#26 pre amp

[Rod Coleman]

MJE15032 measured bad in 300B circuits. Please use another, like 2SC4881.

Please check your power rating for 0.47 resistors. TO-220 thick-film 20W or 35W recommended.

9V is minimum for 300B. You can use 12V or even more, but check your heatsink rating, & test circuit on 4 Ohm resistor (2x2R 25W) to make sure transistors not too hot after 1 hour test.

[coolzero]

Originally Posted by Sheldon

No pardons needed. Some trial and error is good. Some engineering beforehand is even better. A current source fed shunt supply is a good combination. That what Salas's supply is. You can do it with what you have. I'd take the current source you are using before the plate choke and put in in place of R4. Increase the current so that you have about 20mA through the VR tubes. Use two OB3 tubes (will give you about 180V. Note R12 from B+ to the connection between the VR tubes on my schema. That's optional, but can help the tubes fire up). Use something less than 0.1uF cap after the VR tubes, or use a voltage divider and two small caps after the VR tubes (see the schematic I posted). Remove L2 and connect the supply to your plate choke. That should be a nice quiet supply.

If you want to try Salas's supply, just remove everything between C4 (I might use something bigger than 1uF there) and the plate choke and plug in the Salas circuit.

Sheldon

Sheldon,


I tried this and below is what I have ended up with. This gave the best sound overall and quieter B+. Overall sound of the pre improved noticeably and now there is much more resolution and control to the bass and treble. Thanks a lot, it worth the change.

[Sheldon]

Looks nice and simple. You can probably remove R4 also. If the current source is a robust design and can dissipate the heat from the voltage drop, the resistor isn't necessary. You show B+ at the plate choke at 260V. Are you sure that's correct? I'd expect something around 180V (two 90V VR tubes in series).

Sheldon

Checked out the KandK current sources, and your transformer. Looks like you are dropping around 200V. So, on second thought, unless you have a hefty heat sink, I might leave the resistor there to dissipate some of the heat. Not sure of the size though. How much current through the tube? Does this supply one tube or two channels? I'm curious on Rod's take. Would a dropping resistor be better before or after the CCS?

Sheldon

[Salas]

Originally Posted by Sheldon

That what Salas's supply is. You can do it with what you have.

Sheldon

Not exactly. There is no gain error loop active regulation in this approach. Performance isn't on par. Although nice and very ''tube''.

[Sheldon]

Originally Posted by Salas

Not exactly. There is no gain error loop active regulation in this approach. Performance isn't on par. Although nice and very ''tube''.

Didn't mean to imply complete equivalence - just that both are shunt supplies fed by a current source. No question, your shunt regulator is much more sophisticated, and more tightly regulated, with lower Zout. Better "performance"? Certainly, in technical terms. Sounds better in this circuit? I guess it would have to be tried. It won't look as cool, though.

Sheldon

[Salas]

Has been tried by the now elusive DISCO. Remote sensing would be very easy to have and battle the needed long wires too if there is a gain loop.

#26 pre amp

[quanghao]

This circuit Supply for PRe, the sound is bad,
The Simpler Simplistic Design by Salas kill it.

EX; I have been to many the design supply for PRE, but i don't like them.

I think: The Simpler Simplistic Design by Salas:

Advantages: simple design, easily made, low cost, improved sound clarity.

Cons: must use a large heat sink

[Salas]

What is bad? A CCS and a couple of OB3s? Naahh. Sounds silky. Its just lacks resolution and impact but not bad. Have you used that circuit, CCS and tube ''zeners'' exactly?

[quanghao]

Originally Posted by Salas

What is bad? A CCS and a couple of OB3s? Naahh. Sounds silky. Its just lacks resolution and impact but not bad. Have you used that circuit, CCS and tube ''zeners'' exactly?

This is not good. I think OD3 for shunt, the sound is not good!

[coolzero]

Originally Posted by Sheldon

Looks nice and simple. You can probably remove R4 also. If the current source is a robust design and can dissipate the heat from the voltage drop, the resistor isn't necessary. You show B+ at the plate choke at 260V. Are you sure that's correct? I'd expect something around 180V (two 90V VR tubes in series).

Sheldon

Checked out the KandK current sources, and your transformer. Looks like you are dropping around 200V. So, on second thought, unless you have a hefty heat sink, I might leave the resistor there to dissipate some of the heat. Not sure of the size though. How much current through the tube? Does this supply one tube or two channels? I'm curious on Rod's take. Would a dropping resistor be better before or after the CCS?

Sheldon

Your guess is right. I tried it without the dropping resistor but CCS couldn't handle the flow. This way it works. As long as I keep the dropping resistor in place, nothing much of heat generated in CCS. Tested it for couple of hours and it was quite OK with running both the channels together. Current through the tube is close to 6ma. CCS is with a 3W heat sink btw. If my memory is correct, K&K's suggestion was to use the CCS after the VR's initially. I tried both the ways and felt there is slightly more resolution to the sound with latter. What do you think?

Sorry, it was a mistake mentioning both VR tube as OB3. Actually the first tube I am using is an OA2. Voltage is around 240 volts on plate chokes.