[headphonejunkie]

Air wiring means wires in the air. Not on the board. If that is what is implied here.It means wires added form the SOHA to the jisbos bds.

 

[el_matt0]

lol thanks, i feel like an idiot

 

[RLembke]

Good luck with this! I've been considering a similar combo Alien DAC / Mini^3 build so I'm very interested in how the SOHA+ADAC build goes.

 

[MusicallySilent]

Jeff has two sites, they both have some of the same things but each has things the other doesn't

He has

www.glassjaraudio.com
index

 

[heatmizer]

Steichen also mentioned changeing the regulators to 15 volt and using tantalum caps on regulator outputs

 

[el_matt0]

Quote:

Originally Posted by heatmizer /img/forum/go_quote.gif Steichen also mentioned changeing the regulators to 15 volt and using tantalum caps on regulator outputs

as a requirement or simply a recommendation?

 

[tomb]

Quote:

Originally Posted by el_matt0 /img/forum/go_quote.gif as a requirement or simply a recommendation?

It's not a requirement, but rather - an optimization. It must be carefully considered, however.

1. All things equal - especially current through the Jisbos, more voltage will give you more power to the headphone.
2. It puts the Jisbos rails at the same voltage as the potential signal swing coming from the tube (B+ of 55V at load minus the 40V bias - this defines the top/plus portion of the wave, so we have +or-15Vp-p.)
3. The data sheet for the LM7815/7915 lists 17.5V minimum for Vi, the input voltage. Rectifying the 15V secondaries from the SOHA transformer gives approx. 17.6V.

The last item is key - the Jisbos will pull more load from the transformer as all of us attempt to adjust for more current. Over-voltage from a transformer is usually expected, but upsizing the C9 and C10 caps to at least 1000uf is warranted, IMHO. This will help to keep the rectified voltage smoothed to its highest value possible. 1200uf or 1800uf probably wouldn't hurt. I've used 1200uf on my SOHA's and it's nothing but an advantage as far as I can tell, but have yet to try the 15V regs or the Jisbos. Hitting 17.5 may not be as easy as first thought (19V was assumed originally). Most likely there's a 0.9 multiplier that's often neglected in the rectifying equation:

((15V x 0.9) x 1.414) - 1.4 = 17.69V (this assumes adequate cap smoothing, however)

As for the output caps, yes - a tantalum would offer the best performance and is often found on the output of most of our favorite Power Supplies, such as the STEPS and TREAD. However, tantalums vs. electrolytics are not sized at one-to-one, more like ten-to-one. So, 1uf - 4.7uf is the range, but smaller is probably faster and optimum. The data sheets for the 7815/7915 show a 1uf tantalum.

 

[el_matt0]

will positioning when i case it up the avel lindberg trafo inbetween my soha pcb and jisbos and alien dac pcbs cause any problems with hum or interference? or is that likely not too realistic of a concern? basically the way ive got it planned out now is SOHA board on one side, trafo smack in the middle, and the 3 smaller boards of the jisbos and alien dac in a line to the right of the trafo. i could potentially see the trafo in the middle being an issue though so i wanted to make sure before i go ahead drilling case holes etc........

 

[tomb]

Quote:

Originally Posted by el_matt0 /img/forum/go_quote.gif will positioning when i case it up the avel lindberg trafo inbetween my soha pcb and jisbos and alien dac pcbs cause any problems with hum or interference? or is that likely not too realistic of a concern? basically the way ive got it planned out now is SOHA board on one side, trafo smack in the middle, and the 3 smaller boards of the jisbos and alien dac in a line to the right of the trafo. i could potentially see the trafo in the middle being an issue though so i wanted to make sure before i go ahead drilling case holes etc........

Put that transformer in a Jameco walwart box and get it out of the amp completely - but that's just my own personal preference. Batman did it first in the original SOHA thread, but Amb also did the same thing with his CKKIII.

Other than that, yes, it's my guess that it's likely to cause a problem the way you describe it.

 

[UserNotFound]

Quote:

Originally Posted by tomb /img/forum/go_quote.gif ((15V x 0.9) x 1.414) - 1.4V = 17.69V (this assumes adequate cap smoothing, however)

E=IT/C

Use the current drawn from the supply (I) to estimate the voltage ripple (E, to subtract from his 17.69V figure) based on the current draw(i) and cap used(C). T should be 1/60hz for a full wave rectified signal.

You could always use an adjustable regulator (LM317) and get the highest voltage output allowed by adding a 1/4 watt resistor and a 10 turn pot.

 

[el_matt0]

i actually DID ironically already order a jameco wallwart with just that intent. but, i figured it is a huge hassle because i want my on/off switch on the amp not the wallwart. that means running 2 wires down the umbilical between wallwart and amp just for the switch alone + the 4 secondaries if im correct? so, wouldnt that mean buying some kind of 6 pole connector like a neutrik speakon which are kind of huge? i really would prefer to get the trafo out of the case alltogether but its the matter of the switch and the connector that is really disheartening me....

 

[tomb]

Quote:

Originally Posted by el_matt0 /img/forum/go_quote.gif i actually DID ironically already order a jameco wallwart with just that intent. but, i figured it is a huge hassle because i want my on/off switch on the amp not the wallwart. that means running 2 wires down the umbilical between wallwart and amp just for the switch alone + the 4 secondaries if im correct? so, wouldnt that mean buying some kind of 6 pole connector like a neutrik speakon which are kind of huge? i really would prefer to get the trafo out of the case alltogether but its the matter of the switch and the connector that is really disheartening me....

If you're wanting to switch at the line level, then yes - that could be an issue.

However, almost every amp incorporates a switch even if they use a walwart - but switching the low voltage end. That's no increase in wires - a 2PST switch can be used to switch the two hot legs, while keeping them separate. A DIN or mini-DIN can be had for only about 75 cents each - plug and socket, especially if you purchase the 5-pin kind. No need to limit the pins, even if you don't use them all. The actual amp end of the wiring from the walwart only requires 3 wires, minimum. You can actually wire up four wires to the case and use the same plug/socket arrangement for a CKKIII with the same walwart.

Another option is simply to use two, standard 2.1mm or 2.5mm plug/sockets. This is easy because you can buy the leads already made up at Mouser. The Jameco walwart strain relief is big enough to accept both lead sets. Use the "center positive" as the 15VAC transformer taps. The SOHA board already accepts a four-wire input.

Just some suggestions. You'll have to decide for yourself if it's worth the trouble.

 

[el_matt0]

NVM

 

[el_matt0]

NVM

 

[el_matt0]

ACTUALLY nevermind my last 2 posts. i opened up the box for the silly jameco wall-wart i ordered to find that the avel lindberg trafo i have doesnt come anywhere close to fitting inside it! dang...jeff rossel was the one that gave me the P/N's for the wall-wart so i kind of just assumed that the RoHS non-vented version would fit the trafo. no dice - so i guess im keeping the trafo in my amp case. that in mind, the case im doing is a bit of a custom job, like vixrs beautiful vertical plexiglass CK2III. my question is, im going to keep the trafo at the bottom, not inbetween the SOHA and JISBOS boards (As i imagine that would be the worst possible location for it). so basically with the SOHA and JISBOS boards "above" the trafo, do you guys think it would make any significant difference in the amount of interference (if any) whether i positioned the trafo on the same plexiglass board as the PCB's (on the same axis as it), as opposed to mounting the trafo on the "bottom plate" so that it is essentialy perpendicular to the amp PCB's? I dont really know how interference from a trafo works, but basically im asking if anyone knows whether i would get more coming out the "top" of a transformer versus radiating from the "sides" of it. obviously if one is more than the other the best move is to get the PCB's out of the line of fire of it!