Search This Blog

Tuesday, April 2, 2013

Etymotic Research HF-3: An ER-4 equivalent?

Disclaimer: This is Inks' pair!

Etymotic Research has been best known for their extremely strict quality control, and that is why their flagship IEM, ER-4 is pair-matched by their engineers. Although the official tolerance limit is 2 dB, almost all of ER-4's interchannel deviation is usually within ~1dB from 20 Hz to 10 kHz. Such accuracy is absolutely critical, especially when the IEM is used for professional/research purposes.

However, that is also the reason why ER-4 is quite steep in price, keeping regular users unapproachable to Etymotic Research's great quality & fidelity. In order to mediate the situation, in January 2002, the manufacturer re-branded Star Micronic PH-001A, and sold it as ER-6. ER-6 was as accurate as ER-4S, and was very affordable too with an MSRP of $140.

And in 2008, as a successor to ER-6, the HF series was introduced. Because its design and official measurement data were almost identical to ER-4P, it was considered to be ER-4P equivalent. While some said HF's core element, a balanced armature driver, is identical to ER-4's Knowles ED-9689, others argued that it was actually ED-29689.

And recently, some of my fellow Twitterian named @Fontek_Research reported that Knowles ED-29689 measured very different to what's installed in ER-4. Can this IEM possibly be an economical version of ER-4?






Note: measured with a pair of ER38-18 sleeves
PRO:  Great aesthetics, improved microphonics. Frequency response is not as extended as ER-4 or ER-6, but very linear overall. EXTREMELY well-matched drivers, possibly the BEST matching I've measured so far: the interchannel deviation does not exceed more than 0.4 dB from 20 Hz to 10 kHz, which is better than that of my ER-4PT and 4B.

CON: Lacks a good amount of sub-bass, but the distortion figure is definitely improved from ER-4 & ER-6, which makes HF series an Etymotic Research product with the least amount of distortion.

ON SECOND THOUGHT #1: Just like ER-4, insertion depth plays a vital role. Even a tiny displacement of 3 mm causes harshness & disconnection in high frequency bandwidth.

ON SECOND THOUGHT #2: Adding a serial resistance turns HF-3 close to ER-6, but not as accurate as ER-4S.

ON SECOND THOUGHT #3: Other factors are identical to those of ER-4, so please refer to ER-4B analysis for further details: 12, and 3

ON SECOND THOUGHT #4: Is HF series same as ER-4? They are very close to one another, but I don't honestly think they have same drivers within due to two factors: lack of sub-bass and improved distortion. Nonetheless, I can say HF series is a great IEM, possibly better than its successor, ER-6, with a serial resistance added.

23 comments:

  1. Couldn't the lack of sub-bass extension be due to the the housing?

    I think they use the same drivers but are not matched like the ER-4 so I'm wondering if you got a lucky hf3.

    Also, I can't tell a difference between my ER-4P and my hf3.

    Anyway, I have a question that has always bothered me, why do some graphs of the ER-4P show a sub-bass roll off and others don't? I've seen the roll off more apparent at high SPL levels.

    ReplyDelete
    Replies
    1. You can't utilize air compliance within the housing with Knowles ED-9689 & 29689, as the transducer is not vented at all. So the answer is no.

      Yup, I can't tell the difference between ER-4P and HF-3 either. That is why I think HF-3 is a better deal for common listeners! And the channel matching is extremely good..

      There's always a sub-bass roll off with ER-4 series, whatever the version it is. Even though the manufacturer says ER-4 is of Hi-Fidelity, as a matter of fact, it is not only missing a substantial amount of sub-bass response, but also the frequency response above 18 kHz.

      Delete
    2. Thanks for responding.

      I see your first point. However to determine whether the channel matching is really good, I think another hf series should be measured.

      I think the treble response of the ER-4S and even the ER-4P is great, it's not as extended the limit of our hearing range... then again recordings contain little information past 18kHz.

      On the bass roll off, I've always been confused by this. Looking at measurements from various websites, I can't seem to understand why it occurs:

      http://www.innerfidelity.com/images/ComplyTip_graph_EtyER4PLeft.jpg

      http://sonove.angry.jp/ER4P/ER4P_distortion.gif
      http://sonove.angry.jp/FitEar334/FR_FitEarToGO334.gif

      Delete
    3. Simply check out the y-axis scale on those graphs, you can magically make the slight attenuation in the sub-bass disappear by adjusting it. :p

      Delete
    4. -2dB at 20Hz. Negligible is an understatement ;)

      Delete
  2. The difference between the ER4 and the HF5 is not noticeable without some significant listening time with both. I'm not certain I hear a lot of frequency response difference, but I do hear a "coherence" difference -- placement of instruments, interplay of instruments makes the ER4 more "musical" to me.

    ReplyDelete
  3. Hello Rin, how do these compare to the LG Quadbeat or mh1c? I am enjoying your mods to the mh1c and wonder if these unknown IEMs compare.

    ReplyDelete
    Replies
    1. While HF-3 is leaned towards more critical listening, LG Quadbeat and MH1C are targeted for casual usage. I'll post up some data of LG Quadbeat very soon.

      Delete
  4. I believe that if you split apart an HF3 and an ER4P, you'll see the same Knowles ED-29689 driver inside. The "29689" model number is a publicly available number, and thus isn't a custom product made for a specific client. They will therefore be subject to the exact same QC standards.

    As far as I know, there is no "9689" model number, publicly available or to special order clients. Even if there were, it'd violate Knowles' naming scheme.

    What most likely sets the HF3 and ER4 apart are the relaxed QC tolerances for resistor capacitance figures, which could easily account for the minor loss of overall bandwidth. Whereas an ER4PT's 18 ohm resistors are tightly matched, an HF3's probably doesn't go through the same amount of scrutiny.

    ReplyDelete
    Replies
    1. In order to prove a claim, one needs a proof, and @Fontek_Research's discovery on the transducer unit of ER-4 was good enough to convince me. We'll simply need a better evidence than "AFAIK"

      Delete
    2. I understand about the need for quantifiable and qualifying data, but it doesn't make any rational sense for Etymotic to order another batch of similar, but different drivers with a miniscule performance variation just for the sake of product differentiation.

      I don't use Twitter, so I can't view Fontek_Research's tweets.

      If anything, they're attempting to leverage Knowles' production capacity of the ED driver. Globally, the ED driver is Knowles' best seller, so large order quantities would give Etymotic an economic advantage to streamline their HF2/3/5/ER4P/S/B to the same driver.

      The only reason why the ER6/6i originated was that Etymotic found Star Micronics to be a reliable partner in suppressing costs when BAs were very expensive still, around eight years ago. Once volume costs became viable with Knowles, I'm sure Etymotic switched back because of their long-standing relationship with Knowles, beginning with the KEMAR. Even the microphones used in their cables are sourced from Knowles. It makes sense for a company whose true interest is hearing research and protection to stick with a reliable partner for a long period.

      Delete
    3. The only possible way your argument would hold water is to open up ER4 & HF, and look inside. Since @Fontek_Research and I both did a precise impedance measurement, your counter argument shall only remain rhetorical.

      Delete
    4. This comment has been removed by the author.

      Delete
  5. Would utilizing red dampers give it a perceived 10 db bass boost? Or is it the green one? I can't tell with this graph http://4.bp.blogspot.com/-P8U6gTFbfZU/UC4CBO0aQ2I/AAAAAAAABFo/6Pn2vF0qSHc/s1600/3.png

    ReplyDelete
    Replies
    1. Green is the stock. Orange dampers will bring down the mid range about 10 dB.

      Delete
  6. Is green stock for the hf5 aswell? and orange brings it down 10 db compared to unfiltered or stock because it looks like the unfiltered. Sorry for the noob questions :p

    ReplyDelete
    Replies
    1. Sorry for causing confusion, should've organized the data more intuitively.. Green is hf5's stock too.

      BTW my favorite is red for ER-4P, and orange for ER-4B.

      Delete
    2. So orange brings the midrange down 10db right? and would that cause you to percieve a 10db bass increase from neutral?

      Delete
    3. That would likely be the right assumption.

      Delete
  7. Could you upload the insertion depth graph containing the entire dB range to 15k?

    ReplyDelete
    Replies
    1. Of course, Mark! No sweat.
      http://i.imgur.com/sub7zsN.jpg

      Delete
  8. 안녕하세요 산미천님 리뷰 잘 보고 있습니다.^^
    ER4S음색과 근접하려면 저항은 얼마나 추가하는게 적당할까요??
    33옴은 효과가 적을 것 같은데...궁금하네요

    ReplyDelete