Solution Finder Get Tech Support

DER-286 with EER28-CF139

Posted by: valecapaldi on

Hello, 

 

i have designed a 30W Led driver (33V@1.05A)  based on DER-286 and i am having some issues to make it work properly.

 

In the first prototype, i have used a transformer based in an old EER28L NICERA core i have at home, but for production those cores are not very easy to find here in Argentina; so i decided to use the COSMO CF-139.

The issue is that the first prototype (exactly the same pcb as DER-286, but only RFB changed to 100K) was working with the prototype transformer and now the same pcb is not working with the final version of it.

On the other hand, the second prototype (attached image, only the layout was changed to fit inside our box) seems to have something different since the RFB needed is 151K instead of 100K and also it is harder to make it work avoiding the protection status.

 

I have used the tunning seccion in the PIXLS spreadsheet trying to adjust RFB, RV1 and RV2 without success.

 

With RFB=151K is not protecting and the measured values are:
Vout=28.6V

Iout=0.36A

Vbias=13.3V 

 

Playing with the RFB values the better i can get is Iout=0.78A but i can't reach the 1.05 Amps needed.

  

Has it something to do with the differences in the losses or permeability between Nicera and CF139?

Is there any way to change the core material properties and recalculate the transformer in PIXLs?

What am i missing?  What do you reccomend me to do?

 

 

Attached are the PIXLs design, the schematic, the pcb, CF139 datasheet and some pictures of the second prototype board.

 

Thank you very much in advance,

 

best regards,

 

Valentin. 

Files

Comments

Submitted by PI-Dunwell on 02/03/2014

Hi,

 

The output current as you know is dictated by IFB and IV. The RFB resistor in the spreadsheet calls for ~73k and not 150k. This RFB is for a bias voltage of 17V. Double check the Vbias voltage and calculate your actual IFB, this should match with the IFB on line 29 of the spreadsheet to have the correct output current.

 

Regards 

Submitted by valecapaldi on 02/17/2014

Hi, thank you very much for the answer.

 

I have been reading the section "Fine tuning" in the AN49 and as it explains there, a procedure for fine tuning Rv1 and Rv2 that may be necessary.

I have made some little changes in the spreadsheet and circuit, and now i have RFB=117K. With that value, the output current is 0.77A instead of 1.05A, so i decided to go on with the fine tuning iterations.

I have measured and entered the IO1 and IO2 (for VIn1=190V and Vin2=230V) and the RV1 and RV2 recalculation gives me different values (RV1=5.1Mohm  instead if 4 and (RB2 = 360Kohm instead of 1.4Mohm). I replaced the values in the circuit and it seems that the output current is higher (near to 0.97A) but after less than a second it protects and i have no time to measure the values to fine tune the RFB, RV1 or RV2 again.

 

 The question is: How do i know which is the reason of the protection state?. I can't iterate with the values since i can't measure any voltage or current before the protection ocurrs.

 

Thank you very much in advance.

 

best regards.

Submitted by PI-Dunwell on 02/18/2014

The protection that will turn-off the IC is the overtemperature protection. Make sure that the transformer doesnt saturate at the required output power and the secondary diode doesnt cause too much current spike in the primary current (due to reverse recovery) that will heat up the device causing it to trigger the OTP.

Submitted by valecapaldi on 02/19/2014

Dear Dunwell, 

 

Sorry about all the questions, this is my first design and i don't have too much experience with SMPS design.

 I have checked the transformer saturation and as i don't have a way to see the Iout waveform i have made a small experiment leaving a small air gap in the transformer core. Now the OTP is not triggering before 1 or 2 minutes. So, what do you recomend me to do?

could you just correct the PIXLs spreadsheet (or make a new PIExpert for the DER-286 schematic) for me using the EER28 core please?.

I just want to be sure that it is correct and that the transformer turns are calculated correctly before build another one.

 

Thank you very much in advance.

 

Regards. 

 

Submitted by PI-Dunwell on 02/24/2014

Please see attached spreadsheet. Note that I change the operation from reduced mode to full mode so the IFB is center at ~150uA.

Submitted by valecapaldi on 02/26/2014

I have checked everything using your design and i realized that the problem was in the air gap of the ferrite core (mine has 1mm and for the required ALG it should be 2mm).

 

On the other hand, now i have one board working perfectly and 2 others where the led is blinking. I have re-checked everything and even i have used the same transformer and LNK IC without success.

As I know, blinking may be a problem related to the IFB making the ic to miss cycles, but actually i am not sure if that is my problem or something else.

Any suggestions to test?

 

Thank you vey much again. 

Great that you have them working.

 

For blinking, one of the most common cause is Ifb hitting the ifbskip limit. Check that the operating ifb matches with the spreadsheet. Make sure also that the bp pin voltage is ~6.4v.

 

regards 

Submitted by valecapaldi on 03/05/2014

Everything seems to be working fine except for an overheating in the output diode. It is mounted with a heatsink but i estimate the heat is too high. It seems very strange to me since the original DER-286 doesn't need the heatsink. Do you know which could be the reason? which measurements should i do?

 

thanks. 

Submitted by PI-Dunwell on 03/10/2014

You can check the peak inverse voltage of the diode and make sure it doesnt exceed the maximum specified on the data sheet. You can also monitor the current on this diode and make sure the specs are met.