ECP Torpedo III [indexed in first post]

Might be better off with Kiwame since they dissipate up to 2W vs the 1W for Takman carbons....

 

Kiwame are to muddy sounding in my opinion. I'm using two 200K, 1W Takman carbon in parallel.

 

Is this a generally accepted approach to soldering two in parallel? Looks like they just twist the leads and fill the gaps with solder?

 

For anyone interested in AUS I'm selling on my T3 and buying another kit I just loved building it

 

It is probably fine, but if you are paralleling resistors to increase heat dissipation, you might want to space them a little more.

 

I just received my ccs board just to clarify I need to remove IC1 & IC2 and then install the ccs board at K and A right?

 

Yes.

 

I've been looking through the forums for impressions, but there are precious few of the kind I'm looking for.

I've currently got a Mjolnir 2, which I use to drive HD800's, HD650's and Paradox Slants. I've also previously owned the Valhalla 2, which was a superb amp for the Senn's. Can anyone tell me what kind of an upgrade I'd be looking at by getting a Torpedo 3 for the Senn's, and how well it would do for the Paradox Slants (or planars in general)? I've read mixed opinions on how well the T3 fares with planars in particular.

I'd be considering taking yours off your hands @Alaarx

 

I installed the CCS boards last night. I think I noticed better bass quality, there might be a darker quieter background as well. T3 noise has always been inaudible to me, dropping the already low noise a few db is hard to notice. I also installed the small smds capacitors that I messed up before and that made a bigger difference.

edit: Deeper soundstage as well.

 

Time for a small update... My T3 kit cleared the Danish customs at a cost of around USD 350... Ouch... But it's time to build now.

I'm still waiting for my Takman carbon resistors and a few other small goodies to arrive, but I have assembled the most. Right now I'm working on the implementation of the TDK pot and the solution with better RCA connectors + quality wires running to the pot.

I had enough of the DH LABS 18AWG solid core silver hook up wire, so that's what I'm using for L and R signal. http://www.partsconnexion.com/wire_hookup_dhlabs.html This is quite stiff hook up wire, so it stays in place once you position it inside the box, very nice. The ground wire is a good quality teflon insulated wire. The close up picture shows, how the soldering point for the Alps pot is now used for the ground wire from the RCA connectors.

Not a lot of space to work on, but using the small size TDK pot means the wires can run under the pot as the close up picture shows. The pot has the legs sticking out to the side for easy access to soldering of the wires.

Under the pot I cut the PCB traces so that the traces are no longer connected to the soldering points. This means that the long PCB traces running from the RCA's to the pot are not connected to anything now.

 

OK, the work with the RCA's, the wiring and the TDK pot is done...

One thing to be REALLY careful about if you decide to implement a solution similar to this is to screen / seal off the connecting points on the primary side of the output transformers. I know that we have a capacitor coupled output before the transformer, so not that high voltages, but still the sharp ends of the pins at the transformer can potentially cut holes in the insulation of the wires. What I did was to cut the unused pins short, and then isolate the pins that are used with some shrink tubing. To make sure it will not fall off, I sealed it with a bit of silicone glue. This will keep a safe distance between the pins on the transformer and the wires.

The final picture shows the hook up wires running under the small TDK pot. I think it proved to be a good choice to use this smaller TDK ...

 

Final picture for now... I think this turned out well, and not too much of a mess to the beautiful layout. The nice RCA connectors also fit in the holes, so no drilling or filing needed.

A bit of filing is needed to fit the TDK pot though, for that small ''pin'' that goes into the front panel.

 

You should braid the 3 rca wires to reject hum and noise. Just twisting all 3 together isn't going to do it.

 

The traces on the PCB running from the back panel to the front are not braided, not even twisted, they just run in parallel all the way - and no hum.

This has worked well for me during 35 years of work with audio, I think it will work here as well Also the wires are close to the ground plane of the PCB = there is ground on two sides (the ground plane and the chassis).

 

Yeah it probably will work since the pcb are just exposed, just extra security.

 

When pcb traces running the signal the entire length of the PCB is good enough to avoid any noise or hum, then this is more than good enough.

But of course braiding would also work..!

 

I just put in brimar 6060 yellow T and they are amazing! Wow I am getting the tingles.

I am getting tears in my eyes such beautiful music. I wish I can freeze this moment forever as I know my ears will start to adapt to this sound.

 

Argumentum ad verecundiam aside I do think it's not a problem as well. The twisting probably neither helps nor hurts. Case in point, while I haven't ever uncased my ZDS, in looking at pics it does appear as though the ZDS just sends RCA wires in parallel from front to rear along the chassis side.

And, just thinking aloud, what are the potential noise sources inside the case? Most of the time you're trying to shield from external noise, hence the metal box. Internal noise would seem less common and circumstantial. Shielding could be used inside, but I doubt necessary. Think about the noise inside a computer, for example. And they are mostly unshielded PCBs and unshielded wires and cables.

 

The primary noise source is the power transformer which radiates a big em field. The primary place noise is picked up is the input traces from the RCA jacks to the pot (the rest of the amp is differential and thus rejects this as common mode noise). The ground plane, on the top of the board, shields the input traces. On one of the prototypes the layers were reversed (ground plane on the bottom, input traces on top) resulting in considerably more noise. I tried using twisted wire on that one (both above and below the board), and the production version with traces under the ground plane is quieter. YMMV.