SMPs question

[Gaber Mohamed Boraey]

Hello everyone

I think it’s easy for you professionals here, and i am just not so familiar with electronics like you

Something I don’t understand and first time see for smps design, why use two transformer here?, as shown in attachment ?

Often I see in all smps only one transformer, is it for high loads?, and how work?, can you little tell?

Thanks I’m advance

 

[Rahul_d]

The left side transformer is called as proportional base drive transformer, this transformer creates the floating base drive current proportional to the load for half bridge transistors. Right side transformer is the main power transformer.

 

[Gaber Mohamed Boraey]

How the pwm signal goes to the main transformer?, through the drive transformer?, I mean first pulses to drive then again to main?

 

[D.A.(Tony)Stewart]

Did you want 85V to 365 50/60 Hz universal input?

Do you know it has to be active PFC=1 for > 100W?

How much surge current on soft start?

What efficiency min:max at rated load? min load? 90% nominal 95% max?

What ripple ? mVpp
What step load error at 90% to 10% or 0 to 100%?
What budget? volume?
What cooling method? active or passive?

It can be done with only 1 transformer.

What are your detailed specs? or expectations? for everything!

 

[Rahul_d]

How the pwm signal goes to the main transformer?, through the drive transformer?, I mean first pulses to drive then again to main?

The PWM signal through gate drive transformer drives power transistors only, power transistors & 2 X 470uF capacitor form half bridge converter (please read on half bridge converter) which in turn drives the main transformer.

 

[D.A.(Tony)Stewart]

It is a low budget Chinese design on a 1 sided PCB. Mfg cost ~ < $10 in volume.
Do you think you can make it cheaper? AC switch for 120<>240 and not auto-universal reduces cost. Hotspot 50'C rise. I wonder what the inrush and PFC is?
I suspect no active PFC so they were being unloaded cheap on Ebay. $12.

PSU > 100W must be high PFC now.

Small transformers are about <$0.25 in high volume (1k/day)

was on board ( off spec units go cheap)

(LOL)

But hey, it has gold lettering. (disgruntled slave employee)

--- Updated Jan 10, 2024 ---

TI suggests this solution $20 BOM cost not include EMI filter PWB , heatsink and case, socket cable.

 

[Gaber Mohamed Boraey]

Th

The PWM signal through gate drive transformer drives power transistors only, power transistors & 2 X 470uF capacitor form half bridge converter (please read on half bridge converter) which in turn drives the main transformer.

Those are the power transistors?

 

[FvM]

The design is probably at least 30 years old, created before high voltage MOSFET have been widely available. Even 20 years ago, non-PFC PC power supplies, still used BJT half briges with base drive transformer. Today it's legacy.

--- Updated Jan 10, 2024 ---

Those are the power transistors?

Yes. High voltage BJT have relative low current gain. Transformer is an efficient way to provide the drive power.

 

[D.A.(Tony)Stewart]

In my photo you can see the 2 power transistors and diode bridge on heatsink, which you can get on alibaba for $3.52 Cdn for 10 pcs

Yes it looks very much like a 20 yr old design, yet not as good as Lambda's 180 W or PowerOne's 48V PSU 1U rack with VOIP hardware which you can buy on Ebay cheap inside obsolete rack. I designed the rack and added twin 1U high temperature controlled fans to reduce hotspot on power transfer from 50'C rise to 20'C, which only turn on at full power with elevated ambient. UL FCC approved.

PSU's like these are dirt cheap or robust with OTC, OCP, UVP, OVP , PFC, and universal Vin. But the TL494 does have foldback. Don't know about hickup retry.
Take your pick from Bangood. Or get enough experience to work at Murata in Markham, ON or the HV kW PS shops around town.


hFE is not that critical when using them as switches controlled by Ic/Ib=10 but these need Ic/Ib=5 which is normally 10% of linear high hFE, but in these fast HVce types Vce(sat) needs <=50% of linear hFE.

Unlike mohm FETs these lossy BJTs have an Rce = 1/Pmax as switches which for 50 to 100 W makes them well damped for step response.
Whereas ultralow RdsON Fets need lots of compensation for no load high Q resonance with 4 methods of ripple reduction. (PFM, PWM, Current feedback and Type III filter voltage feedback) as well as ultralow ESR caps. Keep in mind there is more energy stored in L&C than delivered to load in order to survive 1 grid cycle loss and stay within regulation limits. So the high series Q is regulated with deadtime as shown in plot

TL494 released JANUARY 1983 https://www.ti.com/lit/an/slva001e/slva001e.pdf?ts=1704901780178

2SC2625 Datasheet, Equivalent, Cross Reference Search​

Type Designator: 2SC2625
Material of Transistor: Si
Polarity: NPN
Maximum Collector Power Dissipation (Pc): 80 W
Maximum Collector-Base Voltage |Vcb|: 450 V
Maximum Emitter-Base Voltage |Veb|: 7 V
Maximum Collector Current |Ic max|: 10 A
Max. Operating Junction Temperature (Tj): 150 °C
Forward Current Transfer Ratio (hFE), MIN: 10

 

[KlausST]

Hi,

Those are the power transistors?

The schematic shows the part number (too small for me to read).
But I´m sure you can read. So do an internet search for the according datasheet and you will get all informations about the BJTs.

Klaus

 

[D.A.(Tony)Stewart]

Dear Klaus, I posted both clear schematics and BJT link. FWIW https://alltransistors.com/adv/pdfview.php?doc=2sc2625.pdf

 

[KlausST]

Hi,

Dear Klaus, I posted both clear schematics and BJT link. FWIW https://alltransistors.com/adv/pdfview.php?doc=2sc2625.pdf

Dear Tony. It's not me who needs the information. ;-) The OP asked about this.

And I just wanted to point out that he has all the necessary information to answer the question himself. With less effort for him than for us.

And yes, I see your efforts to find and post a clear schematic and links to datasheets. That's very kind of you... but I think it's the OP's job.
I think these are rather basic and simple tasks. One does not need to be an engineer.
And it´s not difficult to read the first words in the datasheet that say "POWER TRANSISTOR".

On the other hand I don´t want you to stop your behaviour.

Btw: Just tried both your datasheet links. Both say "404 File not found". Did not investigate why.

Klaus

 

[BradtheRad]

Web links in posts 9 & 11 worked for me (four total).
I got no 'caution' alert from any.

Hold it... Four links opened websites. However the specific all-transistors data sheets do not appear... Instead there's a dark grey area. Just readable is '404 not found'.

 

[D.A.(Tony)Stewart]

Web links in posts 9 & 11 worked for me (four total).
I got no 'caution' alert from any.

Hold it... Four links opened websites. However the specific all-transistors data sheets do not appear... Instead there's a dark grey area. Just readable is '404 not found'.

My bad. https://alltransistors.com/adv/pdfview.php?doc=2sc2625.pdf&dire=_mospec

Normally, I truncate links after .pdf as I thought it was superfluous tracking info. not here. It's essential php script info.

 

[D.A.(Tony)Stewart]

Hi,


Dear Tony. It's not me who needs the information. ;-) The OP asked about this.

And I just wanted to point out that he has all the necessary information to answer the question himself. With less effort for him than for us.

And yes, I see your efforts to find and post a clear schematic and links to datasheets. That's very kind of you... but I think it's the OP's job.
I think these are rather basic and simple tasks. One does not need to be an engineer.
And it´s not difficult to read the first words in the datasheet that say "POWER TRANSISTOR".

On the other hand I don´t want you to stop your behaviour.

Btw: Just tried both your datasheet links. Both say "404 File not found". Did not investigate why.

Klaus

I also got this error
So I reentered the part number and then clicked on the 1st result

https://alltransistors.com/transistor.php?transistor=13455

Then I tried some things again and got it, so it must be related to history or cookies or meh !