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.

Offline 200W Flyback App note shows wrong waveforms?

treez

Advanced Member level 5
Joined
Sep 22, 2008
Messages
7,980
Helped
586
Reputation
1,175
Reaction score
566
Trophy points
1,393
Location
cambridge
Activity points
78,635
Hi,
The below is an App Note for a offline flyback of power 48V at 4.17A. (90-265VAC input)
The controller is the UCC28710. It is a BCM controller.
Would you agree that the drain voltage waveforms on page 12 must be wrong? These waveforms cannot be at full-load as told..they must be at much lighter load, surely?
These waveforms show the converter in deep DCM at full load and this would be a mistake. You would never design for deep DCM at lowest VAC input and full load, surely you agree?

App note (PM10816)

UCC28710 datasheet
 

treez

Advanced Member level 5
Joined
Sep 22, 2008
Messages
7,980
Helped
586
Reputation
1,175
Reaction score
566
Trophy points
1,393
Location
cambridge
Activity points
78,635
Hi,
....Actually......It turns out that the scope waveforms from above are probably correct...and that this 200W , 90-264VAC Flyback has been designed in very deep DCM.... for some unknown reason.....do you know why?

....The UCC28710 controller is said to be a Quasi Resonant controller, so its a mystery how they have got it to operate at the same frequency from 90-264VAC.
After some phorensic "digging", it looks like the attached is the schem and design calcs for this 200w flyback SMPS. (LTspice sim representation is also attached should you wish).
The biq question is...why have they done it like this?
They could so easily have used more turns on the secondary and got it to work in less discontinuos mode. As it is now, at max load, the primary peak current is fixed at 13.3A from 90-264VAC...and the secondary peak current is a whopping 33 Amps......all very needless. (it switches at 28700Hz)
The dissipation in the RCD clamp is going to be 7.9W, which they manage with twelve 1W, 2512 resistors in parallel.
They use a ETD39 core gapped to 1.17mm. The peak B is 0.31mT.
The transformer is made by Feryster and is part number TI-ETD39-26V-136W
The primary is split and uses just two parallel strands of 0.5mm TIW. There appears to be no "packing" to keep the primary away from the gap.....so you would actually assume that the Litz secondary is wound first on the bobbin.....but then why is the primary shown to be split...another mystery of this PM10816 design.

ti.com forum says that (concerning the RLTI-1131 transformer)....
RLTI-1131 was made by renco electronics. You can reach out to them using their contact page to get information on RLTI-1131
...however, searching for this on the web takes you to the Feryster transformer of part number TI-ETD39-26V-136W
I emailed Renco but obviously no reply yet.
 

Attachments

  • Schem _Flyback 200W PM10816.pdf
    173.2 KB · Views: 3
  • CALCS_200w flybak _PM10816.zip
    13.7 KB · Views: 2
  • SIM_Flyback _200W DCM_PM10816.zip
    1.8 KB · Views: 2
Last edited:

treez

Advanced Member level 5
Joined
Sep 22, 2008
Messages
7,980
Helped
586
Reputation
1,175
Reaction score
566
Trophy points
1,393
Location
cambridge
Activity points
78,635
BTW, i believe the attached PDF transformer is the one used by the 200W flyback above.
--- Updated ---

Pg 5 of the test report shows the RCD clamp resistors being at 144degc
These are CRCW2512 resistors which are rated up to 155degc
The 144degc was recorded in open air (no enclosure) with ambient = 29degc.

Surely this does not bode well.....when placed in an enclosure these resistors are clearly going to overheat... would you agree?

200W flyback Test report

CRCW2512 resistor
 

Attachments

  • Transformer RLTI-1131.pdf
    2.4 MB · Views: 4
Last edited:

Easy peasy

Advanced Member level 5
Joined
Aug 15, 2015
Messages
3,043
Helped
1,116
Reputation
2,232
Reaction score
1,155
Trophy points
113
Location
Melbourne
Activity points
16,831
like many app notes carried out in western pacific countries - they are done in a hurry and under pressure - so the wary engineer should not expect perfection - as you say at Vin-min and full load the Vds should ring down to near gnd and then the mosfet should turn on - deep DCM is not really beneficial - some designers believe they are saving I^2R losses on the pri and diode drop losses on the sec ( longer diode cond time ) but I can't completely agree

temps above 100C in a 25 deg C ambient are to be avoided for a psu with a reasonable life expectancy ...
 

treez

Advanced Member level 5
Joined
Sep 22, 2008
Messages
7,980
Helped
586
Reputation
1,175
Reaction score
566
Trophy points
1,393
Location
cambridge
Activity points
78,635
Hi,
In the following schem of a 200W Flyback (from above), the diode running into the RCD clamp is 2 paralleled SMC diodes (ES3J)....


Why two in parallel? Why not just one?....and why not just one SMB (instead of SMC) ultra fast diode?
Peak primary current is 13A, well under the surge current maximum. The average current is tiny, way under the 3A limit of the ES3J diode.

ES3J diode
 

Easy peasy

Advanced Member level 5
Joined
Aug 15, 2015
Messages
3,043
Helped
1,116
Reputation
2,232
Reaction score
1,155
Trophy points
113
Location
Melbourne
Activity points
16,831
the ES3J / 600V is not that utrafast at high fwd currents, the next one down 200V would be better if the volts fit - it may well be that a single SMC would fit the bill - but depending on freq - assuming 100kHz, nothing is certain until the testing is performed, 2 x SMC side by side next to the snubber R's may well help overall dissipation - just due to their bigger size - assuming lots of copper track ... basically the designer is covering their ass as the peak current if just a wee bit high for a single ES3J ...
 

LaTeX Commands Quick-Menu:

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top