Eddie Current Ultralinear 6L6

Discussion in 'Headphone Amplifiers and Combo (DAC/Amp) Units' started by purr1n, Aug 5, 2020.

  1. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    I was trying to decide where to stick this, whether it belonged in this thread (https://www.superbestaudiofriends.o...hd-n-or-why-both-tell-us-almost-nothing.9544/) or deserved it's own thread. I decided on the latter. I dislike reviewing gear that will otherwise never make it into production, but there are several reader curiosity angles here: this amp is one-off from Eddie Current that actually isn't a SET design - it's SE ultralinear; I wanted to analyze the distortion patterns of said ultralinear design; other than servicing existing amps in the field, EC is no more, unless Craig finds the energy and time to resurrect it and offer new designs.

    IMG_20200629_111116 (1).jpg

    The amp started off as a 445, but getting easily bored, Craig wanted to try something different. He rigged it up to use 6L6 power pentodes wired up for ultralinear operation. There are x2 pentodes per channel, so one may think this would be push-pull, but it isn't. It's single-ended all the way. The original OPTs were for single-ended designs anyway, so Craig had to keep this single ended. The outputs of the 6L6 are paralleled. This amp doesn't have any interstage coupling caps either. It uses interstage transformers.

    I took a set of steady state 1kHz sine wave measurements at various output levels from -20dBu to +10dbU and hope to compare them to a SET (SE triode) design in the very near future. One of the two 6D22S rectifiers died on my SET amp. It's going to take a while for the replacement to arrive. I'm curious to see how these amps will be different in terms of distortion measurements if any. It won't be apples to apples because the power levels will be different between the amps, but maybe we will find something interesting. For more proper apples to apples comparison, the pentodes on this could have been strapped as triode, but it shouldn't matter here much since the SET amp I will compare to is of the same topology (albeit with different interstage and output transformers). One thing to note is that ultralinear and triodes do sound different. Very different. Will the measurements look different? The discovery process of science is fun!

    Ultralinear is more powerful, punchy, with sharper and stronger attacks. Triode is softer, more rounded, but offers space, texture, and plankton. Ultralinear is clean, with straight lines, maybe even a bit simplified. Which one is more true? I'll be a weasel like the audio magazine reviewers and say that I do not know and that the listener must decide! My bias is obvious, but it's not nearly so black and white, it depends.

    EC Studio 45 (custom interstage and output transformers) at left and EC Ultralinear 6L6 at right.
    Both amps employ an OPT with a tertiary winding used for feedback and 5842Q drivers. Their topology is the same other than the 6L6s wired for ultralinear operation.
    IMG_20200805_142959.jpg

    BTW, that is a JAR HD600 there on top of the Verum and ZMF Verite Closed. @Hands: you owe me an explanation. The JAR HD600 proto you sent me did not sound like this!

    @ChaChaRealSmooth's JAR HD600 is fu@#^@# fantastically awesome! We also had a Focal Utopia around. The JAR HD600 and Verite Closed (and Open) worked really well with the EC Ultralinear 6L6 (or ECU)! However, the JAR HD600 was a match made in heaven. (As does the Audeze LCD2 and LCD3, and by extension the LCD4). All the Sennheisers have this veil (the HD600 a bit less than HD650, but it's still there). The ECU was able to pierce this veil and tighten up on the lows. Shockingly, the highs were also extremely well behaved. The ECU with its sharper and stronger attacks can exacerbate transducer issues in the highs. Not sure here - I wasn't sure if the amp was bright (this still could be a little so), or if the amp was merely more faithfully transmitting what was inherent in the transducers.

    We didn't try the Utopia on the ECU for obvious reasons, but maybe @ChaChaRealSmooth will get a chance sometime later. I have the Utopias on loan in for the burst measurements (super curious about this). The Utopias were fantastic from the Studio 45.

    Measurements coming up...
     
    • Like Like x 30
    • Epic Epic x 1
    • List
    Last edited: Aug 5, 2020
  2. DEATHxMACHINE

    DEATHxMACHINE Friend

    Pyrate MZR
    Joined:
    May 6, 2018
    Likes Received:
    1,000
    Trophy Points:
    93
    Location:
    Washington
    I'd really like to know how it handles higher impedance planars (compared to the Verums). If you want any older Hifiman's to mess around with, just shoot me a pm.
     
  3. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    I didn't get as much time for measurements as I wanted, but I think I have enough. For most of us who have been studying the latest chip amp measurements, this performance is "horrible" relative to those. I still stand by this in terms of what is good enough: https://www.superbestaudiofriends.o...ically-incorrect-audio-blog.7261/#post-249242

    0.55Vrms or -3dBu into 32-ohms. I used 32-ohms because I expected the ECU to be used with Audezes.
    1kHz steady state sine wave
    EC ultralinear.png

    Lots of AC mains noise and related harmonics, so something to be careful of. With the super low Z Verum I didn't hear any hum. Still at this rather low voltage output level, 60Hz is a good 70db below. The second harmonic is about 80db below and third even lower. I'll present FFTs with smaller window sizes so we can make out the harmonics better in the next post.

    Well Verum at 8 to 12-ohms worked well with a great sense of power and control. Audeze LCD2r2 also did which I think is 30-ohms. The HD600 at 300-ohms also worked. I don't think you will have an issue with older HFMs which are near 50-ohms in terms of drive. The thing I would worry about most with the older HFMs is their tendency to sound etchy unless properly modded. The ECU won't do etchy sounding transducers any favors.
     
  4. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Here are the FFT using smaller window sizes to examine the pattern of the distortion harmonics.

    -20dBu 1kHz into 32-ohms
    Steady state sine wave
    We can't make anything out here.
    -20dbu.png


    -10dBu 1kHz into 32-ohms
    Steady state sine wave
    A little bloop on the second harmonic
    -10dbu.png

    -5dBu 1kHz into 32-ohms
    Steady state sine wave
    A little bloop on the third harmonic too, less than the second harmonic
    -5dBu.png
     
    • Like Like x 2
    • Epic Epic x 1
    • List
  5. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    0dBu 1kHz into 32-ohms
    Steady state sine wave
    Ahh, we see a 4th harmonic now. Each successive harmonic is lower than the rest
    -0dbu.png

    5dBu 1kHz into 32-ohms
    Steady state sine wave
    A little bit higher output now, we've lost that nice triangle shape to the harmonic pattern
    5dbu.png

    10dBu 1kHz into 32-ohms
    Steady state sine wave
    Now we are starting to clip with the addition of 5th and 7th order along with increased 4th and 6th.
    10dbu.png

    15dBu 1kHz into 32-ohms
    Steady state sine wave
    Up higher in output level a bit, but the trend is reversing a bit. The even orders are now taking the lead!
    15dbu.png
     
    • Like Like x 3
    • Epic Epic x 1
    • List
  6. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    That's it for now. Would be curious to compare to a SET design in a few days.

    Here's the head scratcher. Before @ChaChaRealSmooth came by I had left the volume knob cranked up (for bench testing) by accident while the Verums were plugged in. Allow me to pull off an inverse-Amir: the output was clean. I could turn out the volume until the 3pm position, and the output was clean or at least as clean as from one of those zero point zero zero zero solid state chip amps. No difference.

    How is this possible? The distortion of the transducer, especially when cranked up, dominates over the distortion of any amp, even a "crappy"* measuring tube amp of old school design. While steady state measurements can perhaps infer some goings on with what we hear (likely in the transient behaviors), don't let anyone tell you that measurements matter that much after they reach 0.1% to 0.01% THD, especially if they are predisposed to confirmation bias that small amounts do actually make an audible difference.

    *As in relative to what a modern AP, Prizm, AverLAB, QuantAsylum, or whatever analyzer can do today.
     
    Last edited: Aug 5, 2020
  7. Hands

    Hands Overzealous Auto Flusher - Measurbator

    Staff Member Pyrate MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    12,285
    Trophy Points:
    113
    Location:
    Colorado
    Home Page:
    @purr1n LOL, I think I've theorized before about the JAR HD600 you heard, but it's a puzzle that will never be truly solved.

    The pair you heard originally may have been missing some damping on the front side of the enclosure, underneath the pads, according to @ext1. This makes a bit of a difference...but just a bit. Shouldn't have been enough to dramatically sway opinions.

    I measured my pair and the first pair you heard way back when:

    [​IMG]

    Subjectively, they sounded pretty darn close to me. There will always be subtle differences on Sennheisers, stock or otherwise. I can't say I heard that pair as hotter in the upper mids, but measurements may or may not have been outside the realm of normal variance and margin of error. (Pair 2, red, is what you heard.)

    Several months down the road, @ChaChaRealSmooth got a fresh JAR HD600 in for a loaner tour, which I think he ended up purchasing for himself. Some similarly found it too bright on the tour. Again, one of those mysteries.

    If I had to wager, it's partially the (maybe) missing damping on the first pair you heard, and probably some amount of pads and other components breaking in. It's the sort of headphone that seems to settle in nicely with worn pads, whereas the HD650 can get overly warm.

    Or maybe it just really goes well with this equipment? Some other issue? I did have a fresh baby in the house, so I was definitely not at the top of my mental game. Maybe I broke it accidentally without knowing. The world will never know!

    I just feel relieved that it must have been some sort of fluke. Feels good to not think I've gone crazy. I love my JAR HD600.

    Sorry for the detour.

    That looks like a pretty sick amp! I enjoy one-off projects like this, whether custom modded from a known designer and engineer, properly modded by an owner, or a unique DIY project. Makes me value those things that much more.
     
  8. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    I'm betting the above (missing pieces) and @ChaChaRealSmooth more worn pads probably resulted in -3.5db less from 1.5kHz to 4kHz and -3.5db less in a narrow spot at 7kHz. Those little things can be enough to do it. Those two areas are the bane of my existence. You can crank up 5kHz, maybe 6k, or 9-10kHz slightly, and I would be OK. But emphasized 1.5kHz to 4kHz of the beautiful female vocal school of shrillness causes me pain. 7kHz is nasty.

    I can't stand HD600s with brand new pads. Craig's HD600 with brown pads from bleach wash sounded pretty good even though of the all the internal foam over the baffle disintegrated.
     
    Last edited: Aug 5, 2020
  9. ChaChaRealSmooth

    ChaChaRealSmooth SBAF's Mr. Bean

    Staff Member Pyrate Gearmaster
    Joined:
    Dec 6, 2018
    Likes Received:
    10,827
    Trophy Points:
    113
    Location:
    The Complex
    20200807_202453.jpg

    @E_Schaaf kindly has lent me both his HE4 and LFF Code-6. This was more of a curiosity experiment since Marv suspected that the Ultralinear would be really good with planars (this won't be a formal review of any sort since this Ultralinear is an one-off).

    With either the HE4 or the Code-6, the Ultralinear can finally flex and show off its muscles. The sound is visceral; I described it to Marv and some others that it's like getting put in a ring with Mike Tyson in his prime. Very punchy, ultra dynamic, and very tight-sounding, especially in the bass. It's clear that the Ultralinear has the juice to run these and grabs the drivers here by the balls, and the overall character suits these very well.

    Things to be concerned about: even with Code-6, the highs can potentially be a bit bitey and crunchy. This does not bother me one bit, but no, these won't do highs like the Utopia. I think the Verite is better in this regard as well. However, overall timbre of things sound pretty good. And did I mention the phat planar bass? THIS is what I think people are trying to get at when they talk about the phat bass with crazy extension; these extend so damn low and with the Ultralinear it's very well-controlled. I haven't heard many power amps with planars and stuff, but I'll be damned if Ultralinear doesn't turn out to be one of the very best amps to try planars with.

    Compared to Code-6, the modded HE4 sounds thicker, much more fluid in the mids with Ultralinear. I'd say that for my tastes, it's a little bit mid-centric, but it's undeniably much more fluid in that region. It actually sounds really good in that region. Initially I thought I liked the HE4 better based on previous listening at a meet, but to my surprise I prefer the Code-6.

    A note here on the pot usage. I'm at about 10:30 on the pot, and it's definitely head-banging loud for me. I imagine that 12'o'clock would be quite excessive. This is partly because the Yggdrasil's balanced outs (run through a Jensen transformer) is a bit hot, but it's worth mentioning in case anyone has a case of the usable pot range nervosa.

    On dynamic drivers. It still pairs very well, but I would be cautious depending on the headphone. Thoroughly enjoyed JAR HD600 (as Marv pointed out above), Verite with Universe pads, and Auteur. I don't have Utopia with me since Marv has those on loan, but I imagine that it would be too much. The Ultralinear is not tubey at all; it reminds me of the DSHA-3F that happens to have tubes sticking out of it (personally, I feel it's better than the 3F in the areas of layering and stage, but I could be biased and overly excited).

    Finally, versus Starlett. In terms of resolve, I think the nod goes to Starlett, but not by much; if the Starlett gets an A, the Ultralinear gets an A- (no surprise becuase Starlett is a SET amp). The Starlett's stage is more diffuse and nebulous in the sense that the stage doesn't seem to have defined boundaries, whereas the Ultralinear definitely does. It's hard to say which one stages larger, but both amps have fantastic imaging.

    However, what this really boils down to is a massive case of preference. The Ultralinear is slamming and powerful, the Starlett is stately, nuanced, and gentle. The Ultralinear has zero tube bloom, whereas the Starlett definitely has very tasteful tube bloom. Which one is right? I don't know. Which one is better? Depends on your chain; we've said many times that it's a game of synergy.

    The Ultralinear is a fantastic design with fantastic sound. Real shame it's the only one of its kind.
     
    • Like Like x 13
    • Epic Epic x 1
    • List
    Last edited: Aug 8, 2020
  10. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Pyrate BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    89,778
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    I knew it on the HE-5/HE-6!

    I wasn't kidding when I said this might be the best planar amp on the planet. The 6L6 sound like what you described. You can try the Mullard 6V6 for something more easygoing with less bite. The difference isn't huge, it's still the EC Ultralinear.

    JAN 6L6 also work as do 6BQ5 (but will need adapter). The two options above are probably best.
     

Share This Page