Hi Folks please see table
Please confirm whether the data in the table is something you'd expect? . The data is what we had calculated and corroborated in our test setups as we learn about the LT8316 behavior.
This 92W configuration uses SSR via a secondary||primary isolation amplifier. ADuM4195-1. (using Laux to drive for primary side supply of 5V)
Added type 2 compensator for FB pin feedback (to stabilize loop) . eliminated TC pin.
See below at 90W output. yellow is VdrainQ1. Magenta Vin. and blue is Vfb (1.22V target). It is at the upper end of the frequency band.
there is ground noise/ringing on Vfb pin present because i use a secondary board for generating the Vfb signal. (There is no LC filter installed yet. this is for testing purposes only)
The controller holds 24.15V constant from 1W all the way to Pmax. I have not tested the transient behavior. 25% to 75% of Pmax transition.
We found that a minimum of 300-700uF HV input capacitance is warranted to reduce excessive Vin swings. Which translate into 120Hz output ripple.
Output capacitance is 1400uF. The final version will have an LC filter added which should eliminate most ripple.
Works quite well. However, given the 4uH leakage inductance the snubber is tasked w dissipation in excess of 2.2W (at max power).
The snubber Z and RC types and the transformer (mainly the secondary side) constitute the majority of energy loss in the system.
All this really means is that the configuration will not reach 90% since there are input, output, and control stage losses in addition to the power stage. (peak efficiency observed is .83 to .87 or so)
Suffice to say the technology, while somewhat simple and cheap has limits. (No active clamp, no half bridge input rect, high magnetics losses etc.)
Lprim = either 122uH or 160uH
Rsen=25m
Vin=165V
Vout=24V
Pout = 1-92W peak efficiency is 0.83 - 0.87
N=4
Thank you
Edit Notes
updated table[edited by: Janus523 at 9:14 PM (GMT -5) on 13 Feb 2026]
