#20
Nice. Long time since I saw anything about Lazarus. It was very good but never hit a bigger market. Case work was too bad.
It have some distinctive trademarks from Allen Wright like cascoded 6922 and CCS loads in the power supply.
No doubt an Allen Wright design...
Since it is an reverse engineered schematic it's easy to miss something and I'm quite sure there should be a resistor between sources of
the driver mosfets. Around 75 ohm would give approx. give a 20mA current through the driver, which will be sufficient to drive the output mosfets.
The P mosfets is not correctly drawn. See green correction.
Nice. Long time since I saw anything about Lazarus. It was very good but never hit a bigger market. Case work was too bad.
It have some distinctive trademarks from Allen Wright like cascoded 6922 and CCS loads in the power supply.
No doubt an Allen Wright design...
Since it is an reverse engineered schematic it's easy to miss something and I'm quite sure there should be a resistor between sources of
the driver mosfets. Around 75 ohm would give approx. give a 20mA current through the driver, which will be sufficient to drive the output mosfets.
The P mosfets is not correctly drawn. See green correction.
Yes. It’s just a constant current source. There are multiple ways of doing that with tubes, FETs and bipolar transistors. Two BJTs and a couple of resistors will do the trick for less than $1 in component cost. I don’t know much about tubes, but I fail to see what a tube would do for sound in that position.
Here’s what I’m using in the build that’s on my bench right now. The current is roughly 0.7/R19, which ends up being about 12mA in my case. The capacitor is optional and is there to stabilize the current. You may need different transistors based on what voltage you’re operating at.
You did not understand because you dont have how, and you are to busy creating tensed situations, sorry but you will never learn how audio works ! Good luck at what you do 🙂 !
There are plenty of nice amplifiers you can build using those components. As others have pointed out, the one you’re referring to in post #1 makes no sense.
I’ve been thinking of tipping my toes into the tube marsh, and I think I’d start with a preamp. No need for crazy currents or voltages and (depending on who you listen to) that seems to be the best place to introduce the tube sound.
Flez, thanks for the corrections - I thought I had accounted for all the on-board components. P's / N's - all the same to me!!!???
Yes Jacques, a 741 - used as a servo...
Some of the stuff on the drawing was the difference between the 1 and the A. I had both and converted the 1 to an A with the addition of my FET current source.
Yeah and the edges on the heatsinks WILL draw blood! Also added a pair of (series) fans on the top - resistor also in series to slow them down (to inaudible) bypassed by a thermal switch to short and speed them up above maybe 75 degrees. Heats the room nicely on a cold day!
Charles
Yes Jacques, a 741 - used as a servo...
Some of the stuff on the drawing was the difference between the 1 and the A. I had both and converted the 1 to an A with the addition of my FET current source.
Yeah and the edges on the heatsinks WILL draw blood! Also added a pair of (series) fans on the top - resistor also in series to slow them down (to inaudible) bypassed by a thermal switch to short and speed them up above maybe 75 degrees. Heats the room nicely on a cold day!
Charles
agree for the amplifier posted in OP
use a normal current source. no tube needed, DC so dont hear it.
Being a cheesy cheapo amp is what makes it " warm" aka junk distortion
adjusting bias as shown, would seem to make it jump like hell.
Needs source resistors in output section. Bias could be measured more effectively like normal amps.
Not the wild guess method.
with 86 volts could make relatively high power, use better /larger package mosfet.
use a normal current source. no tube needed, DC so dont hear it.
Being a cheesy cheapo amp is what makes it " warm" aka junk distortion
adjusting bias as shown, would seem to make it jump like hell.
Needs source resistors in output section. Bias could be measured more effectively like normal amps.
Not the wild guess method.
with 86 volts could make relatively high power, use better /larger package mosfet.
Note that the CCS is bootstrapped, without which you lose the FET vto output swing, not that it matters much with 86V VCC. Yes, this maintain the voltage on the Tube/CCS but that is also true of a resistor (~1K). Perhaps LV's two wire improved CCS would be a good choice. I'm tempted to post a better circuit using mostly the same parts, but I'll let someone else do that?
Note that the CCS that njswede posted can oscillate, especially if the two transistors are the same part. Recommended is a cap across Q9 or a base resistor.
Note that the CCS that njswede posted can oscillate, especially if the two transistors are the same part. Recommended is a cap across Q9 or a base resistor.
Doesn’t C5 help with that? That’s the way I’ve seen it drawn in most of the literature, including Self’s book.
#30
Doubtful, but not impossible.
The way C5 is connected to the power supply, it really could be connected directly to ground in parallell to R16.
Now it lowers the impedance of the current flow to 4,7kohm from 2 x 4,7kohm. Not likely that it will prevent oscillation, but not impossible.
Strange things have happend before...
I think the purpouse of C5 is to stabalize and lower potential noise if the power supply isn't very good and can be influenced by the current drawn from it.
Impedances around Q9 and Q10 is extremely low so I doubt it can have any significant impact at all.
I have never experienced a CCS of this type to oscillate, but I can understand the reasoning behind it.
A problem that occurs with bigger transistors with higher Miller capacitances?
Can't remember that I have used it on anything bigger than BD139/BD140 though.
Doubtful, but not impossible.
The way C5 is connected to the power supply, it really could be connected directly to ground in parallell to R16.
Now it lowers the impedance of the current flow to 4,7kohm from 2 x 4,7kohm. Not likely that it will prevent oscillation, but not impossible.
Strange things have happend before...
I think the purpouse of C5 is to stabalize and lower potential noise if the power supply isn't very good and can be influenced by the current drawn from it.
Impedances around Q9 and Q10 is extremely low so I doubt it can have any significant impact at all.
I have never experienced a CCS of this type to oscillate, but I can understand the reasoning behind it.
A problem that occurs with bigger transistors with higher Miller capacitances?
Can't remember that I have used it on anything bigger than BD139/BD140 though.
Nope. C5 filters 120/100Hz supply noise, but it is outside the Q9+Q10 feedback loop. As long as Q10 is slower than Q9, ie different parts, there will probably be no problem.