DIY Stax self bias conversion?

[Arthrimus]

Hello all, I'm working on a little project with my old SRD-6 energizer and a cheapo TA2020 amp. I'm trying to put together a compact transportable setup for my SR-5s. I have done a few modifications to this box already, such as wiring the transformers directly to the speaker taps on the back of the box and directly to the Stax socket on the front, completely bypassing the bias board and the loudspeaker switch. Those mods have done great things in terms of sound quality, but what I'm interested in now is modifying the bias circuit to operate as a self biasing one. As I understand it, Stax accomplishes this by tapping some of the output signal from one of the transformers, then passes it through a small transformer to step it up some more before sending it to, more or less, the same bias circuit as in the standard models. 

This seems pretty straightforward, but in my search for information, I've been unable to track down solid details about what type/values of transformer is used, and if there are any other important components in the self bias circuit (resistors, capacitors, diodes, etc.) that I'm not aware of. Any advice would be greatly appreciated. My ultimate goal is to have a system that runs solely off of 12v dc so it can be run off a battery if need be.

Thanks!

[spritzer]

Just use a high ratio 230:12V and tap it off the secondaries of the stepup transformers.  Then you can either put in a voltage doubler with a 115V back to back zener on the input or just run it directly like that. 

[Arthrimus]

Thanks for the info. Any particular advantage to using the voltage doubler?

[nopants]

lower voltage requirement on the secondary

[Arthrimus]

On 10/12/2015 at 4:33 PM, nopants said:

lower voltage requirement on the secondary

 

So it maintains stable bias at lower output volumes I assume. 

Also I just found this schematic for the SRD-6/SB.

EDIT: DO NOT BUILD THIS CIRCUIT. Look at my post at the bottom of this page for a corrected version of this schematic. This version has no zener diode to limit the input voltage and if built as is will create a very different bias voltage than expected. Basically, depending on the input volume of the audio signal, this one can generate bias voltages well in excess of 230V and will damage normal bias headphones.

There's no transformer in this self bias circuit. It looks like the signal is tapped from the output transformer and sent straight to the voltage doubler. How does this approach work?

Edited July 5, 2016 by Arthrimus

[spritzer]

Same as all other voltage multipliers.  Extra parts to keep the circuit in inline but I'd prefer to have the extra transformer.  Makes things a lot easier. 

[Arthrimus]

Same as all other voltage multipliers.  Extra parts to keep the circuit in inline but I'd prefer to have the extra transformer.  Makes things a lot easier. 

Please forgive my ignorance, I only have a loose understanding of these concepts, but how exactly does the above circuit maintain a consistent 230v bias voltage when the input voltage from the step up transformer varies so much with the volume of the music being played. As I understood it the purpose of the extra transformer that is absent in this circuit was to ensure that the voltage would generally be well in excess of what the multiplier needs to make the 230v bias, and that excess input voltage was limited by zener diodes. 

The above circuit doesn't appear to have zeners so how does stay consistent? Or am I reading the schematic wrong?

[JimL]

The way it works is that the same transformer that steps up the voltage to drive the headphones also provides high voltage to the bias circuit.  The bias circuit drains off a little bit of current from the signal voltage to fill up the bias capacitors.  Think of the diodes as being one-way doors for the electrons.  Every time the music voltage peaks higher than the voltage that is already on the bias capacitors, some current flows into those capacitors, gradually topping them up.  The reason this works is that the bias requires voltage but almost no current, except when you first start up, so draining a bit off the music signal will do the trick.  If you were to bias the headphones and then disconnect them, they will hold the bias for a significant length of time.  Usually the first couple diodes are zeners to regulate the voltage in the bias circuit to make sure it doesn't go too high.

 

In my powered Stax converters, which run off 117V, the power cord actually goes directly to the bias circuit with only a 0.3 amp fuse as protection, but using an additional transformer as spritzer suggests is much safer.

 

 

[Arthrimus]

The way it works is that the same transformer that steps up the voltage to drive the headphones also provides high voltage to the bias circuit.  The bias circuit drains off a little bit of current from the signal voltage to fill up the bias capacitors.  Think of the diodes as being one-way doors for the electrons.  Every time the music voltage peaks higher than the voltage that is already on the bias capacitors, some current flows into those capacitors, gradually topping them up.  The reason this works is that the bias requires voltage but almost no current, except when you first start up, so draining a bit off the music signal will do the trick.  If you were to bias the headphones and then disconnect them, they will hold the bias for a significant length of time.  Usually the first couple diodes are zeners to regulate the voltage in the bias circuit to make sure it doesn't go too high.

 

In my powered Stax converters, which run off 117V, the power cord actually goes directly to the bias circuit with only a 0.3 amp fuse as protection, but using an additional transformer as spritzer suggests is much safer.

 

 

I understand how voltage multipliers work, that's not the issue. What I don't understand, and would really like to, is how the SRD-6/SB bias circuit regulates it's voltage to 230V. There don't appear to be any zener diodes in that circuit to limit the voltage before multiplication, so what in that circuit stops it from generating > 230V bias voltages at high input volume? 

Edited October 15, 2015 by Arthrimus

[JimL]

You are correct, there should be a zener diode in there.  My guess is that the schematic you posted is in error, the left-most diode should be a zener.  In my copy of the SRD-7 Mk II that is precisely where there are back-to-back zeners to limit the voltage.

[Arthrimus]

Ok so I have another question, and this one has been bothering me for a while. How do any of the Stax bias circuits produce 230V bias if they are zener limited to 100V on the input side. Shouldn't normal bias circuits produce a 200V bias, and shouldn't pro bias circuits produce a 600V bias? What am I missing?

Edited October 19, 2015 by Arthrimus

[Tinkerer]

I know old normal bias was 200V. They changed it to 230V later. So maybe 200V is what that circuit was made for?

[Arthrimus]

It's strange that every bias supply schematic I've been able to find is set up that way for normal bias. How important is that extra 30V? I've been told that the koss esp-950 sounds better at 620V bias vs 600V, and that is a smaller relative jump in voltage than 200V to 230V. 

[DefQon]

Definitely affects volume gain and sub - bass. I've rebuilt a few old SRA-10/12 and 3S amps back to the 230v NB specification and it makes a difference.

[Arthrimus]

Definitely affects volume gain and sub - bass. I've rebuilt a few old SRA-10/12 and 3S amps back to the 230v NB specification and it makes a difference.

Do you happen to have a schematic for a proper 230v bias supply? Every time I track down a thread that looks like it might have a better bias supply design, the image links are dead.

[sorenb]

Do you happen to have a schematic for a proper 230v bias supply? Every time I track down a thread that looks like it might have a better bias supply design, the image links are dead.

[Arthrimus]

I've seen that schematic. It's a copy of the SRD-7 MK2 schematic and I still don't see how it generates more than 200V normal bias. It's zener limited to 100V on the input and then doubled to 200V for the normal bias tap. Did any of the SRD boxes actually have 230V bias? So far the only 230v bias circuits I've seen have been in stax amps such as the SRM-1/mk2 which doesn't use a voltage doubler at all, but rather takes power directly from the hv secondaries of the power transformer.

[sorenb]

I've seen that schematic. It's a copy of the SRD-7 MK2 schematic and I still don't see how it generates more than 200V normal bias. It's zener limited to 100V on the input and then doubled to 200V for the normal bias tap. Did any of the SRD boxes actually have 230V bias? So far the only 230v bias circuits I've seen have been in stax amps such as the SRM-1/mk2 which doesn't use a voltage doubler at all, but rather takes power directly from the hv secondaries of the power transformer.

have a SRD7mkII that has NB + PRO ....seems to work fine ...don't have a meter that can measure it without dragging it down though

[spritzer]

The circuit is a 6 fold multiplier so limit the input to 100V (with a 5% error in the harmonic suppressor) and you are close enough.  All Stax bias supplies would fall under the "close enough" realm.  There is nothing regulated in any of the Stax amps except the SRM-300. 

[Arthrimus]

The circuit is a 6 fold multiplier so limit the input to 100V (with a 5% error in the harmonic suppressor) and you are close enough.  All Stax bias supplies would fall under the "close enough" realm.  There is nothing regulated in any of the Stax amps except the SRM-300. 

Alright, I'll accept that close enough is good enough and just build the SRD-7 Mk2 circuit, and shut up about it.

Thanks for the help everyone. I'll update this thread when I receive my transformer from china and let you guys know how the self bias conversion goes.

[DefQon]

Do you happen to have a schematic for a proper 230v bias supply? Every time I track down a thread that looks like it might have a better bias supply design, the image links are dead.

For the amps just trace back the bias to the circuit before the ballast resistor. There is always 2 resistors and 1 reference to ground that is a voltage divider off B+ from the HV capacitors.

For the adaptors its a bit different because as Birgir mentioned its a doubler over the AC input.

I`ll post some circuit pictures of my SRD-7sb mk2 as a reference as it has both pro and a proper 230vdc normal bias output.

Sent from my LG-D855 using Tapatalk

[DonnLaurent]

As per my experience it works is that the same transformer that steps up the voltage to drive the headphones also provides high voltage to the bias circuit.  Also the bias circuit drains off a little bit of current from the signal voltage to fill up the bias capacitors.  

And the diodes as being one-way doors for the electrons.  Every time the music voltage peaks higher than the voltage that is already on the bias capacitors, some current flows into those capacitors, gradually topping them up. 

https://www.7pcb.com

[spritzer]

That's pretty much it.  Most of the SRD-7 units used a transformer to further step up the audio signal and isolate the bias supply from the audio signal. 

[Arthrimus]

So I know it's been a while since I posted in this thread, but I just wanted to give a quick update. I recently acquired an SRD-7sb unit and was able to confirm that the circuit that I posted further upthread was indeed incorrect, but only insofar as the diode on the far left was not labeled as a Unirectional zener diode just as JimL said. I modified the schematic and built a test circuit using the new schematic and it works great. 

EDIT: I initially indicated that the Zener Diode should be Bidirectional. This was an error, the Stax circuit uses part # 1Z100 there which is in fact a UNIDIRECTIONAL 100V zener diode. Using a bidirectional diode here causes undesirable effects to the voltage multiplier and will not generate the correct bias voltage. I have updated the schematic below to reflect the change. Sorry about that.

 

Edited July 4, 2016 by Arthrimus
Error on the schematic

[Vlaca]

Hi All!

I have srd6-sb and planning to convert it to AC bias.

If i understand right, i simple cut the self bias “input wire”, and apply AC voltage from little 240V/??V transformer to the “green points”?(input of bias circuit) What secondary voltage transformator needed? (Little more than 100V zener?)

(no 12V AC, just error in my additional drawing)  

Thank you!

Regards,

Laszlo

 

 

 

 

Edited January 25, 2021 by Vlaca
There was an error in text and image.