UCC25600: UCC25600: Peak_Gain(m) versus Qe plot in sluc146 not as in SLUP263

Hi Ben,

It is not a matter of accuracy otherwise we could not follow the proposed design procedure. It may be question of interpretation.

Looking either at figure 13b in SLUP263 (Designing an LLC Resonant Half Bridge Power Converter) and figure 22 in SLUCS846C (spec UCC25600) both give for instance Qe=0.5 for Ln=5 and a Peak Gain of 1.2

While looking at the curve in the sluc146 spread sheet we get Qr>1.25 for m=5 and a Peak_gain of 1.2

Could you confirm that "m" in the spread sheet stands for "ln" in the procedure?

Any way to have a direct support without using E2E?

BR

Daniel

Hi Daniel

 m is Lm/Lr, the ratio of the magnetizing inductance to the resonant inductance (Lr includes the transformer leakage inductance as well as the inductance of any separate resonant inductor)

This is stated on line 18 of the calculator

We try to keep this forum open so that all users of TI devices can benefit from the discussions here. If it happens that a user needs to talk to us about confidential matters then of course we can take the discussion off line.

Regards Colin

Hi Colin & Ben,

Same definition for Ln (Ln = Lm/Lr in equ 21 in SLUP263 ).

Any off-line email for support?

Regards

Dan

Hi Dan,

Please email me at ben.lough@ti.com

Best Regards,
Ben Lough

Hi Dan

The FHA gain versus normalised frequency plots may be calculated from this formula - there are two versions in circulation but they are simply re-arrangements of each other and give the same results. I can't do it now but I will try to get some time tomorrow to verify the various plots and see which conforms to these equations the best. If you have MathCad, I can send you the file that generates the graphs.

Regards
Colin

Hi Dan

I've cross checked the UCC25600 DS (SLUSC846) against the design seminar article (SLU263) and a MathCad file that I used to create the gain curves in the design calculator for the UCC29950 and they all agree with one another. The formulae and curves in these documents are all derived from equation 18 in slu263 (a more explicit version is in the UCC25600 slusc846 data sheet on page 15). The exception is the graphs in the calculator tool for the UCC25600 (sluc146) and I think these should not be used. Instead, I would suggest that you use the peak gain vs Qe curves in the calculator for the UCC29950 sluc576a at www.ti.com/.../toolssoftware.

As you know the FHA method is accurate only near to the resonant frequency of the the power stage. We do have a more accurate, design tool for the LLC at www.ti.com/.../slua733.pdf . This is based on a numerical analysis of the LLC stage and it was originally produced for the UCC29950 controller but it is valid for any LLC stage - including those controlled by the UCC25600.

Regards
Colin

Hi Dan

I thought that Ben and I had addressed your query but we may have been mistaken. Could you clarify to us what needs further attention from us and we'll try to answer your questions. Ben has given you his email already. Mine is colingillmor@ti.com.

Regards
Colin