Patent Application
20070024123
The present application claims priority from Italian Patent Application No. MI2005A 001027 filed Jun. 1, 2005, the disclosure of which is hereby incorporated by reference.
1. Technical Field of the Invention
The present invention relates to an architecture for implementing an integrated capacitance.
The invention particularly, but not exclusively, relates to an architecture for implementing an integrated capacity to be used as tank capacitor in a converter of the Voltage Down type and the following description is made with reference to this field of application by way of illustration only.
2. Description of Related Art
Different architectures for converters of the voltage down type are known in the literature and are widely used.
In particular, in the field of Flash memories, it is known to use converters of the voltage down type or VDC converters (acronym for “Voltage Down Converter”) to allow the use of transistors with gate oxides having thickness lower than 40 Å (Angstrom) in applications requiring conventional supply voltage values, in particular equal to 3V.
In this case, a VDC converter of the known type allows one to obtain a regulated voltage with values equal or lower than 1.8V starting from an external supply voltage in the range 2.4-4.0V. It is also possible, by using a booster circuit inserted in the VDC converter, to obtain a correct regulation also with an external supply voltage with values up to 2V.
In particular, known VDC converters essentially comprise an open or closed loop regulation circuit connected to an output driver, realized for example by means of a MOS transistor with N channel or P channel.
In the case of application to a Flash memory of the NOR type, characterized by extremely quick impulsive consumption (from 0 to tens of mA in less than 10 ns) due to the reading architecture of the asynchronous type with response times in the order of 50 ns, it is also known to use a channel width modulation VDC converter, schematically shown in
The channel modulation VDC converter 1 essentially comprises a regulator 2 of the closed loop series type connected to an output driver 3, in particular realized by means of a Mout transistor of the NMOS type.
The regulator 2 has the task of generating a reference voltage signal VREF having a value equal to a voltage value which is to be present at the output.
In particular, the regulator 2 comprises an operational amplifier A supplied by a first voltage reference, in particular a supply voltage reference Vdd and having a first input terminal connected to an input terminal IN of the channel modulation VDC converter 1, a second input terminal connected, by means of a first impedance element Z1, to a second voltage reference, in particular a ground reference GND, as well as an output terminal connected to a control terminal of the transistor Msb of the output driver 3.
The input terminal IN of the channel modulation converter VDC 1 supplies to the first input terminal of the operational amplifier A of the regulator 2 a stable voltage signal VBGAP.
It is also possible to use, as supply voltage reference of the operational amplifier A, a boosted voltage reference Surv.
The regulator 2 also comprises a second impedance element Z2 connected between the first input terminal of the operational amplifier A and a first conduction terminal of the transistor Msb of the output driver 3, having in turn a second conduction terminal connected to the supply voltage reference Vdd and the control terminal connected to a terminal O3 of the output driver 3.
The channel modulation converter VDC 1 also comprises a stand by driver 4, in particular realized by means of a transistor Msb of the NMOS type inserted between a first voltage reference, in particular a supply voltage reference Vdd and an output terminal OUT of the channel modulation VDC converter 1 and having a control terminal connected to the output terminal O3 of the output driver 3.
Moreover, the channel modulation converter VDC 1 comprises a power block 5, inserted between the supply voltage reference Vdd and the output terminal OUT of the channel modulation converter VDC 1 and realized by means of a plurality of transistors M1 . . . MN, in particular transistors of the NMOS type, inserted, in parallel to each other, between the supply voltage reference Vdd and the output terminal OUT and having the control terminals connected to each other and to the control terminal of the transistor Msb of the stand by driver 4.
In particular, the transistors M1 . . . MN of the power block 5 are connected to the supply voltage reference Vdd by means of a plurality of switches SW1 . . . SWN subjected to a plurality of control signals S1 . . . SN generated by a control block 6, in turn being input an output voltage signal VOUT on the output terminal OUT of the channel modulation converter VDC 1 and the voltage reference signal VREF generated by the regulator 2.
In this way, the power block 5 is able to supply a current value requested by a load connected to the channel modulation converter VDC 1 by means of turning on the transistors M1 . . . MN. In particular, the control block 6 compares the value of the reference voltage signal VREF with the value of the output voltage signal VOUT and drives, by means of the control signals S1 . . . SN the partial or total turn on and off of the transistors M1 . . . MN, which can thus be considered output drivers.
In this way, the channel modulation converter VDC 1 allows to adapt, quickly enough, the transconductance value of the output driver 3 on the basis of the current requests of a load associated with the converter itself. In particular, the power block 5 allows to control this transconductance value through a dynamic modulation of the channel width W thanks to the activation/deactivation of the transistors M1 . . . MN in parallel to each other, thus obtaining a regulated output voltage signal VOUT.
Moreover, the stand by driver 4 realizes an open loop replication of the reference voltage signal VREF, for supplying a weak current value requested in the stand-by mode, thus allowing the turn off of the power block 5 to limit the overall consumption of the channel modulation converter VDC 1 in this mode.
The channel modulation converter VDC 1 finally comprises a so called tank capacitor CTANK inserted between its output terminal OUT and the ground reference GND.
This tank capacitor CTANK allows to sustain the output voltage signal VOUT in the brief transient necessary for the channel modulation converter VDC 1 to reach an activation state of the regulation. In other words, the role of the tank capacitor CTANK is that of addressing currents of the impulse type dissipated by a load connected to the channel modulation converter VDC 1 and thus avoid a temporary “breakdown” of the regulated output voltage signal VOUT.
To do this it is possible to use a capacitor having a value equal to only 2 nF, making the channel modulation converter VDC 1 compatible with an integrated solution.
In reality, in sub-micrometric technologies with oxide thickness in the order of 40 Angstrom, to avoid relatively low performances due to the faultiness of the oxide layers, this tank capacitor CTANK is to be realized through the series of two capacitors of double value, as shown in
It is to be remembered in fact that, in general, using a series of capacitors is very expensive in terms of area occupation on the silicon and thus unadvised in integrated applications. In the case at issue, having to use capacitors with high value (in the order of nano-farads), and to avoid having to “eliminate” a device which, in the testing step, has a fault on these capacitors the series of two capacitors is used so that, if one of these capacitors has a “leakage” current the presence of the other capacitor allows to ensure the regular operation, the possibility that both the capacitors have leakage currents being very low.
This solution however implies a great waste of integration area on silicon and thus makes the channel modulation converter VDC 1 less attractive for integrated applications in these technologies.
There is accordingly a need for providing an integrated capacitor, in particular to be used as tank capacitor in a channel modulation converter VDC, having reduced integration area occupation on silicon so as to allow to overcome the limits and the drawbacks still affecting the solutions used by the prior art.
In accordance with the present invention an architecture is presented for spitting an integrated capacitor in a plurality of modular capacitors which can be activated in a selective way.
In an embodiment, an architecture for implementing an integrated capacitance, comprises a capacitive block inserted between a first and a second voltage reference comprising a plurality of elementary capacitive modules. An enable block is inserted between said first voltage reference and said capacitive block and comprises a plurality of switches connected to said elementary capacitive modules, the switches driven on control terminals by a plurality of control signals, each switch of said enable block being inserted between said first voltage reference and a first end of a corresponding elementary capacitive module, a second end of the elementary capacitive module connected to said second voltage reference. A verify and enable circuit is connected to said first voltage reference and to the input of said first end of said elementary capacitive modules and to the output of said control terminals of said switches of said enable block, said verify and enable circuit detecting the presence of a current value in each elementary capacitive module and, if said current is detected, disabling said elementary capacitive module of said capacitive block by means of said corresponding switch of said enable block.
In an embodiment, the foregoing architecture is utilized to for a tank capacitor for a channel modulation converter.
In another embodiment, an integrated circuit comprises a first capacitor and a second capacitor. A first transistor switch is coupled in series with the first capacitor at a first node, and a second transistor switch is coupled in series with the second capacitor at a second node. A first sensor circuit is coupled to the first node and operable to detect leakage current with respect to the first capacitor and in response thereto disable the first transistor switch, and a second sensor circuit is coupled to the second node and operable to detect leakage current with respect to the second capacitor and in response thereto disable the second transistor switch.
A more complete understanding of the method and apparatus of the present invention may be acquired by reference to the following Detailed Description when taken in conjunction with the accompanying Drawings wherein:
With reference to these figures, and in particular to
The architecture 10 comprises, in particular, a capacitive block 11 inserted between a first voltage reference, in particular an inner supply voltage reference VOUT, and a second voltage reference, in particular a ground reference GND.
The capacitive block 11 comprises in turn a plurality of elementary capacitive modules, indicated as single capacitors C1 . . . CN, N being the number of elementary capacitive modules comprised in the capacitive block 11.
Advantageously according to the invention, the architecture 10 also comprises an enable block 12, inserted between the inner supply voltage reference VOUT and the capacitive block 11.
The enable block 12 comprises a plurality of switches, realized by means of PMOS transistors and indicated with MC1 . . . MCN, driven by means of a plurality of control signals VO1 . . . VON suitably applied to their control terminals.
In particular, each elementary capacitive module or capacity Ci of the capacitive block 11 corresponds to a switch or transistor MCi, the transistor MCi being inserted between the inner supply voltage reference VOUT and a first end of the capacitor Ci, having in turn a second end connected to the ground GND.
Advantageously according to the invention, the capacitive block 11 is thus made of N elementary capacitive modules, each of them being apt for being activated/deactivated by means of a corresponding switch comprised in the enable block 12.
These elementary capacitive modules are suitably sized so that the sum of all the respective capacitive values is equal to a capacitance maximum value which is to be obtained.
Moreover, the PMOS transistors which realize the switches of the enable block 12 are preferably of the low voltage type with low turn on resistance value, RON.
When the architecture 10 is integrated on silicon, these transistors MCi allow to easily deactivate the corresponding elementary capacitive modules Ci in case of faultiness of the oxide which realizes these elementary capacitive modules Ci.
Moreover, the transistors MCi are sized, in a way subordinate to the turn on resistance value RON, to minimize a value of total resistance seen in series to the capacitive block 11 not to degrade the response of the capacitor implemented by the architecture 10 and thus the performance of a system such as a channel modulation converter VDC this architecture would be associated with.
In a simulation, it has been verified that a turn on resistance value RON equal to about 5 Ohms per elementary capacitive module of value equal to 20 Pico-farad does not affect the performance of a channel modulation converter VDC using the architecture 10 for implementing a tank capacitor CTANK, as seen in relation to the prior art.
Advantageously according to embodiments of the invention, the architecture 10 also comprises a verify and enable circuit 13 able to control, at each turn on of the capacity implemented by the architecture 10, the integrity of the elementary capacitive modules, due to the verification of the presence or not of a current value in these modules and of disabling the modules for which this integrity verification gives a negative result. The verify and enable circuit 13 is connected at the input to an end of the elementary capacitive modules, receiving therefrom suitable detection voltage signals Vii and it is connected at the output to the control terminals of the switches, supplying to them suitable voltage control signals VOi. The verify and enable circuit 13 is also supplied by the inner supply voltage reference VOUT and has an enable terminal receiving a suitable enable signal of the verify operations of the capacitive block 11, indicated with VERIFY.
The enable signal VERIFY is set by a so called Power-on-Reset or POR signal, usually in electronic devices and in particular in the memories and rest by one of the signals of end of the power on steps of the device.
In particular, it is to be remembered that if a faultiness of the oxide realizing a given elementary capacitive module is present, this module will show a “leakage” current, usually in the order of the tens of micro-ampere (μA). In other words, the presence of the leakage current is index of a faultiness in the oxide and thus in the elementary capacitive module realized thereby.
Advantageously according to the invention, the verify and enable circuit 13 thus provides to verify the presence of a leakage current and to disable the elementary capacitive modules which show such a current, i.e. a faultiness of the oxide, everything already in a preliminary testing step of the architecture 10 or of a device using it.
In particular, the verify and enable circuit 13 suitably ascertains if a “leakage” current of each elementary capacitive module is present or not, if this “leakage” current is present it excludes the module under examination by means of the corresponding switch.
To do this, the verify and enable circuit 13 comprises a plurality of comparison stages 14, connected to the inner supply voltage reference VOUT and indicated with AD1 . . . ADN, as schematically shown in
A generic comparison stage 14, as schematically shown in
Advantageously according to the invention, a comparison stage 14 is configured so as to detect the presence of a “leakage” current in the elementary capacitive module Ci connected thereto and, on the basis of the presence or not of this “leakage” current, to generate a suitable enable voltage signal VOi for the corresponding switch MCi.
A first embodiment of a comparison stage 14 according to the invention is schematically shown in
The comparison stage 14 comprises a comparator AD0, in particular a comparator with offset, connected to the inner supply voltage reference VOUT by means of an activation transistor M14 and to the ground GND, having a first input terminal, in particular a non inverting input terminal (+), connected to the inner supply voltage reference VOUT, a second input terminal, in particular an inverting input terminal (−) connected to an end of a corresponding elementary capacitive module Ci and receiving therefrom the detection voltage signal Vii and an output terminal connected to a first input terminal of a logic gate ND. It is to be noted that the first input terminal of the comparator ASD0 is connected to a source terminal Si of the switch MCi and the second input terminal of the comparator AD0 is connected to the drain terminal Di of the switch MCi.
In turn, the logic gate ND has a second input terminal receiving a negate value VERIFY_N of the enable signal VERIFY and an output terminal connected to a control terminal of the switch MCi corresponding to the elementary capacitive module Ci and supplying to it the control voltage signal VOi.
Moreover, the negate value VERIFY_N of the enable signal VERIFY is also applied to the control terminal of the activation transistor M14.
In the example shown in
Let's now see the operation of the comparison stage 14 according to the invention. It is to be noted that each comparison stage 14 is only active during the power-on steps of the architecture 10, the enable signal VERIFY being active during the power-on or during the testing step of the device.
The comparison step 14 carries out, by means of the comparator with offset AD0, a comparison of the voltage values on the drain Di and source Si terminals of the switch MCi, to ascertain the presence or not of a leakage current of the module Ci under examination, this comparison being carried out when this transistor switch MCi is off. In fact, advantageously according to the invention, the negate value VERIFY_N of the enable signal VERIFY activates the verification of the module Ci by enabling the comparator with offset AD0 and turning off the transistor MCi.
At this point, if a leakage current is present, the value of the detection voltage signal Vii of the capacity Ci is decreased, causing the releasing of the comparator with offset AD0 and thus the automatic exclusion of the module Ci itself.
In other words, the choice of this offset of the comparator with offset AD0 allows to set the threshold value of the leakage current therefore exclude a determined elementary capacitive module, i.e. consider this module as faulty.
It is also possible to realize the comparison stage 14 according to a further embodiment of the invention schematically shown in
The comparison stage 14 comprises a bistable circuit or flip-flop 15, inserted between the inner supply voltage reference VOUT, i.e. the source terminal Si of the switch MCi, and the ground GND and having a first set terminal S connected to an end of the module Ci, i.e. to the drain terminal Di of the switch MCi, and receiving therefrom the detection voltage signal Vii, a second reset terminal R receiving the Power-on-Reset POR signal and an output terminal O.
In particular, the flip-flop 15 comprises a first Ma and a second transistor Mb, inserted, in series to each other, between the inner supply voltage reference VOUT and the ground, the first transistor Ma having a control terminal connected to the first set terminal S of the flip-flop 15 and the second transistor Mb having a control terminal connected to the second reset terminal R of the flip-flop 15.
The flip-flop 15 also comprises a first logic inverter IV1 connected between the output terminal O of the flip-flop 15 and a drain terminal of the first transistor Ma, as well as a second logic inverter IV2 connected between the output terminal O of the flip-flop 15 and a source terminal of the second transistor Mb, these first IV1 and second logic inverter IV2 being connected to the output terminal O of the flip-flop 15 in opposite directions with respect to each other.
The comparison stage 14 also comprises a logic gate ND having a first input terminal connected to a third logic inverter IV3, in turn being input the enable signal VERIFY, a second input terminal connected to the output terminal O of the flip-flop 15, as well as an output terminal connected to the control terminal of the switch MCi and supplying to it the control voltage signal VOi.
Let's now see the operation of this further embodiment of the comparison stage 14 according to the invention.
The enable signal VERIFY determines the start of the verify step of a given module Ci, powering off the corresponding switch MCi. The detection voltage signal Vii biases the control terminal of the first comparison transistor Ma which represents the set terminal S of the flip-flop 15.
The possible leakage current in the module Ci decreases the detection voltage signal Vii until the second transistor Mb is powered on and thus the setting of the flip-flop 15 is carried out. This flip-flop 15, changing state, thus causes the deactivation of the corresponding module Ci.
In other words, to verify the integrity of the module Ci, the switch MCi is turned off and, in a time range ΔT, the detection voltage signal Vii is measured on the elementary capacitive module Ci. This time range ΔT is determined by the duration of the enable signal VERIFY.
Moreover, the Power-on-Reset POR signal resets the flip-flop 15 in correspondence with each power-on step of the device.
It is to be noted that a pondered choice of this time range ΔT, i.e. of the duration of the verify step, determined by the time range in which the enable signal VERIFY is active, allows to discriminate leakage currents also of some micro-ampere.
In fact, during the verify step the elementary capacitive module Ci of the capacitive module 11, if a leakage current is present, is discharged. The variation of potential at its ends depends on the value of this leakage current and on the duration ΔT of the verify step. The choice of this duration ΔT thus allows to set the threshold value of the leakage current therefore exclude a determined elementary capacitive value, i.e. consider this module as faulty.
Also in this case, the control of the integrity of the modules Ci is carried out by the comparison stage 14 at each power-on step of the architecture 10.
Moreover, advantageously according to the invention, it is possible to implement a verification of the number of active modules of the architecture 10 used inside a memory device of the Flash type.
To do this, the output terminals of the comparator AD0 or of the flip-flop 15 are connected to the DBUS of the Flash memory (the DBUS being a bus of at least 32 bits) and, with the help of the micro-controller inside the Flash memory itself, it is possible to read the values present on these output terminals or also address them onto the DQPAD terminals (output pad) of the Flash memory.
The architecture 10 is advantageously applied in the VDC converters (Voltage Down Converter), in particular in the integration of the tank capacity CTANK of a channel modulation converter VDC.
It is in fact immediate to verify the effective saving in terms of area of the silicon obtained with the use of the architecture 10 for implementing this tank capacity CTANK.
By way of verification, the Applicant has carried out integration simulations of a channel modulation converter VDC in 0.13 mm technology according to the following hypothesis: a tank capacity value CTANK being fixed at 1.28 nF, an architecture 10 is realized comprising a capacitive block 11 of N=64 elementary capacitive modules, each module Ci having value equal to 20 pF and being connected to a MCi p-channel enable transistor having size W=120 μm and L=0.18 μm.
Under typical operation conditions, the transistor MCi being simulated shows a power on resistance value RON equal to 4 Ohms, such as not to degrade the performances of the channel modulation converter VDC.
For each comparison stage realized according to the example of
It is thus verified how the architecture proposed implies the 60% saving of area on the silicon.
It is also immediate to verify that the percentage relative to the area saving depends on the value chosen as single capacitive module 11 of the architecture 10 used for realizing the tank capacity CTANK, as well as on the sizes of the switches MCi. The example used for the simulation represents a compromise between area saving and performances of the converter as a whole.
It is also to be highlighted that this advantage in terms of area saving is even more evident in case the channel modulation converter VDC is used for supplying small Flash memories where the portion of circuitry is strongly affecting in terms of occupied area with respect to the portion of matrixes.
Obviously, a skilled in the art will be allowed to bring several modifications and changes to the above described architecture so as to meet contingent and specific needs, all those modifications and changes falling within the scope of protection of the invention as defined by the following claims.
In particular, it is possible to consider applications wherein it is suitable to carry our an activation/deactivation of the elementary capacitive modules comprised in the capacitive block 11 in a cyclical way, arranging a number N of elementary capacitive modules higher than that necessary for realizing the desired capacity value. In this way, it is in fact possible to reduce the stress accumulated by the elementary capacitive modules and to decrease the breakdown risks of the same.
Although preferred embodiments of the device of the present invention have been illustrated in the accompanying Drawings and described in the foregoing Detailed Description, it will be understood that the invention is not limited to the embodiments disclosed, but is capable of numerous rearrangements, modifications and substitutions without departing from the spirit of the invention as set forth and defined by the following claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| MI2005A 001027 | Jun 2005 | IT | national |
