A quasi-resonant arrangement buys one specific thing: it turns the primary switch on when the voltage across it has fallen to a valley, which reduces the loss that occurs at that instant. Everything else about the design follows from that single behaviour, including the variable switching frequency that makes the electromagnetic work different.
Quasi-resonant chargers: what the arrangement changes
Four consequences: lower turn-on loss, a variable switching frequency, a different transformer design and a different electromagnetic signature.
The four are connected. Valley switching reduces the loss at the switching instant, which allows a higher effective frequency and therefore a smaller transformer; the frequency varies with load because the valley position moves, which changes the filter design; and the softer switching edge changes the emissions profile.
Reading them together is what makes the arrangement a design decision rather than a feature name, and it is also what determines whether the saving in the magnetics justifies the additional work.
Consequence 1: lower turn-on loss
The gain is at the switching instant, and it is largest at high line.
Turning on when the voltage across the switch is at a minimum reduces the energy dissipated in that transition, which is why the efficiency improvement is most visible at the top of the input range. The gain does not apply to the conduction loss, so the arrangement is a targeted improvement rather than a general one, and the measurement that proves it is a comparison at the same load and input.
Consequence 2: a variable switching frequency
The frequency moves with the load because the valley position moves.
Valley detection means the switching instant depends on the resonant ring after the switch turns off, so the frequency changes with the load and with the input voltage. The practical consequences are a wider frequency band for the filter to handle and a control behaviour that differs from a fixed-frequency design at light load. Both belong in the specification rather than in the laboratory notes.
Consequence 3: the transformer
The magnetics are designed for the operating range, not for a single frequency.
The transformer has to be designed for the variation in frequency and for the flux excursion that follows, which is more demanding than a design at a fixed point. In exchange, the higher effective frequency allows a smaller core, which is where the volume saving appears. Where the volume is not constrained, the additional design work may not be justified by the saving.
Consequence 4: the electromagnetic signature
A softer turn-on usually means a lower signature, within a wider band.
The reduced voltage transition at turn-on lowers the energy in the emissions, and the variable frequency spreads them rather than concentrating them at one harmonic. The pre-scan margin is often better than a hard-switched equivalent, which can reduce the filter size; the filter still has to be designed across the frequency band the control produces.
| Consequence | Where it helps | What it costs | How to verify it |
|---|---|---|---|
| Lower turn-on loss | Efficiency at high line | Valley detection and control | Comparison at the same load and input |
| Variable frequency | A smaller transformer | A wider filter band and control behaviour | The frequency range stated |
| Transformer design | Volume and cost of the core | Design effort across the range | The magnetics specification |
| Electromagnetic profile | A lower pre-scan signature | Filter design across the band | The measured margin |
How to specify it in an RFQ
Four items, and each produces a measurement.
The efficiency at defined load points and both ends of the input range; the switching frequency range the control produces; the transformer specification with its core and its intended operating range; and the pre-scan margin across the band. An RFQ that asks for those four can compare two arrangements, and a supplier that answers them is describing a design rather than a topology name.
When the arrangement is the right answer
Three conditions that justify the additional work.
A moderate output where the simple arrangement’s losses are becoming difficult in the available volume; an electromagnetic requirement that needs margin without a large filter; and a volume constraint that makes the smaller transformer valuable. Where none of the three applies, the simpler arrangement with a well-designed layout is usually cheaper and faster to certify.
What to watch in production
Two parameters that vary with the arrangement.
The switching behaviour, because valley detection depends on component tolerances, and the efficiency at the load point that characterisation identified as the sensitive one. Neither can be verified on every unit with a simple fixture, which is why the production plan states which are checked per unit and which are maintained through sample data and incoming verification.
FAQ
Does valley switching improve efficiency at every load?
It improves the turn-on loss, which is largest at high line and higher load. At light load the behaviour changes and the gain may be smaller, so the comparison belongs at defined load points rather than at a single one. The gain is largest at high line and higher load, while at light load the behaviour changes and the advantage may be smaller.
Why is the switching frequency variable?
Because the switching instant follows the valley of the resonant ring, and the position of that valley changes with the load and the input voltage. The resulting frequency range has to be stated, because the filter and the magnetics are designed across it. State the resulting frequency range, because the filter and the magnetics are designed across it and the position of the valley moves with the load.
Is the arrangement cheaper than a resonant architecture?
It is usually simpler and cheaper than a full resonant stage at moderate output, and it costs more than the simplest arrangement. The comparison turns on the volume and the electromagnetic requirement rather than on the power alone. It is usually simpler and cheaper than a full resonant stage at moderate output, and it costs more than the simplest arrangement, so the comparison turns on the volume.
What does it change for certification?
The switching behaviour changes the electromagnetic profile, so the pre-scan and sometimes the submission follow. The construction and safety evidence still follow the product and the market. The switching behaviour changes the electromagnetic profile, so the pre-scan and sometimes the submission follow, while the construction and safety evidence still follow the product and the market rather than the topology. and the topology is not part of that evidence.
Should the topology name appear in the marketing?
Only as far as it explains a customer-visible difference such as size or efficiency at a stated condition. The name by itself does not describe a behaviour the customer can measure. Use the name only where it explains a customer-visible difference such as size or efficiency at a stated condition, because the name alone describes nothing a customer can measure.
Next step
Send the output range, the volume target and the electromagnetic requirement to Wecent Group through the contact page. The reply states the efficiency at the required load points, the switching frequency range, the transformer specification and the pre-scan margin. Related reading: GaN charger applications, OEM and ODM charger manufacturing and charger quality control. External references: IEC standards, IEC Webstore, EU energy-efficient products, USB Power Delivery specification, UL standards and certification and TUV testing services.
{“@context”:”https://schema.org”,”@type”:”FAQPage”,”mainEntity”:[{“@type”:”Question”,”name”:”Does valley switching improve efficiency at every load?”,”acceptedAnswer”:{“@type”:”Answer”,”text”:”It improves the turn-on loss, which is largest at high line and higher load. At light load the behaviour changes and the gain may be smaller, so the comparison belongs at defined load points rather than at a single one.”}},{“@type”:”Question”,”name”:”Why is the switching frequency variable?”,”acceptedAnswer”:{“@type”:”Answer”,”text”:”Because the switching instant follows the valley of the resonant ring, and the position of that valley changes with the load and the input voltage. The resulting frequency range has to be stated, because the filter and the magnetics are designed across it.”}},{“@type”:”Question”,”name”:”Is the arrangement cheaper than a resonant architecture?”,”acceptedAnswer”:{“@type”:”Answer”,”text”:”It is usually simpler and cheaper than a full resonant stage at moderate output, and it costs more than the simplest arrangement. The comparison turns on the volume and the electromagnetic requirement rather than on the power alone.”}},{“@type”:”Question”,”name”:”What does it change for certification?”,”acceptedAnswer”:{“@type”:”Answer”,”text”:”The switching behaviour changes the electromagnetic profile, so the pre-scan and sometimes the submission follow. The construction and safety evidence still follow the product and the market.”}},{“@type”:”Question”,”name”:”Should the topology name appear in the marketing?”,”acceptedAnswer”:{“@type”:”Answer”,”text”:”Only as far as it explains a customer-visible difference such as size or efficiency at a stated condition. The name by itself does not describe a behaviour the customer can measure.”}}]}