The Mjolnir 2 thread

Oops, sorry my bad.

 

Question - and I have done a fair bit of searching on this, so hoping I didn’t miss something obvious.

Can you use 12AU7s in the MJ2? My understanding is that the 12AU7 -> 6922 adapters bridge a couple of pins making the voltage sufficient, but not sure if that’s the case. I just picked up a pair of Raytheon 12AU7a’s as part of a lot I bought, and I picked up a couple of Chinese adapters on eBay that I have inbound. People say great things about the tubes in general, but I just wanted to see if they’ll work in the MJ2, and if folks have a point of view on the sound.

Only thread I could find on the topic was one dude on HF who was driving Verite’s from a MJ2 with 12AU7’s and as much as I wanted to trust, I also wanted to confirm.

Thanks.

EDIT: Actually I found lots more content on HF - a bunch of people warning off others that it’s a 12V and not to use it in a Lyr because that’s a 6V heater. But then someone else saying that the adapters they sell on eBay bridge the 4 & 5 pin, so it’s functionally the same. Anyone with any experience here would be worth a LOT more.

 

12Ax7 class filaments are split, so you can feed the 2 segments in series for 12.6V, or in parallel to use 6.3V.

I've used 12AT7 class tubes (CV4033) in all my MJ2, VH2, Lyr3 and Vali2, with adaptors. Works fine.

 

That’s fantastic. Don’t know what to expect, but the tubes seem to be much more affordable than a lot of the 6922/E88CC bunch.

did you have any 12AT7 favorites with the MJ2?

 

Brimar's CV4033 is my odds on favorite in MJ2, just beautifully balanced, quiet and transparent, and they're mil.spec, so rugged and long lasting besides. IIRC, the lower spec CV4024 sounds much the same.

As issued, they're wire lead, so need carriers, but TubeMonger has had the mod'ed tubes available, and high quality adapters. I've seen them on eBay, too.

They work a treat in VH2 as well, although lately I've been playing a few 6BQ7A (Brimar, RCA, Sylvania JAN) and may actually prefer them there.

 

I've spent a few weeks in quarantine lately, and a lot of that playing around with my Gungnir Multibit / MJ2, a few dozen tubes and a bunch of other audio equipment. Most of the time my source was Tidal through Volumio and a USB connection or through a Chromecast Audio and SPDIF connection. Most of the time my headphones were Focal Clears.


So here are a few insights. I'll focus on aspects where I was surprised or had a very different experience to what others describe.

* The differences between Gungnir Multibit / MJ2 and entry levels amplifiers/ DACs are not stark. I used a mixer to gradually or rapidly fade between Gungnir Multibit / MJ2 and Magni 3+ / Modi 3, as well as switching around DAC/Amps individually. I also used a Dragonfly Cobalt as a DAC or DAC/Amplifier. Most of the time, it was hard to tell the difference.

* However, I do feel that over time the Gungnir Multibit / MJ2 setup feels more satisfying and is a significant step up.


Where the difference was most noticable:

* Sounds that are harsh on other DACs sound musical via the Gungnir Multibit. A great example is King Curtis playing A Whiter Shade of Pale. What sounded like geese honking on my other DACs became controlled and expressive jazz. The contribution from the Gungnir Multibit makes it the star of this pair.

* The MJ2 makes percussion sound more real. A great example is Skin Deep (Live at Cote d'Azur) by Duke Ellington, where the drumbeats and especially the drum solo from 1 min in sound punchy, detailed and rich. Is this what people mean by bass slam?


I read a lot of recommendations about tubes for the MJ2, and bought a wide variety of NOS tubes to try. I read more about them as I experimented.

* It's hard to assess tubes against each other because switching tubes takes time. So I generally compared MJ2 + tubes to the Magni 3+ as a stable reference point.

* Tubes that were highly praised in forums rarely sounded different or better to me. The quality difference between tubes was mostly that some tubes sound defective or were microphonic. I was open to the benefits of tube rolling, but this experience has made me sceptical.

* Different MJ2 tubes didn't make much difference to sound vs Modi 3 with one consistent exception - bass is noticably rolled-off on some tubes.

* Focussing on 5670 style tubes, I found the Russian military tubes were the standouts, especially the 6n3p-DRs. They sounded a little bit more controlled than the WE 396A / Ericsson LM / GE 5-star / Sylvania GBs. The only tubes that stood up to the DRs were Raytheons. I like the 6n3p-Es in the Mjolnir 2 because the bass is a little rolled off compared to the DRs, and that matches my preference for casual listening. Readily available, and at less than $2 per tube, these end worries about the cost of leaving the amplifier powered up.

* I'm puzzled at the worries expressed about the limited lifespan of tubes. Most of these tubes are rated for 5,000 to 10,000 hours which is around the same as modern electrolytic capacitors and basic power supplies. For 1960s tubes, maybe they have degraded over time so the original specifications won't be met. I'm more optimistic that 1980s tubes will maintain their quality for thousands of hours.

 

Have you tried the Electro Harmonix 6922's? They roll off in the bass but I think have a more natural presentation (voices in particular) with less distortion than the Soviet tubes. I've tried a number of 6N3P's (Reflektor, Foton, Oktabyr) in MJ2 and while the first two of those tubes were dark-ish they all had timbre issues. There's just something weird about them in the mids and the highs can feel disconnected, at least to what I hear.

 

Thanks for the suggestion @roshambo123 !

I actually have a pair of new production Electro Harmonix 6922's here and haven't taken them out of their boxes. With two days of quarantine to go, there's a great opportunity to test them out. I'll let you know!

One other question I wanted to put out there -- as far as I can tell, the MJ2 has two separate signal paths with the left channel going through the right tube, and the right channel going through the left tube. That what I inferred from the microphonic tubes. Does that mean there is less concern about 'matched pair' tubes than with other tube amps? Can I run MJ2 with two different tubes to do an a-b (mono) comparison?

 

@Woland where did you find these for $2/tube? Including shipping on eBay closest I can find is almost $10. user-tubes.com has some that cost in that range that are similar, but I don’t think that the non-DR are supposed to be nearly as good.

 

After deciding I liked the first pair, I bought a lot of 8 on ebay for $10 + $8 shipping. The per-tube cost is likely to be higher if you just source a pair and have to average out that kind of shipping cost!

The DRs sounded more neutral, giving the MJ2's a punchy bass. When I'm listening to vocal / acoustic music before falling asleep, I'm not likely to need that punchy bass and am really happy with the 6n3p-Es.

 

If you want the same performance and levels in your right ear as your left you'll still probably want a close match. I've never obsessed about the matched tubes point but I assume with a bad dice roll on unmatched tubes you could have a poor result and you'd have been better off simply buying matched pairs. That being said I've been lucky.

As for different tube makes in each socket and A/B'ing, give that a try and you'll quickly see the problem. It's hard to get a sense of a tube in mono, especially stuff like staging and just how it feels emotionally. I'd recommend buying a reasonably close matched pair of anything and testing it in both ears.

 

I'm not noticing anything weird from 6n3ps. The Electro Harmonix 6922's sound very similar to my ears. Are they made by the same factories / processes as the old soviet ones?

 

Yup. EH tubes are made in the old Reflektor factory in Russia. Process wise, I couldn't tell you what is and isn't the same.

 

Back in the days that I owned the Mjolnir 2 the 6n3p was one of my favorites but nothing beats the cheapo 5670w!

 

Just picked up a pair of Amperex Orange Globes from a Head-Fi seller. They're lively, as advertised. Wasn't expecting a detail increase over the EH tubes on MJ2 but it's there. Hair of holographic soundstage and bass extends a bit lower and has enough volume that the low end could be EQ'd and didn't turn into a hairy mess. Even got confused for a moment because I'm used to only the HD6XX's having any bass but I can coax just a little more out of the HD800's now. Hearing more plankton on the 800's too, perhaps the EH's were less detailed than I thought. Treble isn't fatiguing. I can't decide if they're a hair dark or bright, so I assume they're reasonably neutral.

 

Hi Roshambo

Can you connect any of those observations about the Amperex Orange Globes to the tube build? I'm curious about the connection between the physical and measurable characteristics of a tube and its audio characteristics. The closest I've seen was a discussion on this site of Vali 2 with a 386a tube vs another tubes.

If we can measure and the origin of aspects we like in tube sounds, we would be far ahead of the woolly discussions on vendor and subjectivist sites. Some aspects that are easy to measure are frequency response, noise and harmonic distortion. I've made initial efforts at measuring those on some of my collection. I'm not sure how to pick up things like holographic soundstage/plankton/detail though.

 

I think you're going down a dark rabbit hole with that pursuit. Tubes are a moving target to begin with due to wear factors, but also to my knowledge some of the variation in one tube's sound to another can have a lot to do with materials. In vintage tubes especially, some of these can be very Prop 65. For example, I've seen it claimed coating tungsten filaments with radioactive thorium produces a purer sound. Is that true? I don't know. What tubes have thoriated filaments and which don't? I don't know. That being said, I've heard tell there's all kinds of nasty from mercury to lead to godknowswhat in different tubes and that may be a factor in sound. I'm sure construction plays a role too. Getter materials I've read can vary in effectiveness and that might change the sound. It's probably many small things in a recipe, and then tubes change as they burn in, peak and die.

 

I'm kind of hoping to illuminate the way. For example, I previously measured the different harmonic distortion profiles of different tubes I own, and can match that to which is preferable to my ears. More recently, I got a sense the 6n3p-DR has much better bass than the 6n3p-E but is otherwise very similar n a MJ2. I would hope to confirm that with a frequency response graph and maybe link it to the observable and labelled ruggard construction.

Unfortunately, in tube discussions, a lot of the discussion seems to be echo chamber, confirmation bias, shilling etc. A bit more measurement might tame that. And that would seem to fit in well to the SBAF middle ground - not obsessing over sine wave distortion measurements, and not piling on the adjectives and value judgments.

 

The road to hell tends to have decent street lighting before you pass the last exit. I knew someone that had a Hickock, an MJ2, and I think around 3000 tubes, a lot of which were 6N3P. They were exceptionally generous with their tubes, but I don't think for mental health it's a good way to go if you can avoid it. When you're talking about different years of Reflektor where one year is better than another and this getter design is better your nervosa is probably in a very bad place.

And to add to my earlier comments, you can also have batch to batch variations in the same tube with the same year. So, in summary, YMMV

 

Just got a whole pile of tubes from a couple great sellers. This is all Pi2AES -> SPDIF (BNC) -> Gungnir MB (balanced) -> HD-6XX (balanced). I’ll admit I also have tried the SMSL M-500 (balanced) for MQA Tidal recordings at hi-res, which I don’t always love, but there are a few that I think bring out really nice plankton to test these phones with.

A few early thoughts:

• Raytheon 12AU7A - solid and clear mid and bass - losing some plankton on fine hi-res recordings, but treats redbook wonderfully. It seems liquid but clear, if that makes any sense. It’s not rolled off, but it does a really nice job of integrating the mids on treble.
• Valvo 6201 - Adore this tube; was skeptical because I hadn’t heard of anyone using this, and it sure has a whole bunch of Mullard Flash at turn-on; but it is buttery smooth - has all the warmth I’m looking for but no mush. Don’t know if it’s under appreciated, but I appreciate it. Would definitely keep this one.
• Brimar CV2492 - Wow; these are great. But I dont quite get them yet. Maybe I just needed more than a day’s listening to let them settle in. They are both friendly and challenging for the right music; I found Coltrane to be lovely, but Led Zepplin to be kind of meh.
• Sylvania 6922 D-getter - Lovely; people talk about bass slam, and it’s true it’s there, but I would say that it’s the mids that stand out here; more finely characterized than either of the 2 above tubes; just can’t say enough about how well the transients balance being quick and being smooth. Made Zepplin 3 remaster a much more enjoyable album
• Genelex Gold Lion 6922 - Woof. I don’t know about ears bleeding (a phrase I’ve heard), but it’s just not fun for most of the music I love. The treble is edgy - maybe not etched, but it is fatiguing. For a bit of music that I would not have expected (like Mendelssohn chamber music) it is actually lovely - but anything that has percussion is unfun. I’ll keep these.
• ElectroHarmonix 6922 - dunno, but thus far I’ve found them, umm, boring. Maybe I haven’t matched them well. They don’t really stand out on anything. Maybe they feel kid of similar to LISST to me - like I’m not really feeling the character of the tubes. Not sure I’ll keep these - know some people absolutely love them.
• Soviet 6N1P - actually like these surprisingly; maybe just because they’re so different; not sure what to make of them yet. They make the MJ2 run hot, unlike most of these others. I think these are keepers ‘cause they’re cheap, and they’re fun for a change.