Putting regulators on the main board means extra work for Jens to change the extended Leach board, or more work if he will do an entire new pcb for the Super Duper.
Those who desire to build the Double Barrel with regulation( and not care much about pcb size) may favor this.
Others, who build the Leach as is, end up with boards that have empty space they pay for, and a larger board.
I recall people asking for smaller !
Using perforated boards for regulation is easier.
Just a power separation possibility at the main board would be sufficient ( or a few added solder pads to the layout)
Others may produce a proper regulator pcb themselves.
An option could even be to do a group buy universal regulator pcb.
For those who like a designed regulator pcb and do not have the skills i can make a couple extra (0.1" thickness if desired)
Those who desire to build the Double Barrel with regulation( and not care much about pcb size) may favor this.
Others, who build the Leach as is, end up with boards that have empty space they pay for, and a larger board.
I recall people asking for smaller !
Using perforated boards for regulation is easier.
Just a power separation possibility at the main board would be sufficient ( or a few added solder pads to the layout)
Others may produce a proper regulator pcb themselves.
An option could even be to do a group buy universal regulator pcb.
For those who like a designed regulator pcb and do not have the skills i can make a couple extra (0.1" thickness if desired)
Jacco, Yes Yes, we are with you. I already have a diy power amp and would wish to have regulation on that too, so you can count me in.
However how can you have regulation on say 60 v when the maximum regulated voltage of the devices you quoted are less than 40v?
I guess the extra current capacity can be obtained with extra power transistors yes?
However how can you have regulation on say 60 v when the maximum regulated voltage of the devices you quoted are less than 40v?
I guess the extra current capacity can be obtained with extra power transistors yes?
The LM's are not specified for voltage level but voltage difference.
A LM337 can handle a 40 volt difference between voltage In and voltage Out.
If the reference pin of the LM has 40 volts on the ref pin it can have 80 volts coming in and deliver 40 volts if you like it to do that.
Even with a voltage drop of 40 volts an LM regulator can deliver 150 mA.
By placing something between zero ground and the reference pin the LM is tricked in believing that reference is still 0 volts.
Like you said, by placing a high power device on the output of a regulator like the LM337 the regulator delivers the voltage, the power transistor the current.
Rod Elliot's regulator for output stages is an example of that, i forgot the name he uses again(blame my age)
Actually, a regulator like the LM is sort of an amplifier.
http://aes.sdsu.edu/documents/LM337-D.pdf
As the data sheet of the LM337 shows, minimal ripple reduction is 60 dB.
-60 dB is 0.001 times the value of voltage ripple unregulated.
Suppose you have 1 volt ripple on the powerline without a regulator.
After the regulator there is only 0.001 volts left, and that can be lowered even further by placing a cap to build a filter.
Open loop voltage gain of an audio amplifier can be tremendous, depending on design.
That means a very small voltage, say 1mV, on the entrance could deliver many many volts on the output.
To amplify voltage correctly one might suppose that the voltage a transistor works on needs to be very stable.
You can also imagine that the output stage of a 300 watt amplifier can draw so much current at one time that it sucks power from your power supply caps.
They then need to be recharged by the power transformer, which leads to high power supply ripple.
If the front stage of an amplifier is connected directly to the output power supply it gets the same ripple on its line.
Common in amplifier designs is placing a ripple filter on supply lines for front ends, RC filters.
In a way that is a regulator, a simple one.
A LM337 can handle a 40 volt difference between voltage In and voltage Out.
If the reference pin of the LM has 40 volts on the ref pin it can have 80 volts coming in and deliver 40 volts if you like it to do that.
Even with a voltage drop of 40 volts an LM regulator can deliver 150 mA.
By placing something between zero ground and the reference pin the LM is tricked in believing that reference is still 0 volts.
Like you said, by placing a high power device on the output of a regulator like the LM337 the regulator delivers the voltage, the power transistor the current.
Rod Elliot's regulator for output stages is an example of that, i forgot the name he uses again(blame my age)
Actually, a regulator like the LM is sort of an amplifier.
http://aes.sdsu.edu/documents/LM337-D.pdf
As the data sheet of the LM337 shows, minimal ripple reduction is 60 dB.
-60 dB is 0.001 times the value of voltage ripple unregulated.
Suppose you have 1 volt ripple on the powerline without a regulator.
After the regulator there is only 0.001 volts left, and that can be lowered even further by placing a cap to build a filter.
Open loop voltage gain of an audio amplifier can be tremendous, depending on design.
That means a very small voltage, say 1mV, on the entrance could deliver many many volts on the output.
To amplify voltage correctly one might suppose that the voltage a transistor works on needs to be very stable.
You can also imagine that the output stage of a 300 watt amplifier can draw so much current at one time that it sucks power from your power supply caps.
They then need to be recharged by the power transformer, which leads to high power supply ripple.
If the front stage of an amplifier is connected directly to the output power supply it gets the same ripple on its line.
Common in amplifier designs is placing a ripple filter on supply lines for front ends, RC filters.
In a way that is a regulator, a simple one.
As an outside observer, it would make much more sense to me to start another thread for a high power regulated PSU group buy, rather than build it into the Leach pcb. If you want, I can split out the regulator posts here to a new thread. 🙂
Hello Jacco, in your comments regarding the rectified voltage off a transformer with 55VAC secondaries with 120VAC input, I calculate the maximum theoretical rail voltage to be +/- 83 VDC. Even though the rectifed voltage is 166 VDC the capacitors only see half that voltage in a +/- supply. I n this case 100 VDC rating on the caps should be fine, why do you say you would need 125 VDC rated caps minimum?
Regards
Anthony
Regards
Anthony
Hi Anthony,
the 125 volts refers to the 15 volt and 55 volt secondaries put in series on Terry's toroid.
15 + 55 makes 70VAC
70 VAC = 99 VDC
add 5 % toroid regulation makes 104 VDC
add 10 % surge voltage from the electric company makes 114 VDC
Put 120 volts on a toroid designed for 110 makes 125 VDC.
A toroid may be labelled 120, it could still be a 110 model.
Nice Aleph-X job, btw, saw the pics somewhere.
the 125 volts refers to the 15 volt and 55 volt secondaries put in series on Terry's toroid.
15 + 55 makes 70VAC
70 VAC = 99 VDC
add 5 % toroid regulation makes 104 VDC
add 10 % surge voltage from the electric company makes 114 VDC
Put 120 volts on a toroid designed for 110 makes 125 VDC.
A toroid may be labelled 120, it could still be a 110 model.
Nice Aleph-X job, btw, saw the pics somewhere.
Make the PCB so that a regulated power supply can be daughtered onto the amp board with very little wiring. That way there is the option of regulation or no regulation .
So provide some empty spaces and holes so that standoffs can be added to stack another PCB onto it. Better yet, some headers and pins!
That would allow an easy way to evolve an amp without tearing it apart.
I wish this had been done on the standard Leach redesign.
So provide some empty spaces and holes so that standoffs can be added to stack another PCB onto it. Better yet, some headers and pins!
That would allow an easy way to evolve an amp without tearing it apart.
I wish this had been done on the standard Leach redesign.
Question is where on the main board a regulator pcb should piggyback?
Any suggestion, with the regular Leach board as reference?
Any suggestion, with the regular Leach board as reference?
I'm thinking the attachment points should be four spots. Negative and Positive of voltage and current gain. That way, if someone wants to just do front end only, it's there. If you want fully regulated it's there too.
Here's a more general but I think close to on-topic question.
With the regular leach design (standard components) does anyone think there would be a problem or benefit to running the driver board at 63.1V and the outputs at 58.4V?
I had been planning on running the whole amp at 58.4 (close to recommended voltage), but my transformer has a second set of taps (previously used to seperately power a driver board at 63.1). I could simply use them (unregulated).
Should I bother? I would need to rework and re-etch my driver boards, though it would also give me the opportunity to make another layout improvement I spotted since making the last ones, so not a total wash.
With the regular leach design (standard components) does anyone think there would be a problem or benefit to running the driver board at 63.1V and the outputs at 58.4V?
I had been planning on running the whole amp at 58.4 (close to recommended voltage), but my transformer has a second set of taps (previously used to seperately power a driver board at 63.1). I could simply use them (unregulated).
Should I bother? I would need to rework and re-etch my driver boards, though it would also give me the opportunity to make another layout improvement I spotted since making the last ones, so not a total wash.
WorkingAtHome said:
If you choose to do this, you will have to change the values of R13 and R14 in the driver board. Here is a quote from the webpage:
"If you do not use the specified power supply voltages, you can calculate the values for R13 and R14 from the formula R13 = R14 = (V - 40)/8.2, where V is the power supply voltage. For example, for V = 58 V, the formula gives R13 = R14 = 2.2 kohm. Use the nearest 5% resistor value. (For the Ver. 4.3 amplifier, calculate the values for R13 and R14 from the formula R13 = R14 = (V - 38.2)/5.42, where V is the power supply voltage. For example, for V = 57.7 V, the formula gives R13 = R14 = 3.6 kohm.)"
"Any suggestion, with the regular Leach board as reference?"
R32, R33 is where the split needs to be made.
When using a high voltage tier for the front end, a Baker Clamp needs to be added between Q12, 14 and Q13, 15
The stock Leach was designed so as not to be able to saturate any of the output stage. Diode clamps will restore this.
A simple emiter follower with a zener in the base is all that is really needed to drive the front end, Electrocompaniet did it this way on their Ampliwire series. The load current does not change, so why is anything fancy needed?
R32, R33 is where the split needs to be made.
When using a high voltage tier for the front end, a Baker Clamp needs to be added between Q12, 14 and Q13, 15
The stock Leach was designed so as not to be able to saturate any of the output stage. Diode clamps will restore this.
A simple emiter follower with a zener in the base is all that is really needed to drive the front end, Electrocompaniet did it this way on their Ampliwire series. The load current does not change, so why is anything fancy needed?
djk said:
Exactly.
Make it cascoded and add one more device with a trimpot and you can take me to the ball.
Hello
I hope someone will draw a temporary schematic before I start any serious work on the PCB.
I have no idea what the end result of all this is going to be, so if someone would be kind enough to sort of make a list or drawing that explains the changes to me. If not I will have to make a lot of posts and schematics before the layout work can begin,
Thank you
\Jens
I hope someone will draw a temporary schematic before I start any serious work on the PCB.
I have no idea what the end result of all this is going to be, so if someone would be kind enough to sort of make a list or drawing that explains the changes to me. If not I will have to make a lot of posts and schematics before the layout work can begin,
Thank you
\Jens
For some reason, cascoding the front does not do the trick guys. Been there done that. I know it doesn't make "intellectual sense". I can't explain. Oh well, another one of those amps again.
Yup, The Prof. designed a pretty good amp. Stood the test of time. Improveing it is a real challenge.
Prosit
Prosit
- Status
- Not open for further replies.