Opinion

This appeal is from a speeding conviction resulting from radar surveillance. The appellant contends that (1) he was entitled to introduce the testimony of an expert witness to show that the radar did not accurately record his automobile’s speed and (2) calibration of the radar with tuning forks was insufficient to permit the. introduction of the radar measurement into evidence. We conclude that the testimony of the expert -witness was admissible and that the calibration of the radar by tuning forks alone was insufficient without proper corroboration. For these reasons, we reverse.

At the time of the alleged offense, the police officer was operating a stationary radar unit located in a church parking lot facing *165oncoming traffic. The officer testified that her radar unit recorded that the appellant’s automobile was traveling fifty-three miles per hour. The speed limit was forty miles per hour. The trial court convicted the appellant of speeding and imposed a $50 fine.

During the course of the trial, the appellant testified and attempted to offer the testimony of an expert witness. The appellant testified that he was driving a Pontiac Fierro, a small two-door sports car, made of “indura-flex material . . . essentially plastic.” Fie said that he saw the arresting officer parked in the church parking lot talking with another police officer in another car parked parallel to hers and that the arresting officer looked up when he passed. He said that the arresting officer’s automobile was parked at an angle to the highway of approximately twenty-five to thirty degrees. The traffic was light, and, although there was no traffic in his “immediate vicinity,” there were several cars in back of him, including a large tractor-trailer truck, that were closing on him.

The appellant also proffered testimony of an expert witness. After the witness described his credentials and appellant’s counsel offered him as an expert regarding the training and operation of speed measuring devices, his testimony was proffered outside of the presence of the jury. The purpose of the testimony was to show that the radar did not record the appellant’s vehicle but, instead, recorded that of another vehicle, possibly the tractor-trailer truck behind him. The witness testified that, because of its size and material, the appellant’s vehicle was extremely difficult to pick up on radar, that the thirty degree angle of the police officer’s car to the highway would have caused an inaccuracy in the speed recorded by the radar, that if the radar had skipped over the appellant’s vehicle it would have probably picked up the more reflective vehicle such as the tractor-trailer truck and that the likelihood of the radar picking up the appellant’s vehicle was further reduced by the fact that the police vehicle was parked on a lot approximately eighteen inches to two feet above the road. Finally, he would also have testified that the audio doppler from the radar should not have remained constant as the police officer testified that it did, but should have dropped off as the appellant’s vehicle got closer, unless the radar was reading a vehicle further away, such as the tractor-trailer truck.

*166The result of the use of radar to check the speed of a motor vehicle is prima facie evidence of the speed of that motor vehicle. Code § 46.2-882.1 This follows from the “natural and rational evidentiary relation” between the results of a radar check of speeding and the actual speed of the motor vehicle. Dooley v. Commonwealth, 198 Va. 32, 35, 92 S.E.2d 348, 350 (1956), appeal dismissed, 354 U.S. 915 (1957). Expert testimony is not admissible to contest the reliability of the use of radar for this purpose. Thomas v. City of Norfolk, 207 Va. 12, 14, 147 S.E.2d 727, 729 (1966). However, evidence may be introduced to show that a particular radar reading may have improperly recorded the speed of a vehicle which it had been used to measure. Dooley, 198 Va. at 35, 92 S.E.2d at 350. Such evidence is admissible to rebut the prima facie presumption created by the statute. Id.

In this case, the expert’s testimony was not offered to contest the legislative decision that radar readings provided prima facie evidence of a motor vehicle’s speed; it was introduced, instead, to rebut the prima facie case by showing that the particular radar reading in this instance was not a reliable indication of the speed of the appellant’s automobile. The reflective qualities of the appellant’s motor vehicle, the presence of the tractor-trailer truck behind the appellant’s vehicle, the angle of the radar unit with respect to the highway, and its height above the highway are not factors challenging the radar unit’s ability to measure motor vehicular speed. They are factors, however, that challenge the accuracy of the radar reading of the appellant’s speed on this occasion. For this reason, we hold that the trial court erred in refusing to admit the expert’s testimony into evidence.2

The appellant also contests the sufficiency of the evidence of the radar’s accuracy with the use of tuning forks. Since the proceeding must be remanded for a new trial, we address that issue.

The only evidence that the radar had been tested for accuracy was the officer’s testimony that she had tested it with tuning forks, which in turn had been certified to be accurate by “Southeastern *167Communications.” The officer testified that she used two tuning forks, one for thirty-five miles per hour and one for sixty-five miles per hour and, using standard procedure, the radar readings indicated that the unit was functioning accurately. In addition, the court admitted into evidence, over the appellant’s objection, certificates signed by a person on behalf of Southeastern Communications certifying that ten months earlier the radar and the tuning forks had been tested and were found to be operating properly.3

The General Assembly statutorily has exempted from the bar of the hearsay rule certain certificates used as evidence of a radar device’s calibration; however, the exemption extends only to certificates involving calibration by use of a speedometer, not tuning forks. A certificate is admissible over a hearsay objection when signed by an officer who has calibrated a radar device against the speedometer and when accompanied by a further certificate reflecting the accuracy of the speedometer used to calibrate the radar device. Code § Í6.2-882.4 The certificates must conform to the requirements of the statute, including one which limits their validity to six months. Id. There is no statutory authority overcoming the hearsay rule with regard to certificates of the accuracy of tuning forks. The certificates in this case, therefore, were not exempt from the hearsay rule and should not have been admitted into evidence.

Without the certificates, the evidence of the accuracy of the radar device was limited to the testimony of the officer that the radar device had responded appropriately to two tuning forks, the accuracy of which was not established. Thus, there was no evidence upon which to find that the radar device appropriately responded to the tuning forks. Biesser v. Town of Holland, 208 Va. 167, 169, 156 S.E.2d 792, 794 (1967); see also Crosby v. Commonwealth, 204 Va. 266, 268, 130 S.E.2d 467, 468 (1963); Royals v. Commonwealth, 198 Va. 883, 884, 96 S.E.2d 816, 817 (1957). The evidence of the accuracy of the radar device used in *168this case was, therefore, insufficient.

In summary, we hold that the appellant was entitled to present expert testimony tending to impeach the accuracy of the radar reading upon which the Commonwealth’s case was based, so long as it did not challenge the use of radar as a device for measuring the speed of a moving vehicle. In addition, we hold that a radar unit’s accuracy is not established simply by proof that the unit was tested by tuning forks without proper evidence of the accuracy of the forks.

The judgment of conviction is reversed and this proceeding is remanded for a new trial.

Reversed and remanded.

Willis, J., concurred.

dissenting.

I respectfully disagree with the finding of the majority that all of the proffered testimony of the expert witness was admissible and that the calibration of the radar unit by tuning forks was insufficient to support the conviction.

Police Officer Elma Leap was the only witness for the Commonwealth. No question has been raised concerning her qualification as a radar operator. She has been a police officer for five years and assigned to the traffic division for three years. She testified that on the morning in question she ran tests on the radar unit prior to setting it up. She first checked the internal system in the radar itself and determined that the unit was functioning properly. She next struck the tuning forks on the padded steering wheel and held each one separately in front of the radar field of influence to indicate the speed reflected on the radar. When she struck the thirty-five miles per hour tuning fork, the radar unit registered thirty-five miles per hour. When she struck the sixty-five miles per hour tuning fork, the radar unit registered sixty-five miles per hour. According to her testimony, the readings she obtained indicated the unit was functioning accurately.

After testing the radar unit, Officer Leap put the radar unit in operation on Broad Rock Road facing and clocking eastbound traffic. She testified that she visually observed the defendant ap*169preaching and in her mind estimated his speed to be over the speed limit. She testified that the reading on the radar verified her visual observation of the defendant’s speed. The radar read-out showed the defendant was traveling fifty-three miles per hour in a forty miles per hour posted zone. She pushed a button manually and the defendant’s speed was locked in at fifty-three miles per hour. Officer Leap further testified that after each violation that day she checked the system internally to make sure the unit was operating properly. When she went off duty, she calibrated the radar unit again with the factory certified tuning forks. She testified that the unit operated properly throughout the day, the same as when it started that morning.

Under this set of facts, the majority concludes that “[t]he only evidence that the radar had been tested for accuracy was the officer’s testimony that she had tested it with tuning forks which in turn had been certified to be accurate by Southeastern Communications.” The majority further concludes: “Without the certificates,5 the evidence of the accuracy of the radar device was limited to the testimony of the officer that the radar device had responded appropriately to two forks, the accuracy of which had not been established.” I disagree. In my opinion, it does not follow that because the certificate was inadmissible there was no corroboration of the accuracy of the radar unit in the record. In my opinion, the tests with the two tuning forks constitute sufficient corroboration of the accuracy of the radar unit.

The majority, in effect, holds that the failure to prove the accuracy of the tuning forks is fatal to the Commonwealth’s case. They would require the state to prove not only the correctness of the radar unit which clocked the defendant’s speed, but also demand that the Commonwealth prove the accuracy of the tuning forks used to check the accuracy of the radar unit. I would not cast the burden upon the Commonwealth in a speeding case to do more than prove that a comparative tuning fork analysis was *170made showing the accuracy of the radar unit. The vast majority of cases in the United States do not require any corroboration of the accuracy of the radar unit except comparison with the tuning forks.

This case involves the question of what evidence is required to establish the accuracy of a radar speed-measuring device. Since 1954 the General Assembly has authorized the use of radar machines to measure the speed of motor vehicles. See Code § 46.2-882. The statute contemplates that speed may be checked by radar and that the results of the radar reading will be accepted as prima facie evidence. Sweeny v. Commonwealth, 211 Va. 668, 670, 179 S.E.2d 509, 512 (1971). Proof of the accuracy of the particular radar speed-measuring machine used, however, is required as a prerequisite to the admissibility of the results obtained therefrom.

It is well established that the Commonwealth must prove that the radar machine “had been properly set up and adjusted” and “must have been recently and accurately tested.” Royals v. Commonwealth, 198 Va. 876, 882, 96 S.E.2d 812, 816 (1957). In Thomas v. City of Norfolk, 207 Va. 12, 147 S.E.2d 727 (1966), one issue involved was whether there was sufficient evidence in the record to show that the radar machine had been recently tested for accuracy. The officers testified that they had used two methods to test the radar machine. First, a car driven by Officer Jackson passed through the radar beam at thirty, forty and fifty miles per hour as indicated on his speedometer and these speeds accorded with those indicated on the radar machine and noted by Officer Bennett. Second, the accuracy of the speedometer was corroborated by tests which the officers made with factory pre-tested tuning forks. One fork, when struck and held in front of the radar set, indicated or registered on the latter instrument a speed of thirty-five miles per hour. The second tuning fork did the same at fifty miles per hour. Officer Bennett testified that the tuning fork tests showed that the radar set accurately recorded these respective speeds.

Upon these facts the Supreme Court stated:

While the matter has not heretofore been presented to us it is well recognized in other jurisdictions that evidence of a *171test of radar equipment by tuning forks in this manner is admissible as tending to prove the accuracy of the equipment.

In the absence of evidence to the contrary in the present case, the trial court accepted, as it had a right to do, the evidence of these testings as adequate proof of the accuracy of the radar set.

Id. at 14, 147 S.E.2d at 728 (citations omitted). This case clearly stands for the proposition that testing a radar machine for reliability by tuning forks is sufficient to prove the accuracy of the equipment. See also Howell v. Commonwealth, 213 Va. 590, 592, 194 S.E.2d 758, 760 (1973) (holding that were two speedometers agreed independently with the radar device in every test, it is improbable that all three speed measuring instruments would have been inaccurate to precisely the same degree).

The majority relies upon the case of Biesser v. Town of Holland, 208 Va. 167, 156 S.E.2d 792 (1967), a case which is distinguishable on its facts. In Biesser, the only test of the accuracy of the radar was by the use of “a tuning fork,” which was not otherwise described. Moreover, no evidence was presented concerning the manner of its use. The Court noted that the only witnesses in the case spoke of “a” tuning fork and “his” tuning fork, as if only one was used. The Court indicated there was no explanation in the record how the officer checked the radar speeds of thirty-five, forty-five and fifty-five miles per hour by the use of only one tuning fork. One tuning fork is generally not considered sufficient corroboration to establish the reliability of the radar unit. See People v. Walker, 199 Colo. 475, 479-80, 610 P.2d 496, 500 (1980). Thus, in Biesser, the government’s case failed, not because tuning fork evidence was found insufficient, but because the tuning fork evidence in that case was found deficient.

Our sister states are almost unanimous in holding that when two tuning forks are used to ascertain the accuracy of a radar unit, no additional proof of the accuracy of the tuning forks is necessary because each tuning fork corroborates the accuracy of the other. Without attempting to list all of the cases, I will cite a few as examples of many others: People v. Abdallah, 82 Ill. App. 2d 312, 226 N.E.2d 408 (1967) (holding that where the officer made several tests with the tuning fork prior to and subsequent to *172the time the defendant passed into the radar field, and the officer was of the opinion the radar was working properly, there was sufficient proof of the accuracy of the unit to find the defendant guilty); People v. Lynch, 61 Misc. 2d 117, 304 N.Y.S.2d 985 (1969) (holding that the accuracy of a radar unit was established by internal test calibration in the machine itself, and by means of thirty and fifty mile per hour tuning forks); State v. Overton, 135 N.J. Super. 443, 343 A.2d 516 (1975) (holding that tests made with four tuning forks demonstrated the accuracy of the radar unit); State v. McDonough, 302 Minn. 468, 225 N.W.2d 259 (1975) (holding that the internal tuning fork in the radar unit calibrated as it was to the same speed as the external tuning fork served as an adequate check on the accuracy of the external tuning fork and the prosecution satisfied the requirement that the external tuning fork was reliably calibrated); State v. Shimon, 243 N.W.2d 571 (Iowa 1976) (holding that it is “inconceivable” that a reading of the radar unit prior to the readout taken of defendant’s speed, a subsequent testing of the radar unit with a tuning fork and the testing of the radar unit against the speedometer of the police vehicle all were inaccurate); State v. Primm, 4 Kan. App. 2d 314, 606 P.2d 112 (1980) (holding that the accuracy of the radar unit was adequately shown by the officer’s testimony regarding the internal calibrations and the tuning fork tests); People v. Walker, 199 Colo. 475, 610 P.2d 496 (1980) (holding that where a tuning fork test is used to calibrate a radar device, the prosecution must show that two tuning forks have been used, or, alternatively, that the single tuning fork used has been certified as accurate within one year of the test); State v. Kramer, 99 Wis. 2d 700, 299 N.W.2d 882 (1981) (holding that where the trooper tested the radar unit before and after the arrest, using both the device’s own internal calibration mechanism and two external tuning forks, the tests were sufficient to raise a rebuttable presumption that the radar device was functioning accurately); State v. Trantolo, 37 Conn. Supp. 601, 430 A.2d 465 (1981) (holding that there is no more reason to require independent proof of a tuning fork’s accuracy as a prerequisite to admissibility than that of a ruler); State v. Ahern, 122 N.H. 744, 449 A.2d 1224 (1982) (holding that the use of two tuning forks provided sufficient assurance that the tuning forks themselves were accurate because each tuning fork corroborated the accuracy of the other); Shears v. State, 648 P.2d 841 (Okla. Crim. App. 1982) (holding that the testimony of officer that he had performed both an external test of *173accuracy of the radar unit, using calibrated tuning forks, and an internal test, using the component unit within the radar itself, was sufficient to establish radar accuracy at the time of the arrest for speeding); State v. Bechtel, 24 Ohio App. 3d 72, 493 N.E.2d 318 (1985) (holding that where two tuning forks are used to ascertain the accuracy of the radar unit, no additional proof of the accuracy of the tuning forks is necessary because each tuning fork corroborates the accuracy of the other); State v. Tailo, 779 P.2d 11 (Hawaii 1989) (holding that once the State puts in evidence that the police conducted a tuning fork test indicating the radar gun was properly calibrated, this evidence creates a prima facie presumption that the tuning fork itself was accurately calibrated).

One authoritative writer says this about the potential for error in radar testings:

Some of these potential sources of error can be minimized or excluded by careful operating procedures and on-site tests. These include the use of tuning forks vibrating at frequencies such that their linear motions will cause the speedometer to register particular speeds if it is receiving properly, use of an internal, electronically activated tong for the same purpose, and simply checking that, when aimed at another police car, the radar reading corresponds with that car’s speedometer reading. ... At least on the questions of admissibility, however, most courts recognize that independent errors are unlikely to be identical. They tend to hold that some combination of these methods is sufficient to warrant admissibility.

E. Cleary, McCormick on Evidence § 204 (3d ed. 1984) (footnote omitted). I would follow the overwhelming weight of authority in the United States and hold that a radar unit’s accuracy is established by proof that two tuning forks and the internal system of the radar unit were used as corroborating devices, as was done in this case.

Concerning the evidentiary issue, defense counsel called David Stopper as an expert witness to explain to the jury “that radar doesn’t pick up the closest vehicle, but that a number of things are involved such as the type of vehicle, what’s in the vehicle, and the size of the vehicle.” The defendant contested the fact that his car was picked up on radar, claiming that there was a tractor-trailer truck half a mile down the road. The defense attempted to elicit *174this type of evidence from Stopper by asking him hypothetical questions. The trial court did not permit this evidence to be introduced because it erroneously thought that the radar statute prevented it. However, in my opinion, general rules of evidence governing the use of expert testimony, not the radar statute, precluded the introduction of this evidence. The expert witness in this case had no personal knowledge of the facts. Therefore, he was required to give his opinion based upon facts in evidence assumed in hypothetical questions. C. Friend, The Law of Evidence in Virginia § 217 (3d ed. 1988). In my opinion, the proffered questions did not contain all of the facts in evidence and a proper foundation was not laid for most of them. There is no need to refer to them in more than general terms, and to suggest that on remand the questions should be phrased appropriately and based upon facts proven in the evidence.

For the above reasons, I do not concur in the majority opinion.