John Tomkins sent a series of threatening letters to investment firms and their employees and then mailed packages to two investment managers containing what appeared to be pipe bombs. The homemade devices consisted of a plastic pipe holding gunpowder, lead pellets, and an igniter connected to live batteries. Letters in each package warned that the recipients were only alive because Tomkins left one wire on each device unattached. After a two-week trial, a jury found Tom-kins guilty of mailing threatening communications, 18 U.S.C. § 876(b), illegally possessing firearms, 26 U.S.C. § 5861(d), and using a firearm in connection with a crime of violence, 18 U.S.C. §§ 924(c)(1)(A), (c)(l)(B)(ii). The district court imposed a prison sentence of 37 years. Tomkins now seeks a new trial, arguing that the district court erred by (1) barring his defense that the devices were meant as hoaxes, (2) admitting an x-ray of one of his devices that the government failed to turn over until mid-trial, and (3) refusing to suppress evidence from a search of his home and storage lockers. We affirm.
I. BACKGROUND
Tomkins began sending his threatening letters in 2005, demanding that his targets purchase sufficient shares of certain stock he owned to drive up the price. If his demands were not met, Tomkins threatened that something “very tragic” would happen to his victims’ loved ones, cautioning that “it is so easy to hurt somebody it is almost scary” and “it could be as simple as mailing a package just like The Unibomber [sic] use[d] to do.” As the return address on some of these letters, Tomkins included his victims’ home addresses. At least one letter also included a photograph of the victim’s home. He signed each letter as “The BISHOP.”
In January 2007, Tomkins mailed the packages containing his homemade devices. One went to Kansas City; the other went to Colorado but was rerouted to Chicago. Letters in each package stated “BANG!! YOU’RE DEAD” and warned: “The only reason you are still alive is because I did not attach one wire. If you do not believe me then go ahead and touch that red wire to the top of the battery pack. There is enough gunpowder and steel shot in that tube to kill anyone in a ten foot radius when it goes off.” Investment firm employees opened the Chicago package and contacted the police, and Officer Danny McGuire, an explosives specialist with the Chicago Police Bomb Squad, came to the office, cut the wires connecting the batteries to the pipe, and recovered the device for further analysis.
In the following months, investigators identified Tomkins as a suspect using purchasing records for the stocks referred to in his letters, and postal inspectors obtained search warrants for his home and storage lockers. The searches revealed two additional pipe bombs (one ready to be mailed), drafts of the threat letters, bomb-making materials, information about Tom-kins’s targets and their residences, and financial records related to the stocks mentioned in his threats. Tomkins was then arrested and charged with ten counts of mailing threatening communications, two counts of illegally possessing firearms (with each count corresponding to one of the two mailed pipe bombs), and one count of using a firearm in connection with a crime of violence. The final charge, Count *34213, related to the Chicago device alone; the government did not charge Tomkins with violating § 924(c)(1)(A) in regard to the Kansas City device, even though it was the subject of one of the § 5861(d) charges.
The case was heavily litigated leading up to trial, with one point of contention being Tomkins’s effort to suppress the evidence from the searches. The district court acknowledged that the warrants had failed to impose a time limit for financial records to be seized and concluded that seizure of a filing cabinet containing documents related to Tomkins’s role as treasurer of his local union exceeded the scope of the warrant. The court reasoned, however, that the lack of particularity in some parts of the warrants did not turn them into impermissible general warrants because attachments to the warrants contained detailed categories of evidence and the warrants were supported by thorough affidavits. Moreover, the court concluded that the good-faith exception permitted seizure of the items at issue.
Another pretrial concern was Tomkins’s request to represent himself. In 2010, the court granted that request and allowed Tomkins to proceed pro se, with stand-by counsel, during pretrial litigation and trial.
Tomkins’s trial began on April 23, 2012, five years after his arrest. The government’s case-in-chief took up the first week and a half of trial and included nearly forty witnesses, most of whom were either employees of investment firms that had received Tomkins’s mailings or government investigators who had a role in tracking him down. The government also introduced the threatening letters, handwriting samples, the draft letters and devices recovered from Tomkins’s property, and photographs of his devices. All of the photographic evidence (save for the x-ray discussed later) showed the devices after investigators had broken open the pipes by blasting them with a water cannon, rendering them safe and revealing the gunpowder and lead shot inside.
Other government witnesses testified about the design of Tomkins’s devices. Raymond Voorhees, an explosives expert, determined that the devices had the components of improvised explosive devices but that he could not be certain whether the devices would have exploded if the loose wires were attached because he had examined the devices after they had been rendered safe. In his opinion, the Chicago device was not functional without the wire attached. Officer McGuire of the Chicago Police Bomb Squad also testified that the Chicago device would not function with the wire left unattached. But another explosives expert, John Winslow, testified that the devices would have functioned if the unattached wire had made contact with the positive terminal of the battery in the package. Even if the wire never connected to the battery, he added, the devices could have ignited due to physical shock, friction, heat, or static electricity — even possibly as a result of being improperly handled during shipping. He described each device as a “dangerous weapon” and “explosive bomb.”
During Officer McGuire’s testimony, he mentioned that he had taken an x-ray of the Chicago device before the pipe had been broken open. The government then displayed a copy of the x-ray (shown below), on a monitor in front of McGuire. Tomkins objected that this was the first time he had seen the exhibit. After a side bar discussion, the district court allowed McGuire to discuss having taken the x-ray but forbade admission of the x-ray at that timé. The government later conceded that it could not find the x-ray in the exhibits given to Tomkins, and the court prohibited the government from introducing the x-ray *343during its case-in-chief. The court warned Tomkins, however, that the x-ray might come in as rebuttal evidence.
On the morning of May 2, 2012, the seventh day of trial, the government moved before resting its case to bar Tom-kins from testifying about his subjective intent in creating his devices. The government noted the similarity of this case to United States v. Johnson, 152 F.3d 618 (7th Cir.1998), which affirmed the preclusion of evidence that a defendant intended pipe bombs he created to serve only as hoaxes when the devices lacked any legitimate social purpose. Tomkins objected that the government was raising the issue “at the last-minute” and argued that he could show that his devices were useful for purposes other than as weapons. Stand-by counsel pointed to United States v. Fleischli 305 F.3d 643, 656-57 (7th Cir. 2002), in which a defendant was allowed to present the defense that his devices were fireworks rather than destructive devices.
After the government rested, but before Tomkins called any witnesses, the district court informed the parties of its conclusion that “Johnson is controlling, Fleischli is distinguishable, and therefore, the ruling in this case will be that subjective evidence of intent is irrelevant and therefore inadmissible.” The court later issued a written decision emphasizing that Johnson is squarely on point.
Tomkins then presented three character witnesses before taking the stand himself. During his testimony, Tomkins admitted to creating and sending the threatening letters and packages containing explosive materials but testified that his devices were not designed to explode. He asserted that “certain design features” in each device “made them nondestructive devices,” including that there was a gap between the batteries in the devices and that in some of his devices — he could not remember which ones — the lead shot “was loaded into the end where the electric match was and the gunpowder was clear on the other end.” Additionally, he testified that the unattached wire was fully insulated so that it could not connect with the battery and that he tested each device with a voltmeter “to make sure that no electricity was flow*344ing through them.” The prosecutor objected to this testimony, citing Johnson, but the district court overruled the objection.
After Tomkins’s testimony, the government moved to introduce the x-ray as rebuttal to Tomkins’s statements about separating the gunpowder from the igniter. The district court allowed admission of the x-ray, but instructed the government not to say that it was being admitted to show Tomkins lied, since the court could not say for certain whether that was a fair characterization of his testimony. The court agreed, however, that the government could characterize the testimony how it wished during closing arguments. Indeed, during closing, the government argued that Tomkins “was not credible” in his testimony about separating the gunpowder from the igniter — an assertion that the government called “ridiculous” given the relative sizes of the two substances in the pipe and the fact that the packages had traveled through the mail.
In instructing the jury, the district court explained that the definition of “ ‘firearm’ includes any destructive device,” and that the term “destructive device ... means any explosive bomb or any combination of parts either designed or intended for use in converting any device into a destructive device and from which a destructive device may be readily assembled.” The court refused to grant Tomkins’s request to instruct the jury that “ ‘destructive device’ shall not include any device which is neither designed nor redesigned for use as a weapon.” The court reasoned that this phrase was just “a flip side” of the included definition for “destructive device,” and that the added instruction was unnecessary because, under Johnson, the device was “a weapon as a matter of law.”
After being found guilty, Tomkins moved for a mistrial and a new trial, raising concerns about the delayed production of the x-ray and the jury instructions that were based on Johnson. The district court denied both motions. The court acknowledged that the government admitted to learning of the x-ray on April 9, 2012, fourteen days before trial, and obtaining a copy on April 17, six days before trial. But the court noted that Tomkins was warned that the x-ray could come in as rebuttal evidence and had five days after learning about the x-ray to prepare his defense. Therefore, the court reasoned, Tomkins’s defense was not prejudiced by the court’s ruling on the x-ray. Moreover, the court observed that, even apart from the x-ray, the government “presented overwhelming evidence regarding the configuration of the devices in this case.” As for the jury instructions, the court noted that, although Tomkins argued that Johnson was wrongly decided, it remains binding precedent.
On May 21, 2013, a year after the trial, the district court sentenced Tomkins to a total prison term of 37 years, including concurrent terms of 7 years on Counts 1 through 6 and 8 through 12,1 plus a mandatory minimum of 30 years for using a firearm in relation to a crime of violence. That minimum applies only if the firearm at issue was a “destructive device.” See 18 U.S.C. § 924(c)(1)(B)(ii). The court rejected an enhancement for obstruction of justice on the basis of Tomkins lying at trial because, in the court’s view, Tomkins deserved “the benefit of the doubt” because it was not certain that he “told a bald*345faced lie about the separation between explosive powder and the engine.”
II. DISCUSSION
A. EVIDENCE OF SUBJECTIVE INTENT
On appeal, Tomkins first argues that, in barring his testimony that the devices were hoaxes, the district court applied Johnson in an overly broad way that unfairly prejudiced his defense. He emphasizes that the government’s witnesses were permitted to testify about how his devices were designed and intended to function, and contends that this uneven treatment violates principles of due process.
The definition of “destructive device,” which is essentially identical in 18 U.S.C. § 921(a)(4) and 26 U.S.C. § 5845(f), includes “any explosive ... bomb,” 18 U.S.C. § 921(a)(4)(A)(i); 26 U.S.C. § 5845(f)(1)(A), and “any combination of parts either designed or intended for use in converting any device into a destructive device ... and from which a destructive device may be readily assembled,” 18 U.S.C. § 921(a)(4)(C); 26 U.S.C. § 5845(f)(3). Based on this language, we held in Johnson that the analysis whether a certain apparatus is a “destructive device” must look first at the objective design of the device:
If the objective design of the device or component parts indicates that the object may only be used as a weapon, i.e., for no legitimate social or commercial purpose, then the inquiry is at an end and subjective intent is not relevant. However, if the objective design inquiry is not dispositive because the assembled device or unassembled parts may form an object with both a legitimate and an illegitimate use, then subjective intent is an appropriate consideration in determining whether the device or parts at issue constitute a destructive device....
152 F.3d at 628. We then applied this analysis to uphold exclusion of evidence that Johnson intended to make hoax devices. Id. As opposed to a device like a firecracker, we explained, Johnson’s devices had “all of the properties of a destructive device, including shrapnel,” indicating “that they were useful only as weapons.” Id. at 627-28; see also United States v. Saunders, 166 F.3d 907, 914 (7th Cir.1999) (“If the objective design of the device indicates that the object serves no legitimate social or commercial purpose, subjective intent is not relevant to the analysis.”). Later, in Fleischli, we approved a jury instruction following the Johnson analysis but noted that it still allowed the defendant “to proceed with his defense that the objects were actually fireworks, not destructive devices.” 305 F.3d at 656-57.
This case is on all fours with Johnson. Unlike in Fleischli, Tomkins did not seek to present evidence that his devices were intended for a benign purpose; rather, he sought to present the same “hoax” defense rejected in Johnson. Additionally, as in Johnson, the lead shot in the devices shows that they were not useful for any legitimate purpose other than use as a weapon. Tomkins argues that the individual components of his devices had legitimate social purposes, but this argument is undermined by the “combination of parts” language in the relevant statutes. See Saunders, 166 F.3d at 914-15 (upholding “destructive device” conviction when detective testified that “all of the components necessary to make the device explode were present in the device”); United States v. Copus, 93 F.3d 269, 273 (7th Cir.1996) (upholding “destructive device” conviction for homemade apparatus made of sealed metal casing containing explosive powder and fuse).
*346At least two circuits have expressly adopted the same approach to evidence of subjective intent in this context as we did in Johnson. See United States v. Urban, 140 F.3d 229, 234 (3d Cir.1998) (holding that intent is irrelevant when it is clear that components, when combined, would create destructive device); United States v. Posnjak, 457 F.2d 1110, 1119 (2d Cir. 1972) (same); see also United States v. Lussier, 128 F.3d 1312, 1317 (9th Cir.1997) (concluding that “no auxiliary evidence concerning [defendant’s intent to use [his devices] as weapons was required” when “their nature and characteristics convincingly demonstrated that they were designed as weapons”). Tomkins does not point to any circuit that disagrees.
Tomkins seeks to distinguish Johnson on the grounds that, unlike here, it did not involve a statutory minimum prison sentence that hinged on the definition of “destructive device.” The minimum is critical, Tomkins argues, because under Alleyne v. United States, — U.S. -, 133 S.Ct. 2151, 2155, 186 L.Ed.2d 314 (2013), “any fact that increases the mandatory minimum is an ‘element’ that must be submitted to the jury.” See United States v. Zuniga, 767 F.3d 712, 718 (7th Cir. 2014), cert. denied, — U.S. -, 135 S.Ct. 1018, 190 L.Ed.2d 886 (2015). Tomkins thus argues that the court “usurped the jury’s role in declaring the devices firearms and destructive devices as a matter of law.” He contends that, as part of the jury instructions, the jury should have been required to specifically find that his devices were “destructive devices” versus some other type of firearm. Cf. United States v. Jones, 763 F.3d 777, 817 (7th Cir.2014) (finding Alleyne error when jury did not make specific finding on drug quantities that increased minimum), vacated on other grounds by United States v. Drake, 774 F.3d 1104 (7th Cir.2014).
But a specific jury instruction was not necessary here because the instructions made clear that, to find Tomkins guilty, the jurors needed to conclude that the device he constructed was a destructive device. Not only did the instructions give no other definition of “firearm,” they also advised the jury that the charge for Count 13 was “knowingly using and carrying a destructive device during and in relation to, and possession of a destructive device in furtherance of, a crime of violence” (emphasis added). Additionally, the instructions for the two counts of illegally possessing a firearm directed that Tomkins, to be guilty, must have known “that the firearm possessed the characteristics that qualified it as a destructive device.” Finally, a copy of the indictment provided to the jury lists the charge as possession of “a firearm, namely, a destructive device as defined at Title 18, United States Code, Sections 921(a)(4)” (emphasis added). Thus, even without a specific jury finding, in finding Tomkins guilty, we are convinced that the jury necessarily found that his devices constituted destructive devices.
Tomkins also argues that Johnson was wrongfully decided “because of the impact of the exclusion of state of mind evidence.” He emphasizes that, after we decided Johnson, the Supreme Court in Apprendi v. New Jersey, 530 U.S. 466, 493, 120 S.Ct. 2348, 147 L.Ed.2d 435 (2000), recognized that a “defendant’s intent in committing a crime is perhaps as close as one might hope to come to a core criminal offense ‘element.’ ” But Apprendi addressed a statute that on its face required examination of the defendant’s state of mind. Id. at 492-93, 120 S.Ct. 2348. In contrast, as discussed in Johnson, 152 F.3d at 627-28, the definition of “destructive device” in the statutes at issue here only requires examination of a defendant’s mental state in certain circumstances. Ignoring the sec*347ond subpart of the definition, which is irrelevant to this case, the statutes establish three separate ways in which a device may be a destructive device: if it is (1) an explosive, like a bomb, (2) a combination of parts designed for use in converting a device into a destructive device, or (3) a combination of parts intended for use in converting a device into a destructive device. As a result of the district court’s ruling that Tomkins’s intent was irrelevant because his device was a “weapon as a matter, of law,” the last definition was off the table. For that reason, Apprendi does not undermine the jury’s verdict.
That being said, once the district court decided that Tomkins’s subjective intent was irrelevant to the jury’s analysis, the court should not have asked the jury to determine whether he “intended” to create a bomb. At least one district court, citing Johnson and Saunders, has addressed this situation and concluded that it constitutes error not to drop the term “intended” from the “combination of parts” provision when it is clear that the device at issue does not have any legitimate or socially beneficial purpose. United States v. Sheehan, No. 13-er-0186, 2014 WL 3490323, at *20-22 (E.D.N.Y. July 11, 2014) (holding that the error was ultimately harmless). We agree. There are some situations where it is appropriate to provide the jury with alternative definitions of “destructive device,” as recognized in Johnson, which held that the district court “acted well within its discretion” by instructing the jury on a theory of liability under both subsections 1 and 3 of § 5845(f). 152 F.3d at 628. But when, as here, a defendant is excluded from presenting evidence of subjective intent, courts should refrain from issuing a “intended for use” jury instruction, particularly when a “designed for use” instruction will suffice.
We agree with the government, however, that any error with the jury instructions’ definition of “destructive device” was harmless because there was ample evidence to prove beyond a reasonable doubt that the devices were destructive devices. Tomkins argues that the errors here were structural in nature and thus are not subject to harmless error review, relying for support on Frost v. Van Boening, 757 F.3d 910, 915 (9th Cir.2014). But the Supreme Court recently reversed that decision, explaining that harmless error review is only inapplicable to the rare type of error that infects the entire trial process and renders the trial fundamentally unfair. Glebe v. Frost, — U.S. -, 135 S.Ct. 429, 430-31, 190 L.Ed.2d 317 (2014); see Neder v. United States, 527 U.S. 1, 7, 119 S.Ct. 1827, 144 L.Ed.2d 35 (1999) (“[W]e have recognized a limited class of fundamental constitutional errors that defy analysis by ‘harmless error’ standards.”) (internal quotations omitted). We are not convinced that the district court’s error in not removing the term “intended” from the jury instructions’ definition of destructive device rendered Tomkins’s trial fundamentally unfair. Moreover, in light of the unanimous testimony that Tomkins’s device contained the components of an explosive bomb, we are persuaded that the error did not affect the ultimate outcome of trial.
B. INTRODUCTION OF X-RAY
Tomkins next argues that the government violated Federal Rule of Criminal Procedure 16(a) through its delay in turning over the x-ray that Officer McGuire took of the Chicago device before it had been rendered safe, and that the district court should have granted a mistrial on that basis. He emphasizes that the government admits that it received the x-ray the day before the final pretrial conference and yet did not mention it during the *348conference or turn it over at the start of the trial.2
Under Rule 16, the government must permit the defendant to inspect or copy-items within the government’s control that (1) are material to preparing the defense, (2) the government intends to use in its case-in-chief, or (3) were obtained from or belong to the defendant. Fed. R.Crim.P. 16(a)(1)(E); see United States v. Baker, 453 F.3d 419, 424 (7th Cir.2006). The government offers a meager defense of its compliance with this rule, contending that the record does not establish whether Officer McGuire was a member of the prosecution team and that Rule 16 does not impose a duty on federal prosecutors to obtain documents in possession of state police. See United States v. Hamilton, 107 F.3d 499, 509 n. 5 (7th Cir.1997). But as Tomkins notes, even if McGuire was not part of the prosecution team, that does not excuse the government’s failure to provide the x-ray at the pretrial conference or start of trial.
But even if the government did not comply with Rule 16, that does not mean that the district court erred in admitting the x-ray as rebuttal evidence. Trial courts have “discretion to fashion a remedy for the government’s noncompliance with Rule 16, and this court will not disturb the district court’s decision absent a showing of abuse of discretion and prejudice.” United States v. Warren, 454 F.3d 752, 760 (7th Cir.2006) (citations omitted); see United States v. Breland, 356 F.3d 787, 797 (7th Cir.2004). Tomkins has not made that showing here. Even though Tomkins was pro se, the court’s refusal to allow the x-ray during the government’s case-in-chief and clear warning that the x-ray could come in as rebuttal evidence provided a reasonable remedy for any Rule 16 violation. The court’s rulings allowed Tomkins adequate opportunity to adjust his defense, especially since Tomkins must have known how the contents of the device had been arranged. Although Tomkins sought a new trial, that remedy is appropriate “only when ‘all other, less drastic remedies are inadequate.’ ” Warren, 454 F.3d at 760 (quoting United States v. De La Rosa, 196 F.3d 712, 715 (7th Cir.1999)). Tomkins has failed to convince us that the less drastic remedy adopted by the district court was unreasonable.
Moreover, the district court’s exercise of discretion in regard to Rule 16 is reviewed for harmless error, see United States v. Hurt, 574 F.3d 439, 442 (7th Cir.2009), and we are not persuaded that the alleged error here affected the outcome of trial. In particular, as the district court noted, the government introduced photographic evidence showing that the relative sizes of the lead pellets and gunpowder would have made it highly unlikely that they did not mix together and contact the igniter. Further, explosive experts confirmed that Tomkins’s devices had all the elements of explosive bombs, and expert Winslow maintained that the devices could have exploded if mishandled during shipping. In light of this testimony, we are not convinced that the x-ray made a critical difference in the jury’s decision.
C. DENIAL OF MOTION TO SUPPRESS
Finally, Tomkins argues that search warrants allowing seizure of materials from his home and storage lockers were not adequately constrained in time *349and scope. He particularly objects to the warrants’ permission to seize “all financial records and documents,” which resulted in seizure of a file cabinet and box containing numerous records related to his union, where he was treasurer. He argues that the government should have narrowed the warrants by providing information in the affidavits about his union activities.
We are not persuaded by Tom-kins’s arguments. First, we agree with the government that the authorization to seize financial records was not unreasonable given that Tomkins’s threats related to stock manipulation, and his role as treasurer of his union made the union records potentially relevant to the financial aspect of his scheme. Tomkins argues that there was no evidence of a connection between his crimes and his union activity, and that the warrants were flawed by omission of that information. But although the omission of certain information, such as evidence discrediting an informant, may result in a deficient warrant by undermining probable cause, e.g., United States v. Glover, 755 F.3d 811, 818 (7th Cir.2014), nothing about Tomkins’s union activities would have undermined the probable cause to seize his financial records given the financial aspect of his crimes, see United States v. Clark, 668 F.3d 934, 942 (7th Cir.2012) (holding that omission of information about computer registration did not undermine probable cause for computer search).
Moreover, even if the warrants were deficient, we agree with the district court that the good-faith exception applies. The good-faith exception precludes application of the exclusionary rule when law enforcement reasonably and in good faith believed that a search was lawful. United States v. Leon, 468 U.S. 897, 922, 104 S.Ct. 3405, 82 L.Ed.2d 677 (1984); United States v. Gutierrez, 760 F.3d 750, 753 (7th Cir.), cert. denied, — U.S. -, 135 S.Ct. 735, 190 L.Ed.2d 459 (2014). Under that analysis, it is prima facie evidence of good faith that the government obtained a search warrant before searching Tomkins’s property. See United States v. Garcia, 528 F.3d 481, 487 (7th Cir.2008). Tomkins asserts that the warrants were plainly deficient, but the warrants here contained enough detail — discussing the series of threatening letters, explosive devices, and apparent motivation of stock manipulation — that an officer could reasonably rely on them. Tomkins thus fails to overcome the presumption of good faith that attached to the officers’ reliance on an issued warrant. See United States v. Miller, 673 F.3d 688, 694 (7th Cir.2012); Clark, 668 F.3d at 941-42.
Accordingly, the judgment of the district court is AFFIRMED. .