Patent Grant
8717168
Alarm units generally employ a microcontroller with an optocoupler (micro/opto design) to provide various features of the alarm units. Alarm units based on the micro/opto design have been proven to be reliable while providing excellent performance. Examples of such alarm units are disclosed in U.S. Pat. Nos. 6,369,696 and 6,311,021, which are assigned to the present assignee and are herein incorporated by reference.
However, in attempting to further improve alarm units based on the micro/opto design, it has been found that the micro/opto design has certain constraints. These constraints affect performance and the overall cost of the alarm unit.
Therefore, a need exists in the art for an alarm unit that is not based on the micro/opto design, thereby removing constraints that affect performance and the overall cost of the alarm unit that are attributable to the micro/opto design.
The present invention is an ASIC-controlled alarm unit. The ASIC circuit performs all the necessary control functions to provide audible and visual signaling when used with external horn and strobe circuits. Several illustrative advantages of the ASIC-controlled alarm unit are disclosed below.
In one embodiment, the strobe circuit with the ASIC operates at a constant frequency, e.g., 16 kHz as compared to the micro/opto circuit which operates at approximately 7 kHz. The faster switching speed allows for the use of a smaller inductor, thereby allowing the strobe circuit to operate more quietly because any magnetostriction caused by the inductor is at the upper threshold of the human hearing response.
In one embodiment, the new ASIC circuit has a more advanced peak current limiting circuit. The micro/opto circuit limited the initial peak current only during the initial power-up stage. The new circuit continuously senses the input current level and will limit the current any time it rises above a set level. The clamp level is determined by the voltage level on a resistor which is sensed by the ASIC, and the level can be changed by changing the sense resistor. This is an actively controlled current-limiter compared to other current-limiting schemes that use a passive foldback-type configuration.
In one embodiment, the ASIC circuit has improved MOSFET driving capability built into it. For example, it can drive a MOSFET at ten volts (or within an approximate range of 7.3-10.25 volts) with a faster on and off switching time (less than 400 nanoseconds), compared to the micro/opto circuit which drives the MOSFET at five volts and has a much slower switching speed (several microseconds). This improvement helps to reduce losses and makes the circuit efficiency better.
In one embodiment, the ASIC has two input pins which are used to set the candela setting for the strobe circuit. The pins are connected to a slide switch and can be a logic high (+5V) or a logic low (0V) depending on the switch position. Setting the candela sets an internal voltage reference level that is compared to the input on the ISENSE input pin. The old circuit had the candela switch on the input side of the circuit and it switched the sense resistances directly. The input current flowed directly through the switch. In the new circuit the input current does not flow through the switch.
In one embodiment, the ASIC offers more precise control of the strobe circuit. The energy level of the strobe is controlled by the voltage level on the sense resistor that goes to the ISENSE pin on the chip. This level is trimmed during the chip manufacturing process and is set within a much tighter tolerance limit compared to the micro/opto circuit. The micro/opto circuit relies on the tolerance of the forward voltage of the diode in the optocoupler and is less precise.
The teachings of the present invention can be readily understood by considering the following detailed description in conjunction with the accompanying drawings, in which:
To facilitate understanding, identical reference numerals have been used, where possible, to designate identical elements that are common to the figures.
In brief, the alarm unit 100 is generally powered by a supply voltage of 12 volts or 16-33 volts, and such supply voltage may be either D.C. supplied by a battery or a full-wave rectified voltage. In one embodiment of the present invention, the ASIC 110 functions as a controller and serves to control and regulate various functions of the alarm unit.
For example, the ASIC 110 serves to control the audio circuit 150 for generating an audio warning, e.g., via a horn, buzzer and the like. The ASIC 110 can control and regulate various audible features such as the frequency of the audio warning, e.g., to generate a Code 3 audio pattern. It should be noted that the audio circuit 150 shown in a dashed box can be optionally omitted if the alarm unit is implemented as a strobe only alarm unit.
The inrush filter (or current limiting circuit) 130 serves to limit the effect of an inrush condition. Inrush is a condition that may occur upon initial power-on, where a higher than average current is present in the alarm unit when power is applied to the power terminals for the first time to start alarm notification. Inrush can cause a momentary overload in the power supply and may cause the overcurrent protection in the panel to activate which can prevent the alarm units from operating. The overload may also damage relay contacts located in the panel which switch the loop to an alarm condition. Similarly, the inrush filter 130 shown in a dashed box can be optionally omitted if the inrush condition is not present or is addressed outside of the alarm unit.
The current sensing circuit 140 assists in detecting peak current condition. This circuit assists in converting the input voltage, e.g., 24 volts, to a voltage, e.g., 125-250 volts, sufficient to fire the flashtube within the flash circuit 170.
In one embodiment of the present invention, the alarm unit incorporates a switch having a plurality of positions, e.g., four positions that are representative of a plurality of intensity settings. By setting the switch to a particular position, the alarm unit will produce a predefined intensity level associated with that particular switch position. For example, setting the switch to a 110 candela setting will cause the alarm unit to produce a flash having a light output intensity of at least 110 candela upon activation of the alarm unit. The switch is coupled to an actuator assembly (not shown) and disposed within the alarm unit housing such that the switch is tamper resistant after installation, while the selected intensity setting is still clearly visible for inspection. The novel actuator assembly and associated display or menu is disclosed in U.S. Pat. No. 6,411,201, which is herein incorporated by reference.
In turn, the flash circuit 170 includes the voltage doubler 160 that serves the function of presenting a voltage across the flashtube that is twice the actual voltage that is stored in a storage capacitor, thereby allowing the flashtube to reliably fire at lower voltages. The importance of the voltage doubler 160 is due to the fact that the alarm unit may provide the selectable multi-candela feature. This feature places a difficult constraint on the circuitry of the alarm unit in that different voltages must be presented across the flashtube. Namely, the flashtube will be fired by a voltage that is dictated by a particular intensity level setting. As such, since the alarm unit is expected to produce intensity levels ranging widely from 15-110 candela, the alarm unit must reliably operate with relatively low voltages stored on a single storage capacitor. Without the reliability provided by the voltage doubler 160, multiple storage capacitors with additional switching will be required, especially when the selectable multi-candela feature offers a wide range of intensity levels. More specifically, the voltage doubler 160 allows the alarm unit of the present invention to reliably offer a selectable multi-candela feature that offers four (4) candela settings that widely ranges from 15 to 110 candela. The ability to offer a wide range of candela settings serves to eliminate more models of alarm units.
Additionally, the various circuits described in
The application circuit 200 employs an ASIC 110 as a controller. Several embodiments of the ASIC are disclosed below, e.g., an 18-pin package (as shown in
Electrical Specifications:
2Parametric condition for test, not result of ASIC behavior; ASIC has no control of this input parameter.
3Additional conditions: Auto Mode or Sync Mode with Tspa3 condition.
Functional description is now provided for the application circuit 200. The function description discloses several operational advantages offered by the ASIC 110.
Startup FETG Hold-Off
At startup, the FETG output is held low for 50 ms. This creates a 20 ms window of time after the Trgate (initially high) is turned off and the inrush clamp circuit is turned on, and before the FETG output is turned on. The 20 ms allows the storage cap (C3) to charge up before operating the dc-dc voltage booster.
Sync Pulse Detection
The sync pulse detection and control circuit detects sync pulses, and controls and synchronizes strobe and horn function. The Sync Pulse detection circuit will recognize a Sync Pulse if the voltage drops to a logic low on Vsdet for more than 6 ms.
Sound Control
The sound control circuit controls whether the horn is silent, running continuously, or operating in code 3 mode. The horn operates in code 3 mode whenever either the C3_HB input is low (with jumper plug installed) or a sync pulse has been detected within the last 1 second. When in Code 3 mode, the horn is silent 20 ms before to 480 ms after the strobe pulse; the horn will sound 480 ms after the Strobe Pulse, and be silenced again either when a sync pulse is detected or 20 ms before the next strobe pulse. It will sound for about ½ second, with ½ second pause, three times; then it will remain silent for an additional second, and then repeat the pattern.
If Code 3 is low, the horn will always run in code 3 mode. At initial power on, there is a delay of approximately 0.5 seconds before the first horn burst.
If the Code 3 input is high, the horn will only run continuously when no sync pulses are sent. At initial power on, the horn will start within 25 milliseconds.
If a second Sync Pulse is sent between 60 ms and 140 ms after the first, the horn will be silenced. This will also halt the count of the Code 3 pattern, so that when sound is re-enabled the pattern will pick up where it left off. If a second Sync Pulse is sent between 160 ms and 240 ms after the first, the horn will sound again and silence will end. The horn defaults to sounding on power-up.
Strobe Control
The strobe is fired when the ASIC receives a strobe sync pulse, or automatically every 975 ms when operating under auto mode. The auto mode causes the strobe to flash between predefined time intervals without the need to receive a strobe sync pulse. The auto mode can be entered in the event that a synchronization module fails to provide strobe sync pulses to the alarm units.
The strobe is also re-triggered if the strobe capacitor is still high after the strobe is turned off. With each strobe trigger, the current limiting transistor is switched off to protect against “after-glow” of the flashtube.
A sync pulse is recognized as a strobe sync pulse if it is either the first sync pulse, or if more than 500 ms has elapsed since the last strobe sync pulse. When a strobe sync pulse is received the strobe is fired after a delay of 20 ms. Additionally the ASIC goes into sync mode. In sync mode, the ASIC waits for another strobe sync pulse for up to 1.1 seconds. After 1.1 seconds the ASIC automatically strobes and falls in auto mode. Upon receiving a sync pulse, the strobe charging circuit (oscillator) is switched off to conserve power while the input voltage is low.
In auto mode, the ASIC automatically strobes every 975 ms. A sync pulse at any time in the cycle will cause the part to strobe and go into sync mode. This is the default mode if no sync pulses are detected.
In sync mode or auto mode, if the strobe capacitor is still charged after the first strobe output has gone high and low, then the strobe output will be re-triggered after 60 ms.
Over-Voltage Protection
The over-voltage protection circuit detects whether the strobe capacitor has been discharged after a trigger pulse. If the strobe is not discharged, FETG is held off to prevent further charging. In a normal cycle, the strobe capacitor (signal Vstcap) is checked during a window of 10-20 ms after the strobe is triggered. If the capacitor is still charged at this point, then a second trigger pulse will occur 60 ms after the first strobe trigger goes low. If, after the second pulse, the strobe capacitor is still charged, the ASIC enters an over-voltage condition.
The over-voltage condition ends when the strobe capacitor is discharged, when Vstcap is low. This condition only becomes effective during the silence pulse window (20-120 ms after the first strobe, regardless of sync or auto mode). This allows nearly a full cycle to charge up the strobe capacitor.
One important advantage of the ASIC-controlled alarm unit is that it provides better voltage and current monitoring functions. For example, the ASIC offers more precise control of the strobe circuit. In one embodiment, the energy level of the strobe is controlled by the voltage level on the sense resistor R1 that goes to the ISENSE pin on the chip. This level is trimmed during the chip manufacturing process and is set within a much tighter tolerance limit compared to the micro/opto circuit. The micro/opto circuit relies on the tolerance of the forward voltage of the diode in the optocoupler and is less precise.
In another embodiment, the ASIC circuit has a more advanced peak current limiting circuit. Micro/opto circuit generally limits the initial peak current only during the initial power-up stage. The present ASIC circuit continuously senses the input current level and will limit the current any time it rises above a set level. The clamp level is determined by the voltage level on a resistor R42 which is sensed by the ASIC, and the level can be changed by changing the sense resistor. This is an actively controlled current-limiter compared to other current-limiting schemes that use a passive foldback-type configuration.
Horn Tone Generation
The horn tone (on the PHORN output) is generated by producing two cycles of each frequency specified in either the NS or AS table shown below. The tone starts at the highest frequency and after two cycles is decremented until the minimum frequency is reached, producing two cycles at each frequency. The frequency is then incremented until the maximum frequency is reached again producing 2 cycles at each frequency. This sweep frequency is then repeated as long as the horn tone is enabled. This results in a sweep frequency of 117 HZ for the NS tone and 109.5 for the AS tone.
For the AS horn tone the on time at each frequency is fixed at 120 uS. For the NS horn tone the on time at each frequency is fixed at 115 uS.
The logic state of the pin NS_ASB determines which tone is selected. If NS_ASB is high or open the NS tone is selected. The AS tone is selected if NS_ASB is low. Thus, the ASIC-based architecture allows the selection of either NS tone or AS tone, i.e., providing the ability to select a particular horn tone frequency.
This approach in implementing the horn tone generation via an ASIC provides a reduction in the number of components that are deployed. For example, prior implementations deploy two integrated circuits to provide this function.
Horn Frequency Table
Test Mode
The ASIC has a special mode for measuring the ASIC during fabrication testing. In this test mode, the strobe cycle is sped up by a factor of 4000, such that 1 ms is reduced to a single ¼ μs clock. The entrance into this test mode has been designed to avoid accidental triggering. The entrance algorithm requires cycling through a count of 0-3 on MC0 and MC1 (where MC1 is the MSB) twice. This must be done in 4 μs steps and must match precisely to a ¼μs clock. As a result, the entrance algorithm requires 32 μs of precisely matching inputs on MC0 and MC1 for each and every ¼ μs clock, making accidental entrance very unlikely. This entrance algorithm is synchronous; moving the MC0 and MC1 inputs will not bypass any steps to the entrance algorithm. Furthermore, a timeout has been added such that if the part does accidentally enter test mode it will time out in at most 7 ms (29 strobe cycle timeouts in test mode), as denoted by the spec parameter Ttmto. At this point, it will resume operation in auto mode.
Non-ASIC Specifications:
Notes:
Inrush Current Limiting Circuit
The inrush limiting circuit 130 limits input current through Q7. The current is sensed across resistor R42. When the voltage across R42 matches an internal voltage reference V1, transistor Q7 is turned off by an operational amplifier. The voltage reference has 4 settings selectable by 2 digital inputs MC0 and MC1, such that each candela energy setting has a different inrush limit. It should be noted that although not shown in the diagram, at initial power up the inrush is limited to the lowest setting, to reduce power loading transistor Q7.
Additionally, strobe afterglow is prohibited by turning off the transistor Q7 during a strobe. One common method to prevent strobe afterglow is by using a limiting resistor, but such approach creates efficiency losses in that same resistor. As a result, the present novel ASIC-based approach of controlling/disabling inrush current improves strobe efficiency by removing losses of a limiting resistor and preventing flash tube afterglow.
Strobe DC-DC Boost Converter Circuit
The DC-DC boost converter circuit allows for accurate energy charging of a storage capacitor. Typically high voltage capacitors are not very accurate in terms of capacitance value (e.g., ±20%). As such, measuring the voltage on the capacitor is not an accurate method of determining the energy stored on it. Alternatively, another method to measure stored energy is to put a fixed amount of energy in. Since inductors and resistors are more accurately specified, they can be used to more accurately quantify the energy stored.
The DC-DC boost converter accurately stores energy based on a fixed inductance (L1), and a precisely set peak current. The inductor charge cycle begins every 60 μs, by turning on transistor MQ4. Current and energy increase through the inductor L1. When, the voltage across the sense resistor R1 reaches and equals the internal voltage reference, the transistor MQ4 is latched off, until the next charge cycle. The voltage reference has 4 settings for 4 energy levels, controlled by 2 digital inputs MC0 and MC1. This voltage reference is trimmed for accuracy, so as to set a peak voltage/current accurate to ±2%.
Further, the sense voltage (and therefore the inductor L1 peak current) required is adjusted based on the supply voltage so as to keep the energy charged constant over supply voltage. This is accomplished by means of the resistor dividers R19/R20 and R2a/b. The ASIC also detects a DC or full wave rectified power supply and adjusts the energy charged accordingly. The resistor divider R2a/b has 4 settings to correspond with the 4 energy settings, such that energy is kept flat over supply voltage on each energy setting.
The 60 μs (˜16 kHz) charge cycle is faster than the typical strobe charge cycle (8 kHz or less). This results in the benefits of a strobe that is quieter (16 kHz is not typically audible), and a boost inductor has a lower inductance (and is therefore smaller and cheaper).
Another improvement is the driver for the gate of transistor MQ4. This driver is high voltage, and runs at 9.5 v typically, which provides a greater Vgs to MQ4 so that it has a lower effective RON, and therefore providing greater drive current than a typical 5V logic output. The result is faster switching times (>200 ns vs.˜1 μs for an IRF710). Both of these improvements increase the efficiency of the DC-DC boost conversion by reducing losses in the transistor MQ4.
The DC-DC converter also has an over voltage protection feature. In the case that the strobe capacitor C9 does not discharge after a strobe signal is enabled, the DC-DC boost converter is turned off (MQ4 is held off) until the strobe capacitor is discharged and prevents an over voltage condition on the strobe capacitor.
In one embodiment, the value of R20 (26.7 k) has been adjusted for optimal Vsply operation. The value (e.g., from 20 k-27K) should be selected to allow for 8 v operation to still register as a high on Vsdet(>1.2 v, for safety margins). However, the resistor divider voltage should not get too high so as to exceed the maximum input voltage of Vdd1+0.3 v (typically 9.8 v). The values selected were chosen to go as high as is safe for the Vsdet input.
If there is still difficulty at low voltage operation, a forward biased diode (0.3 v<Vd<=0.75 v) could be added in series with R20, and R20 can be adjusted to 20 k. This is to ensure that the ASIC detects the power supply to be on normally, and off during a sync pulse.
In one embodiment, the non-sync implementation removes the Vsdet resistor divider. It should be noted that the ASIC 110 will not operate with full-wave rectified supply unless Vdd1 is decoupled: 24 v (16-33 v) operation requires at least 2 μF, 12 v (8-17 v) operation requires at least 3 μF of decoupling.
Although various embodiments which incorporate the teachings of the present invention have been shown and described in detail herein, those skilled in the art can readily devise many other varied embodiments that still incorporate these teachings.
This application is a continuation of U.S. patent application Ser. No. 10/650,464 filed on Aug. 28, 2003, which claims the benefit of U.S. Provisional Application No. 60/495,305 filed on Aug. 14, 2003, where each of the above cited applications is herein incorporated by reference. The present invention generally relates to an alarm unit. More particularly, the invention is a strobe alarm unit, a horn unit and/or a strobe and horn unit that is controlled by an ASIC (application specific integrated circuit) to provide audible and/or visual alarm notification.
| Number | Name | Date | Kind |
|---|---|---|---|
| 3792309 | McDonald | Feb 1974 | A |
| 3959688 | Bonazoli et al. | May 1976 | A |
| 4189718 | Carson et al. | Feb 1980 | A |
| 4477796 | Kearsley | Oct 1984 | A |
| 4578586 | Preston | Mar 1986 | A |
| 4625151 | Kataoka | Nov 1986 | A |
| 5126078 | Steiner et al. | Jun 1992 | A |
| 5347200 | De Nardis | Sep 1994 | A |
| 5595075 | Chen | Jan 1997 | A |
| 5675311 | Burnett et al. | Oct 1997 | A |
| 5691704 | Wong | Nov 1997 | A |
| 5694118 | Park et al. | Dec 1997 | A |
| 5896092 | Bechtel | Apr 1999 | A |
| 6078269 | Markwell et al. | Jun 2000 | A |
| 6166647 | Wong | Dec 2000 | A |
| 6243001 | Kodaka | Jun 2001 | B1 |
| 6307328 | Ko et al. | Oct 2001 | B1 |
| 6311021 | Kosich | Oct 2001 | B1 |
| 6313744 | Capowski et al. | Nov 2001 | B1 |
| 6369696 | Curran et al. | Apr 2002 | B2 |
| 6411201 | Hur et al. | Jun 2002 | B1 |
| 6437698 | Byrne et al. | Aug 2002 | B1 |
| 6522261 | Scheffler et al. | Feb 2003 | B2 |
| 6532406 | Schmedding et al. | Mar 2003 | B1 |
| 6556132 | Pattok et al. | Apr 2003 | B1 |
| 6617967 | Baldwin et al. | Sep 2003 | B2 |
| 7830252 | Keeney et al. | Nov 2010 | B2 |
| 20040169585 | Smith et al. | Sep 2004 | A1 |
| Number | Date | Country | |
|---|---|---|---|
| 20110050448 A1 | Mar 2011 | US |
| Number | Date | Country | |
|---|---|---|---|
| 60495305 | Aug 2003 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10650464 | Aug 2003 | US |
| Child | 12941897 | US |
