Thesis Pte Ltd

icon

One thesis

Copyright © 

2025

 Thesis Pte. Ltd. All Rights Reserved

OFFICE

34 Boon Leat Terrace #04-09A Singapore 119866

GET IN TOUCH

One thesis

Copyright © 

2025

 Thesis Pte. Ltd. All Rights Reserved

Categories
Uncategorized

Zero-voltage switching (ZVS) inductive heating

Here we show the waveform of a zero-voltage switching (ZVS) inductive heating circuit and the thermal profile in operation.

Snapshot of the switching voltages.

A typical buck regulator DC-DC is challenging to design when there is a significant voltage difference between the input and the output voltages. A significant voltage difference typically increases switching losses and limits the device’s switching frequency. As a result, a non-isolated buck-voltage regulator would experience high switching losses because high currents and voltages simultaneously load the MOSFET during the on and off phases. To mitigate this, designers may use more than one MOSFET stage to achieve the desired performance. For example, 19 V to 1.8 V is a voltage drop that may require two regulation stages, which would demand more board space and force designers to use larger filtering components.

Zero voltage switching

One solution that allows a return to a faster switching frequency at a higher input voltage and a lower voltage drop is zero voltage switching (ZVS). This technique uses pulse width modulation (PWM) or resonant technique but with an additional separate phase for the switching signal to enable ZVS operation. ZVS allows the voltage regulator to switch “smoothly”, avoiding the switching losses that typically occur with conventional PWM operation and timing.

The coil heating up.

Zero-voltage switching (ZVS) inductive heating

One application of a zero-voltage switching is to act as an inductive heater, aka Zero-voltage switching (ZVS) inductive heating. In this project, our client provided their custom coil for us to measure the efficiency of the ZVS circuit with their custom coil. As ZVS can also be formed using a resonant circuit, the inductance of the coil directly affects the resonant frequency.

How to calculate resonant frequency

To calculate the resonant frequency of such a circuit, one can use the following formula:

Resonant frequency (Hz) = 1 / (2 x pi x (inductance x capacitance)^0.5). Inductance is the coil inductance, and capacitance is the capacitor designed for the ZVS circuit.

Our evaluation

In the above circuit, the coil has a measured inductance of ~1uH at 100KHz, and the ZVS has two parallel 0.33uF capacitors, which equals 0.66uF. As a result, the resonant frequency is estimated around 196KHz. This matches quite closely with the resonant frequency being probed on the circuit.

Yellow = Inductive coil current
Green = Inductive coil voltage at one end
Blue = System Current
Red = Inductive coil voltage at the other end

The above diagram makes it easy to appreciate why it is called zero voltage switching because the switch only occurs when one end of the voltage goes to 0V (See Green and Red). A typical buck-switching MOSFET-based regulator would have higher overlapping during switching, which is power loss.

Gate switching voltages

Yellow = Inductive coil current
Green = Mosfet 1 gate pin
Blue = System Current
Red = Mosfet 2 gate pin

The above snapshot displays the gate switching voltages, and we can assume that the MOSFET may not have a chance to enter saturation mode. We speculated here because the circuit is rated for higher current, and we also do not have the part number for the MOSFET. Back to the topic, without entering the saturation region, the MOSFET would not be able to allow the maximum amount of current to flow into the coil and limit its heating capability. However, by operating in linear mode, the MOSFET remains relatively “cool” and prevents thermal runaway. To tune such a circuit, the designer must tradeoff between current capability and heat buildup. This is also why designers should have a heatsink placed on the MOSFET component. As for this case, the solution is for the client to increase coil inductance to reduce resonant frequency and to improve heating capability by allowing the MOSFET to operate in saturation mode longer.

As the MOSFETs have heatsinks installed, so they remained relatively cool as compared to the capacitor which is heating up due to the power load and switching.

Have a design challenge? Talk to us at contact@onethesis.com.

For some of our past projects, please visit www.onethesis.com/works/.

If you like our content, please consider following our LinkedIn page. Thank you.

Also, visit our Instagram page https://www.instagram.com/onethesis/.

Categories
News Uncategorized

Ingress Protection Codes and Ratings

Ingress Protection Codes and Ratings

Regardless of how smart your electronics are, they need protection against the elements and daily handling, which comes in the form of an enclosure that provides mechanical and/or water protection. Some recent notable devices are the water-resistant Samsung Galaxy S8, iPhone X and wearables such as the Apple Watch Series 3 and most, if not all, modern smartwatches.

The term “water-resistant” does not necessarily mean that the device is “waterproof” or submergible, and the standard now used to measure the “waterproof-ness” of a device is known as the IP Code or Ingress Protection marking, which is based on ANSI-IEC60529 standards that classify and rate the degree of protection provided against intrusion such as dust particles, and water by mechanical casings and electrical enclosures. The standard is published by the International Electrotechnical Commission (IEC). The equivalent European standard is EN 60529.

IP codes are useful references to quantify a device’s water resistance, and mobile phone manufacturers have been marketing them to demonstrate the water resistance of their flagships. For example, Apple’s iPhone X is IP67 rated and the Samsung S8 has a rating of IP68. So what does IP68 or IP67 mean? Here’s a useful breakdown of those ratings.

Protection against Solids

The first numeral represents the level of protection against ingress of solids such as dust or sand, the ratings range from 0 to 6. The levels represent the size of particles that the device can keep out. The larger the particle, the lower the rating. As the device becomes increasingly capable of keeping out tiny particles, the rating gets higher.

Protection against liquids

Levels of ingress protection against liquids are 0-9K. These levels denote the movement, depth, and pressure of water the device is capable of withstanding. The higher the number, the greater the water resistance. In mobile technology, we generally see ratings 0-8, without any “K” designations, which denote increased water pressure.

Some manufacturers may go beyond the code itself and publish further specifications on the duration and depth of the water-resistant rating as part of differentiating in their marketing and this is usually done for adventure/outdoor types of devices such as fitness trackers.

An “X” put in place of the solid or liquid numeral denotes that the device is not rated for solid- or liquid- ingress protection. This is different from a complete lack of protection (which would be a zero). For example, an “IPX6” rating represents that the device is not rated for solid-ingress protection, but has level ‘6’ liquid-ingress protection.

In a nutshell, products that are designed to withstand environmental elements are given an IP rating. This will be the letter IP followed by two numbers. The first of these numbers indicates how well a device can withstand dust and solid objects, the second number indicates how well the device can withstand water. Any water-resistant phone that you buy will have an IP rating mentioned somewhere and depending on this rating will be able to withstand different movements, depths, and pressure of water. The most common IP water ratings for phones are 6, 7 and 8 (remember that we’re looking at the second number, so that’s IPX6, IPX7, and IPX8, where the X is a different number indicating dust resistance).

A device or smartphone with an IPX6 rating can withstand strong jets of water from any direction for 3 minutes (for example, a shower) but cannot be immersed in water whilst an IPX7 device can be immersed in water anywhere from 15cm to 1-m in depth for up to 30 minutes. An IPX8 rating means that the device can be immersed in water over 1 m in depth for an extended period.

What is IP69K?
The IP69K rating is the highest protection available and is a protection provision for high-temperature and high-pressure water which is prescribed by Germany’s standard DIN 40050-9 and is not a standard in IEC 60529. IP69K rating specifies a spray nozzle that is fed with 80°C water at 80 to 100 Bar and a flow rate of 14 to 16 L/min – making products with this certification ideal for use in conditions where equipment must be carefully sanitized such as devices in industries such as food processing, where hygiene and cleanliness are paramount, and equipment must be able to withstand rigorous high pressure, high-temperature washing procedures.

Water resistance does not apply to all liquids.

It is important to note that water resistance does not mean that the device is indestructible. For example, a common scenario is where a smartphone is accidentally thrown into the washing machine along with regular laundry. Whilst the water-resistant conditions of IP7 or IP8 may be met, the constant tumbling action may crack the waterproof O-ring or seal and the water resistance is lost. Another factor is the presence of chemicals. The IP rating specifies only resistance to water, not chemicals. Chlorine in swimming pools, corrosive seawater, acids present in liquid foods, surfactants in detergents, alkalis in household cleaners and alcohols in beer or wine could damage the device despite its water resistance.

Design Principles

Designing environmental protection or water resistance for devices is now more important than ever with IoT sensors or smart devices being placed into a new environment or for a new application where the device is subjected to environmental and weather conditions. By far the simplest and most common method of waterproofing is the addition of an O-ring between enclosure joints or barriers.

However, that approach may not be that straightforward when there are ports or connectors on the device, such as a power connector, a USB charging socket or a headphone socket. Or when the device emits a considerable amount of heat from its electronics that must be vented away, precluding the option to seal a device completely.

As each application varies, the design of new smart devices or IoT systems will need careful planning and engineering. Do contact us for queries on your next project!

Categories
News Uncategorized

When USB lights get too hot

USB powered lights are convenient for reading, off-center lighting for mobile photo-taking and the like and on the market are some low-cost and simple USB-powered reading lights such as the S$2 Xiaomi Mi LED light.

Unfortunately, we found that the light output is not sufficient for certain conditions. A quick tear-down revealed a SAM semiconductor S8101 LED driver and what looks to be a row of six PLCC-2 3020 Package LEDs. The Mi LED light is rated at 1.2W and draws about 240mA @ 5V, that gives about 40mA/200mW power dissipation per LED.

While low-cost, the flexible-neck USB light puts out a measly light. Inspired, it was time to quickly hack our own.

We found another low-cost LED-powered lamp, rated at 5W and 450lumens on the packaging, which is supposed to be four times brighter. It has ten 5730 COB LEDs. The 5730s are rated to output more light than a 3020 module (~45 vs 5.4 lumens), see comparison here. The 5730 LED module also has a greater efficiency than the 3020 (90 vs 80 lumens/watt), however, note that lumen brightness varies greatly for different colour temperatures and LED binning, and this is just a simple gauge.

We removed the LED-PCB from its plastic housing and, with a little sawing (literally), soldered it to a flexible-neck USB extender and voila! We’ve now got our 450-lumen flexible USB-LED light. It’s blazingly bright, looks good and performs well!

Until we touched it to remove it from the USB port and YIKES! IT WAS HOT. It wasn’t warm, but HOT.

The thermal camera showed the LEDs and the schottky diodes heating up to a blazing 105°C. Insane!

Not to worry, we had a couple of cooling fans meant for cooling PC motherboard northbridges lying around. Those fans usually run at 12V from a motherboard fan header. USB voltage is rated at 5V and adding a 5V to 12V step-up converter is neither practical nor a 5-minute solution. So instead, we got a small 5V 40x40x10mm cooling fan from the hardware room and swapped out the 12V 40mm fan with a 5V one. No fuss!

Fortunately, the mounting holes of the northbridge cooler and the LED PCB were conveniently aligned! So with some thermal paste and trusty cable ties, the heatsink fan assembly nestled nicely onto the PCB and the fan wires were soldered to the 5V and USB ground. The assembly runs at a cool 34.7°C now!

A quick measurement with the light-meter shows 115.7 kLux of brightness, (converting that over 60mm2 or 0.0036m2 area to lumens is 416.52 lumens, close to the rated 450 lumens of the product.

It also seems to be drawing 960mA, or 4.82watts at 5.03 volts, which is expected.

A note of caution: a standard USB2.0 port on your computer is capable of delivering up to 500mA (0.5A); with USB 3.0, it moves up to 900mA (0.9A), so this guy isn’t exactly USB-hub friendly, but plugging it into a standard powerbank or a dedicated USB wall-adapter isn’t an issue since they can provide upwards of 1A.

With that, now we have a 416lumen USB-powered flexible lamp for various uses in the lab. This project took a little less than a lunch break to hack together, we hope you like it!

Build the future.