ECP Torpedo III [indexed in first post]

As I understand it the yellow and white label Brimars are identical, but the yellow is much more loved and hyped. If searching for a full and tubey sound, then these and the standard Mullards are good options (not the military version, they are cleaner and tighter sounding as Fluffy mention in his excellent review).

Interesting with Brent Jesse, he set's the price at $499 for a tightly matched pair of Telefunken ECC801s... I'm glad that I only paid $100 for the matched pair I just bought

Here is another source of British tubes, he is really good at Mullard and Brimar, and is very knowledgeable. He gives excellent advise on tubes to choose, if you search for a specific sound. He usually has more tubes in stock, than the few listed on his site:
http://www.mullardmagic.co.uk/ecc81...811cc-twin-triode-valve-various-manufacturers

 

Two tubes I'd really like to hear in this amp that nobody has mentioned:

• GEC A2900
• Valvo 6201 triple mica blue

These are the Mullards that many consider the nicest - M8162 gold pin - but they're far too price prohibitive for me.

 

GEC A2900 would be the trump card for showing off, but much too expensive. I am curious about how they would sound in this implementation. Also those pictured ones are made in Holland.

 

Yeah, the A2900 are probably a bit sparkly in this amp... But I'm not sure. I'd like to hear them before ever considering purchase!!

If I could find the M8162 square getter for a reasonable price, I'd definitely take a chance on them.... Mine are these, which I consider pretty nice but definitely warm like British tubes tend to be, and a bit restricted in soundstage vs Holland 7062s...

 

Hi @dsavitsk

I think you once commented on carbon film resistors favorably and I'd be curious what you think about using them in a T3 instead of the metal film. Something like Kiwame / KOA Speer SPR.

Thanks!

 

Maybe it's OK if I give my impression on resistors as well... I have a small stash of top quality resistors and have tried a lot of different ones in my constructions over time.

First of all: In my opinion the difference in sound between resistors is not as big as the difference in sound as different capacitors or tubes. But then again, resistors are much cheaper, so it makes sense to experiment with these also.

Carbon resistors such as Kiwame, Allen Bradley, Riken etc are to the warm side in tonality, and they are not that transparent. Think in the direction of a warm ''tubey sound''. I notice that some find the Torpedo III a bit too neutral sounding and would like more warmth. Then these ''old school'' carbon resistors can be a great thing.

For me some of the best carbon resistors are the Takman carbon. They have a slightly warm tonality, very organic, but also transparent. And they are not very expensive. These are the orange / pink ones in the middle of the picture. Takman also make nice metal film resistors. They are very neutral in tonality and transparent.

My personal favorite resistors are the Shinkoh tantalium. Very dynamic, organic, transparent and ''real'' sounding. They are not made anymore, a bit expensive, and can be hard to find in the value you need. So sometimes two resistors in parallel is needed to get the value right, adding to the cost. They are the blue ones at the bottom of the picture.

If I have the correct values I will use Shinkoh tantalium resistors on the critical locations in the signal path of my Torpedo III (that is now in the hands of the Danish customs - man that tend to be expensive)...

 

Yes it looks to me like carbon film resistors add distortion. Nowhere near the level of distortion from tubes, but new distortion nonetheless and thus the resulting warmer signature. I have heard from someone I do trust that the Kiwame do sound very natural juxtaposed against the slightly tinny sound of many metal film resistors. Since Kiwame seem to be re-branded (and overpriced) KOA Speer SPR - they can be had from the likes of Mouser at little added cost.

This was a post from @dsavitsk that led me to wonder if they'd be a good choice in the T3...

 

You are digging up some very old posts. The linked circuit seems to have disappeared, but it was just a CCS loaded emitter follower. But yes, my experience is that in certain positions, the type of resistor seems to make a difference. Those positions tend to be in an I to V sort of spot, like in a DAC, or loading the plate of a tube. So with the T3, the place to experiment is probably R23, R24, R25, & R26 - I don't think any others would make any difference. Those are 100K 2W (1W is probably OK) and need to be rated to 350V.

 

I believe it's a myth that carbon resistors add distortion. I have never been able to measure that or seen measurements from others documenting distortion. They do have higher noise figures though, this is documented in the data sheets for most resistors. It's of no real importance in circuits like this one though, but if designing an active MC step up for instance then metal resistors might be a better choice.

Choose what you think sounds the best...

 

Thanks for your input on this.

I would think that R5, R6, R7, R8, R9 and R10 would also make a sonic difference..?

For power resistors Mills are very good sounding in my experience, these could be interesting to test for R23, R24, R25 and R26. Or the Kiwame if a warmer sound is desired.

EDIT, I just checked the Mills, and they biggest value is 27K, so they are not an option.

 

R9,10,11,12 were the first I found connected to the pot just looking at the PCB traces... I haven't looked further yet.

 

Look at the schematic, much easier than trying to figure out the circuit by following pcb traces... http://www.diyforums.org/Torpedo-III/TorpedoIII-schematic.php

 

This is from Douglas Self's book, Audio Power Amplifier Design. Distortion from metal film resistors in the same setup fell below measurement rig's noise floor.

Since distortion rises with level, it will be much harder to measure in the typical solid state amp. Tube amps generally develop much higher voltage signals across single resistors.

 

Whoops, that's not the plot you want. This one is.

Not nearly as bad as carbon composition, but carbon film still shows more level-dependant distortion than metal film.

 

Yeah, I'd seen reference to this but never a plot.

The distortion difference is notable, but the absolute level of distortion still seems rather low. Agree?

 

Maybe, maybe not. THD is an almost meaningless measure for determining sound quality. 10% 2nd harmonic distortion is benign, 0.01% 7th harmonic is not. Even by that plot you have no idea which is better. All you can really learn is that different resistors do indeed behave differently. But how they will influence sound quality is still anybody's guess.

 

I agree, well put...

 

OK, English as second language. I see that I did not manage to express myself well enough...

What I was trying to say is, that I believe it's a myth, that carbon resistors add distortion that is similar / comparable with tubes when implemented in a circuit. I have never seen a passive component (resistor in this case) add distortion / sonic signature similar to an active component like a tube... Don't know I a managed to explain myself better this time.

Thanks for the picture, this is similar to what you can see in component data sheets also. You will also see more noise and more drift in value from carbon resistors when looking at data sheets.

So basically, based on data carbon resistors are not as good as metal film resistors. But listening often tells something else. I would choose tantalum or carbon over metal film for most purposes when it comes to the sound.

 

10% is rather heavy even for 2nd harmonic, imo. I can be disturbed by 1% in midrange.

Tantalum is a metal too. I did not know resistors were made out of tantalum, rather expensive too.

 

Yes the Shinkoh Tantalum resistors are expensive. Not as expensive as something like Vishay Z foil resistors though... And not as expensive as the tubes and coupling capacitors we buy for our amps.

I will use the Takman carbon resistors on the locations mentioned above. Very good sound for little money. I would have considered the Shinkoh tantalum for the 4 resistors loading the plates (see dsavitsk input above), but the values are no longer available. So it's the Takman carbons for me... I did also order a set of TDK metal film for the plate loading resistors - to check if I can hear a difference.