US2848612A

US2848612A - Complex noise generator

Info

Publication number: US2848612A
Application number: US459969A
Authority: US
Inventors: Ralph E Allison
Original assignee: Patent Management Inc
Current assignee: Patent Management Inc
Priority date: 1954-10-04
Filing date: 1954-10-04
Publication date: 1958-08-19
Anticipated expiration: 1975-08-19

Description

Filed Oct. 4. 1954 04/7907 VOL 7465 @znsezucs Q LDA 5. fiLL/SOU INVENTOR- By W COMPLEX NOISE GENERATOR Application October 4, 1954, Serial No. 459,969

1 Claim. (Cl. 25036) This invention relates to a complex noise generator or to a generator of masking tone used in audiometric testmg.

Explanatory of the present invention, when testing the hearing loss of a patient and finding that his ears have a different degree of impairment, it is possible that the tone being applied to the poorer ear is actually being heard by the better ear either through the bones of the head or by air conduction around the head without the patient being aware of this condition. In order to obtain a true evaluation of the hearing loss in the poorer ear, it is necessary to prevent the better ear from receiving the test tone. To do this, a masking signal is applied to the better ear.

Generally speaking, the ear is rather sharp at discriminating between various sounds and hearing the ones that it wishes to hear. As a result, the only satisfactory masking tone is one that contains a frequency or frequencies close to the frequency of the test tone.

The test frequencies cover a range of from 125 cycles to 8000 cycles per second. The conventional test frequencies employed are 125 cycles, 250 cycles, 500 cycles, 1000 cycles, 1500 cycles, 2000 cycles, 3000 cycles, 4000 cycles, 6000 cycles, and 8000 cycles. Frequently, however, the use of 6000 cycles and 8000 cycles is dispensed with. For a satisfactory masking tone generator for audiometric testing the noise generated thereby should contain components very near the test frequencies over this relatively wide range that is used for testing.

There has been a great need for a low cost masking tone generator. In laboratory and research installations resistance noise amplified by a high gain amplifier has been used and found satisfactory to produce desired masking effects. This, however, is a relatively expensive piece of equipment and is too expensive for use in the ordinary portable type of audiometer. For this reason a number of make-shift devices have been employed. One such make-shift has been to use a 60 cycle power line signal which is conducted through a network to remove most of the 60 cycle signal and leave only the noise components as a masking tone. This had the disadvantage that the noise that is available is dependent upon the condition of the power line. This condition of the power line may vary from one installation or setup to another. Another objection is that practically the maximum frequency that is normally present is the harmonics of the tooth frequency of the generator which usually will be up in the order of 1000 cycles.

In slightly better installations the 60 cycle tone has been applied to a neon bulb in a sawtooth generator circuit, and this signal has been used as a masking tone. This arrangement has the advantage of having more of the higher harmonics present than the power line source. However, it still does not have enough of the frequencies above 1000 cycles to be of any substantial value as a masking tone above 1000 cycles.

Since neon bulbs are relatively inexpensive, it is an object of the present invention to provide a circuit that tes Patent "ice would use a plurality of neon bulbs and generate a large number of frequencies or harmonics so that there would be available for masking tones frequencies over the entire range or most of the range of test frequencies.

Another object of the present invention is to provide an inexpensive complex noise generator or masking tone generator that can be advantageously used in portable audiometers which is relatively simple and inexpensive and which will generate frequencies over a substantial portion of the range of test frequencies conventionally employed.

With the foregoing and other objects in view, which will be made manifest in the following detailed description and specifically pointed out in the appended claim, reference is had to the accompanying drawings for an illustrative embodiment of the invention, wherein:

Figure 1 is a wiring diagram of the complex noise generator or masking tone generator embodying the present invention; and

Fig. 2 is a graph showing the wave form of the output of the generator.

Referring to the accompanying drawings wherein similar reference characters designate similar parts throughout, I employ a plurality of neon tubes, three being preferable, although the number may vary. Such neon tubes are designated, respectively, by the reference characters A, B and C.

One side of the tube A is connected through a low resistance 10 directed to one side of an alternating current source of supply so that this tube is supplied with current in phase with the current of the supply lines. Tube B is supplied with current through a network composed of the resistance 11 and the condenser 12 so that this tube is supplied with current from a supply line but out of phase with the current supplied to tube A. In a similar manner tube C is supplied with current through a network including condenser 13 and resistance 14 so that this tube is supplied with current out of phase with the current supplied to either of tubes A or B. In effect, therefore, the three tubes are supplied with three phase current in that the phase supplied to each of the tubes is out of phase with that of the others.

The output sides of the neon tubes A, B and C are connected in parallel through the primary winding of a transformer 15 which is connected to a selenium rectifier 16. A resistance 17 is connected across the primary of the transformer.

As 35 ma. rectifiers are readily available on the market at low expense and these rectifiers are very inefiicient when operating at low current, a relatively heavy resistance 18 is added across the power line to put the rectifier in a good rectifying position or, in other words, preload or artificially load the rectifier. The secondary of the transformer 15 is usually connected through an attenuator to an earphone (neither of which is shown).

The purpose of the rectifier 16 is to reduce the 60 cycle component in the output from the three neon tubes A, B and C.

On Fig. 2 I have illustrated the wave form out of the circuit above described. The lower graph is that of one complete cycle of reference voltage or voltage supplied through the supply line. As will be noted, as the voltage builds up during the first half of the cycle the output from'the transformer is of a relatively negligible amount. This occurs while the voltage is building up to the voltage required to fire the neon tube A. When this occurs and the tube fires current can readily pass therethrough with the result that the output voltage from the transformer is represented by the peak A on this figure. As the current supplied to the tube B is out of phase with that supplied to tube A, the peak B occurs a little later in the cycle of reference voltage. Similarly, the peak C occurs 3 at a later time during the first half cycle of reference voltage. During the start of the second half of the cycle, I find that the tube A fired the second time but tubes B and C ordinarily do not when the condensers and resistances have the values assigned to them on the drawing.

If the output of the masking tone generator above described is analyzed on a wave analyzer, it will be found that the comparative signal level from this masking tone at the various peaks as compared with the pure tone output of an audiometer measured electrically at the various frequencies is very favorable. In other words, there are peaks near practically all of the test frequencies of 250 cycles through 4000 cycles. As these are the frequencies that are the most commonly used as test frequencies in audiometric work in making subjective tests, it is found that the dial of the attenuator for the masking tone can be calibrated in decibels of masking effect and will be accurate at all frequencies between 250 cycles and 4000 cycles within plus or minus three decibels. This degree of accuracy is considered very satisfactory for masking tone equipment for audiometric hearing test use.

It will be appreciated from the above described construction that a relatively inexpensive masking tone generator has been provided which has the advantages of being operated with a conventional alternating current supply and which will generate masking tones ofiadequate level close to most, if not all, of the test frequencies conventionally employed. The generator in its action is independent of line characteristics and generator noise.

Various changes may be made in the details of construction without departing from the spirit and scope of the invention as defined by the appended claim.

I claim:

A complex noise generator comprising a circuit including a single phase power source, two or more phase shifting networks connected to said power source, gas tubes conductive in both directions with one each connected to the output of each of said phase shifting networks, a half wave rectifier in one side of said circuit, and the outputs of said gas tubes being in parallel and connected in series with the output of said circuit and with said half wave rectifier.

References Cited in the file of this patent Mahoney et al. Aug. 19, 1947