JeffYoung's DIY Torpedo 3 Build

V2 based on Tom's feedback.

Power transformer moved down slightly to keep it farther from input signals.
Also moved the upper rectifier back down. (I had moved it up to center the mounting post, but since rectifiers tend to be noisy I decided to err on the side of caution.)
Input signal traces added to board at least as far as the crowded center section.

 

Some of these have already been reported one-by-one, but I thought it might be useful to have them in one place:

• 2SC4793 BJT NPNs: http://www.bdent.com/2sc4793-toshiba-audio-power-transistor.html
• Power transformer: www.edcorusa.com, #XPWR274 (note: not in online catalog)
• Output transformers: www.cinemag.biz, CM-15615PC (note: not in online catalog)

I don't claim any of the following are "right"; they're just what I'm using:

• Case: http://www.enclosuresandcasesinc.com/split-body---specs.html (4012 x 14.5”)
• Z switch: http://www.guitarnucleus.net/guitarparts-electronics-switch-minitoggle.html (WDE9B)
• Volume pot: https://www.hificollective.co.uk/catalog/2cp601-dual-taper-potentiometer-p-8595.html
• Volume knob: http://www.goodcomponent.com/produc...-control-knobblack-anodizedgdak0162-1228.html
• 22uF electrolytic behind the top output CCS board: substitute UBT2W220MHD8 (125°C) for UPW2G220MHD (105°C) http://www.mouser.ie/Search/Product...D8virtualkey64700000virtualkey647-UBT2W220MHD

BTW, I accidentally ended up with 4 volume knobs, so if you need one just PM me and I'll send one your way.

 

Since you mentioned the rectifiers, remember we used a couple of 100pf SMD (1206) snubber caps for the heaters rectifier. They don't show up on the schematic, but you will see them in the PCB layouts on the bottom of the PCB. You are correct to assume that they could be noisy. We had a ripple noise on the T1 that we finally discovered was the result of the rectifiers oscillating and the PT picking up the noise and reflecting it through the amp. It was a really bedeviling problem that had us stumped for a while. The "ripple" (it was actually centered around 180Hz) would not manifest itself until we placed the case lid on the amp, but it was quiet without it. The snubber caps, along with some slower rectifiers, finally removed all noise from the T1. Doug put them on the T3 from the very beginning and we never had an issue.

Also no secret here, but the knobs Beezar use come from PartsPipe (hongkongsuperseller) on the 'Bay:
http://www.ebay.com/itm/2PCS-30mmDI...LUME-TONE-ROTARY-CONTROL-KNOB-B-/291271329287
Theirs is usually less than $10 USD for two, with free shipping. They exaggerate on the "CNC-machined solid aluminum," though. I think they are actually sintered, with a bit of machining for the shaft hole and set screw. Still, hard to beat for an inexpensive, heavy chunk of metal volume knob. Your selection is identical, but maybe a tiny bit more costly with shipping. Surely, they're all made in the same place somewhere in China.

 

Yeah, they were in the BOM, so I didn't even notice that they weren't on the schematic.

He he. Which is one of the reasons I have 4. I ordered from GDParts first, but after seeing nothing for three weeks and not getting any replies to emails, I ordered another pair from PartsPipe. Of course the GDParts showed up the next day....

 

V3 of my board design.

Mostly it's just spacing adjustments and a few required re-routings based on moving from straw man to production-ready, but I also separated the PSU grounds from the signal grounds.

I also left the output grounds (from the Cinemags) floating. The headphones shouldn't care, and perhaps this will keep any ground noise away from them. Good idea or bad?

[Edit: board design removed pending update with output transformers grounded.]

 

Bad. Ground noise is not a thing, and transformer secondaries in a high voltage circuit must be grounded for safety. Notice, that says "must", not "probably should".

 

Thanks, dsavitsk. I've removed the floating-ground board layout for now; I'll post another once I have it corrected.

 

Grounds returned to output transformer secondaries:

 

I considered a bunch of options for clamping the casework to the table for milling the ventilation slots, but ended up cutting a block of MDF which fit tightly inside so I could just clamp it in the vise. This turned out to be much easier for holding it in different orientations.

I tried to take a picture while milling, but the camera was set wrong and all I got was a horrible-quality video.

Anyway, the ventilated casework:





Cheers,
Jeff.

 

The back plate on the mill getting a hole cut for the power receptacle:



And the front and back plates:

 

Nice casework! Where did you get the metal?

 

Links for case (and other bits) in post #22.

 

Thanks! The link doesn't post prices. How much was your case if you don't mind.

 

That is beautiful Jeff. One thing though, do you think the top openings are sufficient?

What I mean is you may have enough openings, but they're all concentrated in one area.

 

4012H-14.5" Natural Finish: $29.05
Lead time 7 - 10 days.
Shipping was a bit expensive ($12.74).

 

Pretty much all the heat is generated by the tubes, output transistors, and output CCS boards. (See the thermal images posted by @bazelio here.)

I hope it's enough, as once I get the case anodised I'm not going to want to put more holes in it.

 

Having been haunted by these words ever since penning them, I decided on a bit of relief for the next-hottest spot (the bridge rectifiers):

 

I think that is a good idea and it actually looks good. it's best to allow the hot air to rise up and escape the enclosure quickly instead of building up inside due to the long shape.

 

While waiting for the case to be anodised (I never did find anyone in Ireland to do it, so it's been shipped off to England), I drew up the designs for the front/back:

 

While waiting for the anodisers, I fired up Quasimodo (http://www.diyaudio.com/forums/powe...sformer-snubber-using-quasimodo-test-jig.html) on the Edcor power transformer. The following CRC networks will more-or-less critically dampen the ringing produced by the rectifiers:

6.3 V filament secondary
Cx: 0.01uF
Cs: 0.15uF
Rs: 10 - 14 ohms

200V B+ secondary
Cx: 6,800pF
Cs: 0.068uF
Rs: ~ 760 ohms