Simple circuitry for cellular telephone/camera flash illumination

Next generation cellular telephones will include high quality photographic capability. Improved image sensors and optics are readily utilized, but high quality "Flash" illumination requires special attention. Flash lighting is crucial for obtaining good photographic performance and must be quite carefully considered.

Flash illumination alternatives
Two practical choices exist for flash illumination-LEDs (light emitting diode) and flashlamps. Figure 1 compares various performance categories for LED and flashlamp approaches. LEDs feature continuous operating capability and low density support circuitry among other advantages. Flashlamps, however, have some particularly important characteristics for high quality photography. Their line source light output is hundreds of times greater than point source LEDs, resulting in dense, easily diffused light over a wide area. Additionally, the flashlamp color temperature of 5500 to 6000¡K, quite close to natural light, eliminates the color correction necessitated by a white LED's blue peaked output.



Figure 1: Performance characteristics for LED and flashlamp-based illumination. LEDs feature small size, no charge time and continuous operating capability; flashlamps are much brighter with better color temperature.

Flashlamp basics
Figure 2 shows a conceptual flashlamp. The cylindrical glass envelope is filled with Xenon gas. Anode and cathode electrodes directly contact the gas; the trigger electrode, distributed along the lamp's outer surface, does not.



Figure 2: Flashlamp consists of Xenon gas-filled glass cylinder with anode, cathode and trigger electrodes. High voltage trigger ionizes gas, lowering breakdown potential to permit light producing current flow between anode and cathode. Distributed trigger connection along lamp length ensures complete lamp breakdown, resulting in optimal illumination.

Gas breakdown potential is in the multikilovolt range; once breakdown occurs, lamp impedance drops to ²1½. High current flow in the broken down gas produces intense visible light. Practically, the large current necessary requires that the lamp be put into its low impedance state before emitting light. The trigger electrode serves this function. It transmits a high voltage pulse through the glass envelope, ionizing the Xenon gas along the lamp length. This ionization breaks down the gas, placing it into a low impedance state. The low impedance permits large current to flow between anode and cathode, producing intense light. The energy involved is so high that current flow and light output are limited to pulsed operation. Continuous operation would quickly produce extreme temperatures, damaging the lamp. When the current pulse decays, lamp voltage drops to a low point and the lamp reverts to its high impedance state, necessitating another trigger event to initiate conduction.

Support circuitry
Figure 3 diagrams conceptual support circuitry for flashlamp operation. The flashlamp is serviced by a trigger circuit and a storage capacitor that generates the high transient current. In operation, the flash capacitor is typically charged to 300V. Initially, the capacitor cannot discharge because the lamp is in its high impedance state. A command applied to the trigger circuit results in a multikilovolt trigger pulse at the lamp. The lamp breaks down, allowing the capacitor to discharge1. Capacitor, wiring and lamp impedances typically total a few ohms, resulting in transient current flow in the 100A range. This heavy current pulse produces the intense flash of light. The ultimate limitation on flash repetition rate is the lamp's ability to safely dissipate heat. A secondary limitation is the time required for the charging circuit to fully charge the flash capacitor. The large capacitor charging towards a high voltage combines with the charge circuit's finite output impedance, limiting how quickly charging can occur. Charge times of 1 to 5 seconds are realizable, depending upon available input power, capacitor value and charge circuit characteristics.




Figure 3: Conceptual flashlamp circuitry includes charge circuit, storage capacitor, trigger and lamp. Trigger command ionizes lamp gas, allowing capacitor discharge through flashlamp. Capacitor must be recharged before next trigger induced lamp flash can occur.

The scheme shown discharges the capacitor in response to a trigger command. It is sometimes desirable to effect partial discharge, resulting in less intense light flashes. Such operation permits "red-eye" reduction, where the main flash is immediately preceded by one or more reduced intensity flashes2. Figure 4's modifications provide this operation. A driver and a high current switch have been added to Figure 3. These components permit stopping flash capacitor discharge by opening the lamp's conductive path. This arrangement allows the "trigger/ flash command" control line pulse width to set current flow duration, and hence, flash energy. The low energy, partial capacitor discharge allows rapid recharge, permitting several low intensity flashes in rapid succession immediately preceeding the main flash without lamp damage.




Figure 4: Driver/power switch added to Figure 3 permits partial capacitor discharge, resulting in controllable light emission. Capability allows pulsed low level light before main flash, minimizing "red-eye" phenomena.

Flash capacitor charger circuit considerations
The flash capacitor charger (Figure 5) is basically a transformer coupled step-up converter with some special capabilities3. When the "charge" control line goes high, the regulator clocks the power switch, allowing step-up transformer T1 to produce high voltage pulses. These pulses are rectified and filtered, producing the 300V DC output. Conversion efficiency is about 80%. The circuit regulates by stopping drive to the power switch when the desired output is reached. It also pulls the "DONE" line low, indicating that the capacitor is fully charged. Any capacitor leakage-induced loss is compensated by intermittent power switch cycling. Normally, feedback would be provided by resistively dividing down the output voltage. This approach is not acceptable because it would require excessive switch cycling to offset the feedback resistor's constant power drain. While this action would maintain regulation, it would also drain excessive power from the primary source, presumably a battery. Regulation is instead obtained by monitoring T1's flyback pulse characteristic, which reflects T1's secondary amplitude. The output voltage is set by T1's turns ratio4. This feature permits tight capacitor voltage regulation, necessary to ensure consistent flash intensity without exceeding lamp energy or capacitor voltage ratings. Also, flashlamp energy is conveniently determined by the capacitor value without any other circuit alterations.




Figure 5: Flash capacitor charger circuit includes IC regulator, step-up transformer, rectifier and capacitor. Regulator controls capacitor voltage by monitoring T1's flyback pulse, eliminating conventional feedback resistor divider's loss path. Control pins include charge command and charging complete ("DONE") indication.

Detailed circuit discussion
Before proceeding any further, the reader is warned that caution must be used in the construction, testing and use of this circuit. High voltage, lethal potenials are present in this circuit. Extreme caution must be used in working with, and making connections to, this circuit. Repeat: this circuit contains dangerous, high voltage potentials. Use caution.

Figure 6 is a complete flashlamp circuit based on the previous text discussion. The capacitor charging circuit, similar to Figure 5, appears at the upper left. D2 has been added to safely clamp T1-originated reverse transient voltage events. Q1 and Q2 drive high current switch Q3. The high voltage trigger pulse is formed by step-up transformer T2. Assuming C1 is fully charged, when Q1-Q2 turns Q3 on, C2 deposits current into T2's primary. T2's secondary delivers a high voltage trigger pulse to the lamp, ionizing it to permit conduction. C1 discharges through the lamp, producing light.




Figure 6: Complete flashlamp circuit includes capacitor charging components (figure left), flash capacitor C1, trigger (R1, C2, T2), Q1-Q2 driver, Q3 power switch and flashlamp. "TRIGGER" command simultaneously biases Q3 and ionizes lamp via T2. Resultant C1 discharge through lamp produces light.

Figure 7 details the capacitor charging sequence. Trace A, the "charge" input, goes high. This initiates T1 switching, causing C1 to ramp up (trace B). When C1 arrives at the regulation point, switching ceases and the resistively pulled-up "DONE" line drops low (trace C), indicating C1's charged state. The "TRIGGER" command (trace D), resulting in C1's discharge via the lamp-Q3 path, may occur any time (in this case Å600ms) after "DONE" goes low. Note that this figure's trigger command is lengthened for photographic clarity; it normally is 500 to 1000µs in duration for a complete C1 discharge. Low level flash events, such as for "red-eye" reduction, are facilitated by short duration trigger input commands.




Figure 7: Capacitor charging waveforms include charge input (trace A), C1 (trace B), "DONE" output (trace C) and "TRIGGER" input (trace D). C1's charge time depends upon its value and charge circuit output impedance. "TRIGGER" input, widened for figure clarity, may occur any time after "DONE" goes low.

Figure 8 shows high speed detail of the high voltage trigger pulse (trace A) and resultant flashlamp current (trace B). Some amount of time is required for the lamp to ionize and begin conduction after triggering. Here, 10µs after the 8kVP-P trigger pulse, flashlamp current begins its ascent to nearly 100A. The current rises smoothly in 5µs to a well defined peak before beginning its descent. The resultant light pro-duced (Figure 9) rises more slowly, peaking in about 25µs before decaying. Slowing the oscilloscope sweep permits capturing the entire current and light events. Figure 10 shows that light output (trace B) follows lamp current (trace A) profile, although current peaking is more abrupt. Total event duration is Å500µs with most energy expended in the first 200µs. The leading edge's discontinuous presentation is due to oscilloscope chopped display mode operation.




Figure 8: High speed detail of trigger pulse (trace A) and resultant flashlamp current (trace B). Current approaches 100A after trigger pulse ionizes lamp.

Lamp, layout, RFI and related issues:

Lamp considerations
Several lamp related issues require attention. Lamp triggering requirements must be thoroughly understood and adhered to. If this is not done, incomplete or no lamp flash may occur.

Most trigger related problems involve trigger transformer selection, drive and physical location with respect to the lamp. Some lamp manufacturers supply the trigger transformer, lamp and light diffuser as a single, integrated assembly. This obviously implies trigger transformer approval by the lamp vendor, assuming it is driven properly.

In cases where the lamp is triggered with a user-selected transformer and drive scheme, it is essential to obtain lamp vendor approval before going to production.

The lamp's anode and cathode access the lamp's main discharge path. Electrode polarity must be respected or severe lifetime degradation will occur. Similarly, lamp energy dissipation restrictions must be respected or lifetime will suffer. Severe lamp energy overdrive can result in lamp cracking or disintegration.

Energy is easily and reliably controlled by selecting capacitor value and charge voltage and restricting flash repetition rate. As with triggering, lamp flash conditions promoted by the user's circuit require lamp manufacturer approval before production.

Assuming proper triggering and flash energy, lamp lifetimes of Å5000 flashes may be expected. Lifetime for various lamp types differs from this figure, although all are vendor specified. Lifetime is typically defined as the point where lamp luminosity drops to 80% of its original value.

Layout
The high voltages and currents mandate layout planning. Referring back to Figure 6, C1's discharge path is through the lamp, Q3 and back to ground. The Å100A peak current means this discharge path must be maintained at low impedance. Conduction paths between C1, the lamp and Q3 should be short and well below 1½. Additionally, Q3's emitter and C1's negative terminal should be directly connected, the goal being a tight, highly conductive loop between C1's positive terminal, the lamp and Q3's return back to C1. Abrupt trace discontinuities and vias should be avoided as the high current flow can cause conductor erosion in local high resistivity regions. If vias must be employed they should be filled, verified for low resistance or used in multiples. Unavoidable capacitor ESR, lamp and Q3 resistances typically total 1½ to 2.5½, so total trace resistance of 0.5½ or less is adequate. Similarly, the high current's relatively slow risetime (see Figure 8) means trace inductance does not have to be tightly controlled.



Figure 9: Smoothly ascending flashlamp light output peaks in 25µs.



Figure 10: Photograph captures entire current (trace A) and light (trace B) events. Light output follows current profile although peaking is less defined. Leading edges dashed presentation derives from oscilloscope's chopped display operation.

C1 is the largest component in the circuit; space considerations may make remotely mounting it desirable. This can be facilitated with long traces or wires so long as interconnect resistance is maintained within the limits stated above.

Capacitor charger IC layout is similar to conventional switching regulator practice. The electrical path formed by the IC's VIN pin, its bypass capacitor, the transformer primary and the switch pin must be short and highly conductive. The IC's ground pin should directly return to a low impedance, planar ground connection. The transformer's 300V output requires larger than minimum spacing for all high voltage nodes to meet circuit board breakdown requirements. Verify board material breakdown specifications and ensure that board washing procedures do not introduce conductive contaminants. T2's multikilovolt trigger winding must connect directly to the lamp's trigger electrode, preferably with less than 1/4" of conductor. Adequate high voltage spacing must be employed. In general, what little conductor there is should not contact the circuit board. Excessive T2 output length can cause trigger pulse degradation or radio frequency interference (RFI). Modular flashlamp-trigger trans-former assemblies are excellent choices in this regard.



Figure 11: Magnified demonstration layout for Figure 6. High current flows in tight loop from C1 positive terminal, through lamp, into Q3 and back to C1. Lamp connections are wires, not traces. Wide T1 secondary spacing accommodates 300V output.

A demonstration layout for Figure 6 appears in Figure 11. The topside component layer is shown. Power and ground are distributed on internal layers. LT3468 layout is typical of switching regulator practice previously described, although wide trace spacing accomodates T1's 300V output. The Å100A pulsed current flows in a tight, low resistance loop from C1's postive terminal, through the lamp, into Q3 and back to C1. In this case lamp connections are made with wires, although modular flashlamp-trigger transformers allow trace-based connections.



Figure 12 (left): 90A current peak is confined within 70kHz bandwidth by 5µs risetime, minimizing noise concerns.

Figure 13: Trigger pulses high amplitude and fast risetime promote RFI, but energy and path exposure are small, simplifying radiation management.

Radio frequency interference
The flash circuit's pulsed high voltages and currents make RFI a concern. The capacitor's high energy discharge is actually far less offensive than might be supposed. Figure 12 shows the discharge's 90A current peak confined to a 70kHz bandwidth by its 5µs risetime. This means there is little harmonic energy at radio frequencies, easing this concern. Conversely, Figure 13's T2 high voltage output has a 250ns risetime (BW Å 1.5MHz), qualifying it as a potential RFI source. Fortunately, the energy involved and the exposed path length (see "layout") are small, making interference management possible.

The simplest interference management involves placing radiating components away from sensitive circuit nodes or employing shielding. Another option takes advantage of the predictable time when the flash circuit operates. Sensitive circuitry within the telephone can be blanked during flash events, which typically last well under 1ms.

References
1. Perkin Elmer, "Flashtubes."
2. Perkin Elmer, "Everything You Always Wanted to Know About Flashtubes."
3. Linear Technology Corporation, "LT3468/ LT3468-1/LT3468-2 Data Sheet."
4. Wu, Albert, "Photoflash Capacitor Chargers Fit Into Tight Spots," Linear Technology, Vol. XIII, No. 4, December, 2003.
5. Rubycon Corporation. Catalog 2004, "Type FW Photoflash Capacitor," Page 187.