From his conviction of the crime of speeding (General Statutes § 14-219) in a trial to the jury, the defendant has appealed, assigning error in the denial of his motion to set aside the verdict, in the refusal of his requests to charge, in the charge to the jury and in the admission and exclusion of certain evidence. The facts are not in dispute. The only witnesses called by the state were *70the two state troopers involved in the operation of the radar unit — Trooper Johnson, operator of the radar machine itself, and Trooper G-rinold, operator of the “chase car.” The defendant did not testify and the only witness on his behalf was Sergeant Hurlburt of the state police, who testified as an expert witness on radar.
On August 7, 1961, at about 10 a.m., the radar unit was set up on the southerly side of route 20 in Somers to clock eastbound traffic. It was tested for accuracy — before being put into operation and again after the operation was completed at about 11:30 a.m. — by use of 40-mile-an-hour and 60-mile-an-hour tuning forks, and in each instance the visual speedometer and graphic recorder indicated respective readings of 40 and 60 miles per hour. No other test for accuracy was made. The tuning forks had themselves been checked for accuracy on June 15, 1961. The chase car was parked 800 to 1000 feet east of the radar unit. The radar unit and the chase car were not visible to each other. There were no intersections between the location of the radar unit and the chase car. There was a state traffic commission sign, indicating limit of 35 miles an hour, two-tenths of a mile west of the radar unit. At about 11 a.m. on August 7,1961, a green 1961 Comet, proceeding easterly, passed through the radar field and caused both the visual speedometer and the graphic recorder to register 51 miles an hour. Trooper Johnson observed that the car bore a Connecticut registration plate, but owing to its speed he could not get the number. He immediately radioed to the chase car the following: “1961 Comet, 51 miles per hour.” Trooper Crinold, on receiving the radio message, stepped into the road and signaled a green Comet approaching from the west to stop, and it did. The defendant was the operator of the Comet; its registration number was Connecti*71cnt 22235; Trooper Grin old radioed this number to Trooper Johnson, and the latter wrote on the graphic recorder above the tracing showing 51 miles per hour the following: “11:00 A. M. green Comet, Connecticut Registration 22235.”
The state offered evidence to prove, and claimed to have proved through the testimony of Sergeant Hurlburt as a radar expert, that the 40- and 60-mile-per-hour tuning fork tests were sufficient to establish the accuracy of the radar equipment through a range extending from 20 to 70 miles an hour, that the radar equipment used on the day in question was properly and adequately tested for a car traveling at the rate of 51 miles an hour, that running a car with a calibrated speedometer through the radar field is not an absolutely accurate check of the accuracy of the radar unit but may be used to substantiate the tuning fork tests, and that a tracing on the graphic recorder which commences one-half mile above the base line will give a higher reading to the extent of one-half mile an hour. The state also claimed to have proved that the tracing on the graph as to the Comet was defined and identified not only by the written notations thereon but also by the oral testimony of Trooper Johnson.
The defendant offered evidence to prove and claimed to have proved that in order to test the overall accuracy of the radar unit in the range from 20 to 100 miles per hour it is necessary to use three tuning forks, i.e. 40-, 60- and 80-mile per hour forks; that the radar unit should be given an additional test by driving a ear with a calibrated speedometer through the radar field; that the tracing on the graph as to the Comet was neither clearly defined nor identified, nor was it accurate, because the needle commenced its recording one-half mile per hour higher than the base line; that it is preferable *72to have visual contact between the operators of the radar unit and the chase car; and that where there is no such visual contact there would be a question of identification.
No question is raised as to the well-established principles that the court may take judicial notice of the fact that the radar device can accurately record the speed of an automobile and that the speed recorded on the radar unit may be admissible in evidence provided the accuracy of the radar unit at the time in question is established by tests made within a reasonable time before and after the speed was recorded and provided the speed recorded was for the defendant’s automobile. See State v. Dantonio, 18 N.J. 570. What the defendant asserts is that the evidence of tests for accuracy by means of the 40- and 60-mile-an-hour tuning forks was insufficient to justify the admission in evidence of the recorded speed and that tests by the 80-mile-an-hour tuning fork and by running a car with a calibrated speedometer through the radar field were essential. Sergeant Hurlburt, the defendant’s expert witness, testified that to test the overall accuracy of the radar unit in covering a range of 20 to 100 railes per hour all three tuning forks had to be used, but that to test the accuracy over a range of 20 to 70 miles an hour and 40- and 60-mile-an-hour tuning forks were sufficient, and that a reading of 51 miles per hour on a unit tested only by the 40- and 60-mile-an-hour forks was an accurate measurement of speed. The court did not err in excluding testimony as to the 80-mile-an-hour tuning fork. Nor did the court err in excluding evidence concerning the necessity of running a car with a calibrated speedometer through the radar field as a test for accuracy. Sergeant Hurlburt also testified that such a test was not absolutely necessary but should be made; it merely substantiates the tuning fork test. In prac*73tically all the cases cited by the defendant, the testing method used was to run a test car through the radar and no mention was made of any tuning fork test, and the method used was held to be a proper method provided the speedometer of the test ear had been properly calibrated. See State v. Graham, 322 S.W.2d 188, 197 (Mo. App.); People v. Sachs, 1 Misc. 2d (N.Y.) 148, 156; East Cleveland v. Ferell, 168 Ohio St. 298; Wilson v. State, 168 Tex. Crim. 439; Royals v. Commonwealth, 198 Va. 876. We have been unable to find any authority for the proposition that both the tuning fork test and the test car test are necessary. In denying defendant’s requests to charge and in its charge to the jury, the court properly left to the jury the decision of the question whether the accuracy of the radar unit was adequately tested.
On the question of identity, the defendant claims that the graph showing the tracing of the green Comet was improperly admitted into evidence because of inadequate identification of the tracing on the graph. The court admitted the graph when Trooper Johnson testified that the writing he made on the graph, “11:00 A.M. green Comet, Connecticut Registration 22235,” referred to the fifty-one-mile-an-hour tracing. The court properly admitted it. The testimony of Sergeant Hurlburt, in response to a hypothetical question, that in his opinion there was inadequate identification on the graph was for the jury to consider in resolving the question of identity. The court correctly charged the jury, leaving the decision of this question to them. The state is not obligated to make out its case beyond every possible doubt.
In reviewing the trial court’s denial of a motion to set aside the verdict, the appellate court examines and tests the evidence in the same way the jury *74should have done. The verdict will not be disturbed if there is any reasonable ground appearing in the evidence upon which the jury might have acted. The trial court is acting in the exercise of a legal discretion, and great weight must be given to its ruling and all reasonable presumptions resolved in its support. See Butler v. Steck, 146 Conn. 114, 116; Maltbie, Conn. App. Proc. § 190, and eases cited. In the case before us, we conclude that the court below correctly denied the motion to set the verdict aside.
The final claim of error relates to the comments of the prosecutor in his summation to the jury concerning the failure of the defendant to testify. Section 54-84 of the General Statutes specifically forbids comment to the court or jury on the neglect or refusal of an accused person to testify. In the case before us, the prosecutor did make a direct reference, both in his opening argument and in his closing argument to the jury, to the fact that the defendant failed to testify. The exact language used is not of record. In each case, the attorney for the defendant made no objection, did not take exception thereto, did not move for a mistrial, and did not request that the prosecutor be admonished, nor did he request of the court any instruction of any kind to the jury; he remained completely silent. The court made no comment in respect thereto and gave the charge customarily given to the jury when the accused does not testify; to this charge the defendant’s attorney did not except. On the same day as the rendition of the verdict of guilty, the defendant filed a motion to set the verdict aside on the ground that it was contrary to the law and the evidence, and the motion did not contain any reference whatever to the prosecutor’s comments. Nor was any reference made thereto until the attorney for the defendant argued the motion to set aside the verdict about a week later. The court concluded that *75the right of this defendant to a trial by a jury before whom no comment of any kind is made concerning the neglect or refusal of the defendant to testify is one which he may waive and which he in this case did waive by neglecting to move for a discharge of the jury at the time and by proceeding without objection or motion to the adverse ending of the trial. This conclusion is in accord with the general rule. It is only in flagrant cases that the court will grant a new trial. See such cases as State v. Buxton, 79 Conn. 477, 480; State v. Monahan, 96 Conn. 289, 292; State v. Frost, 105 Conn. 326, 338; State v. Ferrone, 96 Conn. 160, 169; State v. Washelesky, 81 Conn. 22, 28. Although the statute appears to have been violated in the instant case, we have no record of the exact words used and the context in which they were used, and we cannot say that the trial court erred in denying a new trial to the defendant.
There is no error.
In this opinion Jacobs and Kinmonth, Js., concurred.