US3686683A - Mass spectrometer electrode gap control - Google Patents

Abstract

In a spark discharge source for a mass spectrometer in which at least one of two electrodes is vibrated to vary the electrode gap across which the discharge takes place, means are provided for adjusting the mean spacing between the electrodes to provide optimum current flow between the electrodes.

Description

United States Patent Powers et a1.

[54] MASS SPECTROMETER ELECTRODE GAP CONTROL [721 Inventors: Patrick Powers; Reginald Graham, both of Cheshire; Richard Albert Bingham, Manchester, all of England [73] Assignee: Associated Electrical Industries Limited, London, England [22] Filed: April 8, 1969 21 Appl. No.: 18,361

[30] Foreign Application Priority Data April 8, 1968 Great Britain..'. ..16,742/68 [52] US. Cl. ..315/357, 250/41.9 SA, 250/4l.9 SR

[51] Int. Cl. ..B0ld 59/44 i [58] Field of Search.250/4l'.9 SA, 41.9 SR; 315/326, 315/357; 356/86 [56] ReferencesCited UNITED STATES PATENTS 2,690,521 9/1954 Turner ..2s0/41.9x

[151 3,686,683 1 51 Aug. 22, 1972 King et a1 ..250/41.9 X Berry ..250/41.9 SA 11/1955 Slepian ..250/4l.9 SA

8/1967 Friedman et al....250/41.9 SA

OTHER PUBLICATIONS Deines et al., Applied Spectroscopy, Vol. 21, No. 1, Jan-Feb. 1967, pages 28- 30.

Primary Examiner-Ronald L. Wibert Assistant Examiner-F. L. Evans Attorney-Watts, Hoffman, Fisher & l-leinke This application filed under rule 47a.

1 1 ABSTRACT in a spark discharge source for a mass spectrometer in which at least one of twoelectrodes is vibrated to varythe electrode gap across which the discharge takes place, means are provided for adjusting the mean spacing between the electrodes to provide optimum current flow between the electrodes.

18 Claims, 9 Drawing Figures- Patented Aug. 22, 1972 3,686,683

.2 Sheets-Sheet l PULSE LE/VGTH FIG. #1 W H6. /5 u A F/G. /c

F/ /E mvsmoxs PA TRICK POWERS BY REGINALD GRAHAM RICHARD A. B/NGHAM Wadi, War/124b,

ATTORNEYS Patented Aug. 22, 1972 3,686,683

.2 Sheets-Sheet 2 vGAP INVENTORS PA TR/CK POWERS BY REGl/VALD GRAHAM RICHARD A. B/NGHAM MAY/41 Y/Iff fl H $724 //c&n,&

ATTURNEYS MASS SPECTROMETER ELECTRODE GAP CONTROL BACKGROUND OF THE INVENTION 1. Field of the Invention This invention relates to ion sources in or for mass spectrometers, and more particularly to such ion sources in which ions are produced by means of an electrical discharge.

2. Description of the Prior Art A spark source is an ion source comprising an ion chamber in which ions of a material to be analyzed are produced by means of an electrical discharge established flip-flop an electrode comprising a sample of said material and at least one other electrode which may be also a sample electrode or merely a counterelectrode. In operating a spark source mass spectrometer, after the electrodes have been initially set and a discharge struck between them, the electrode position must be continually adjusted to maintain the discharge in such manner as to maintain adequate ion beam. Hitherto this adjustment has been performed manually by an operator.

It has been proposed to vibrate at least one of the electrodes relatively to the other electrode in such a manner as to change continually the area of the sample from which the discharge is drawn. The varying electrode gap across which the discharge takes place, with consequent spreading of the discharge over a greater area of the sample, results in the extracted ions being more representative of an inhomogeneous sample material. The varying gap also results in averaging of the mean energy with which each ion species is formed, and this insures that ions of energy level, within the spread which the spectrometer can accept, are produced from any impurity element which, when sparked with fixed electrodes, would have an excessively high mean energy of release. Furthermore, it enables the electrode position adjustment to be left undisturbed for longer periods of time without ill effect. However, in consequence of electrode erosion. which takes place as an analysis proceeds, the electrode gap characteristics eventually change and operator intervention is required to return the conditions to those prevailing at the start of the analysis. Analytical accuracy is therefore highly dependent on operator skill and the procedure is susceptible to the introduction of error.

Accordingly, it is a general object of the present invention to provide a control by means of which the mean spacing between the electrodes is automatically adjusted to an optimum value.

SUMMARY OF THE INVENTION According to one aspect of the invention, a mass spectrometer spark source having means for vibrating at least one of its electrodes includes means for adjusting the mean position of one of its electrodes, sensing means for producing an electrical signal representative of current flow in the discharge between the electrodes, and indicator and/or control means for determining from this signal an optimum adjustment of that mean position. The indicator and/or control means may conveniently provide an indication enabling an operator to remotely control said mean position and preferably provides an automatic control of the mean position.

Where, as will usually be the case, the current is ac or intermittent dc the sensing can be accomplished electromagnetically by means of a sensing coil so placed as to have induced in it a signal representative of the discharge current. The signal can be displaced on an oscilloscope and will show definite changes of pattern as the electrode gap is altered. If the signal is rectified, smoothed and provided to a voltmeter, the meter will indicate a peak when the gap width is optimum. The signal is preferably supplied to control circuitry for automatically adjusting the gap width to its optimum or maximum-current width.

BRIEF DESCRIPTION OF THE DRAWING FIGS. lA-lE illustrate diagrammatically oscilloscope traces obtained under various conditions of electrode spacing,

FIG. 2 shows an idealized waveform representing rectified current flow from the sensing coil as the gap between the electrodes is, varied;

FIG. 3 is a diagram similar to FIG. 2 illustrating how optimum electrode gap width may be determined; and

FIG. 4 is a circuit diagram, with portions shown in block form, of an arrangement for automatically adjusting the electrode gap to optimum width.

DESCRIPTION OF A PREFERRED EMBODIMENT As used hereinafter, the term spark source embraces the following types of ion source:

RF (rf) SPARK SOURCE: In this type of source ions are produced by striking a discharge between the electrodes. The potential used for this is a radio frequency potential of about 500 kilocycles per second (kc) and is variable in value between zero and kilovolts (kv) peak to peak. For a greater degree of control, it is usually pulsed. Typical conditions are with a pulse length of 50-200 microseconds and repetition rates of from to 10,000 cycles per second. Both insulators and conductors can be sparked. DIRECT CURRENT,

(dc): This is a simple source working on the electric bell principle. Two electrodes are vibrated in a manner such that when they make contact current flows through them to create a magnetic field which then separates them. Clearly, only good and mechanically sound conductors (not, for example, compacted graphite) can be analyzed.

TRIGGERED DC ARC SOURCE: A high voltage (approximately 50 kv) pulse from -a pulse transformer breaks down the electrode gap and allows a low dc voltage to use the then existing plasma between the electrodes as a conductor. The relatively heavy arc current then formed produces a heavy ion current. Repetition rates can be varied within certain limits but analysis of insulators is inefficient, because only the trigger pulse is usable for this purpose.

FIG. 1A shows an oscilloscope trace (idealized) of a signal induced in a sensing device so placed as to have induced in it a signal representative of the discharge current between the electrodes of an rf vibrator spark source. In this particular example, the electrodes are so far apart that there is no discharge between them. The rf signal applied to the electrodes is represented by a sine wave F.

FIG. 1B shows a similar induced signal when the electrode gap has been reduced to a width where intermittent spark discharges occur between the electrodes. Peaks S protruding from the rf signal envelope represent the spark discharges.

FIG. 1C illustrates the signal obtained when the electrode gap is such that the electrodes just physically strike one another during vibration. A center line L is just beginning to appear. This is the optimum condition for maximum discharge current flow and maximum ionization of a sample material.

FIG. 1D shows an induced signal obtained when the gap is such that the electrodes are contacting each other heavily and hence remaining in contact for longer times than in the example shown in FIG. 1C. As shown in FIG. 1D, the center line L has become much more pronounced than in the FIG. 1C example.

FIG. 1E illustrates the situation existing when the electrodes are permanently in contact. They are thus shorted out and there is no discharge between the electrodes.

It is easy to recognize the pattern (such as FIG. 1C) which corresponds to an electrode position adjustment providing the best analytical reproducibility in any particular case. It is therefore easier than it was without such optimum position detection to control theelectrode gap characteristics manually.

It has been found that, as illustrated by FIG. 2, if the sensing coil signal is rectified (by a half-wave rectification circuit, for example) and smoothed, the resulting smoothed voltage output waveform V varies as the mean electrode gap is varied and peaks at or near the optimum size of the gap. However, the optimum size being not very critical, satisfactory operating characteristics can be achieved with any gap size within an optimum range G. The shape of the waveform V results from the face that the vibration produces intermittent sparking at wide gaps and intermittent electrode contact at small gaps. Both of these conditions result in low means outputs from the sensing coil, but between these extremes the effects combine to produce a maximum in the sensing coil output just at the point where the optimum gap is attained.

The voltage V can be displayed visually on a voltmeter, for example, in order to provide guidance to an operator in manual control of electrode position. For automatic control, an electronic circuit arrangement can be provided for detecting variations of the voltage output from the maximum value and producing a bias signal for application to suitable means for adjusting the mean electrode position in such manner as to maintain the gap within the optimum range. Since the peak is not sharp and the actual value of the voltage at the peak can vary from one spectrometric analysis to another, the electronic circuit arrangement is preferably such as to cause the bias signal to vary cyclically. Thus, the mean electrode position sweeps repeatedly across the peak voltage position between two points (within the optimum gap range) at which a substantial reduction in the voltage from its peak can be detected and utilized to reverse the electrode movement by reversing the sense of the bias signal variation. The sweep times may be of the order of ten seconds in each direction, i.e., 20 seconds for a complete cycle.

One way in which the end points of the sweep can be detected is illustrated by FIG. 3. The electronic circuit arrangement is arranged to produce a second voltage output V', which is a predetermined fraction (preferably adjustable) of the rectified and smoothed voltage output V, as the latter rises towards its peak P and which, as the latter falls during further movement of the electrode, remains substantially at its maximum value P until both voltages become equal. The reduction of the voltage difference to zero is sensed and utilized to reverse the electrode movement. At the same time the second voltage V is allowed to fall to the predetermined fraction of the full voltage V at that moment, so that the process is repeated during the next sweep of electrode movement in the opposite direction, and soon continuously. In FIG. 3, full lines are used to show the initial build-up of the second voltage V as the electrode gap increases until it enters the sweep range H, the buildup thereof during sweeps in which the gap is increasing to a first end point El and the fall thereof at that point as the electrode movement is reversed. Broken lines show the similar variation of the second voltage V for sweeps in which the gap is decreasing to the second end point E2. The ratio of the voltages during build-up is so chosen as to ensure that the sweep range H falls within the optimum gap range G.

Initial electrode movement, when the apparatus is started up, may be controlled manually to ensure that it takes place in the right direction to bring the gap into the sweep range l-I. This is to avoid electrode contact and consequent welding, and also to avoid a situation in which the gap continually increases from an initial setting larger than the gap corresponding to the end point El. Means may, however, be provided for effecting this initial control automatically by sensing the value of the bias signal which determines the mean electrode position.

A suitable electronic circuit arrangement is illustrated by FIG. 4, which is a diagrammatic representation of an rf spark source in accordance with the invention, in combination with such a circuit arrangement.

Referring to FIG. a, the spark source represented thereby comprises an ion chamber 10 within a source housing represented by the dashed circle 12. It has means, including a vibrator coil 14 for vibrating an electrode 16 of a pair of electrodes 16, 18, the other electrode 18 being held stationary. At least one of the electrodes would comprise a sample of material to be analyzed by means of a mass spectrometer whose ion source is the spark source being described. The electrode vibrating means, which may take the form described in US. patent application Ser. No. 638,857, filed May 16, 1967, is arranged to vibrate the electrode 16 in such manner as to change continually the area of the sample from which an ion-producing electrical discharge between the electrodes will be drawn.

One of the electrodes (the stationary electrode 18, as shown) is connected to a terminal 20 by means of a lead 22 for supplying the spectrometer accelerating potential which may, typically, be 20 kv dc. A suitable dc source (not shown) is connected to the terminal 2t]. The other electrode 16 is connected to the terminal 20 through a secondary winding 24 of an rf Tesla coil for applying between the electrodes a suitable rf voltage which may, typically, be 20 kv at a frequency of 500 kc for creating the ion-producing electrical discharge.

The vibrator coil 14 is connected across two power supplies. A first of these, which may suitably have a frequency of 60 cycles per second, is provided from a secondary winding of a transformer 26 to provide a suitable output voltage across a potentiometer 28 for applying an adjustable fraction of the transformer output to the vibrator coil 14 through a capacitor 30. The capacitor is provided to prevent direct current from the second vibrator coil supply from flowing into the transformer circuit.

The vibrator is so constructed that, when a direct current (of either polarity) is passed through its coil 14, it will impart to the vibratable electrode 16 a definite deflection from a reference position, the magnitude and direction of the deflection being dependent on the magnitude and polarity respectively of the direct current. The purpose of the second power supply is to provide such direct current to the vibrator coil in a controlled manner, which will be described. The electrode vibration caused by the ac supply to the vibrator coil will then take place about a mean electrode position determined by the magnitude and polarity of the direct current from the dc supply.

Around the connection 22 from the stationary electrode 18 to the junction between the terminal 20 and the Tesla coil secondary 24 is placed a ferrite ring 32 constituting the magnetic core of a sensing coil 34. When an rf discharge takes place between the source electrodes 16, 18, a voltage is induced in the coil 34. This voltage is applied to a half-wave rectification and smoothing circuit comprising a diode 36, a capacitor 38 and a potentiometer 40 of suitable values. The diode 36 and the potentiometer 40 are connected in series and the capacitor 38 is connected across the potentiometer 40 so that the capacitor 38 is charged positively. A resistor 73 is connected across the sensing coil to prevent excessive voltage arising when the diode 36 is reversed biassed. The smoothed positive voltage thus produced may be supplied to an indicator 41, such as voltmeter or oscilloscope, and is applied through a lead 42 to one input of a difference amplifier 44. This voltage corresponds to the signal V shown in FIGS. 2 and 3. An adjustable fraction of the smoothed voltage is taken from an adjustable arm of the potentiometer 40 to one side (anode) of a diode 46. This fractional voltage corresponds to the signal V of FIG. 3. The other side (cathode) of the diode 46 is connected to a second input of the difference amplifier 44 and also to one side of a capacitor 48, whose other side is connected to a return lead 50 to the sensing coil 34. The polarity of the diode 46 is such that, when the smoothed voltage falls after rising to a peak and having caused the second capacitor 48 to be charged to a definite fraction of that peak voltage determined by the setting of the potentiometer 40, the voltage on the capacitor 48 is maintained substantially constant until a relay contact 52a, connected across the diode 46, closes to permit the capacitor to discharge to the definite fraction of the full voltage existing at that moment. The relay contact 52a is operated in this way by a relay coil 52, which is energized by a positive output from the difference amplifier 44, which occurs when the difference between the voltages at the two amplifier inputs becomes slightly negative after passing through zero; that is, when the voltage across the capacitor 48 exceeds that across the capacitor 38. The potentiometer 40 is adjusted to bring the points E1 and E2 (FIG. 3) at which this occurs within the optimum gap range G. The amplifier 44 is so connected that it produces a negative output when its input from the capacitor 38 is more positive than its input from the capacitor 48. However, this negative output is prevented from energizing the relay 52 by an appropriately connected diode 54.

Such detection of the end points El and E2 of the sweep H (FIG. 3) is utilized to cause a reversal of electrode travel by means of a bistable flipflop 56. The input of the fiipflop 56 is connected to the output of the difference amplifier 44, and the flipflop 56 is arranged to flip over from one stable state to its other stable state whenever the amplifier output becomes positive as a result of the difference between its input voltages having become negative after passing through zero. An actuating coil 58 .of a relay having a two-position or change-over contact 58a is connected to the bistable circuit output. It causes the contact 58a to change-over from one position to its other position whenever the bistable circuit changes its state.

The change-over contact 58a is arranged, on changing over, to reverse the polarity of a i 15 volt (or other convenient voltage) dc supply connected through the contact to a first input of a dc amplifier 60 via a combination of a series resistance 62 and a shunt capacitance 64. The amplifier output is connected to the base of a transistor 66, whose emitter is connected to the base of a second transistor 68 of similar conductivity type (NPN, as shown). The transistors 66, 68 constitute, respectively, an emitter follower stage and a power amplifier stage, the former being used to provide sufficient current gain to drive the latter. The collectors of the two transistors are connected, respectively. to the ends of the vibrator coil 14. A dc power supply (not shown) for the two transistors is connected to the collector of the first transistor 66. The emitter of the second transistor 68 is connected through a resistor 70 to ground and to a second input to the dc amplifier 60, thus providing negative feedback to define the overall gain of the amplifier system. The emitter of the transistor 68 may also be connected by a lead 72 to a circuit (not shown) for automatically determining and controlling the initial direction of movement of the vibratable electrode 16, when the apparatus is started up, by sensing the voltage across this resistor 70.

The combination of the dc amplifier 60 and the transistors 66, 68 is arranged to be such that the value of the collector current of the power amplifier transistor 68 (which is the direct current in the vibrator coil 14) is dependent on the value of the voltage at the first input of the amplifier 60. The resistor-capacitor combination between this input and the relay contact 58a is designed to have a time constant such that, subsequent to a change-over of the dc supply polarity through the contact 58a, the direct current level in the vibrator coil will change at a suitable rate to provide a sweep of the vibratable electrode 16 over a suitable period, for example, 10 seconds.

Thus, the effect of the circuit arrangement as a whole is to cause the mean position of the vibratable electrode 16 to sweep to and fro as described with reference to FIG. 3, so that the electrode gap is automatically maintained with an optimum range for the conditions existing at the time, electrode erosion being thereby counteracted.

It is envisioned that the aforementioned circuit for determining the direction of electrode movement when the apparatus is started up will be designed to produce, when the potential across the resistor 70 is below a certain threshold level corresponding to a very wide gap, a signal which will be used to set the bistable flipflop S6 to the state which causes the relay 52 to connect one input of the amplifier 60 through the resistor 62 to the positive supply. This will cause the potential across the capacitor 64 to increase and, consequently, the electrode gap to decrease, thus ensuring that when the apparatus is initially switched on the flipflop 56 will not remain in the state which would result in failure of the ion source system to start up.

It would be possible to employ, instead of the vibrator coil described, other means for vibrating or oscillating at least one of the electrodes for example a cam or other mechanical means. Furthermore, the vibrating of at least one of the electrodes could be combined with rotational motion of that or another of the electrodes. Successful operation has been achieved by rotating a sample electrode of disc form and vibrating against its face a suitable counter-electrode. This arrangement is advantageous where the sample electrode has to be made from material in powder form, since it is easier to form discs than it is to form rods by pressing them from powder.

It is also feasible to adjust the mean electrode gap by adjusting the position of the electrode 18, which has heretofore been described as fixed. This can be done by mechanically connecting that electrode to be positioned by a coil similar to the coil 14, which is energized only dc from the power supply comprising the amplifier 60 and the transistors 66, 68. In that case, the vibrating electrode would be driven only from the transformer 26.

The position of the sensing coil 34 is not limited to that heretofore mentioned. There may be other positions in which a suitable signal strength and fonn are achieved. With an rf signal, it is satisfactory to place the sensing coil in a more remote position than with an intermittent dc signal in the case of another type of spark source, since the current to be sensed by the sensing coil is not the rf current but the current due to the vibration frequency. In the case of the r'f spark source described, the core of the sensing coil could equally well be placed around the lead from the other electrode 16 to the Tesla coil secondary 24, but the position shown happens to be the most convenient in the particular apparatus being considered, having regard to a particular type of size of sensing coil.

It is envisioned that an improvement could also be achieved by employing a monostable circuit at the output of the difference amplifier 44 to drive the relay coil 52 and the flipflop 56.

We claim:

1. A mass spectrometer electrode gap control for a spark source having two electrodes, means for electrically energizing said electrodes to provide a spark discharge across a gap therebetween, means for moving at least one of said electrodes cyclically toward and away from the other electrode, and positioning means for relatively positioning said electrodes with a mean spacing therebetween, comprising:

a. sensing means for producing an electrical signal representative of current flow in said discharge between said electrodes; and

b. utilization means coupled to said sensing means and responsive to said electrical signal for controlling the mean spacing between said electrodes within a range wherein optimum discharge current flow occurs.

2. The control of claim 1, wherein said utilization means includes a voltmeter.

3. The control of claim I, wherein said utilization means includes an oscilloscope.

4. The control of claim 1, wherein said utilization means comprises a control circuit including energizing means for energizing said positioning means to vary automatically the mean spacing between said two electrodes between first and second points within said range.

5. The control of claim 4, wherein said energizing means includes means for causing said positioning means to vary said mean spacing cyclically between said first and second points.

6. The control of claim 4, wherein said control circuit includes detecting means for detecting when said mean spacing reaches each of said first and second points.

7. The control of claim 5, wherein said control circuit includes detecting means for detecting when said mean spacing reaches each of said first and second points.

8. The control of claim 6, wherein said detecting means detects when said discharge current falls to predetermined minimum levels after having reached a maximum level, which minimum levels define said first and second points.

9. The control of claim 7, wherein said detecting means detects when said discharge current falls to predetermined minimum levels after having reached a maximum level, which minimum levels define said first and second points.

10. The control of claim 6, wherein said detecting means includes comparison means for comparing said electrical signal with a second signal representing a predetermined fractional portion of a maximum level attained by said electrical signal, and providing a control signal when said electrical signal and said second signal become substantially equal as said electrical signal level is decreasing.

11. The control of claim 7, wherein said detecting means includes comparison means for comparing said electrical signal with a second signal representing a predetermined fractional portion of a maximum level attained by said electrical signal, and providing a control signal when said electrical signal and said second signal become substantially equal as said electrical signal level is decreasing.

12. The control of claim 10, wherein said second signal is provided by a peak detector.

13. The control of claim 11, wherein said second signal is provided by a peak detector.

14. In a mass spectrometer an electrical discharge ion source having a pair of spaced-apart electrodes, circuit means for applying an energizing signal to said pair of electrodes to develop an electrical discharge in the space between said pair of electrodes, drive means for cyclically moving at least one of said pair of electrodes toward and away from the other one of said pair of electrodes so as to define a variable gap between said pair of electrodes, the improvement in an electrode gap control comprising:

a. sensing means for developing an electrical signal having a value representative of the value of current flow in said electrical discharge; and,

b. circuit means coupled to said sensing means for varying a mean gap distance of said variable gapin response to the value of said electrical signal.

16. An apparatus as defined in claim 15 wherein said circuit means includes detecting means for developing a control signal when said electrode reaches either of said end points of travel.

17. An apparatus as defined in claim 16 wherein said circuit means includes means for, upon receipt of a control signal, reversing the direction of travel of said one electrode.

18. An apparatus as defined in claim 17 wherein said reversing means includes bistable switchingmeans.

UNITED STATES PATENT GFFICE fiERTWECATE 9F QQEC'HN Patent No. 6 .683 Dated August 22, 1972 It is certified that error appears in the above-identified patent and that said Letters Patent are hereby corrected as shown below:

Column 1, line 14, "flip-flop" should be -between--.

Column 3, line 39, "face" should be --fact---.

Signed and sealed this 23rd day of January 1973.v

(SEAL) Attest:

EDWARD M.FLETCHER,JR. ROBERT SOTTSCHALK Attesting Officer Commissioner of Patents ORM 1 0-1050 (10-69) USCOMM-DC 6O376-P69 U45. GOVERNMENT PRINTING OFFICE I969 O-366334

Claims (18)

1. A mass spectrometer electrode gap control for a spark source having two electrodes, means for electrically energizing said electrodes to provide a spark discharge across a gap therebetween, means for moving at least one of said electrodes cyclically toward and away from the other electrode, and positioning means for relatively positioning said electrodes with a mean spacing therebetween, comprising: a. sensing means for producing an electrical signal representative of current flow in said discharge between said electrodes; and b. utilization means coupled to said sensing means and responsive to said electrical signal for controlling the mean spacing between said electrodes within a range wherein optimum discharge current flow occurs.

2. The control of claim 1, wherein said utilization means includes a voltmeter.

3. The control of claim 1, wherein said utilization means includes an oscilloscope.

4. The control of claim 1, wherein said utilization means comprises a control circuit including energizing means for energizing said positioning means to vary automatically the mean spacing between said two electrodes between first and second points within said range.

5. The control of claim 4, wherein said energizing means includes means for causing said positioning means to vary said mean spacing cyclically between said first and second points.

6. The control of claim 4, wherein said control circuit includes detecting means for detecting when said mean spacing reaches each of said first and second points.

7. The control of claim 5, wherein said control circuit includes detecting means for detecting when said mean spacing reaches each of said first and second points.

8. The control of claim 6, wherein said detecting means detects when said discharge current falls to predetermined minimum levels after having reached a maximum level, which minimum levels define said first and second points.

9. The control of claim 7, wherein said detecting means detects when said discharge current falls to predetermined minimum levels after having reached a maximum level, which minimum levels define said first and second points.

10. The control of claim 6, wherein said detecting means includes comparison means for comparing said electrical signal with a second signal representing a predetermined fractional portion of a maximum level attained by said electrical signal, and providing a control signal when said electrical signal and said second signal become substantially equal as said electrical signal level is decreasing.

11. The control of claim 7, wherein said detecting means includes comparison means for comparing said electrical signal with a second signal representing a predetermined fractional portion of a maximum level attained by said electrical signal, and providing a control signal when said electrical signal and said second signal become substantially equal as said electrical signal level is decreasing.

12. The control of claim 10, wherein said second signal is provided by a peak detector.

13. The control of claim 11, wherein said second signal is provided by a peak detector.

14. In a mass spectrometer an electrical discharge ion source having a pair of spaced-apart electrodes, circuit means for applying an energizing signal to said pair of electrodes to develop an electrical discharge in the space between said pair of electrodes, drive means for cyclically moving at least one of said pair of electrodes toward and away from the other one of said pair of electrodes so as to define a variable gap between said pair of electrodes, the improvement in an electrode gap control comprising: a. sensing means for developing an electrical signAl having a value representative of the value of current flow in said electrical discharge; and, b. circuit means coupled to said sensing means for varying a mean gap distance of said variable gap between said pair of electrodes in response to the value of said electrical signal in order to optimize the current flow in said electrical discharge.

15. An apparatus as defined in claim 14 wherein said circuit means includes positioning means for varying the end points of travel of said one electrode, said positioning means coupled to said sensing means to receive said electrical signal so that said end points are varied in response to the value of said electrical signal.

16. An apparatus as defined in claim 15 wherein said circuit means includes detecting means for developing a control signal when said electrode reaches either of said end points of travel.

17. An apparatus as defined in claim 16 wherein said circuit means includes means for, upon receipt of a control signal, reversing the direction of travel of said one electrode.

18. An apparatus as defined in claim 17 wherein said reversing means includes bistable switching means.

Images