PMP4334: Vbias Circuitry needed for LED dimming

Hi,

  Since the designer has no longer in TI. I had consulted with the related colleague that all the questions you mention are voltage regulation.

First part, Q1, ZD4 and R17 will form a linear regulator that to keep the voltage regulation within specification of UCC28810D(18V, specify 16.5~20). I think you could keep these parts for more reliability on your design for the voltage working range.

Second Part, I think it is for system bias requirement. No regulation specification info I got but, I think it is 17V above because of the turns ratio(As shown below). For the 12V rail, I think it is 10V if the Zener clamp voltage show 10V on the schematic and the BOM. The Q9 remove reason might be the voltage bias for OPA and photo-couple are both enough.

Regards,

Richard

Try to attached the capture photo.

Hi Richard,

may I ask you how did you calculate the turns ration in order to get the 17V on Vaux? 

Regarding the 12V rail, did you say the Zenner is clamping the level to 10V if I do not include Q9?

Thank you for your time once again.

Best regards, 

Hi Jose,
    Sorry about the confuse on the Vaux rail that I just roughly to saw the turns ratio at N(8-9). Actually, the voltage in normal operation should be 14.44V according the calculation(In real applications, the voltage spike will let the voltage higher than calculation result).
Here is the calculation...
If N(8-10)=N(8-9)+N(9-10)...1 Reflect voltage to output will hold on 50.8V(Regulate 50V + Rectifier Diode 0.8V).
 N(1-2)/N(8-9)= 7.667, N(1-2)/N(9-10)= 3.286.
N(8-9)*7.667 = N(9-10)*3.286.
N(9-10) = 2.333 N(8-9)...2
From 1 and 2,
N(8-10)=3.333 N(8-9).
If V_N(8-10)=50.8, than V_N(8-9) = 50.8/3.333=15.24V.
Vaux=V_N(8-9) - 0.8 = 14.44V.
    Yes, It will be 10V if you do not include Q9.

Regards,

Richard

Hi Richard,

thanks for the detailed information, got it. Sorry for my continuous questions about this detail. 

One last question if you allow me; how did you get the regulated main output voltage of 50V?  If possible, could it be slightly tunned?

Thanks one more time! :)

Best regards,

José

Hi Richard,

thanks for your previous comment, it was quite helpful. Do you happen to know the general Np:Ns of this Transformer? It would help me calculate the VOR and maximum duty cycle for my application. 

Once again thank you for your trully helpful support.

Best regards,

José

Hi Jose,
    Good question. It is the first step of design a flyback converter. We do need to ensure the voltage de-rating is good enough on main switch.
    So usually we will looking to the reflect voltage VOS from Vo to the V_Np(Vo * Np/Ns) plus the Input Voltage Vi also need consider the spike from leakage parameter as the highest voltage on main switch. And  the highest voltage should under 90% of rated voltage on the main swtch to make sure the reliability of that.
    For example from PMP4334 design.
Np/Ns_(8-10)=2.3,
Vo = 50, VOS=115,
Vin_max=265*1.44= 381V,
Vds_off_Q5 = 381+115+100(Note) V = 596 V
Note : Spike voltage when main switch off depends the leakage inductance, Snubber obsorption ability. Not a fix value, means it is case by case.
So the highest voltage stress on Q5 will be 596V. And it's rating is 800V. So this de-rating on Q5 is good enough.
So you can try to calculate the turns ratio with considerate the main switch voltage stress at first. Then go on to the next...

Regards,

Richard

Hi Richard,

thanks for your previous email. It's making things more clear but still I do have some questions for you. If you could help me that would be great!

My first question comes from your Np(1-5)/Ns(8-10) = 2.3 , how did you get the value 2.3. I cannot find on the datasheet the relation between Np and Ns. But indeed if Vo * 2.3 we get 115V to VOR as you stated above. I am just missing here why the 2.3.

In Vin_max why did you multiply the maximum input voltage (265V) by 1.44? I am asking this question once we also need Vin_max to calculate the Duty Cycle_max and I am just wondering if I should use 265V or 381V. 

Once again thanks for your extraordinary help!

Best regards,

José

Hi Jose,
    According to the spec of transformer
Np_(1-2)/Ns_(8-9) = 7.667
Np_(1-2)/Ns_(9-10) = 3.286
Np_(1-2)/Ns_(8-10) = Np_(1-2)/ [Ns_(8-9) + Ns(9-10)]=[Ns_(8-9)/Np_(1-2)+Ns(9-10)/Np_(1-2)]^(-1)=1/[(1/7.667)+(1/3.286)]=2.3001.

The AC input voltage is Vrms. Vi_max should be peak of Vac(if ignor the Vf_rectifier).  So the Vi_max = (2^0.5) * 265 = 374.7.
Sorry for the typo, It should be1.414 not 1.44.

Regards,

Richard

Hi Richard,

Thanks for the previous post. It was helpful to get the Np/Ns overall relation. Still regarding the transformer; for the calculation of turns for the Nd(4-5) I would need the Vcc on this winding. 

If I include the components Not Assembled  (keeping the same layout) in the schematic of the PMP4334 (Q1, R17, ZD4) and keeping R39, what would be the Vcc voltage on Nd(4-5). I am asking this also because there is a label on the schematic called Vcc (right before D4) that goes nowhere. Is there any specific purpose for this Vcc label?

Once again thank you for your help!

Regards,

José