---

[Quoting CTY]

After listening for a while, I have come up with some other questions

I use the SOHA mainly with Sennheiser HD600s - would it be a wise decision to leave out the output resistors?

---

---

[Quote]

What effect does increasing the plate voltage over 40V have, especially on the sound? is there the chance to break anything if I increase it too much?

---

This may be one reason that with the use of the Jisbos buffers, it was suggested by Steinchen to go ahead and use 15V regulators, but that's a subject better left to a Jisbos application discussion. (There are voltage input limitations with most opamps, which is why the 1n4148's are all around the opamp. Those go away with the Jisbos, however.)

At any rate, you can see that there's a potential of clipping the signal through the tube on the positive side if you bias it up much higher - the signal could bump up against the B+ voltage if it had dropped to 55V or so under load.

---

[Quote]

Greetings
-Pascal

---

[Edited by tomb on 12-02-2007 at 01:02 PM.]

---

[Quoting raromachine]

Hrm - could you explain that with a diagram? I'm not clear on how +55 is lower than +40..

---

Ideally for a 12AU7, I think you'd put 200V on it for B+, bias it at 100V, then let it swing +or-100V.

That's why they call it "bias", though, we are artificially raising the zero point of the music signal so that it has room to cycle up and down from there.

EDIT: Sorry there's no diagram, but maybe someone else more expert can provide a better explanation and do one of those. I've probably said enough to get me into trouble already.

[Edited by tomb on 12-03-2007 at 04:46 PM.]

What has been decided about mains ground, connect or leave off ?

What about using shielded cable for the signal wires, should the shield be grounded on only one end as typically done ?

If using the jisbos should the opamp protection diodes be left off?

[Edited by regal on 12-14-2007 at 04:28 PM.]

---

[Quoting regal]

Is it OK to use RCA jacks that ground to the chassis ?

---

---

[Quote]

What has been decided about mains ground, connect or leave off ?

---

---

[Quote]

What about using shielded cable for the signal wires, should the shield be grounded on only one end as typically done ?

---

---

[Quote]

If using the jisbos should the opamp protection diodes be left off?

---

[Edited by tomb on 12-14-2007 at 04:47 PM.]

Grounding - so should the output ground be tied to the chassis ground or left floating? I'm grounding the AC mains ground to the chassis.

Opamp protection diodes - I'm configuring to use the opamp first before putting in the JISBOS - do i need to remove the diodes when using the JISBOS or can i leave it in?

thanks

her shann

Although with AMB's new alternate procedure for hooking up the Jisbos that may not be practical. May just go straight to the jisbos.

---

[Quote]

do i need to remove the diodes when using the JISBOS or can i leave it in?

---

Yes, leave them in.
You wanna guess what happens when the output from the tube exceeds the buffer's supply rail limits?

---

[Quoting Mister X]

---

[Quote]

do i need to remove the diodes when using the JISBOS or can i leave it in?

---

Yes, leave them in.
You wanna guess what happens when the output from the tube exceeds the buffer's supply rail limits?

---

The recommendation came from the first person to ever apply a buffer of any sort to a SOHA:

---

[Quoting dBel84]

Hi Ferds , any schematic to show how you get the B+ ? With the Jisbos output you could drop the protection diodes and finally, that ground plane may not result in a quiet amp. I would make it double sided and try not to break up the ground plane, also be careful of the ground plane on the ps section - probably safer to remove it (ground plane around ps) in an attempt to isolate any noise. I like you final concept for the amp..dB

---

---

[Quoting Runeight]

However, with such high gain it is possible to exceed the input voltage tolerances for the opamp with just 1Vp at the input. The data sheet for the OPA2134 (and many similar opamps) indicates that the maximum input voltage is (V-) - 0.7V to (V+) + 0.7V. This means that the input swing must not exceed the supply voltage by more than one diode drop. The diodes ensure that this does not happen.

If the diodes conduct, the excess current passes into or out of the bipolar power supply. What happens thereafter depends on the ability of the tube to source/sink current and the ability of the PS to sink/source it. In this case, the tube will source/sink in the range of hundreds of micro amps which will find their way to the output caps of the bipolar supply which are in turn supplying current to the opamp V+ and V-. The output of the regulators will fluctuate some, but at this point the amp would not be operating properly anyway.

---

---

[Quote]

I'd rather not guess.

---

---

[Quoting Mister X]

---

[Quote]

I'd rather not guess.

---

So don't... my reply was not directed at you anyhow.

---

Most DAC's are 2V RMS. The tube stage will gain this higher right? I don't understand how we could ever be below +_ 0.7V to the buffer ?

---

[Quoting regal]

--------------------------------------------------------------------------------
[Quoting Runeight]
However, with such high gain it is possible to exceed the input voltage tolerances for the opamp with just 1Vp at the input. The data sheet for the OPA2134 (and many similar opamps) indicates that the maximum input voltage is (V-) - 0.7V to (V+) + 0.7V. This means that the input swing must not exceed the supply voltage by more than one diode drop. The diodes ensure that this does not happen.

Most DAC's are 2V RMS. The tube stage will gain this higher right? I don't understand how we could ever be below +_ 0.7V to the buffer ?

---

For instance, in the basic SOHA setup, the opamp is fed with V+ and V- of +12VDC and -12VDC, respectively (the LM7812 and LM7912). So, the signal music voltage input to the opamp cannot exceed +12.7Vpeak or -12.7Vpeak. The diodes prevent this from happening because they switch on if the signal voltage peak exceeds the voltage supply, minus the drop in the diode.

The danger is there in the basic SOHA because the gain is about 12 coming through the tube, so if a music signal of 1V peak-to-peak hits the tube, it will multiply that signal voltage by 12, so that the swing going to the opamp will be +or- 12V peak-to-peak. If the signal input swing exceeds 1V by only a very small amount, then the opamp restrictions are exceeded. Hence, Runeight put the diodes in.

If you already knew all that and I misinterpreted your post, then I apologize for the redundant explanation.

Anyway, it doesn't seem that this is related to what happens in a discrete buffer such as the Jisbos, but as I suggested (and as Steinchen originally recommended), a +or-15VDC power supply for the Jisbos would seem to take care of all of that.

We've mentioned elsewhere that Runeight states B+ voltage drops under load to about +55VDC. So, if the tube is biased at 40V, that allows a 15V swing on the plus side. So, 15VDC would be the max through the whole circuit if the Jisbos had a +or-15VDC supply.

Someone else will have to explain why not having the diodes would be an issue with a +or-12VDC supply with the Jisbos. I already said I'm not guessing about that one.

[Edited by tomb on 12-15-2007 at 02:36 PM.]

I am planning on using AMB's gain reduction method for integratig the Jisbos so I imagine this alone should provide enough safeguard to ommit the protection diodes.

Here is that maximum rating table that Runeight refers to for the OPA2134:

As you can see in the note at the bottom, "Stresses above these ratings may cause permanent damage." IOW, the end result may be opamp destruction. Hence, "protection diodes" for the SOHA's opamp.

(can't seem to get attachments to work around here anymore, so I uploaded it to one of my servers - probably something I'm doing wrong)

EDIT: Yes, you are correct that attenuation would cut the gain enough to make this a non-issue.

[Edited by tomb on 12-15-2007 at 01:21 PM.]

I recently was able to aquire a finished SOHA less case, including the epsilon12 board and a custom output buffer (in fact, this is the SOHA headwize member GregVDS sold to me).

I test ran it shortly to verify its function, all was okay.
Then, a few weeks later and the SOHA still without case, I wanted to check voltages and that's when disaster struck...
while moving around with my probes and the SOHA not attached firmly to something the transformer slid off the insulating piece of rubber underneath it and shorted to the aluminum plate...the SOHA got sort of electrocuted...

The transformer was toast so I replaced it...but I wonder what else is most likely destroyed...yesterday I powered it up but it smelled "electric" after a few seconds...there was no smoke and I cannot spot any obviously fried parts but for sure it has suffered a good bit.
In the few seconds I was able to check at least one voltage... -12V seems to be okay but I didn't dare to leave it on any longer.
So, how can I proceed to identify what's "kaputt" and what not...?

Thanks and my apologies for being stoopid...

thanks for your help so far.

I just soldered out the R6 and R16 resistors, and I don't notice a difference with the HD600s. What exactly are the output resistors supposed to do?

A question about the LM337: Can I just parallel some of them to reduce the current each of them has to supply? Because I might want to bolt them to a metal chassis and the heat will be transported away better if it's not produced all in one place (at least I hope so )

---

[Quoting CTY]

Hi again,

thanks for your help so far.

I just soldered out the R6 and R16 resistors, and I don't notice a difference with the HD600s. What exactly are the output resistors supposed to do?

---

---

[Quote]

A question about the LM337: Can I just parallel some of them to reduce the current each of them has to supply? Because I might want to bolt them to a metal chassis and the heat will be transported away better if it's not produced all in one place (at least I hope so )

---

If you really want to do this, the thing to do is to place a high-wattage resistor ahead of the regulator: one of those with a heat-sinked body that's meant to be bolted in place. With the right calculation and sizing, you can burn off several volts ahead of a regulator.

That said, something like that is more often used when you have a wide disparity between voltage supplies - like a 6.3V heater with a 24V or greater power supply. In the SOHA, it's only reducing voltage by about ~17.5VDC (rectified 15VAC) down to 12.6VDC. Part of that is the 2 or 3 volts the regulator needs to burn. So, I don't think it's really applicable to the SOHA's tube heater regulator. A small stamped heat sink with some of the longer fins has proven to be sufficient on the ones I've built.

I guess I just have to find an even bigger heatsink for it, it runs really hot now... or another torroid. Does using a big, overrated torroid like I do now help the sound in any way? It's as big in diameter as the SOHA board...