TNY290P - Issue with 12V 2.35 Amp power supply

Transformer is possibly reverse phased. 12V output rectifier anode should go negative when primary switch is on..

Hi,
Thanks for your reply
However, i rechecked the transformer connections and they are correct
I am sure that my schematic is also correct (connection of 12V output rectifier diode)

I request you to try this design on PI Expert and check if I ave made any fundamental mistake
Regards
Atul Bhakay

Hi ,
I have done the calculations for primary inductance and primary turns by the traditional equations method
The same is attached herewith. KIndly confirm if the Lp and Np calculations are correct
Regards
Atul Bhakay

The design pushes the capabilities of the TNY290. The original design used an efficiency of only 79%, resulting in a transformer design that runs highly continuous. This might cause premature termination of switching cycles when one starts getting close to maximum output load. This may be exacerbated by using an ultrafast diode (and not a very good one) for the output rectifier instead of a Schottky rectifier, resulting in a large leading edge current spike..
It's a good idea to probe the drain - a voltage probe might be sufficient, but a current probe would give you more data re leading edge peak current. Look for prematurely terminated pulses as you turn up the load.
I re-ran the design using PIXLS with an estimated efficiency of 85% rather than 79%, resulting in a design which is much less continuous (KP = 0.38 for old design vs. 0.51 for the new). VOR for the new design is 120 V. . PIXLS selected the same transformer core as the old design (EF25). Np = 66t, Ns = 7t, Nb = 12t, Lp = 1.06 mH.nominal.
An appropriate output rectifier would be an SB5100 with added snubber of 1nF in series with 10-15 ohms. It is tempting to use a higher voltage ultrafast device to try and get around the need for a snubber, but the added reverse recovery noise from the un-snubbed diode can result in EMI issues. There might still be PIV issues as well if the supply remains continuous at a relatively high line voltage.

I also tried running the design using a TOP264K. at 132 kHz, and got reasonable results.

Hi PI Wrench Tam,
Thanks a lot for your efforts.
I really appreciate the same and your comments were enlightening to me

In the meantime, I checked the DC FANS which are the actual load for this power supply.

The FANs take 3 amps starting current and this current lasts for 1 second !!!!
So 2 FANS will take a total of 3+3=6 amps as starting current for 1 second
Now my next query is, whether power supply has to be designed for 6 amps ? If yes, which is best suitable device
Regards
Atul Bhakay
9503044791

You might try using a constant current output. The output voltage will collapse while the fans spin up. They may start slower that way, but you also may be able to get away with a lower current rating for the supply. I would also suggest moving to a TOPSwitch-JX for more output power capability.
You can experiment with optimal current limiting by using a bench supply with adjustable current limit.

Thank You
I have redone the design using TOP266EG with output of 12V , 6 Amps and input of 100V to 265VAC
The transformer selected was ETD29/16/10
Regards
Atul Bhakay

That approach will work. Another way of approaching the problem might be to add some circuitry to the output to allow the supply to operate in quasi-constant power mode, where the output current goes up as the output voltage falls. That way, you get a healthy slug of current at lower output voltage to allow the fans to start.