ECP Torpedo III [indexed in first post]

If your T3 has output ccs boards, My vote is to try snag some brimar 12au7 or similar (I forget the military code one, MrRogers had it in his amp I believe).

 

Found a few hits @brencho:

• http://www.superbestaudiofriends.or...-indexed-in-first-post.434/page-10#post-25511
  
• http://www.superbestaudiofriends.or...-indexed-in-first-post.434/page-13#post-30579
  
• http://www.superbestaudiofriends.or...-indexed-in-first-post.434/page-15#post-31659

General gain info here: http://www.diyforums.org/Torpedo-III/TorpedoIII-tubes.php. "12AU7 - probably has too little gain to be very useful, but if you have a particularly hot source and/or tend to listen quietly, it might be worth playing with." Yggdrasil definitely qualifies as a hot source in my book.

Edit: here's a link to what looks like a good option from Tubemonger: http://www.tubemonger.com/Brimar_CV4034_NOS_NIB_1970_Prem_CV4003_STC_England_p/2020m.htm

 

Still like the 12AZ7 in mine ... but it does run at a bit higher current - nothing like a 6829 or 6414, but still a bit higher. I like the dynamics, though. Brands can be quite different. Sylvanias suck so far, but most GE's sound very dynamic with a very wide frequency response. If you're looking for optimum, remember that the amp was designed for the 12AY7. I haven't really seen anyone talk very much about rolling different NOS brands of those tubes. I suspect that there's a lot of potential, there. I shied away from it, because like the 12AT7 and 12AU7, it's a universally recognized audio tube and has the commensurate pricing to go with that reputation.

The T4?
Maybe Doug won't get upset if I share and summarize a few things. I believe all of these have been mentioned before, just not all in one place. IMHO, the final iteration of the T3 is competitive with any amp in the world for the HD800 (and others). That said, it could still be improved significantly. Besides having become too difficult to build for all but a few, the T3 still has a bit of power supply noise infection from the proximity of the signal traces to the PT. This is below the level of audibility, but it shows up in testing. So, there will no longer be a "torpedo" planform. Also, the T3 still uses capacitors in the signal path to connect the parafeed output transformers. It would be nice to get rid of those entirely (no more giant Mundorfs and Dow Corning cement). Finally, we have non-audio quality issues with the Cinemag transformers that is causing us to look elsewhere. We are addressing all of these issues with the T4. The only one that’s not a certainty (AFAIK) is the Cinemag issue. Lundahl is our target, but they will have to create a custom core – just for ECP Audio/Beezar Audio – that they’ve never done before. If they don’t ultimately agree to do that, we may have to fall back to the Cinemags.

Doug and I also discussed that most of the T4 will be SMD-based. Some parts are simply too difficult to source these days outside of SMD. It also cuts down on the casework costs (but introduces higher investment costs). That will put the difficulty factor up pretty high, so there are no plans to offer kits. They will most likely be mass produced in batches, but a lot of this is unforeseen for the time being. Just understand that Doug is working diligently on it and I hope to help very soon.

 

Any update on the potential decision to move away from the 12A*7 in the T4 as well or is that TBD?

 

No. 12A*7 will stay the basis of the T4.

 

Argh was hoping octals. Sad to see it not diy if that's the case, though I definitely can understand why it might not be. Maybe just bare boards?

 

Well, Doug talks like maybe he or I would be hard-pressed to build one by hand, too. We may have to order it already picked-and-placed, or purchase an oven suitable for SMD - even for prototyping. So, even offering the PCBs may be a problem. This is still up in the air, though, so that level of implied difficulty may change. It's hard to say right now.

 

The problem with octals is that there are so few available. The 6sn7 is overpriced, the mu is too low, the current demand is too high, the heater current is too high, and the rp is nothing special. It is a least common denominator tube that excels at nothing. The 6SL7 is virtually useless outside of guitar amps. There are a few others, but nothing special enough to really bother with. The 12A*7 tubes give us lots of options.

As Tom says, the real focus of the T4, beyond increasing the overall power, is on eliminating the coupling caps. That means a mechanism for keeping DC from flowing across the transformer primary. It also means a transformer that can tolerate a voltage differential from the primaries to the secondaries. The T3's Cinemags are excellent transformers (I don't think the few with askew lams is as big of a deal as Tom, and I don't want to besmirch Cinemag's reputation - they have always done right by me) but they do not fit the bill here so we need a redesign. We are pursuing Lundahl though I am not sure it is going to be an option. We'll see.

As for DIY - the issue is not stuffing the boards. That's relatively easy if a little tedious. The issue is diagnosing what's wrong if it does not work right out of the gate.

 

I bow to your awesomeness on which tubes to use...I have an irrational love for octals for some reason.

Very interested to see what method you use for dealing with DC across the transformer.

Yeah, definitely get the stuffing vs. diagnosing - that's why I'd understand. You guys are community saints for all the help you've probably provided over the years for folks projects that don't quite work right.

 

Even if the T4 is coming i think it's kinda cool to have the T3 now it's become unobtanium

Just out of curiosity, if i may ask, how many T3's are there out in the wild?

 

Awesomesauce, the cat's out of the bag. @dsavitsk was so tight lipped at the LA meet despite my prodding and poking. And, wait, didn't I ask about Lundahl and you said 'No way; too big' or something to that effect? ;-) So coy.

Doug how about the PCC88 (7 volt, 7DJ8). They've got a very sweet sound everywhere I've heard them and enough gain to be useful in MC phono stages. Though the 6SL7 works quite well in the Eddie Current ZDS headphone amp and has a big effect on the overall sound stage, which is an area I think T4 could target for improvement over T3. Good point that the good ones (NOS) these days are severely overpriced, however. At least with PCC88s the crowd favorites like Philips are not completely out of reach.

Regarding DIY, maybe take an explicit hands off approach there. I think an instruction set that includes some basic debug checks (node voltages, etc) and a "you're on your own" disclaimer might ease some burden? But, yeah, I can see it both ways too.

 

Had a bit of a nice WTF moment after installing the heatsinks. There was nasty brightness, grain and tonality was off. So i figured well f**k me i've fucked something up right?

Turns out the T3 likes my Yggdrasil just so long as it doesn't get too close. I didn't want to power cycle the Yggdrasil so i just unplugged the mains from the T3 and went on my way and moved the Yggdrasil more to the right so i had some room. So i moved everything back but a bit closer together then they were appearantly....

Moral of the story is transformers don't like to snuggle.

 

How close are/were the chassis from each other?

 

If i had to guess probably 5cm or closer, i reckon i had it closer to 10cm before. Right now i moved it to about 20 just to keep the gremlins at bay.

 

You can pretty easily hear the hum if you get the T3 too close to another transformer.

 

I didn't really notice an increase in hum coming from the T3 but then again i didn't listen for it, the T3 has been quiet since i got the new PT from you.

 

Well, yes - the point is that the T3's PT reacts when it gets too close to another transformer and you can hear it as hum. You can use that to tell how close you can get the T3 to another component. Listen with the volume turned up a bit with no source, if you hear hum, move it further away. It's a good telltale and very consistent.

 

Ok now i'm a bit confused, just to be clear you're talking about hum you hear over your headphones right? The problem i had was mechanical hum a couple of months ago, that got fixed by the new PT. What i described earlier is basically distortion in the upper mids and lower treble, i did not hear hum when it was quiet, but i guess you mean with no source connected?

 

The hum I'm describing is a generic term. The noise can vary, depending on what kind of transformer is in proximity to the T3. And yes - I'm talking about listening through the headphones. Whether a source is connected or not is superfluous. I only meant that you listen without any music going on.

Yeah, if the T3 is close enough to another transformer, it will affect the sound quality. The only point I'm trying to make is that you can hear it over the headphones, first (without music). If you hear it, move the amp.

Maybe this is more obvious to me after listening to dozens of tube amp prototypes and trying to determine if they have hum. That's always a primary consideration in designing/building a tube amp. It becomes second nature to check it. If you hear it, the first thing to do is to see if the reason the amp hums is because it's in proximity to another transformer. In the case of the T3, this is very consistent. If it's too close to a transformer - you can hear it over the headphones and it needs to be moved.

I was not addressing mechanical noise.

 

Aha yeah now i get it, thanks for clearing that up! Easier to understand when sober...
Anyway i didn't hear anything while it was quiet just when music was playing there was distortion, and it was more noticeable on the right channel.
Maybe the PT in the Yggdrasil had some effect on one of the cinemags, in any case it's all good now, still interesting though!