Using crystals in embedded applications - Part 5

What is series resonance and parallel resonance in crystals?

Series resonance crystal does not have any reactive components like capacitor. Parallel resonance crystal is used where there are reactive components involved. Series resonance crystals are resistive in nature where as parallel resonant crystals combine the phase shift properties of reactive components to determine the oscillator frequency. The following circuits explain the difference.

Parallel Resonant circuit

Series Resonant circuit 

In a parallel resonant circuit, load capacitance is a critical specification.Load capacitance is the capacitance which is seen across the terminals of crystal. When we use this crystal in the application, this load capacitance must match the load capacitance requirements of the device to which crystal is connected. 

A same crystal can act as series as well as parallel resonant. When used in an application, the first resonance at which crystal oscillates is the series resonance and later when reactive component like capacitor is added it oscillates at different frequency which is called parallel resonance. So, from this point if i want to use a parallel resonant crystal and we use a series resonant crystal my crystal operates at a higher frequency than desired. In turn, if a parallel resonant crystal is used in place of series resonant requirement, the parallel resonant crystal operates at a lower frequency than specified.

What is the bandwidth of crystal?

The difference between series resonant frequency and parallel resonant frequency is called the bandwidth of the crystal. Lower the bandwidth higher the stability of the crystal. As per the quality factor definition, (Centre Frequency/bandwidth), lower the bandwidth higher the quality of the crystal. So, for your application choose a crystal by checking at the quality factor curve.

What is the crystal equivalent circuit?

In the above circuit,

Co = Holder capacitance (capacitance contributed by leads extending outside the crystal)

L1 = Motional inductance

C1 = Motional Capacitance

R1 = Series resistance

The common term used in crystal specification is hermetically sealed package. What does that mean?

Hermetically sealed meant air tight enclosure with only provision for external connectivity. In our crystals case, the electrodes are extended out.

What is the difference between AT cut and AT strip crystals?

AT cut is the cylindrical version of the crystal and AT cut is the stripped version of cylindrical crystals for miniature crystals.

What constraints can affect the crystal performance?

• Assume that we connected a crystal in our application, the drive provided to the crystal should not exceed the specifications of crystal. If drive provided exceeds the rating, the crystal life time gets reduced.
• Temperature at which the crystal is operated
• Improper load capacitance can vary the crystal frequency from the desired value

What are the fundamental and overtone specification of a crystal?

Fundamental frequency is specified for lower frequency crystals (around 8 MHz, 24 MHz, etc) where as overtone is odd multiple of fundamental frequency. Overtone crystals are not preferred for micro-controller applications.

Using crystals in embedded applications - Part 4

Ceramic resonator (vs) Quartz Crystal (vs) RC Oscillator (vs) Silicon Oscillator:

• Ceramic resonator and quartz crystal are mechanical resonant devices where as oscillator works on electrical phase shift principle (where R-C are the phase shift circuit).
• Power consumption of Ceramic is more than quartz crystal. 
• Ceramic resonators have good rise time than quartz crystals.
• Ceramic resonator and quartz crystal operate on the same principle as the electrical signal applied to them causes mechanical vibrations. 
• Ceramic resonators are more rugged than crystal.
• Quartz crystals are very stable and maintain their frequency under extreme conditions. Quartz crystal may not drift from it's frequency even when the PCB stray capacitance is high. Also, at quick varying temperatures, the quartz crystal remains stable.
• Ceramic resonators are made of ceramic material which are not as stable as quartz. These are piezoelectric in nature.
• Comparatively, Quartz costs more than ceramic resonators. 
• Ceramic requires much higher load capacitance required to quartz crystal.
• An oscillator is addition of feedback to the crystal along with amplification such that oscillations happen. Take the case of micro-controller, on the external clock pins we add a crystal and combined with internal circuitry it forms a oscillator.
• Oscillator is immune to EMI and humid conditions as they come packaged with all components for frequency generation.
• Temperature co-efficient which is a important material property is low for crystals and resonators.
• Ceramic resonators are available in miniature packages also.
• Ceramic resonator can sometimes be of compound material which helps change the characteristics of the crystal as per the material.
• The main disadvantages of oscillators is package size and cost and sensitive to vibration,  

Note: Some micro-controllers have internal capacitors, in this case, there is no need to have external capacitors.

Considering the disadvantages of ceramic resonator compared to Quartz, where do they find application?

Check the age old radio circuits you find resonators. Also, they are used in cost critical application and applications where stability is not important criteria.

What is the main selection criteria for any clock input?

Accuracy is the important criteria. Depending on the application for which the circuit is used, clock must be chosen properly. For example, applications like USB, SATA, PCIe requires a very stable clock with less deviation from the desired frequency.

Note: There are R-C oscillators internal to some ics which have very less stability but is a very cheap implementation. Micro-controller internal RC oscillator is an example.

Using Wi-Fi modules - Part 1

Using wireless technology for connecting to external world is the trend. Gone are the days where people used those Ethernet cables which didn't allow people to move around while browsing and other activities. Now it is very common that you see more tablets, smartphones, mini PCs and even desktops connected using wireless technologies. PCs use wired Ethernet and Wi-Fi for connectivity whereas smartphones rely on Wi-Fi and GPRS connectivity. So, Wi-Fi is one of the common mode of connectivity whether it be any media. So, when you are planning to use Wi-Fi module in your application it is always important you make a right choice of the module you are using.

To those starters, you have to know that a Wi-Fi is enabled in your device using a transceiver chip which has an antenna connected to it. The antenna acts as a transducer and helps in communicating to external world.

What need to be looked at when you are choosing a W-Fi module?

• Encryption type in the module
• WEP, WPA, WPA2 types
• Wi-Fi standard
• IEEE802.11a/b/g/n
• Supports point-to-point or access point connectivity
• Cost perspective
• Range
• In-built antenna present or not
• TLS (Transport layer security) support
• Form Factor
• Can have a web server or not? (HTTPS support)
• Interfacing to controller (UART/SPI/SDIO)
• Operating temperature
• Physical contact of the Wi-Fi module with the mother board
• Certifications
• Power consumption
• Customer support of the vendor
• Module dimensions

Basics: Why Active low signals used?

Are you working on digital circuits? Then you must be using control signals in your design and must be hearing about terminologies like ACTIVE LOW, ACTIVE HIGH. We sometimes just follow the design recommendations and never think of the need to do it. Let us analyse the need for Active Low in our designs.

What is active low?

A general criteria while using any control signals (like Enable) is to make it HIGH from LOW state. But in some cases, like the chip select and reset it is in reverse, the signal will be initially HIGH and when it is turned LOW, it means that particular signal is asserted (when asserted, it changes the connected chip functionality as desired, can say connected chip changes state). Control signals which are used in this manner are termed Active Low. The terminology itself tells that signal is termed active when it is LOW.

How is Active LOW signal differentiated from Active HIGH signal?

The representation of the signal (as per the Active HIGH or Active LOW ) in the datasheet of that particular chip is different. Active LOW is represented by having a dash on top of the signal name. Chip Select if active low is represented as (CS).

Reasons for using Active LOW only:

• Let us assume that you want to choose a device between n-channel MOSFET and p-channel MOSFET for your application. One would always prefer going for n-channel as electrons are the majority carriers in this type of device and have the highest mobility then holes. In this case, when the transistor is ON, the output is driven low. So, a normal condition is that output is high and when transistor in on output is low. This is one of the factor which makes implementation of active low state easy. Same mobility case applies to BJT also where NPN is preferred compared to PNP.
• Active LOW always helps eliminate indeterminate states due to improper supply voltages. 
• A genuine reason is that it is easier to pull down a signal than pulling it up. 
• Under a Active low condition, it is always easy to use wired-or condition and apply common reset to several chips. So, fanout can be increased. Control signal count can be reduced.
• For critical signals like Reset, it is always important that a state is maintained properly, if they are made active high, any noisy transitions may cause improper reset in the circuit. So, a state of high initially and then making it low to reset the chip can be the best option. 
• Consider the sourcing capability for example of a micro controller which doesn't go beyond 10 mA. If you want to source more you may have to use an external driver. In these cases, it is always preferable that we use Active LOW. For digital circuits, sink current capability is more than source current.

Wireless Charging - The emerging trend

Many of us use a wall mount charger for charging out multimedia devices like smartphones, tablets, music devices, etc. These devices have a socket into which the external charger goes in. It is always recommended that you use a suggested charger for a specific device. What if you don't have to use this wired charger and charge your mobile hassle free? Yes, this is what wireless charging is. Just place your mobile on a charging pad and your mobile gets charged, as simple as that.

Wireless charging is based on the principle of magnetic induction. Magnetic induction meant there will be two coils in which the first coil (transmitter) has a current flowing through it which causes magnetic flux. when the second coil (receiver) is placed near to first coil the magnetic flux induces current in the second coil. So, a charging pad on which you place the mobile will have a coil in which flux will be developed. When you place your mobile right next to the charging pad, which has a second coil , it induces current and hence battery charges. Basically, it is the electromagnetic field that plays a role here.

An alternative to magnetic induction is resonant charging. In this the receiver need not be next to the transmitter. Both of them work on the principle of resonance. In this way of charging power is transmitted wireless at a 6.78 MHz rate. In this case, the resonance helps charging.

What are the various wireless charging standards?

• A4WP (Alliance for wireless power)
• PMA (Power Matter Allaince)
• WPC (Wireless power consortium)

WPC developed Qi standard which is the most popular now and is in use. Qi is basically the magnetic induction type of charging. Regulatory approvals are in place for this standard.

A4WP developed resonant type charging. Regulatory approvals not in place.

What are the challenges from the designer point of view?

Let us take a smartphone designer, who wants to implement wireless charging in his device. The challenge for him is to select a chipset which is used as receiver. The chip has to support the desired standard. Also, the charging control circuitry to be implemented to connect to unit to be charged (battery). Chip sets are also available in the market which supports various charging standards. Coupling factor is of primary importance. A higher coupling factor leads to faster charging rate.

What are the various vendors for wireless charging chips?

• IDT
• Broadcom
• NXP
• Texas Instruments
• Freescale
• STMicroelectronics

What are the various domains wireless charging is expected to get exposed to?

• Smartphone charging where the charge pads can be mounted outdoors at public places like bus stops, cafeterias.
• Electric vehicles charging (Battery)

What are the differences between Inductive charging and resonant charging?

• For inductive charging, the receiver must be in the close proximity of transmitter (receiver and transmitter separated by centimeter distance) where as for resonant charging this is not the case.
• Coupling factor is high in inductive charging
• Size of the coils is more in inductive charging than resonant charging
• Cost is also high for inductive charging than resonant charging
• Efficiency is high in inductive charging
• Heat is more in inductive charging

Soldering Techniques

We might have heard about soldering techniques used by manufacturers to place/solder components on a bare PCB. For an experienced guy, these techniques are familiar but for a fresher these all look like hard stuff which add to the additional burden of learning. The two most terminologies you hear from manufacturing industry are wave soldering and re-flow soldering. If PCB is not so complex and if you are getting your PCB s soldered at your local manufacturing locations you may not be hearing more about re-flow soldering. Let us look at the major differences between these two techniques.

Wave soldering:

Wave soldering is used to solder both through hole and SMD also. But this technique may not be suitable for most of the SMD components. Even in this era of surface mount technology, there are many applications which demand through hole components. So, wave soldering is a popularly used technique.

Reflow soldering:

Reflow soldering is mainly used for surface mount components. Even the through hole component soldering can be done with this technology. In this soldering technique, a solder paste is added to the PCB to glue the components upon which is the PCB is exposed to a controlled temperature environment. Check the datasheets before selecting the component and check your manufacturing facilities in parallel to give a manufacturing guideline for maximum reflow temperature to be used. Reflow machines can be of infrared type or conventional type.

What are the latest technologies?

Now-a-days embedded market is growing faster enabling the technology to be used at every step. SMT reflow machines are having lots of inventions incorporated and are coming up with latest technologies. some of the important observations are:

• Graphical view of the temperature using a display attached 
• Touch based screens for control activities
• Minute temperature adjustment 
• Complete process level data logging
• Advanced cooling strategies
• Efficient flux distribution techniques
• Tracking the PCB from insertion to completion
• Power consumption reduction in latest machines
• Alarms for security
• Remote control

What are the advantages of latest innovations?

• Less maintenance costs
• Exceptional thermal performance
• Labour reduction
• Automation
• Quick turnout and hence high throughput
• Controlled environment
• Less chance of failures
• Report generation at every step

Understanding Oscilloscope Specifications - Part 2

Oscilloscopes can be said like heart of electronics testing especially protocol and high speed signals testing. Now a days oscilloscopes come with integrated test benches which help analyze high speed signals like USB, SATA, etc. These inbuilt test benches analyze signal specifications to the minute level and give a complete test report as a package. As a tester, this adds lot of flexibility and helps built confidence on the design. For a tester using oscilloscope initial challenge lies in understanding  specifications of it and selecting a right one for his application. 

Types of oscilloscopes:

The modern oscilloscopes come in various form factors which can be as big as bench-top to as well as wearable type and  middle level as hand-held type/pocket type. Hand held, pocket and wearable types are a very good addition to the field application engineers. That heavy packing, fright charges for a field engineer are no more a burden. The only disadvantage with these portable scopes is that their bandwidth is very less. So, the extent of debugging that can be done by the field engineer with these scopes is very limited. Even with such limitation, it is worth having one with you. Some people always have an eye on the pricing and cost wise they are nothing to worry as most of them are available in the range of $100 to $300. There are oscilloscopes especially designed for debugging phones during development.

Range of bandwidths:

Based on the domain you work on the oscilloscope varies. If you are working on a very low end digital signals and buses like SPI, I2C, SMBUS and UART, you don't need a very high end oscilloscope. The bandwidth requirement for such applications is very less and a 100 MHz bandwidth scope like TDS2014C works very much in such cases. If you are working on high end RF to high end digital signal signals, high bandwidth scopes are a necessity and be careful in choosing bandwidth of scope based on highest frequency to be measured. If you want to measure a frequency of 2.5 GHz, you must have a scope of bandwidth of at least 7.5 GHz. Today oscilloscopes are available up to 100 GHz. 

Waveform Generation:

One of the feature of oscilloscope is you have a output port which outputs fixed frequency signal. Hence, an oscilloscope can also be used in case you need a reference frequency for testing.