High repetition rate transient recorder with automatic integration
- Publication number
- CA1225155A CA1225155A CA 429710 CA429710A CA1225155A CA 1225155 A CA1225155 A CA 1225155A CA 429710 CA429710 CA 429710 CA 429710 A CA429710 A CA 429710A CA 1225155 A CA1225155 A CA 1225155A
- Grant status
- Patent type
- Prior art keywords
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- H01—BASIC ELECTRIC ELEMENTS
- H01J—ELECTRIC DISCHARGE TUBES OR DISCHARGE LAMPS
- H01J49/00—Particle spectrometers or separator tubes
- H01J49/0027—Methods for using particle spectrometers
- H01J49/0036—Step by step routines describing the handling of the data generated during a measurement
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R13/00—Arrangements for displaying electric variables or waveforms
- G01R13/04—Arrangements for displaying electric variables or waveforms for producing permanent records
- G01R13/06—Modifications for recording transient disturbances, e.g. by starting or accelerating a recording medium
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R13/00—Arrangements for displaying electric variables or waveforms
- G01R13/20—Cathode-ray oscilloscopes ; Oscilloscopes using other screens than CRT's, e.g. LCD's
- G01R13/22—Circuits therefor
- G01R13/32—Circuits for displaying non-recurrent functions such as transients; Circuits for triggering; Circuits for synchronisation; Circuits for time-base expansion
A system for high speed acquisition and storage of data from transient electrical signal waveforms sampled in time at a multiplicity of points, and where it is desired to collect and store data in real time in a mass storage unit, but where the data rate is in excess of the apparent rate at which data can be transferred to mass storage. The overall system employs two levels of data reduction, the first being digital integration (summation) and the second being data processing. In the digital summation section memory read and write cycles overlap for higher speed. In one embodiment, a charge-coupled device is employed as an input element to reduce the data rate prior to subsequent digital processing. The system is capable of recording all the intensities of all transients continously without loss of data significance.
~LZ25~5Si RECORDER WITH AUTOMATIC INTEGRATION
Support for this invention was received through 5 Michigan State University, National Institutes of Health, and Office of Naval Research.
The present invention relates to high speed acquisition and storage of data from transient electrical signal waveforms sampled in time at a multiplicity of 10 points. The invention particularly relates to situations where it is desired to collect and store data in real time in a mass storage unit such as a magnetic disk unit, but where the data rate is in excess of the apparent rate at which data can be transferred to mass storage.
The subject transient recorder was developed specifically in the context of recording the output of a tlme-of-flight (TO) mass spectrometer, but has other applications as well. In particular, the subject invention has application in any transient data recording situation 20 where repetitive transients occur at a relatively high rate but the information from one transient to the next changes at a relatively slow rate. As another example, the present invention has application where a chemical system is pulsed with a laser to produce decaying fluorescence output 25 pulses. However, for purposes of example and to facilitate description, the invention is described herein in the context of TO mass spectrometer.
In a time-of-flight mass spectrometer, sample ions are produced and then extracted and accelerated by an 30 accelerating voltage applied to suitable acceleration electrodes. A typical value of accelerating voltage is 3.5 TV. Constant energy and constant-momentum acceleration modes are known.
1 In either case, lighter (lower mass) ions are accelerated to higher velocities than the heavier ions.
The ions then enter a drift region or flight tube which establishes an ion path length 1, and which is followed 5 by an ion detector. In the drift region, the ions separate along the ion path as a function of their individual velocities and thus arrive at the ion detector at different times depending upon their velocities, and therefore, depending upon their mass.
To permit measurement of flight time, ions in a time-of~flight mass spectrometer are bunched, typically by means of a pulsed ion source, and all ions of a given bunch enter the drift region at substantially the same position and time. By correlating ion pulsing or bunching with 15 arrival time of various ions at the ion detector, the time-of-flight of each individual ion or group of identical-mass ions can be determined. Ion velocity follows from the simple relationship:
(Velocity) = (Path Length)/(Time-Of-Flight).
20 From velocity, ion mass can be calculated, taking into account the characteristics of the ion accelerator.
With an ion path or flight length 1 of 1.0 meter, all ions from one pulse of the ion source, ranging from 1 to 1000 mass units, reach the ion detector within 40 25 microseconds. Many individual ions of any given mass may reach the detector at substantially the same time, ranging up to several hundreds of ions. The source pulses are repeated at a rate in the order of 10 to 25 kHz.
The output of -the ion detector is a transient
3 waveform for each source pulse. Each transient waveform has a magnitude which varies as a function of time, with peaks of the waveform along a time axis corresponding Jo different masses of the various sample ions.
3 ~2S~S5 1 It will be appreciated that the data rate is extremely high, much faster than can be stored by any known mass storage unit it a magnetic disk unit). For example, in each transient there may be as many as 16,000 5 relevant sample points in time (or "time bouncily). At a source pulse rate of 10 kHz, these two factors give a data rate of 1.6 x 108 items of information per second.
Moreover, for any given time bin up to several hundred ions may be arriving which can be represented by a data word 10 Of eight binary bits. Optimally, every single ion arriving at the ion detector can be resolved with intensities of 255 ions per bin or less. This last factor increases the potential data rate to 1.3 x 109 bits per second. Clearly this is too fast for known mass storage techniques.
Commonly-available commercial time-of-flight mass spectrometers record detected ion current intensities by sampling -techniques. Ion current is sampled during only one arrival time for each source pulse. A sampling window or time slice (aperture time) is established and the delay 20 from the extraction pulse to this window is slowly scanned over all arrival times of interest while the, source is repetitively pulsed, thereby recording a complete mass spectrum of the sample under study by collecting the ion intensities for each successive arrival time. This 25 technique is known as Time-Slice Detection (TED).
Additionally integrating forms of time-slice detectors have been employed, known as "boxcar integrators".
The boxcar integrator is triggered for each ion pulse r and integrates ion current during the same aperture time at a 3 constant arrival time for a number of pulses. The arrival time can either be constant or be slowly scanned.
Integration itself represents a means of data flow reduction where the information contained in successive pulses is changing slowly, as is the case in many TO mass I ~22~55 1 spectrometer applications. Although the data pulses are occurring at a 10 kHz rate, the spectrum each pulse detects is changing at a much slower rate, and time-resolved data from 10 to 1000 scans can be integrated or averaged with-5 out loss of actual information.
Time-Slice Detection has the disadvantage of losing most of the information available in the ion beam since aperture time is a small fraction of the total time over which ions are arriving at the detector. This creates lo a potential problem where sampling times or sample quantities are limited. Accordingly, various devices for Time Array Detection (TAD) have been proposed, known variously as "transient recorders" or "digital transient recorders". Such recorders, rather than responding to a 15 single time slice relative to the pulsed source, collect the entire output from a single source pulse in a time-of-flight mass spectrometer to produce individual time-resolved data channels for each of a multiplicity of sample points taken serially in time.
For example, Lincoln has constructed a detector system which captures a substantial fraction of the information in a single ion source pulse from a time-of-flight mass spectrometer employing a digital transient recorder having a OK memory (Biomation Model 25 8100). See KIWI. Lincoln, "Data Acquisition Techniques for Exploiting the Uniqueness of the Time-of-Flight Mass Spectrometer: Application to Sampling Pulsed Gas Systems", Dyne Mass Spectrum, 111-119 (1981~; also published as NASA Report Tm-81224.
3 Prior art digital transient recorders, although offering an improvement over time-slice detection, are not capable of measuring ten thousand transients per second I ~225~55 - 1 consistent with the ten thousand per second pulse rate typical in TO mass spectrometer, and thus lye data as a result of spectra not collected. In particular, their data readout time is in the order of milliseconcls,sand is 5 inconsistent with the 10 kHz or greater pulsate of time-of-flight analysis. Moreover, only a limited number of time-resolved channels, for example 2000, ye available in typical prior art instruments.
Just as a boxcar integrator is an integrating lo form of time-slice detector, integrating for~i,of digital transient recorders have been employed, although operating relatively slowly.
One example of such a device is know as a Computer of Average Transients, or "CAT". -, As another example, the Lincoln digital transient recorder, as described in -the li-te~ture cited ,,, above, has its digital output connected to aspirate I, , "Signal Average" which functions as an integrator. For this purpose, Incline employs a Nucleate Muddle with a 20 Model 178 plug-in unit specifically made for digital-to-digital interfacing with the Bohemian Transient Recorder. As Lincoln points out, known "signal averages"
are not fast enough to acquire spectra in retime. In the Lincoln system, the rate-determining rat limiting 25 factor is the approximately three millisecond required -to dump the 2000-word memory of the transient reorder into the signal average. This sequence of eventsrenables only 330 transient pulses to be analyzed each second.
Up to this point, the background oath invention 3 has been described in the context of convent~nal time-of-flight (TO) mass spectrometer Indeed, thellpresent invention provides significant advantages in,~çonventional TO mass spectrometer. Jo ~;:Z5~55 l There is, however, another, completely new form of time-resolved mass spectrometer with which the subject invention may be employed as an element of an overall detector system. Specifically, this new form of time-5 resolved mass spectrometer is disclosed and claimed in commonly-assigned US. Patent No. 4,~72,631, granted September 18, 1984, , by Christie George EKE, John Timothy STUNTS and John Francis HOLLAND, entitled "COMBINATION OF TIP RESOLUTION AND MASS DISPERSIVE
10 TECHNIQUES IN MASS SPECTROMETER". In the instruments disclosed in the above-identified -US. Patent No.
4,472,631,time-of-flight mass spectrometer techniques are simultaneously combined with path-bending spatial dispersion in magnetic- or electric-sector mass spectrometers to 15 improve the mass resolution or, with an ion fragmentation region, to rapidly obtain the same multidimensional mass spectral data previously obtained by tandem mass Spector-metro. The technique may be identified as time-resolved magnetic or electric-sector mass spectrometer. The 20 instrumentation generates data defining relationships between selected parent ions and daughter ions produced by fragmentation (either metastable or induced), data to differentiate stable from metastable ions, and data to improve mass resolution.
In these instruments, it is highly advantageous to rapidly and continuously collect data in real time so that the full benefits of the combined techniques can be achieved. Specifically, it is desirable to acquire and store data at a rate of 200 MHz. Moreover, for greatest 3 sensitivity, particularly where sample quantities are limited, it is desirable that data be continuously collected and recorded, with no pauses during operation.
Ideally, every single ion reaching the ion detector is recorded in its proper time-resolved channel, and no data 35 significance is lost.
Z5~55 - 1 With these time-resolved magnetic- or electric-sector instruments, the information contained in successive transient output pulse changes at rates up to 1000 times per second. Thus, assuming a pulse rate of 10
5 kHz, information could be lost if the information in more than ten pulses is averaged, although in many cases the data of fifty or more pulses might be averaged. Further, it will be appreciated that any pause in data collection (i.e. for readout following integration) can lead to a lo substantial loss of significant data.
The present invention provides a high speed data acquisition and storage system for processing and recording data values from repetitive transient waveforms without loss of data significance with an integrating 15 transient recorder capable of acquiring and integrating transient data at a rate in excess of 200 MHz. the subject invention records all the intensities of all -transients and effectively integrates sampled data as fast as it is acquired without interruption for data read out.
Briefly, in accordance with an overall concept of the invention, a high speed data acquisition and storage system is provided for processing and storing input data values representative of the magnitude of a repetitive transient signal waveform sampled for each occurrence at a 25 plurality of sample points in time. The system includes a data summation system for summing (digitally integrating) the data values in a predetermined number of individual time-resolved time bins over a predetermined number of repetitive transient signal waveforms with each time bin 3 corresponding to a sample point, and for outputting the summed data in the time bins.
1 The system also includes a data reduction system for compacting the summed data from the time bins in real time, and a mass storage unit for receiving compacted data from the data reduction system, also in real time.
The data reduction system performs various operations on the data, including limited calculations, before transferring the data to the mass storage unit. For example, a threshold is established, and all summed data values below this threshold are treated as zero. Any one 10~ of various data compaction techniques is employed. Mass spectrum peaks are identified through calculation and their position and magnitude recorded, without necessity for the raw data defining the peak to be recorded.
In a more particular aspect of the invention, 15 there is provided a data summation system for receiving successive input data values representative of the magnitude of a repetitive transient signal waveform sampled for each occurrence at a plurality of sample points in time, for summing the data values in a predetermined number of 20 repetitive transient signal waveforms with each time bin corresponding to a sample point, and for enabling output of the summation results in each time bin to occur while continuing to receive and process incoming data. The data summation system includes at least one data summation 25 subsystem, which in turn includes an input latch for temporarily storing each successive input data value, and a set of three summation sub circuits organized such that one is a common summation sub circuit capable of being paired with each of the others. Each summation sub circuit 3 includes an adder with a pair of inputs, one adder input being connected to receive data from the input latch; an intermediate data latch connected for temporarily storing Lowe g 1 the output of the adder, and a memory having an individual memory location corresponding to each time bin, the memory being connected to the intermediate data latch for receiving data to be stored. The other adder input of each 5 of the other summation suhcircuits is connected to receive data from the memory of the common summation sub circuit.
The data summation subsystem also includes an input multiplexer and an output multiplexer arranged to interconnect the summation sub circuits for summation 10 operation in pairs alternately selected from the common summation sub circuit and one of the other summation sobriquets, with the unselected summation Sybarite enabled for read out of completed summation data. The.
input multiplexer connects the output of the memory of the 15 selected one of the other summation sobriquets to the other adder input of the common summation Sybarite, and the output multiplexer connects the output of the memory of the unsalted one of the other summation sub circuits to an output of the data summation subsystem.
In operation, the data summation subsystem functions during the occurrence of each transient signal waveform for each successive input data value received to store the data value in the input latch, to read previously-stored cumulative sum data for the particular 25 time bin eorrespondin~ to the input data value from the memory of one of the selected summation sobriquets, to add in the adder of the other of the selected summation sobriquets the previously stored cumulative sum data to the new data value stored in said input latch to provide a 3 new cumulative sum for the particular time bin corresponding to the input data value, and to temporarily store the new cumulative sum in the intermediate latch of the other of the selected summation sub circuits. Memory read and write operations overlap, with new cumulative sum data for the 35 particular time bin corresponding to the previous input data I, Jo .
:~2~5~L55 1 value stored in the intermediate latch of the other of the selected summation sub circuits being stored in the memory of the other of -the selected summation sub circuits at the same time previously-stored cumulative new data is read 5 from the memory of the one of the selected summation sub-circuits. Following the occurrence of each complete transient signal wave form the roles of the two selected summation sub circuits are reversed. Following the pro-determined number of repetitive transient signal waveforms, 10 final cumulative sum data for each time bin is left in the memory of whichever of the selected summation sub circuits is not the common summation sub circuit.
Preferably, -the data summation system includes a plurality of data summation subsystems, each including a 15 subset of the individual -time-bin, with input data being circulated to the individual data summation subsystems.
FIG. 1 is an overall block diagram of a time-of-flight mass spectrometer including an integrating 20 transient recorder in accordance with the present invention;
FIG. 2 is a memory map depicting the manner in which banks of random access memories are organized into "time bins" corresponding to individual time-resolved channels;
FIG. 3 is an overall block diagram of a data acquisition system in accordance with the invention;
FIG. 4 is a block diagram depicting a representative one of eight summation systems in the FIG.
3 data acquisition system;
3 FIG. 5 is a block diagram of a data reduction subsystem; and FIG. 6 is a block diagram of an embodiment employing charge-coupled devices as an analog summing and storage element.
--if--1 Referring first to FIG. 1, a time-of flight mass spectrometer, generally designated 10, comprises a pulsed ion source 12 for producing and accelerating ions to velocities inversely related to their mass, an ion path 5 14 for allowing separation or dispersion in time in accordance with ion velocity, and an ion detector 16 for detecting ion current and providing an output as a function of time as depicted by the transient analog waveform 18. The pulsed ion source 12 is controlled by a pulse generator 20.
In operation, the ion source 12 provides ions in discrete packets or bunches, such as represented at 21, and accelerates the ions to velocities as an inverse function of mass. Thus, heavier ions have lower velocities than lighter ions. An exemplary ion pulse width is 10 15 nanoseconds, with a repetition rate of 10 kHz being -typical.
Along the ion path 14, the ions disperse as a function of their velocities, and therefore of their masses. The fighter, faster ions arrive first at the ion 20 detector 16, followed by the heavier, slower ions. Thus, ion detector current as a function of time indicates ion mass, and the waveform 18 represents the mass spectrum of the ions comprising a single source pulse, e.g., for the source pulse (not shown) immediately proceeding the 25 depicted pulse 21. The transient waveforms 18 then repeat at a 10 kHz rate.
The ion detector 16 is any suitable -type, for example an electron multiplier. The ion detector 16 has its output connected to an amplifier 22 suitably 3 configured as a current-to-voltage converter.
An integrating transient recorder in accordance with the invention is generally designated 24, and represents . .
1 the first of two levels of data reduction in the overall FIG. 1 system prior to mass storage.
More particularly, the integrating transient recorder 24 comprises a high-speed analog--to-digital 5 converter 26 connected to a summation system 28 which samples the output of the ion detector 16 at a 200 MHz rate and digitally integrates, in individual time-resolved channels or "time bins', a plurality (e.g. any number from about 10 to 1000) of individual source pulses. In this lo example the individual ion pulses occur at a 10 kHz rate.
By way of example, there are 16384 discrete time bins, each storing a representation of sampled ion current representative of a 5 nanosecond slice of time. The individual time bins collectively digitally represent 15 the ion detector 16 output waveform integrated (digitally summed) over a plurality of individual source pulses. So long as the information contained in the individual transient pulses changes at a rate slower than the lo kHz source pulse rate and is consistent with the number of 20 pulses summed in individual time-resolved channels, no data significance is lost by this process.
To enable ion pulsing time to be correlated with arrival time of ions at the ion detector 16, the pulse generator 20 which controls the ion source 12 is 25 also connected to the summation and store section 28.
Although not specifically shown, it will be appreciated that a suitable delay element is preferably included so that, following each ion pulse from the source 20, data acquisi-lion does not commence until the arrival time of the 3 lightest fastest) ions of each particular pulse. In the subject system, this time delay is included in the control and timing circuitry (not shown) for the integrating transient recorder 24.
issue 1 Following the summation system 28 is a data reduction subsystem 30, described hereinafter with reference to FIG. 5. The output of the data reduction system is connected to a disk interface 32 which represents 5 a mass storage unit. The-data reduction system 30 thus represents the second of two levels of data reduction in the overall FIG. 1 system.
As is described hereinafter in detail with reference to FIGS. 3 and 4, the summation system 28 lo comprises an exemplary eight summation subsystems. Each summation subsystem includes three random-access Memories (Rams). For convenience of illustration only in FIG. 2, the three Rams of each summation subsystem are representated as a single RAY bank. Thus there are eight repr~sen-ta-ti~e 15 RAM banks in total, desicJn~ted RUM BANK O through RAM BANK 7.
Each of the RAM's comprises an eighteen-bit x OK emitter coupled logic (EEL) random access memory. As will be understood by those skilled in the art, a "2X"
memory actually includes 2048 individually-addressed memory 20 locations or words designated WORD O through WORD 2047.
An ll-bit address is required for this memory size.
As depicted in FIG. 2, the first time bin, i.e., BIN 1, corresponds to WORD O of RAM BANK O. Time BIN 2 corresponds to WORD O of RAM BANK 1. This sequence 25 continues, as illustrated, with time BIN 16384 corresponding to WORD 2047 of RAM BANK 7. As will be apparent from the description hereinafter with reference to FIGS. 3 and 4, this arrangement allows data to be summed (digitally integrated) in real time at much faster rates than would 3 apparently be permitted by the total cycle time required to perform a memory read operation, an addition, and a memory write operation, typically 40 nanoseconds.
isles 1 With reference now to FIG. 3, the integrating transient recorder 24 of the invention is shown in greater detail.
As described above with reference to FIG. l, 5 the ion detector 16 has its output connected to the amplifier 22 configured as a current-to-voltage converter.
The output of the current-to-voltage converter 22 is connected to a buffer amplifier 32, the output of which is connected to the input of the analog-to-diyital converter lo 26.
A noise generator 36 is connected to the analog signal path between the current-to-voltage converter 22 and the buffer amplifier 32. The output level of the noise generator 36 is set Jo assure that -the noise level is 15 greater than the least significant bit of the subsequent eiyht-bit analog-to-digital converter 26 so that the summing (digital integration) increases the dynamic range.
The analog-to-digital converter 26 is an eight-bit flash converter having a 5 x Lowe second 20 conversion time, and operates in the subject system at a data rate of 200 x 106 samples per second. A suitable device for the analog-to-digital converter 26 is a Leroy Instrumentation Type No. TRY.
Following the buffer amplifier 32 and the 25 analog-to-digital converter 26 are three major subsystems, the summation system 28, the data reduction system 30, and the disk interface subsystem 32.
The output of the analog-to-digital converter 26 supplies an eight bit ADO DATA BUS, in turn connected 3 to the inputs of eight individual summation subsystems 41 through 48, also designated BANK O through BANK 7.
1 Individual output busses of the summation subsystems are an exemplary 18 bits each. Since the output of the analog-to-digital converter 26 is only eight bits, this ensures that 1000 summations can be accommodated 5 without danger of overflow, seven if for a particular time bin corresponding to a particular sample point in time the analog-to-digital converter 26 is always outputting a maximum data value, i.e., 11111111.
The data reduction subsystem 30, described 10 below with reference to FIG. 5, performs further data processing, depending on the particular application. The disk interface subsystem 32 then provides high speed storage of the data on a disk, where it can be processed subsequently by a host computer (not shown) to determine 15 and plot various spectra in accordance with the use to which the instrument is being put.
FIG. 4 depicts a representative one of the FIG.
3 summation subsystem 41 in greater detail. Except as noted, the various components in FIG. 4 are standard EEL
20 digital logic circuits of the look EEL logic family manufactured by Fairchild Semiconductor.
In FIG. 4, an eight bit input latch 50 is connected to the eight-bit ADO DATA BUS, and latches in data upon receipt of a STROBE command from control logic 25 circuitry (not shown). During operation of the overall FIG. 3 system, individual STROBE pulses are sequentially applied to the individual input latches in the summation subsystems in recirculating fashion such that data values from successive sample intervals are in turn latched into 30 the eight-bit latches 50 in each of the summation subsystems, with the sequence then repeating.
Within the summation subsystem 41 are -three summation sub circuits AYE, 52B and 52C organized such that 1 one, specifically 52B, is a common summation sub circuit capable of being selectively paired with each of the other summation sub circuits AYE and 52C. Each of the summation sub circuits includes an adder AYE, 54B or 54C with a pair 5 of inputs. One input of each of the adders AYE, 54B and 54C
is connected to receive data from the input latch 50. Since the adders AYE, 54B and 54C are 18-Bit adders, the 8-bit output of the input latch 50 is connected to the least-significant data bit inputs, and the unused adder input 10 bits are tied to logic "0".
Each of the summation sub circuits AYE, 52B and 52C also includes an intermediate data latch AYE, 56B or 56C connected for temporarily storing the output of -the respective adder AYE, 54B or 5~C.
Finally, each of -the summation sub circuits AYE, 52B and 52C includes arespecti~e random access memory RAM
lay lo and lo. Each of the Rams lay lo and lo is an 18-bit x 2X emitter coupled logic (EEL) device having a twenty nanosecond access time.
In operation, the summation sub circuits AYE, 52B and 52C are paired as either pair A or pair BY and the summed data are stored alternately between the two Rams of the selected pair until a complete integrated scan is obtained by repeated summing. More particularly, the 25 sequence is to read out of one RAM, add, and then write into the other RAM of the selected pair for each input point of a particular transient waveform. At the conclusion of a complete integrating cycle, the aggregate sum point by points stored in the outer RAM, e.g., RAY lo 3 or RAM lo. Then the other pair is selected, and a new integrated scan is commenced. This allows the final sum which remains in the now-unselected outer RAM to be read 5~55 l out address by address and transferred to the data reduction subsystem 30 during the next subsequent integrated scan time. Further, this arrangement wherein data is read out of one RAM, added to incoming data, and 5 stored in another RAM allows the operation to proceed at a faster rate because the READ/WRITE cycles of the two active RAMS occur at the same time.
To accomplish this operation, the other inputs of the adders AYE and 54C are connected to receive data 10 from RAM lo of the common summation sub circuit 52s. An input multiplexer 60 is arranged to correct the output of the memory RAM lo or RAM lo of the selected one of -the sub circuits AYE or 52C to the other input of adder 54B.
An output multiplexer 62 connects the output of -the memory 15 RAM lo or lo of the unselected one of the date summation subsystems to the EEL to TTL converter.
Also, depicted in FIG. 4 are address inputs to the Rams lay lo and lo, and READ/WRITE ROY) control inputs. The RAM addresses and READ/WRITE control signals 20 are generated by the control logic snot shown).
The operation of the FIG. 4 summation subsystem 41 will now be described by way of a detailed example. In this example, it is assumed that sub circuits AYE and 52B
are the selected active sub circuits, i.e., the input 25 multiplexer 60 is selecting the output of the RAM lo to apply to the input of the adder 54B. The other sub circuit 52C is enabled for data readout, the output multiplexer 62 selecting the RAM lo for this purpose.
At the very beginning of a transient signal 3 waveform, the first input data value intended for the summation subsystem 41, (for example, corresponding to time BIN 1) is latched into the input latch 50. Previously-.
~5~5~:i 1 stored cumulative sum data is read from WORD 0 of RAM lay corresponding to time BIN l. however, for the first transient waveform in each integrated scan, the initial read for each time bin from I lo is a dummy read because 5 the beginning cumulative sup must, by definition, be zero.
To accomplish this, advantageously a "master reset" command is applied to the input multiplexer 60 causing its output applied -to the adder 54B to be zero, regardless of the input data.
lo The adder 54B then adds the data value from the multiplexer 60 output (zero for the first transient) to the data value stored in the input latch 50 -to calculate a new cumulative sum for the particular -time bin, and -this new cumulative sum is temporarily stored in the intermediate 15 latch 56B.
Upon receipt of the next input sample point intended for the summation subsystem 41B, this new point is latched into the input latch 50 as before. Again, the previously-stored cumulative sum is read from the RAM lo 20 and applied to the adder 54B through the multiplexer 60.
Again, in the case of the first transient of a particular integration, the multiplexer 60 output is selected to be zero.
A significant feature of the FIG. 4 summation 25 subsystem 41 is what happens to the cumulative sum data for the last previous time bin stored in the intermediate latch 56B. In particular, while previously-stored cumulative sum data is being read from RAM lo for the current time bin, the new cumulative sum data for the preceding time bin is 3 read from the intermediate latch 56B into the RAM lo.
Thus, the read and write memory cycles overlap, significantly speeding up the overall operation.
~:2~5~i 1 Operation for each input data point is Dodd into two phases with each phase being 20 nanoseconds.
During the first phase, data is read from RAM lay while data stored in the latch 54B is simultaneously written into 5 the RAM lo. During the second phase, addition occurs in the adder 54B, and the results stored in the latch 56B.
Operation continues in this fashion until all 16384 time bins are filled. At this point, the roles of the sub circuits AYE and 52B reverse. In the reverse 10 operation, as incoming data is latched into the input latch 50, previously-stored cumulative sum data for the~curr~nt time bin is read from RAM lo and summed in the adder I
with the incoming data in the input latch 50, and the new cumulative sum is temporarily stored in the latch AYE. As 15 Lo the case above, at the same time the RAM lo is bring read, the new cumulative sum corresponding to the previous time bin which was temporarily stored in the latch 56~ is written into the RAM lay This operation continues in alternating fashion 20 until a predetermined number of transients have been integrated in individual time-resolved channels.
Significantly, this arrangement allows collection and integration of incoming data to proceed substantially continuously, with no time lost for data 25 readout-It will be appreciated the FIG. 4 summation subsystem 41 requires a control pulse generator to generate appropriately-timed control pulses, as well as to generate 3 ll-bit memory addresses. This control circuitry (not shown may be entirely conventional and implemented, for example, with continously-running digital counters with suitably-decoded outputs.
1 FIG. 5 is a generalized block diagram of the data reduction system 30. The data reduction system 30 is a significant element in reducing the volume of data, without losing significance prior to storage on the mass 5 storage disk via the disk interface 32. The data reduction system 30 includes a set 60 of input buffers, connected respectively to receive data from the individual summation subsystems as depicted in FIG. 3.
The data reduction system 30 includes a processor 10 64 which performs data compaction, as well as data analysis.
The processor 64 is implemented employing bit-slice micro-processor elements executing appropriate algorithm. A
relatively simple data reduction algorithm is to establish a threshold, and to treat all data values below this 15 threshold as zero values. In accordance with any one of a variety of known data compaction techniques, the data are compressed with appropriate coding to indicate the position of the various zero and non-zero values.
Advantageously, the processor 64 performs a 20 limited amount of data analysis, for example, determining the presence of mass peaks, and outputting summary data, i.e., the mass position and magnitude of a given peak.
This, then, eliminates any need to store individual ion current data comprising the mass peak.
The final element in the date reduction system 30 is an output buffer 66, in turn connected to the disk interface 32.
The input buffers 60 and the output buffer 66 comprise standard digital logic memory.
3 Finally, the disk interface 32 is a modified standard disk controller connected to one port of a large, high-speed, dual port disk memory. The other port of the .
~5~5~i 1 dual port disk memory is connected to the host computer (not shown). The disk interface 32 is arranged so that the summation system 28 and data reduction system 30 can take control of the high speed disk unit 32 and load data at 5 high speed. After the data aye written on a disk, control is transferred back to the host computer so that the data can be processed.
Referring finally to FIG. 6, there is depicted in block diagram form an alternative arrangement of a high 10 speed data acquisition and storage system. The FIG. 6 system employs charge coupled devices Cuds 80 and 82 as analog integration and storage devices to receive and initially integrate data from the ion detector 16.
Charge-coupled devices are comparable to long shift 15 registers, having a plurality ox memory locations or cells.
They have the capability of storing analog signals in the form of charge, and shifting the stored charge from one cell to the other in shift register fashion as the device is clocked. While charge-coupled devices are potentially 20 capable of relatively high input data rates, and can -thus lessen the speed requirements (and cost) of a subsequent digital summation system such as the summation system 28, presently-available charge-coupled devices are relatively limited in bandwidth and in the magnitude of the charge or 25 signal which can be stored in each cell before saturation occurs. In particular, for the present application it is believed that an upper limit to integration in a charged coupled device would be in the order of 10 scans.
Referring to FIG. 6 in detail, the output of 3 -the ion detector 16 is connected through a pair of respective summing amplifiers 84 and 86 to serial inputs 88 and 90 of the Cuds 80 and 82. The summing amplifiers are part of a feedback path for each COD to enable new in-coming Tut be summed with previously-stored cumulative sum 35 data.
l Respective serial outputs 92 and 94 of the Cuds 80 and 82 are connected through respective analog switches 96 and 98 for data readout to sample and hold circuits 100 and 102. The outputs of the sample and hold 5 circuits lo and 102 are illustratively alternatively selected by an analog multiplexer 104 connected to the input of the analog-to-digital converter 26. It will, however, be appreciated that other arrangements may be employed, such as arrangements which avoid the need for the lo output multiplexer 104 through suitable operation of the analog switches 96 and 98.
For feedback of previously-stored cumulative sum data, respective analog switches 106 and 108 and amplifiers 110 and 112 are connected from -the COD outputs 15 92 and 94 to the summing amplifiers 84 and 86. Finally, in order to provide a zero initial "previous" cumulative sun during the first transient of each plurality to be summed in time-resolved channels, analog switches 114 and 116 are provided to selectively ground the feedback paths.
In operation, assuming integration is to occur in COD 80, for each incoming transient from the ion detector 16 the COD 80 is clocked at high speed completely through as sampled data is serially applied to its input 88 via the summing amplifier 84. For the first transient, the -25 switch 114 is closed, and so the incoming data is summed with zero. At the end of the first transient, the individual cells of the COD 80 have stored in them the sampled values, with the first sample point of the transient being stored at the output 92 end of the COD
3 device 80, and the last sample point of the transient stored at the input end 88 of the COD device 80. For the next nine input transients, as new sample points are received, the prior summed or integrated value for the 1 particular time channel is read from the COD 80 via output 92 and analog switch 106, summed with the incoming analog value in summing amplifier 84, and loaded back into the COD
80 via input 88.
Following the time-resolved summation of an exemplary ten transients in COD 80, the summation operation immediately switches to COD 82 without pause and without loss of incoming data. Then while summation is occurring in COD 82 in the same manner, COD 80 is read out via analog lo switch 96. During readout, the clock rate applied to the COD 80 is only one-tenth the high-speed rate for data acquisition.
In another embodiment (not shown), a single COD
series can be utilized. As the incoming signal is added to 15 the prior sum for each cell and the resulting new sum is stored, a read out and clear process can be activated at a slower rate, reading and clearing, for example, every tenth cell. By this means, for every ten scans one summed scan can be converted and stored. The ration of the input rate 20 to the output rate determines the number of individual transients summed prior to digitization.
a. a strobed input latch for latching therein, upon receipt of a strobe signal, an input data value, and said means for directing input data values directs strobe signals sequen-tially to the input latches in all of the data summation sub-systems to sequentially direct input data values to all of the data summation subsystems;
b. each data summation subcircuit including:
i. an adder with a pair of inputs, one adder input being connected to receive data from said input latch, ii. an intermediate data latch connected to the adder for temporarily storing the output of said adder, and iii.said memory bank having individual memory loca-tions corresponding to individual time bins, said memory bank being connected to said intermediate data latch for receiving data to be stored;
c. the other adder input of each of the first and second summation subcircuits being connected to receive data from the memory of the common third summation subcircuit; and d. said input switching means comprising an input multi-plexer having its output coupled to the other adder input of said third common summation subcircuit, said input multiplexer being coupled to the memory outputs of said first and second summation subcircuits.
a. a strobe input latch for latching therein, upon receipt of a strobe signal, an input data value, and said means for directing input data values directs strobe signals sequent-ially to the input latches in all of the data summation subsystems to sequentially direct input data values to all of the data summation subsystems;
b. a third, common summation subcircuit having a memory bank therein which is paired with each of the first and second summation subcircuits in data summing operations, each data summation subcircuit including:
i. an adder with a pair of inputs, one adder input being connected to receive data from said input latch;
ii. an intermediate data latch connected to the adder for temporarily storing the output of said adder, and iii.said memory bank having individual memory locations corresponding to individual time bins, said memory bank being connected to said intermediate data latch for receiv-ing data to be stored;
c. the other adder input of each of the first and second summation subcircuits being connected to receive data from the memory of the common third summation subcircuit; and d. said input switching means comprising an input multi-plexer having its output coupled to the other adder input or said third common summation subcircuit, said input multiplexer being coupled to the memory outputs of said first and second common summation subcircuits.
Priority Applications (2)
|Application Number||Priority Date||Filing Date||Title|
|US06385115 US4490806A (en)||1982-06-04||1982-06-04||High repetition rate transient recorder with automatic integration|
|Publication Number||Publication Date|
|CA1225155A true CA1225155A (en)||1987-08-04|
Family Applications (1)
|Application Number||Title||Priority Date||Filing Date|
|CA 429710 Expired CA1225155A (en)||1982-06-04||1983-06-03||High repetition rate transient recorder with automatic integration|
Country Status (6)
|US (1)||US4490806A (en)|
|JP (1)||JPS59501563A (en)|
|CA (1)||CA1225155A (en)|
|DK (1)||DK49484D0 (en)|
|EP (1)||EP0110981A4 (en)|
|WO (1)||WO1983004326A1 (en)|
Families Citing this family (33)
|Publication number||Priority date||Publication date||Assignee||Title|
|US4787029A (en) *||1986-09-29||1988-11-22||Gte Communication Systems Corporation||Level converting bus extender with subsystem selection signal decoding enabling connection to microprocessor|
|US4849930A (en) *||1987-02-25||1989-07-18||Westinghouse Electric Corp.||Method of compactly storing digital data|
|DE4029177C2 (en) *||1990-09-14||1993-06-17||Bergwerksverband Gmbh, 4300 Essen, De|
|US5175430A (en) *||1991-05-17||1992-12-29||Meridian Instruments, Inc.||Time-compressed chromatography in mass spectrometry|
|US5367162A (en) *||1993-06-23||1994-11-22||Meridian Instruments, Inc.||Integrating transient recorder apparatus for time array detection in time-of-flight mass spectrometry|
|US5712480A (en) *||1995-11-16||1998-01-27||Leco Corporation||Time-of-flight data acquisition system|
|US6627877B1 (en)||1997-03-12||2003-09-30||Gbc Scientific Equipment Pty Ltd.||Time of flight analysis device|
|US6094627A (en) *||1997-05-30||2000-07-25||Perkinelmer Instruments, Inc.||High-performance digital signal averager|
|US6028543A (en) *||1997-10-03||2000-02-22||Eg&G Instruments, Inc.||Apparatus for improvement of the speed of convergence to sub-least-significant-bit accuracy and precision in a digital signal averager and method of use|
|DE69921900T2 (en)||1998-01-23||2005-03-17||Micromass Uk Ltd.||Flight mass spectrometer and detector for double fortification|
|US5995989A (en) *||1998-04-24||1999-11-30||Eg&G Instruments, Inc.||Method and apparatus for compression and filtering of data associated with spectrometry|
|US20020009394A1 (en)||1999-04-02||2002-01-24||Hubert Koster||Automated process line|
|US6526355B1 (en) *||2000-03-30||2003-02-25||Lam Research Corporation||Integrated full wavelength spectrometer for wafer processing|
|US7356580B1 (en)||2000-03-30||2008-04-08||Lam Research Corporation||Plug and play sensor integration for a process module|
|US7672747B2 (en) *||2000-03-30||2010-03-02||Lam Research Corporation||Recipe-and-component control module and methods thereof|
|US7917301B1 (en)||2000-09-19||2011-03-29||Sequenom, Inc.||Method and device for identifying a biological sample|
|US6791815B1 (en) *||2000-10-27||2004-09-14||Ion Systems||Dynamic air ionizer and method|
|DE10206173B4 (en) *||2002-02-14||2006-08-31||Bruker Daltonik Gmbh||High-resolution detection of flight mass spectrometer|
|DE60319213T2 (en) *||2002-04-23||2009-05-28||Analytica of Branford, Inc., Branford||Devices and procedures designed to reduce coherent noise in a digital signal mid|
|US7333918B2 (en) *||2002-09-05||2008-02-19||Strategic Power Systems, Inc.||System and method for calculating part life|
|GB0403535D0 (en) *||2004-02-18||2004-03-24||Hoffman Andrew||Data acquisition and processing mass spectrometer|
|US7109475B1 (en)||2005-04-28||2006-09-19||Thermo Finnigan Llc||Leading edge/trailing edge TOF detection|
|US8101368B2 (en) *||2006-02-13||2012-01-24||Dvs Sciences Inc.||Quantitation of cellular DNA and cell numbers using element labeling|
|US20070255749A1 (en) *||2006-04-27||2007-11-01||Vrba Joseph A||Selective extractions of data from large transient data sets|
|US7501621B2 (en) *||2006-07-12||2009-03-10||Leco Corporation||Data acquisition system for a spectrometer using an adaptive threshold|
|CA2658787C (en) *||2006-08-15||2013-04-09||Alexei Antonov||Apparatus and method for elemental analysis of particles by mass spectrometry|
|US7565220B2 (en) *||2006-09-28||2009-07-21||Lam Research Corporation||Targeted data collection architecture|
|US7814046B2 (en) *||2006-09-29||2010-10-12||Lam Research Corporation||Dynamic component-tracking system and methods therefor|
|GB0709799D0 (en)||2007-05-22||2007-06-27||Micromass Ltd||Mass spectrometer|
|US8553994B2 (en) *||2008-02-05||2013-10-08||Futurewei Technologies, Inc.||Compressive sampling for multimedia coding|
|US20120041695A1 (en)||2010-08-16||2012-02-16||Csi Technology, Inc.||Integrated vibration measurement and analysis system|
|US8942943B2 (en) *||2011-07-15||2015-01-27||Bruker Daltonics, Inc.||Dynamic range improvement for mass spectrometry|
|US9218949B2 (en)||2013-06-04||2015-12-22||Fluidigm Canada, Inc.||Strategic dynamic range control for time-of-flight mass spectrometry|
Family Cites Families (6)
|Publication number||Priority date||Publication date||Assignee||Title|
|NL291827A (en) *||1961-03-17|
|US3182181A (en) *||1963-03-27||1965-05-04||Nuclear Data Inc||Method and apparatus for averaging a series of electrical transients|
|FR2188873A5 (en) *||1972-06-08||1974-01-18||Commissariat Energie Atomique|
|US3840861A (en) *||1972-10-30||1974-10-08||Amdahl Corp||Data processing system having an instruction pipeline for concurrently processing a plurality of instructions|
|DE2347632B2 (en) *||1973-09-21||1976-02-19||to reduce assembly and process of the access time in a memory system|
|US4054786A (en) *||1973-09-24||1977-10-18||The United States Of America As Represented By The Secretary Of The Navy||Running average computer|
Also Published As
|Publication number||Publication date||Type|
|US3506813A (en)||Signal-to-noise ratio enhancement methods and means|
|US5901348A (en)||Apparatus for enhancing sensitivity in compressive receivers and method for the same|
|Adinolfi et al.||The trigger system of the KLOE experiment|
|Amoretti et al.||The ATHENA antihydrogen apparatus|
|US4864140A (en)||Coincidence detection system for positron emission tomography|
|Katzenstein et al.||New Time‐of‐Flight Mass Spectrometer|
|US6809674B1 (en)||Analog-to-digital converters|
|Adamczyk et al.||The BRAHMS experiment at RHIC|
|US7084393B2 (en)||Fast time-of-flight mass spectrometer with improved data acquisition system|
|Lazarus et al.||The GREAT triggerless total data readout method|
|Grzywacz||Applications of digital pulse processing in nuclear spectroscopy|
|Cardarelli et al.||Avalanche and streamer mode operation of resistive plate chambers|
|US7501621B2 (en)||Data acquisition system for a spectrometer using an adaptive threshold|
|US6680476B1 (en)||Summed time-of-flight mass spectrometry utilizing thresholding to reduce noise|
|US4262258A (en)||CCD Correlated quadruple sampling processor|
|US4495586A (en)||Waveform acquisition apparatus and method|
|Mankel||A Concurrent track evolution algorithm for pattern recognition in the HERA-B main tracking system|
|US6822227B1 (en)||Time-of-flight mass spectrometry utilizing finite impulse response filters to improve resolution and reduce noise|
|US5712480A (en)||Time-of-flight data acquisition system|
|US4472631A (en)||Combination of time resolution and mass dispersive techniques in mass spectrometry|
|Neugebauer||Initial deceleration of solar wind positive ions in the earth's bow shock|
|US3972031A (en)||Variable length shift register alternately operable to store and recirculate data and addressing circuit therefor|
|US5367162A (en)||Integrating transient recorder apparatus for time array detection in time-of-flight mass spectrometry|
|US3297860A (en)||Spectroscopic computer|
|US6094627A (en)||High-performance digital signal averager|