Here is a link to their tube socket page.

www.partsconnex...ubesockets.html

Which is the correct socket? I believe it is either:

Chassis Mount (with solder lugs)
SOCKET-55293/SOCKET 59000 (silver/gold) - 9 pin miniature Ceramic, cup style contacts, 2 pieces, top mount

PCB Mount
Socket-52640/Socket 55292 - 9 pin Ceramic, cup style contacts, 2 piece

I think the chassis mount is correct.

Should I go gold or silver contacts?

Thanks super.

[Edited by searleb on 12-04-2006 at 07:29 PM.]

---

[Quoting xelion]

Okay, just to clear it up, as I didn't realize I was being ambigious eariler.

plug->switch->fuse->transformer->ac1

right?

---

I almost have it done, but I broke my pot, so I'm going to have to get another one from Radio Shack or something. Sadly I have no money, so another 2 week wait until my next paycheck. :/

[Edited by tomb on 12-04-2006 at 08:20 PM.]

Yeah, I figure for $1 I can go for the gold contacts. I'm sure I won't hear a difference.

I'm happy to hear these temp measurments, though I was realeved too to know it was not celsius.

could you guys give temps measurments for Intersils too? Or explain how you measure this.

All the best,

GregVDS

Regards

---

[Quoting vixr]

I'm sorry, I forgot some of you guys are metric...
LM337T = 62.2 C, BUF634s = 76.7 C, L7812/L7912 = 66.7 C

---

Are these values measured when the enclosure is closed?
If not, it's reasonable to assume that the temperature can be much higher than these values. Of course, it's also depending on how well the enclosure is ventilated.

I don't use on board heatsinks, but fixed the LM337T on the side plate (heatsink) of my Hifi-2000 enclosure and the the 7812/7912 are mounted on the metal bottom plate of the enclosure. By doing this way, the enclosure becomes a kind of (big) heatsink. It becomes (acceptable) warm but never gets hot!

My only concern is the temperature of the (non-stacked) BUF634's/OPA627. These IC's become very warm comparing to the voltage regulators. Hopefully I can find a kind of top mount heatsink for these IC's.

Not forget to notice that the sound is a big improvement over the stock OPA2134 .

[Edited by Ferrari on 12-05-2006 at 04:14 PM.]

Regards

---

[Quoting GregVDS]

Very nice Build, Vixr!

I'm happy to hear these temp measurments, though I was realeved too to know it was not celsius.

could you guys give temps measurments for Intersils too? Or explain how you measure this.

All the best,

GregVDS

---

YES - Vixr's photos are great, too!

Greg - about the temperature measurement:

Gewa's infrared is nice. However, many Digital MultiMeters come with temperature measurement capability: a thermocouple with a tip that can be pressed onto the surface of the chip. I got mine as part of a DMM for $19.99 on sale at Harbor Freight.

For what it's worth (maybe not much), the Intersil stacks in my Millett ran cooler than the BUF's.

[Edited by vixr on 12-14-2006 at 08:34 AM.]

I will soon be at the point of putting into the Hammond 1455 case, the one with metal endplates. Digi's PC cards fit the slots perfectly. My question is, is the PC card secured by the hardware on the Alps pot to the front metal panel only? The holes along the edge of the card are slightly eclipsed by the extruded edges of the case in case I want to add screws etc to the bottom of the case (which slides out). I don't know if additional fasteners are needed since there really won't be any forces pulling on the card - the external power line will have strain relief and is not attached directly to the card anyway. I thought about a dab of epoxy but that is hard to "undo" if I decide to pull out the board for opamp swaps or other changes.

I am also mulling over how best to mount the encapsulated version of the transformer. Perhaps I can mount it to the sliding bottom cover, or trim some perfboard to fit the grooves in the side of the case and mount the xformer to the perfboard.

BTW, I plan to check out the following chips: opa2134 (stock), AD843 (on Browndog adaptor), ad 8620, ad 8610 (on Browndog adaptor), and I also have an ad823 but it does not have much current capability and it also doesn't resolve as well as the 8620. But it sure sounds great with female vocals on one of my SACD players! What are some of the other opamps people have tried?

For tubes I have the JJ and ElectroHarminix

Thanks for your ideas!

g e

[Edited by G E on 12-06-2006 at 09:02 AM.]

IMHO, the OPA551 is probably the best single chipset (Brown Dog, as you say) to put in this amp. That's because the OPA551 can deliver up to 200ma output current. That's over 4 times the current capability of the standard OPA2134 recommended for this design. It's also unity gain stable. With a jumper on the socket pins, it can even serve as an outstanding buffer on the Millett, as championed by Amb.

The OPA551 was confirmed as the best performing opamp in the original SOHA p2p thread by one of the early builders: theDoctor.

That said, it's not as fast as some of the chips you mention, and lacks some of the detail of an OPA627, for instance. That's why this effort by Greg, Gewa and others to attach buffers.

As for the Hammond case - I think you will find that most DIYer's re-define Hammond's intent by using the sliding panel as the TOP, not the bottom. This is especially helpful in tube designs, since one is almost always into the top of the board adjusting bias. It also simplifies drilling the huge hole for the tube because you are dealing with pretty much a flat plate - easier mounting in the drill press or on your back porch, in other words. As a matter of fact, any 1/16" thick sheet of metal, clear plastic, etc. will slide into the slots. That may leave a little hump where the sides bend over, but this is not objectionable - most eyes are drawn right to the tube. In the case of a clear top, no one will ever notice that it sits slightly lower - they'll be too busy looking at the board. Regardless, use the Hammond plate as the TOP, if you use it.

As far as the board sliding around, most just cut a rectangle out of perfboard or similar material (maybe the scraps from your new clear top) to slide in the empty space. This locks the board in place. Alternately, you could simply drill/thread a couple of screws into the slot. Again, using the sliding plate as the top makes this all possible.

You didn't mention how long your case is, but by chance if you have enough height - you could kill two birds by mounting the xfmr on your perfboard slot-stuffer.

Good luck.

[Edited by tomb on 12-06-2006 at 11:38 AM.]

[Edited by vixr on 12-06-2006 at 11:24 AM.]

I picked some up last night at RS. While there I also picked up some hook up wire. Three festively colored spools of 20 gauge solid core. I plan to use this for wiring on the secondary side of the transformer, although it looks like it also has capacity amp and voltage wize to be used on the primary side (according to severak websites I checked). I may see if I can find 16 gauge for the 120VAC side, just for an added margin.

With regard to signal carrying wire, is there a preference for going with shielded wire or braiding unshielded wire? If braiding, should L-G be braided separately from R-G or can can it all get braided together L-R-G ? How many twists per inch? I may open up my pre-amp (Audio by Van Alstine) to see what he did. I've had it open before but I don't recall how that is set up.

And a sheet metal question - what have people done to cut the hole for the tube? I don't have a hole saw but I do have a drills and dremels. If I cut this "generously" large, giving a 1/4 or 3/8 gap around tube, would I need to cut ventilation hole above the area of VR3? Or should I just plan to make the ventilation holes?

Thanks again for advice.

G E

[Edited by G E on 12-07-2006 at 08:43 AM.]

---

[Quoting xelion]

Okay, just to clear it up, as I didn't realize I was being ambigious eariler.

plug->switch->fuse->transformer->ac1

right?

---

Its right.for this connection,the fuse is 0.15-0.2A 110VAC fast blow.

Zang

[Edited by digi01 on 12-07-2006 at 08:58 PM.]

---

[Quoting xelion]

Alright, thanks for all the help tomb.

I almost have it done, but I broke my pot, so I'm going to have to get another one from Radio Shack or something. Sadly I have no money, so another 2 week wait until my next paycheck. :/

---

FYI,
for under 0dB with volume control source,such as pc soundcard or ipod player.you can bypass the pot.

Zang

[Edited by digi01 on 12-07-2006 at 09:03 PM.]

My build -
v3.0 Digi board
LND 150 based with TO220 case regulators.
The LND mosfets are 180 degrees rotated from silkscreen on v3 of SOHA board as advised
R17/R17 = 357 ohm, R8/R18 1K Ohm as advised
70053 Toroid transformer (encapsulated) with center tap going to ground on board
Alps pot NOT grounded,mains NOT grounded, RCA line in NOT installed, output jack NOT installed

Voltages:
Pot, tube and opamp installed. Center tap to ground, the two pads close to AC1 and AC2

The one that is off is for 60 vdc. I get 69 - 70 vdc.
Both 40 vdc points are spot on after trimming
-12.6 my actual is -12.54
-12 my actual is -12.06
12 my actual is 12.06

Mains voltage is 123VAC, 60 Hz
Voltages of the secondaries: 19.17 vac (19.7 with no load) on either side to center tap.
Paralleled Primaries - pins 4 + 6 tied, pins 3 + 5 tied
serial secondaries - pin 14 and pin 11 to AC1 and AC2, pins 12+13 tied (center tap) and go to ground

I reviewed my soldering using a magnified light - a very helpful accessory! - and everything looks OK

Tube elements are lighted, LED is lighted. VR3 runs pretty hot - too hot to leave finger on for more than a moment or two, but that sounds typical from what I have read in the threads.

I have tried 3 different tubes with simikar results except that I have to readjust trimmers to get to 40 vdc.

Any ideas are most welcome.

G E

[Edited by G E on 12-09-2006 at 08:47 PM.]

---

[Quoting vixr]

GE, my board is also at 70 volts at the 60 volt test point... It doesn't seem to hurt anything.

---

Even with load? Interesting