Other Parts Discussed in Thread: UCC25800
We are developing a low-power- 250mW, dual-output isolated bias supply using the UCC25800-Q1 in a Secondary-Side Resonant LLC topology, targeting regulated outputs of +20V and -5V.
Based on simulation and calculation for our low-power design (max operational load current < 15mA), we have encountered a persistent issue:
- Discontinuous Conduction Mode (DCM) at Light Load: We want to use some off the shelf solution such as HTX7045C-124ERC
- .However at light loads approx 10mA or R_Load > 1500, the converter consistently operates in Discontinuous Conduction Mode (DCM), even when operating exactly at the resonant frequency f_SW = f_r.
Question: Is the UCC25800-Q1 designed to tolerate operation in DCM at light load as an intended behavior? Or are we expected to force Continuous Conduction Mode (CCM) by consuming excessive power via a bleeder resistor?
2. EMI Signature Concern: My main concern relates to the EMI signature of this DCM operation. Since the resonant current (I_Lsec) and primary current (I_Lprim) waveforms exhibit sharp, non-sinusoidal spikes during DCM (as shown in design guide examples):
- Do you have any available EMI spectral data showing the difference between CCM and DCM operation?
- Specifically, can you provide or reference data for the EMI signature (e.g., conducted emissions) for a low-power design, such as the TIDT255 or TIDT393 references, when operating at an extremely light load (e.g., 10 mA @ 25 V raw output)?
