ACOBA, Judge.
Defendant-Appellant Gary R. Dibenedetto (Defendant) was charged in an August 13, 1991 Complaint with Driving Under the Influence of Intoxicating Liquor (DUI) in violation of either Hawaii Revised Statutes (HRS) § 291-4(a)(l) (Supp.1992) (Count I) or HRS § 291-4(a)(2) (Supp.1992)1 (Count II) and with Non-Compliance with Speed Limit in violation of HRS § 291C-102 (1993) (Count III). After a jury trial on February [141]*1412, 1993, Defendant was found guilty on all counts. The court sentenced Defendant and entered judgment on February 5, 1993. We affirm.
I.
On April 18, 1991, Officer Lester Hite (Officer Hite) was on duty on the midnight shift, driving on South King Street in the Koko Head direction2 on the island of 0‘ahu. Defendant’s vehicle “quickly” passed him on the left and moved in front of the officer. The officer was traveling thirty miles per hour, the speed limit for that area. Officer Hite then “sped up and maintained a distance of about four car lengths [from Defendant’s vehicle] and monitored [Defendant’s] speed ... [at] approximately thirty-eight miles an hour.” After “clocking” Defendant in this manner for “[a]pproximately three-tenths of a mile[,]” Officer Hite followed Defendant into a parking lot at approximately 11:50 p.m.
In the parking lot, Officer Hite observed Defendant’s vehicle turn “quickly” into an open space, “str[ike] [a] raised curb[,] ... [move] up on the curb and eventually roll[ ] back down into the [parking] space.” The officer stopped behind Defendant’s car. When Defendant exited, Officer Hite approached him. Standing within three feet of Defendant, the officer “detected a moderate odor of alcohol [on Defendant’s] breath” and observed that Defendant’s “eyes appeared to be bloodshot and glassy.” Officer Hite also noticed that Defendant’s “speech appeared ... to be slurred somewhat.”
Officer Hite then requested Defendant to perform a field sobriety test which consisted of three parts—the “horizontal walk and turn,” the “one-leg stand,” and the “count to thirty” during the one-leg stand. In the officer’s opinion, Defendant failed the tests, and he arrested Defendant for “speeding and driving under the influence of alcohol[.]”
During cross-examination, Officer Hite testified, in part, as follows:
Q. [ (Defense counsel) ] ... earlier you testified that you don’t remember all the particulars of this field sobriety test that you refreshed your memory ... using this sheet and the other sheets that you filled out?
A. [(Officer Hite)] Yes.
[[Image here]]
Q. ... how big [was] the gap [between Defendant’s heel and toe] on that fourth step ... [in the “horizontal walk and turn”]?
A. I don’t recall exactly the distance of the gap, no.
Q. You don’t recall the distance of the gap so you are basically testifying to that distance on that gap from ... this sheet of paper....
A. Yes.
Q. Without memory of what actually happened?
A. Yes.
[[Image here]]
Q. ... is it fair to say that your memory as to what the actual events are is pretty cloudy ... what you have is memory of your recently reviewing this field sobriety test [document]?
A. Yes.
(Emphases added.)
Defense counsel argued that the officer was not testifying from his “independent recollection” but “testifying off of the written [police] report” and moved to strike the officer’s testimony. The court ruled that the matter was for the jury to determine.
[Defense counsel]: Your Honor, it seems that he’s not testify [sic] of his independent recollection of the night in question and I move to strike all of his testimony since he’s not able to remember. He’s just testifying off of the written report which I don’t think satisfies the standard of admissible [sic] for evidence in the trial.
[Deputy Prosecutor]: The State would argue that he said on the field sobriety test, he doesn’t have a clear memory of all the facts of the field sobriety test and he says he does not [sic] have an independent [142]*142recollection of the arrest of the speeding as well as talking to the defendant.
THE COURT: All right. I note the objection for the record. It’s for the jury to make a determination on that basis. I think that’s entirely appropriate for the jury to determine this initial testimony.
(Emphasis added.)
Officer Hite testified that another officer arrived at their location to transport Defendant to the police station. At the station, Officer Hite explained Hawaii’s “implied consent law” to Defendant. After the officer explained the implied consent form (State’s Exhibit No. 2) to Defendant, Defendant signed the form, electing to take the breathalyzer or intoxilyzer test.
The results of the intoxilyzer test administered at 12:45 a.m. on April 19, 1991 indicated that Defendant’s blood alcohol concentration (BAC) by weight was .113% (State’s Exhibit No. 5). At trial, defense counsel cross-examined Joanne Furuya and Claire Chun, both criminalists for the Honolulu Police Department on the accuracy of the intox-ilyzer machine. Ms. Furuya testified that the recognized “margin of error” for the intoxilyzer machine was “[pjoint 0 one [.01].” She explained that because the intoxilyzer’s margin of error was in the hundredths of one percent, the thousandth digit was “irrelevant.”
Q. So the machine is not reading completely accurately that there is no alcohol in that air blank sample?
A. That’s why the margin of error is at the hundredth place. That’s why it’s not at the thousandth place.
Q. And so when we get the number in the thousands [sic] column, there is a question as to just how accurate the machine is reading at that point?
A. No. Because as I mentioned before if one gets a point one seven three reading it’s plus or minus point zero one which [sic] a hundredth place it can vary between one six three and one eight three so that third digit is irrelevant.
Q. The third digit is irrelevant?
A. For the reading as proposed here.
Q. It only become[s] relevant as applied to the defendant?
A. Yes, it’s—you are mistaken for what I am saying. When you apply the hundredths place margin of error, your third digit becomes irrelevant because you are moving the range in the hundredths place which is a less accurate value. Now you understand what I am saying?
According to Ms. Furuya, the thousandth place digit is not rounded off or considered:
A. If your margin of error is plus, minus point 0 one, what we do is if there’s a reading of point 0 nine five, that five is not automatically brought up to the point one 0, it’s totally intricate [sic].
Ms. Chun, who testified after Ms. Furuya, reiterated the fact that the thousandth place digit was not relevant:
Q. [ (Defense counsel) ] ... So if the intoxilyzer is reading, it cannot only be accurate within a plus or minus one-one hundredth [.01] of a percentage, how is it that the intoxilyzer machine can measure accurately to the thousandths of a percent?
A. [ (Ms. Chun) ] The instrument mil read to a thousandths of a percent but the manufacturer stated the margin of error in accuracy terms of point 0 one [.01] which is one-one hundredth of a percent.
The reason that we record the third digit shown in the display was because of requests by, I believe, the judiciary and the prosecutor’s office some years ago to record the third digit. However, the statute in the state are [sic] two digits and we feel that the two-digit readout on the printout can suffice. The reading or the logging of the third digit is just for confirmation and per our [sic] request to do that.
The evidence implied that the intoxilyzer was only warranted by the manufacturer to be accurate to .01 percent.
Q. [ (Defense counsel) ] Now what other sources of error have created this margin of error of one-one hundredth of a percentage?
[143]*143A. [ (Ms. Chun) ] I’m not certain [sic] the exact sources, but in warrantee [sic] of the design of an instrument is some sort of margin of error, basically on the mechanical design of the instrument.
Q. So the plus or minus one-one hundredth is the mechanical error, it’s the accuracy of the machine?
A. Of the instrument itself, yes.3
Moving to a different subject, defense counsel cross-examined Ms. Chun on the subject of a “twenty-one hundred to one ... partition ratio.” Ms. Chun explained that “the partition ratio value states that for twenty-one hundred[] part[s] of a person’s breath, that breath would contain an equivalent amount of alcohol on [sic] one part [of] the person’s blood and I guess that would be a multiplication of the value.” She stated:
Basically the partition ratio states that for a certain amount of alcohol in the breath, there is a mathematical correlation with the amount of alcohol in a person’s blood and the instrument reads the alcohol in the person’s breath, deep lung breath, and then processes that value into what’s called a BAC, or blood alcohol concentration.
Defense counsel questioned the relevancy of an “average” partition ratio in calculating Defendant’s actual BAC level:
Q. ... If a person had a lower partition ratio than the average ratio which is put into these machines, would that provide for a falsely [sic] high reading of blood alcohol content from the breath sample blown?
A. If a person had a lower partition ratio then the reading would be higher, that’s correct.
[[Image here]]
Q. So did you examine [Defendant] to find out that his blood alcohol—I mean his partition ratio was the average, twenty-one [hundred] to one?
A. No....
Ms. Chun explained that “there’s only one partition ratio accepted for instruments and the State of Hawaii [Hawaii] approved this instrument which was designed to [a] twenty-one hundred to one value.” She further stated that “[t]he Federal Government, the National Highway Traffic Safety Administration only approves instruments manufactured through [sic] this partition ratio....” The defense did not present any evidence of Defendant’s personal “partition ratio” or any evidence challenging governmental approval of the “partition ratio” applicable to the machine used.
At the conclusion of the State’s case, the defense moved for a judgment of acquittal, arguing, inter alia, that the State had failed to prove beyond a reasonable doubt that the “twenty-one hundred to one partition ratio” employed by the intoxilyzer would accurately reflect Defendant’s personal “partition ratio.” The court denied the motion.
During the settlement of jury instructions, defense counsel objected to the court giving Court’s Special Instruction No. 5 (Court’s Instruction No. 5), instead of Defendant’s Requested Jury Instruction No. 1 (Requested Instruction No. 1). Defendant’s Requested Instruction No. 1 read:
The .01 margin of error of the Intoxilyzer machine must be construed in favor of the defendant.
The court rejected Requested Instruction No. 1 as “redundant” of Court’s Instruction No. 5, which read as follows:
The Intoxilyzer has a margin of error of one one-hundredth (0.01) per cent, which means that in order to find the Defendant guilty under Count II, you must find that any Intoxilyzer test result introduced as evidence actually showed a blood alcohol content by weight of eleven one-hundredth (0.11) per cent or more.
Following the reading of Court’s Instruction No. 5 to the jury, defense counsel again objected to the instruction, on the ground that “it leads the jury to believe that a BAC, point one ... one [.11] [reading] on the machine ... is sufficient to convict.”
[144]*144On appeal, Defendant contends as error that (1) the arresting police officer lacked a present recollection “as to material matters” in his testimony; (2) the intoxilyzer result as measured to the thousandth digit should not have been received in evidence; (3) the jury should not have been instructed under Count II that Defendant was guilty if the “evidence actually showed a blood alcohol weight of eleven one-hundreds [sic] (0.11) percent or more”; and (4) Court’s Instruction No. 5 foreclosed the jury from considering the fact that the partition ratio may vary among individuals.
II.
In arguing that Officer Hite had no present recollection of “material events,” Defendant does not specifically challenge the speeding violation but, as our following discussion indicates, focuses on the DUI charge. Thus, there is no apparent dispute with respect to the speeding conviction.
Defendant claims the officer lacked a present recollection of the intoxilyzer examination and field sobriety test.4 Because Officer Hite did not administer the intoxilyzer examination, his recollection of the examination is immaterial. We turn, then, to the officer’s testimony on the field sobriety test.
Although defense counsel incorrectly asked the officer if he had an “independent recollection” of the field test rather than “a present recollection” of the test, the officer’s candid testimony leads us to the conclusion that the officer did not have a “present recollection” of the test at the time he testified. As set forth supra, the officer indicated that his testimony was based on what he had recently read in his report. Thus, we are left with the question of whether the court erred in failing to strike Officer Hite’s testimony regarding the field sobriety test.
Hawaii Rules of Evidence (HRE) Rule 612 indicates that “a witness [may] use[ ] a writing to refresh [his] memory for the purpose of testifying....” A writing, such as a police report, used to refresh a witness’s memory is ordinarily not submitted into evidence. 3 J. Wigmore, Evidence in Trial at Common Law § 763, at 142 (Chad-bourn rev. ed. 1970). When used to refresh the witness’s present recollection, a writing is solely employed to jog the memory of the testifying witness. 1 J. Strong, McCormick on Evidence § 9, at 29 (4th ed. 1992). Accordingly, when a writing is used to refresh a witness’s recollection, the witness should testify from “a memory thus revived,” resulting in testimony from present recollection, not a memory of the writing itself. Id. “A witnesses] recollection must be revived after he [or she] consults the particular writing or object offered as a stimulus so that ... [the] resulting testimony relates to a present recollection.” 3 J. Weinstein & M. Berger, Weinstein’s Evidence ¶ 612[01], at 612-16 (1995). If the writing fails to rekindle the witness’s memory, the witness cannot be permitted to testify as to the contents of the writing unless the writing is otherwise admitted into evidence. 28 C. Wright & V. Gold, Federal Practice and Procedure: Evidence § 6183, at 463 (1993).
On redirect examination, the State did not attempt to establish that Officer Hite was in fact testifying from present recollection. But the court declined to rule on Defendant’s motion to strike, treating the issue as one for the jury to determine.
The question of whether Officer Hite was properly allowed to testify about the field sobriety test was not a question of credibility for the jury to decide, as the State maintains, but one of admissibility for the judge to determine. HRE Rule 104 mandates that “[pjreliminary questions concerning the qualification of a person to be a witness ... or the admissibility of evidence shall be determined by the cowi[.]” (Emphasis added.) HRE Rule 601 provides that “[e]very person is competent to be a witness except as otherwise provided in [the HRE].” In that connection, HRE Rule 602 declares that a “witness may not testify to a matter [145]*145unless evidence is introduced sufficient to support a finding that the witness has personal knowledge of the matter. Evidence to prove personal knowledge may ... consist of the witness’s own testimony....” “Personal knowledge” of a witness, under HRE Rule 602, “means that the witness perceived the event about which [the witness] testifies and [the witness] has a present recollection of that perception.” Commentary to HRE Rule 602 (1993) (emphasis added). Thus, the question of whether the evidence was sufficient to support a finding that Officer Hite had a present recollection of the field sobriety test was to be determined by the court as a question of the officer’s qualifications to testify as a witness on that matter.
The standard of review on appeals from evidentiary rulings depends on the particular rule of evidence at issue. Kealoha v. County of Hawaii, 74 Haw. 308, 315, 844 P.2d 670, 674 (1993); Kam Fui Trust v. Brandhorst, 77 Hawai'i 320, 326, 884 P.2d 383, 389 (Haw.App.1994). See State v. Robe, 5 Haw.App. 251, 687 P.2d 554 (1984). Errors on evidentiary rulings are subject to the abuse of discretion standard unless the application of the rule can produce only one correct answer, in which case, then, the review falls under a “right/wrong standard.” Kam Fui Trust, 77 Hawai'i at 326, 884 P.2d at 389 (quoting Kealoha, 74 Haw. at 319, 844 P.2d at 676).
Because a witness cannot be permitted to testify if the witness has no present recollection, there can be only one correct answer to the question of whether the witness had a present recollection of the material events after reviewing the writing and setting it aside. Consequently, we apply the “righVwrong” standard in determining the correctness of a ruling that the witness’s attempt to refresh his or her memory resulted in a present recollection of the subject events.
Based on our review of the officer’s testimony, we conclude that the court’s ruling was wrong, and the officer’s testimony relating to the field sobriety test should have been stricken and the jury instructed to disregard such testimony.
III.
As the foregoing indicates, Officer Hite’s testimony about the field sobriety test should not have been considered in determining guilt under HRS § 291-4(a)(l). We examine, then, Defendant’s arguments against his conviction under HRS § 291-4(a)(2), which prohibits the operation of any vehicle by a person with .10% or more, by weight of alcohol in his or her blood.
A.
Defendant first asserts that the court erred in admitting the intoxilyzer’s “three digit” test result of “.113%,” recorded on State’s Exhibit No. 5. He maintains that “the machine cannot accurately measure the [breath] sample to the thousandth place” and the result would “erroneously lead the jury to believe that the Intoxilyzer is much more accurate than it actually- is.” As recounted previously, Ms. Furuya indicated that the “third digit” was “irrelevant” for purposes of this case, and Ms. Chun explained that the “third digit” was only recorded “because of requests by ... the judiciary and the prosecutor’s office some years ago.... ” Ms. Chun additionally testified that although “[t]he instrument will read to a thousandth of a percent[,]” the “manufacturer stated [that] the margin of error in accuracy terms ... is one-hundredth of a percent.”
Because the testimony concerning the thousandth place reading was that it was irrelevant, the court should have ordered the “3” in the “.113%” reading, in State’s Exhibit No. 5, redacted. Nevertheless, Ms. Furuya and Ms. Chun had testified, in essence, that the “third digit” was irrelevant. Thus, the jury would not have been misled into believing the intoxilyzer reading was material beyond a margin of error of .01%. Furthermore, the court’s instructions properly focused the jury on the test result necessary for conviction and the margin of error to be considered. See discussion infra. Hence, under the evidence received, the thousandth place reading could not have prejudicially affected Defendant. Even if allowing the “th[ird] digit” of the BAC reading into evi[146]*146dence was error, it amounted to harmless error. Hawaii Rules of Penal Procedure Rule 52(a).5 We hold, then, that where the intoxilyzer measured the BAC level to a thousandth of one percent, the error, if any, in receiving the thousandth reading in evidence is harmless beyond a reasonable doubt when the evidence itself indicates the thousandth digit reading was irrelevant. See State v. Holbron, 80 Hawai'i 27, 31 n. 12, 904 P.2d 912, 916 n. 12 (1995).
B.
Defendant next challenges Court’s Instruction No. 5. Accounting for the intoxilyzer’s .01% margin of error, this instruction required the prosecution to prove Defendant’s intoxilyzer result was at least .11%. This instruction was correct under State v. Boehmer, 1 Haw.App. 44, 613 P.2d 916 (1980). Boehmer indicated that a margin of error of .0165% on an intoxilyzer or breathalyzer test meant that the defendants’ BAC readings could be below the requisite .10% BAC threshold, which at that time, established a presumption that an accused was under the influence of intoxicating liquor. Id. at 46, 613 P.2d at 918. This court reasoned that, “[t]he margin of error of the breathalyzer test means that on any given breathalyzer test a defendant’s actual blood alcohol content could be a 0.0165% more or less than the reading shown by the breathalyzer test.” Id. (emphasis added). Accordingly, “the results of such a test when taken together with its tolerance for error, must equal or exceed the statutory level” of .10%. Id. at 48, 613 P.2d at 919. Because “the inherent margin of error could put [a defendant’s] actual blood alcohol level below the level necessary for the presumption to arise[,]” it was held that the “prosecution [failed] to establish beyond a reasonable doubt that the actual weight of alcohol in defendant’s blood was at least 0.10%.” Id. at 47, 613 P.2d at 918.
While Boehmer concerned a statutory presumption of intoxication, HRS § 291-4(a)(2) per se establishes illegal intoxication upon a BAC reading of .10. Consequently, the BAC reading here would have a decisive impact on the determination of guilt. That “the inherent margin of error could put [a defendant’s] actual blood alcohol concentration” below .10 BAC is no less true where .10 BAC establishes a presumption of intoxication or per se establishes intoxication. It follows, then, and we hold that under HRS § 291^4(a)(2), the prosecution must establish that the result of the chemical test involved “when taken together with its tolerance for error must equal or exceed the statutory” threshold of .10 BAC to prove “beyond a reasonable doubt that the actual weight of alcohol in [a defendant’s] blood was at least O.10%.” Id. at 47, 613 P.2d at 918.
Because Court’s Instruction No. 5 based guilt on a BAC reading of .11% or more, it adequately accounted for the .01% margin of error inherent in the test. The court’s instruction was more accurate and complete than Defendant’s requested instruction that the .01% margin of error be “construed” in Defendant’s “favor.” A trial judge’s discretion in refusing to give a jury instruction is “defined by the obligation to give sufficient instructions....” Tittle v. Hurlbutt, 53 Haw. 526, 530, 497 P.2d 1354, 1357 (1972). In general, “ ‘[r]epetitious instructions should be avoided[.]’ ” DiCenzo v. Izawa, 68 Haw. 528, 543, 723 P.2d 171, 180 (1986) (quoting Tittle, 53 Haw. at 531, 497 P.2d at 1357). The court did not err in its refusal to give a proposed instruction that was “not an accurate statement of the law” and had been “adequately covered by the instructions given....” Housing Fin. & Dev. Corp. v. Castle Found., 79 Hawai‘i 321, 329, 901 P.2d 1300, 1308 (App.1995). We agree, then, that Defendant’s Requested Instruction No. 1 was unnecessary.
C.
Defendant’s last objection to Court’s Instruction No. 5 is that “it [mis]-leads the jury [in]to believ[ing] that a reading of 0.11% BAC or higher is sufficient to convict Defendant” and deprives the “jury [of [147]*147the discretion] to decide how much weight to give to the reading based on their [sic] belief as to how accurately the machine reflects the actual BAC of the defendant.” The apparent premise for this particular contention was the testimony elicited from Ms. Chun on cross-examination that the “partition ratio,” the mathematical correlation between the amount of alcohol in a person’s breath and the amount of alcohol in a person’s blood, may vary among individuals. But this court has held that the employment of a 2,100 to 1 partition ratio “goes to the weight the jury should accord the Intoxilyzer test result with respect its accuracy, not the admissibility of the test result.” State v. Gates, 7 Haw.App. 440, 446, 777 P.2d 717, 721 (1989); State v. Young, 8 Haw.App. 145, 155, 795 P.2d 285, 291, cert. denied, 71 Haw. 669, 833 P.2d 901 (1990). In this respect, Court’s General Instruction No. 8 properly informed the jurors that they were the “sole and exclusive judges of the effect and value of the evidence and of the credibility of the witnesses.” By its verdict in Count II, the jury, as the judge of “the effect and value of the evidence,” apparently believed the intoxilyzer reliably measured Defendant’s BAC level.6 Therefore, we perceive no error in Court’s Instruction No. 5 as it may have related to testimony concerning the partition ratio.
IV.
For the foregoing reasons, the judgment of February 5, 1993 is affirmed.