one single resistor in series, in this case 200 Ohm.
The datasheet for this component is here:

**broken link removed**

I would like to calculate the peak clamp for a tri-phase powered industrial power supply with a bus of 535 VCC, exactly the same circuit as the example below. I have the data for the transformer.
I don't find very easy the way ST suggest its calculation.
Could some friend give me a hand with this particular issue ? I would appreciate that very much.
Best regards to all

Roberto
 

Hello Dear Roberto
How much power do you need and with what voltage and current ?
And what is your real ( clear) problem with them ?
Best Wishes
Goldsmith
 

Good day and thank you Goldsmith.
Well you have the complete circuit schematic.
This is a 24 V @ 10 A output power supply.
Here are the parameters from the design program PI Expert
One of the problems is that this program is not able to design for the StackFET configuration

TRANSFORMER PRIMARY DESIGN PARAMETERS
LP 330 uHenries Primary Inductance
LP Tolerance 10 Tolerance of Primary Inductance
NP 85 Primary Winding Number of Turns
NB 12 Bias Winding Number of Turns
ALG 45 nH/T^2 Gapped Core Effective Inductance
BM 1736 Gauss Maximum Flux Density at PO, VMIN (BM<3000)
BP 2422 Gauss Peak Flux Density (BP<4200) at ILIMITMAX and LP_MAX. Note: Recommended values for adapters and external power supplies <=3600 Gauss
BAC 521 Gauss AC Flux Density for Core Loss Curves (0.5 X Peak to Peak)
ur 1847 Relative Permeability of Ungapped Core
LG 3,42 mm Gap Length (Lg > 0.1 mm)
BWE 77,1 mm Effective Bobbin Width
OD 0,90 mm Maximum Primary Wire Diameter including insulation
INS 0,08 mm Estimated Total Insulation Thickness (= 2 * film thickness)
DIA 0,82 mm Bare conductor diameter
AWG 20 AWG Primary Wire Gauge (Rounded to next smaller standard AWG value)
CM 1024 Cmils Bare conductor effective area in circular mils
CMA 435 Cmils/Amp Primary Winding Current Capacity (200 < CMA < 500)
Primary Current Density (J) 4,53 Amps/mm^2 Primary Winding Current density (3.8 < J < 9.75)


TRANSFORMER SECONDARY DESIGN PARAMETERS (SINGLE OUTPUT EQUIVALENT)
Lumped parameters
ISP 25,22 Amps Peak Secondary Current
ISRMS 14,81 Amps Secondary RMS Current
IO_PEAK 10,00 Amps Secondary Peak Output Current
IO 10,00 Amps Average Power Supply Output Current
IRIPPLE 10,92 Amps Output Capacitor RMS Ripple Current
CMS 2962 Cmils Secondary Bare Conductor minimum circular mils
AWGS 15 AWG Secondary Wire Gauge (Rounded up to next larger standard AWG value)
DIAS 1,45 mm Secondary Minimum Bare Conductor Diameter
ODS 1,35 mm Secondary Maximum Outside Diameter for Triple Insulated Wire
INSS -0,05 mm Maximum Secondary Insulation Wall Thickness


TRANSFORMER CORE/CONSTRUCTION VARIABLES
Core Type ETD39 ETD39 Core Type
Core ETD39 P/N: PC40ETD39-Z
Bobbin ETD39_BOBBIN P/N: *
AE 1,25 cm^2 Core Effective Cross Sectional Area
LE 9,21 cm Core Effective Path Length
AL 3150 nH/T^2 Ungapped Core Effective Inductance
BW 25,7 mm Bobbin Physical Winding Width
M 0,00 mm Safety Margin Width (Half the Primary to Secondary Creepage Distance)
L 3,00 Number of Primary Layers
NS 19 19 Number of Secondary Turns


DC INPUT VOLTAGE PARAMETERS
VMIN 226 Volts Minimum DC Input Voltage
VMAX 636 Volts Maximum DC Input Voltage





So the power output is 240 W continuous.

Thank you !
 

Again Hi
As i know , the fly back converters , aren't good selections , for the powers that are more than 150w and the out put wave , has bad spikes ! so why fly back ? why not forward converter ? it hasn't disadvantages of fly back arrangement .
Good luck
Goldsmith
 

My dear friend, I thank your intention in replying.
I have some experience in low frequency switching power supplies up to 50 KHz.
I have designed some of them using the old and fantastic TL494, and built my own tranformers.
This device works at 132 KHz and need much more experience.
Why flyback converter ? because this chip is economic and solves many problems.
With my respect for your opinion, I think that you need some update about the flyback converter and abou this kind of chips.
Flyback converters can work perfectly well up to 500 W. After this, the best option could be half-bridge and even full-bridge at a much higher cost.
I suggest you to take a look at Welcome To Power Integrations | Power Integrations
You will find some surprising material and reference designs.
These company is one of the leaders now and the products are superb.
My only problem is calculating a snubber network when the device has industrial 3 x 380 V mains.

I still wait for somebody with much more experience than mine to give me some advice :-D.

Very kind regards
 

Again Hi My friend
I know that they can work at higher powers , but as i told you those specifications at higher powers are not good .
And about forward converter : it is similar to a fly back converter , really ( my mean is single ended forward converter ).
it just has an auxiliary winding on the transformer core with some turns ! and a simple inductor at out put ! and it's quality is very high , and it will be able to give powers that are more than 1KW but upper than 1KW , it's specifications , will going to become poor !
Kind Wishes
Goldsmith
 

:smile: Thank you again
I know what you are talking about.

Best wishes for you too

Roberto
 

So what is your selection ? fly back or forward ?
I can help you at both of them .
Regards
Goldsmith
 

For the reasons exposed about the chip ( please I suggest to read the information ) I will choose flyback.
Only if you share the same information that I am talking about we could work on this issue.
I appreciate your opinion but for many reasons I need to use this line of products.
Also, maybe you would like to download the excellent free design programs called PI Expert.
Cheers
Roberto
 

Dear Roberto
No problem with your choice , it is respectful in my mind .
And about design programs : i didn't use them until now , because i love design all things with myself ! :idea:
So why you didn't try to design them simply and with yourself ?! is it hard ? no , i shared a simple way that i currently using for all of my SMPS designs ! see below please:
https://www.edaboard.com/threads/229615/
I hope that can help you !
Good luck
Goldsmith
 

My friend once more I thank your good will.
It is always a good opportunity for learning.
In my case I am a logic / digital / microprocessor / industrial control designer 35 years now. I forgot about ferrites along time ago, but I don't loose the hope of learning.
So it is easy to use one of these excellent programs because they can deal with ignorance as mine.
If the problem is C language, state machines, parsers, P.I.D. controls, and the like, I can manage.
But calculating transformers by hand ... for sure is the best way but I don't know if I'm good on that.
Anyway I will take a look at your papers in some more minutes.
Believe me that we have commercial reasons to use these chips in important quantities, it isd not just a matter of hobbysts.
Let me see what you are talking in your calculation methos and I will comment if I am able or not to follow you.
As I said before, my only problem is a diode and a resistor and I am finished.
But I don't like to close the door to the opportunity of learning, so I will check your papers.
If you could help me with my diode, please do and I promise to make a try with manual method.

Best regards
Roberto

 

Sorry my friend ! with that diode i can't help you , because , all of the snubber systems and protection systems that are in parallel with windings at SMPS , that i used until now , are RCD , or some of non dissipative snubber . so sorry , because i can't help you at your purpose . :sad:
Good luck
Goldsmith
 

No problem. I will read the 4 pages document. Seems to be interesting but frequency is too high for me. :???:
 

you can select your desired frequency and , don't fear from high frequencies ! in my idea , the high frequencies are beautiful and interesting , and easier to use !!!!!!!! ( i designed a push pull regulator with 500KHZ frequency !):-D
I hope that you earn success at your project .
Wish you the best
Goldsmith
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top