Headphone enthusiasts accuse the KZ headphone company of dishonest marketing. Measurements indicate that multi-hybrid and tribrid drivers are inaudible in many models.
A new scandal is brewing in the IEM world. This time, rather than questionable driver material claims, headphone enthusiasts are tearing apart KZ’s multi-driver IEMs and disconnecting the balanced armature (BA) and magnetostatic drivers, and measuring the IEM’s frequency response.
To their surprise, they are measuring little to no difference between ONLY the dynamic driver connected (BA and magnetostatic drivers disabled) and with all drivers working.
The initial interpretation was that the smaller drivers, supposedly tasked with midrange and treble frequencies, were fake or non-functional. Additional enthusiasts have jumped in, started tearing apart their IEMs, and refined the claims that the additional drivers are real and connected, but simply dampened in volume via resistors on the crossover board to the point of inaudibility.
We all know that more isn’t necessarily better. And perhaps more isn’t actually more at all.
KZ Company Overview
KZ Acoustics was founded in 2008 and generated a huge amount of industry buzz by creating products that redefined the ratio of price to performance. There is perpetual hype over new KZ products as although seldom considered the best, they are typically very competitive at their low price point.
As a result, KZ is a very popular brand, and is the introduction for many people to the world of Chi-Fi IEMs.
KZ is known for releasing many different models, so it gets convoluted when trying to figure out their hierarchy of IEMs. To further complicate things, KZ also releases IEMs under other brand names as well. Clear Concept Audio (CCA) and Tripowin are sister brands and have released very similar IEMs (sharing components, technology, design, ear tips, etc. with KZ models).
KZ x Crinacle CRN
KZ has received a lot of attention lately for its collaboration with headphone reviewer/influencer Crinacle (Corin Ako). Famous in headphone circles for his comprehensive comparative IEM frequency graph database at In-Ear Fidelity, he’s recently made a successful dive into the more lucrative waters of YouTube. His hard work has been recognized by more than just enthusiasts, as more and more IEM companies are tapping him to help tune new IEM revisions or models.
What makes the KZ CRN especially interesting is that they are marketed as true tribrid IEMs. That is, they contain a 6.8mm low voltage electrostatic unit, a 10mm double magnetic dynamic driver, and a 30095 balanced armature driver per side. All directed by a fixed crossover board to split the frequencies to the appropriate driver. It’s a first for KZ, and a first at this price point.
The CRN also has the honor of being the least expensive tribrid on the market – by a wide margin. I reviewed the CRN and mused how I can’t understand how KZ can do all this for less than USD$40. I find the CRN to sound quite decent for their low price, with a tamer and more balanced tuning than is the typical deep v-shaped KZ house sound.
On March 5, 2022, an enthusiast who goes by the moniker Delta Fyre, known for his IEM teardowns, posted the following on Facebook.
“Crinacle CRN- did KZ just jam drivers here so they could call it a tribrid?
I opened my pair of CRN (Zex Pro) and wanted to see if the EST and BA drivers were real, i carefully disconnected the EST and measured the sound after, there isn’t a difference. Ok, maybe the BA is real at least, and we can lower the 9k peak.
Well no appearantly [sic] the BA isn’t real too..
i unsoldered the BA expecting to see the treble fall, but even then the sound remained the EXACT SAME!
i even measured with just the BA and EST connected and there is no sound output from the CRN. i can get the drivers to play a very faint hiss but if i play music through them it makes virtually no sound.
The crinacle CRN Zex Pro turns out to be a 1DD earphone. It even graphs the same as QKZ VK4 if you notice.”
This, and his subsequent post on Reddit, set off a firestorm of outrage and concern.
Other KZ Models
Jeffrey Fries and Delta Fyre have individually tested several KZ IEMs, including the CA16 Pro, CA24, DQ6, DQ6S, NRA, ZES, ZEX, and ZEX Pro/CRN, and have concluded that the issue of inaudible or redundant drivers is widespread.
“The KZ ZEX “EST” driver – Is it Fake??
So what happens if i unsolder the EST driver and measure the frequency response after?
It turns out, disconnecting the EST driver BARELY changes the sound, as a matter of fact, the treble response actually INCREASES after disconnecting the EST driver!
This tells me that the EST driver possibly has inverted polarity, and after testing It is like this on both the left and right sides.
It seems the EST driver is not “fake”, it affects the frequency range >4Khz, but for some reason it is working in reverse to the DD. (it also doesn’t do very much) – Delta Fyre
KZ x HBB DQ6S
Another YouTube reviewer Hawaii Bad Boy (HBB) of Bad Guy Good Audio Reviews (BGGAR) has also recently collaborated with KZ to produce the DQ6S. The DQ6S is a 3 dynamic driver array design, with a 10mm dual magnetic dynamic unit and two 6mm single magnetic dynamic drivers.
Delta Fyre reported “HBB x KZ DQ6S – Triple drivers, but only one of them is actually working? i unsoldered the connections going to the tweeters one by one expecting to see a drop in the treble, but the sound quality is EXACTLY the same after doing this. Is KZ putting decorative drivers in iems as a scam?? #ExposeKZ”
Who Does This Reflect Badly Upon?
Clearly, KZ is going to lose consumer trust. Many of their customers have joined the cry, outraged by an apparent lack of transparency and dishonest marketing claims.
Is it just as simple as “we should stop buying from companies that lie to us?” as one commenter proposed on FB?
Of course, regardless of the function of the drivers inside, this doesn’t change the overall sound of the model. Others have proposed that KZ would be ahead of the game if they marketed their products as a single dynamic driver, saving the cost of the superfluous extra drivers. KZ has built much of its brand and reputation on bargain-priced multi-driver IEMs, often using driver count and type as a differentiator between products and reason for higher price tags.
Although KZ has yet to reply directly to the claims, another Chinese IEM manufacturer, DUNU, has released a response to distance their products from this situation.
Subsequently, some suspicious responses on Facebook defending KZ (now deleted) are starting to make the rounds on social media. The presumably fake accounts and humorously awkward comments are being interpreted in a negative light by many.
The celebrity tuners
Crinacle has directly responded to the CRN concerns on Reddit and in a post on his website, while HBB released a video as his response after unsuccessfully waiting for KZ to make an official response.
“That I didn’t discover that these are not actually doing anything, I’ll say sorry for, but I didn’t imagine I’m going to have to check each driver independently and individually…
This set should be taken down and not for sale. End of story… this is false advertising is what this is… fraud… to say that something is one thing and to actually be knowingly delivering another thing… and I’m a part of that.” – HBB
“…tuning the CRN was a very FR-first approach and the project was greenlit once the target FR (or close enough) was achieved.
KZ didn’t provide any crossover FR breakdowns (but to be fair, I didn’t request for any either) since the primary goal had been achieved, so I wasn’t aware of [sic] that the magnetostats and BAs are duds. But now we know, seems like KZ effectively wasted production costs on some extra drivers when that single dynamic driver was enough. Maybe could’ve slashed the MSRP a little with a swap, but hindsight and all that.
Anyways the problem is bigger than the CRN and seems like KZ is inflating driver count with dud drivers, perhaps for marketing clout. Which would also be a very convenient explanation as to why groups of KZ models seem to sound the same despite having different drivers and whatnot.” – Crinacle
This leads to questions as to what the responsibility is of the tuner before they give their name to a product. There is a financial incentive for participating with a company to tune a pair of IEMs, but is the tuner responsible for verifying the physical construction of the product beyond taking the manufacturer at their word?
What are the responsibilities of the celebrity tuner? To only provide a target frequency graph, and the manufacturer to make whatever changes and compromises to meet that target at a certain price point?
“But still, I know I’m not totally blameless.
I acknowledge that my branding and endorsement can almost be seen as a “mark of excellence” for many, and issues like these only serve to disappoint the ones who have put faith in my work. It was my responsibility to ensure that issues like this should not happen especially with my name riding on it, and on this front I have failed.” – Crinacle
Regardless, there is little question that Crinacle and HBB are unhappy with the recent revelations. It’s never fun to feel deceived, and they likely fear negative impacts to their personal brands as a result.
Reviewers of these models may feel a certain amount of egg on their faces as well. The controversial concepts of distinctive ‘BA timbre’ and incoherency that some claim to hear (while others dispute) in hybrid IEMs come under question if the Balanced Armature drivers in these hybrid models are inaudible.
One reviewer, user name Precogvision, was called out on Reddit for his review of the CRN and use of terms like “characteristically gritty decay” to describe the sound of the BA driver. They handled the criticism gracefully and their responses likely echo many reviewers’ feelings (and are similar to my own).
“…While I’d like to say I have golden ears and that I’m able to attest to things with 100% certainty, the reality is that I simply don’t and I can’t. Ambiguity is inherent given the subjective nature of the hobby, and I think that leaving room for supposition is important for getting a feel for how a reviewer might be biased.
Of course, that doesn’t mean that we can’t try to mitigate some of it. I agree with your comment that giving impressions – at least sound related ones – regardless of the technology used in the system is a noble aim, and I appreciate the high bar you hold for me even if I don’t think I’m necessarily deserving of it!” – Precogvision
Is there something more going on here that we simply don’t understand? Is there some interplay between the low output BA drivers and the dynamic drivers that is adding to the sound, even if it isn’t immediately obvious from a frequency response graph? More investigation is warranted.
Before we vilify a company, it behooves us to progress slowly and carefully and to make sure we all have the facts before jumping to conclusions. Certainly, the initial impression is damning, and in the end, we may find that marketing has won over transparency or integrity. Or, perhaps further information will be uncovered that better explains things. We shall see.
Is KZ’s time as a popular entry-level IEM choice over? Or will they weather this storm, either with continued silence, clear explanation, or a revision of policies?
This isn’t the last we’ve heard of this issue, and stay tuned as new developments are uncovered. Thanks to all the dedicated enthusiasts out there for keeping everyone honest and informed!
March 8, 2022
KZ has posted an official response to the issue, apologizing for production issues with the DQ6S but apparently discounting all concerns with BA and EST drivers in other models.
Click the link to read in entirety. Grammatical issues are as per the original document.
“We have noticed recently negative public opinion of our IEM models, it’s a heated discussion and cause some of customers are concerns about our products. We deeply apologize for this.
Due to our design defects of model DQ6S, there will be a percentage of potential risk that it is blocked by glue during the mass production process… we will improve our follow-up product quality inspection process…
We need to explain to everyone that all of our products with BA, EST do work, and this can be proved by any testing. Even if BA and EST with low sound pressure, it does not mean they are useless… Meanwhile, there is no academic certification that can prove whose sounds quality standard can be judged completely good or not only by relying on the graph…
We recommend that you can test our IEM details… We are very confident in our patented dynamic technical technology, and our original design intention is keen to improve the sound style timbre of hybrid products with BA…
However, we are not good at brand promotion and so on, that’s the reason why we keep silent and take action slowly to respond to the public’s doubt’s about the product…
All in all, we apologize for the inconvenience caused by the recent product problems, and we will definitely figure out them as soon as possible…” – KZ
March 11, 2022
“Firstly, the 2 drivers does produce sound, secondly the glue that was shown on the photos where the glue is blocking is at the wrong position… it is not that it does not produce sound but during our quality check it might be a wiring issue…
The method of removing the main DD and testing the 2 smaller DD individually is not correct. Reason being they are sharing the same cross over and if the main DD is removed the circuit becomes a incomplete circuit. The 2 smaller DD will thus unable to produce sound due to the influx of impedance thus unable to produce sound through the nozzle. So the proper method to test the 2 smaller DD is to disconnect the 2 smaller DD from the main DD to test if it produce sound…
In fact we do admit that our quality check does have some issues…
Our concept is that the bigger the size of the DD the higher the frequency so we want the smaller DD to provide the timbre and the tone…
… we acknowledge that the red wire and the blue wire has been connected wrongly during DQ6S production…
To us, the purpose of the EST is to provide the tone and timbre and not to increase the volume so high that it is audible…
With regards to customers who feel that what they bought is not what was advertised, first we will explain clearly… We will accept return or exchange or refund of the product and to compensate the customer.
…this incident is a misunderstanding and it is not done intentionally…our product specs and details are correct.” – Mr. He – KZ representative
March 14, 2022
KZ has released a new statement via Facebook channels to clarify the return process for dissatisfied customers. Somewhat contrary to the statement made in the video above, returns are limited to only the DQ6S model purchased between February 20 to March 8, 2022. Customers are not required to send back the defective IEMs.
At time of writing, KZ has not clarified how they should be contacted to initiate a return, or what retail locations are eligible.
“In addition, due to the large number of people discussing this incident, it has led to a lot of people deliberately making up news about product problems of our other models in order to fire it in. We hope you can look at this kind of news from a rational perspective.” – KZ Official
March 26, 2022
Delta Fyre has released a new Facebook post that appears to contradict his previous conclusions regarding the KZ drivers.
“Doing the same exact thing i did to KZ CRN, disconnecting the positive wire going to the EST (Sonion EST65DA01 Dual-Tweeter) and measuring the frequency response after shows… no changes? not even a SLIGHT drop to those treble peaks around 10k??Doing the same process to the BA, (Knowles 29689 Unit) disconnecting the BA results in an INCREASE to the frequencies 2K and beyond..
The first thing i did with these troubling results was share it with our friend at DUNU, and asked why the Sonion unit has seemingly negative SPL. They use the same exact Sonion unit in their EST112, and explained that these EST units are simply very low-output drivers in basically EVERY implementation. The sound output of the Sonion EST is about 90dB with proper measurements. The DD in the BAX has an output of 118dB, which is over 100x LOUDER than the EST. These EST drivers are for simply altering timbre. It’s for giving an iem that extremely subtle crispiness. And if you’ve ever listened, you would get the impression that it’s working.
So why then are my meaurements [sic] so DECEIVING? it’s quite simple really, these frequency graphs are a TERRIBLE, horrible way of representing sums…
All this time I’m using this IEC mic like it’s the scroll of truth, forgetting just how MISLEADING statistics can actually be.” – Delta Fyre