Tubehybridmaker, I mounted the tube preamp board on the sx780 hratsink.
The amplifier PCB connects to the main board via where the old amp vas drivers were.
It kept all wiring tidy.
The sound is amazing. The tube warmth is addictive as is the 110db s/n silence.
B+ is 125v from capacitive regulator delivered from a voltage tripler. Heater comes from the 8v tap and dropping resistor.
I used 6922 tubes here
The amplifier PCB connects to the main board via where the old amp vas drivers were.
It kept all wiring tidy.
The sound is amazing. The tube warmth is addictive as is the 110db s/n silence.
B+ is 125v from capacitive regulator delivered from a voltage tripler. Heater comes from the 8v tap and dropping resistor.
I used 6922 tubes here
stocktrader200, that's a great idea. Sounds complicated, but pays off in ridiculously low noise for a tube amp!
I was reading over the power calculations earlier, and noticed an error. AI thought we were doing push-pull for some reason. Here is the correction:
Your calculations are correct for power output, but there's a key detail we need to address because this is a Class A single-ended amplifier.
In a Class A single-ended amp, the output device (the MOSFETs in your case) conducts all the time, and the current never shuts off. This affects the efficiency and voltage swing of the amplifier. While your general power output formula is correct, we need to make a few adjustments based on how Class A single-ended amps work.
### Key Adjustments for Class A Single-Ended:
1. Voltage Swing in Single-Ended Class A:
- For single-ended Class A, the voltage swing is typically less than 70% of the supply voltage. It's often closer to 50%-60% of the supply voltage, because the output device needs to stay in its active region and can't swing all the way to 0V or the full supply voltage.
- Let's assume the swing is about 50-60% for a more conservative estimate, based on the nature of single-ended Class A operation.
### Recalculating the Voltage Swing:
#### 1. 18V Supply (50% swing):
- Peak-to-peak voltage swing: 18V × 0.5 = 9V.
- RMS voltage: \( \frac{9V}{\sqrt{2}} \approx 6.36V \).
#### 2. 24V Supply (50% swing):
- Peak-to-peak voltage swing: 24V × 0.5 = 12V.
- RMS voltage: \( \frac{12V}{\sqrt{2}} \approx 8.49V \).
#### 3. 30V Supply (50% swing):
- Peak-to-peak voltage swing: 30V × 0.5 = 15V.
- RMS voltage: \( \frac{15V}{\sqrt{2}} \approx 10.61V \).
### Revised Power Output Calculations:
#### For 32Ω Load:
1. 18V Supply (RMS Voltage = 6.36V):
\[
P = \frac{6.36^2}{32} = \frac{40.47}{32} \approx 1.26W
\]
2. 24V Supply (RMS Voltage = 8.49V):
\[
P = \frac{8.49^2}{32} = \frac{72.05}{32} \approx 2.25W
\]
3. 30V Supply (RMS Voltage = 10.61V):
\[
P = \frac{10.61^2}{32} = \frac{112.58}{32} \approx 3.52W
\]
#### For 300Ω Load:
1. 18V Supply (RMS Voltage = 6.36V):
\[
P = \frac{6.36^2}{300} = \frac{40.47}{300} \approx 0.13W \text{ (130mW)}
\]
2. 24V Supply (RMS Voltage = 8.49V):
\[
P = \frac{8.49^2}{300} = \frac{72.05}{300} \approx 0.24W \text{ (240mW)}
\]
3. 30V Supply (RMS Voltage = 10.61V):
\[
P = \frac{10.61^2}{300} = \frac{112.58}{300} \approx 0.38W \text{ (380mW)}
\]
---
### Summary of Revised Power Output Estimates for Class A Single-Ended:
| Supply Voltage | Power Output (32Ω) | Power Output (300Ω) |
|--------------------|------------------------|-------------------------|
| 18V DC | 1.26W | 130mW |
| 24V DC | 2.25W | 240mW |
| 30V DC | 3.52W | 380mW |
### Key Differences:
---
### Conclusion:
Your original formula and basic process were correct, but for Class A single-ended operation, the voltage swing is typically smaller, which results in slightly lower power output than you originally calculated. With these adjustments, you can now confidently estimate the actual power output based on your design!
Let me know if you need further clarification or have any additional questions!
I was reading over the power calculations earlier, and noticed an error. AI thought we were doing push-pull for some reason. Here is the correction:
Your calculations are correct for power output, but there's a key detail we need to address because this is a Class A single-ended amplifier.
In a Class A single-ended amp, the output device (the MOSFETs in your case) conducts all the time, and the current never shuts off. This affects the efficiency and voltage swing of the amplifier. While your general power output formula is correct, we need to make a few adjustments based on how Class A single-ended amps work.
### Key Adjustments for Class A Single-Ended:
1. Voltage Swing in Single-Ended Class A:
- For single-ended Class A, the voltage swing is typically less than 70% of the supply voltage. It's often closer to 50%-60% of the supply voltage, because the output device needs to stay in its active region and can't swing all the way to 0V or the full supply voltage.
- Let's assume the swing is about 50-60% for a more conservative estimate, based on the nature of single-ended Class A operation.
### Recalculating the Voltage Swing:
#### 1. 18V Supply (50% swing):
- Peak-to-peak voltage swing: 18V × 0.5 = 9V.
- RMS voltage: \( \frac{9V}{\sqrt{2}} \approx 6.36V \).
#### 2. 24V Supply (50% swing):
- Peak-to-peak voltage swing: 24V × 0.5 = 12V.
- RMS voltage: \( \frac{12V}{\sqrt{2}} \approx 8.49V \).
#### 3. 30V Supply (50% swing):
- Peak-to-peak voltage swing: 30V × 0.5 = 15V.
- RMS voltage: \( \frac{15V}{\sqrt{2}} \approx 10.61V \).
### Revised Power Output Calculations:
#### For 32Ω Load:
1. 18V Supply (RMS Voltage = 6.36V):
\[
P = \frac{6.36^2}{32} = \frac{40.47}{32} \approx 1.26W
\]
2. 24V Supply (RMS Voltage = 8.49V):
\[
P = \frac{8.49^2}{32} = \frac{72.05}{32} \approx 2.25W
\]
3. 30V Supply (RMS Voltage = 10.61V):
\[
P = \frac{10.61^2}{32} = \frac{112.58}{32} \approx 3.52W
\]
#### For 300Ω Load:
1. 18V Supply (RMS Voltage = 6.36V):
\[
P = \frac{6.36^2}{300} = \frac{40.47}{300} \approx 0.13W \text{ (130mW)}
\]
2. 24V Supply (RMS Voltage = 8.49V):
\[
P = \frac{8.49^2}{300} = \frac{72.05}{300} \approx 0.24W \text{ (240mW)}
\]
3. 30V Supply (RMS Voltage = 10.61V):
\[
P = \frac{10.61^2}{300} = \frac{112.58}{300} \approx 0.38W \text{ (380mW)}
\]
---
### Summary of Revised Power Output Estimates for Class A Single-Ended:
| Supply Voltage | Power Output (32Ω) | Power Output (300Ω) |
|--------------------|------------------------|-------------------------|
| 18V DC | 1.26W | 130mW |
| 24V DC | 2.25W | 240mW |
| 30V DC | 3.52W | 380mW |
### Key Differences:
- The power output is lower than in your original calculations because Class A single-ended amps typically have less voltage swing (about 50-60%) compared to a push-pull or Class B amplifier.
- This makes 24V or 30V DC a much better choice, especially for driving high-impedance headphones (300Ω), where the power output increases significantly with higher supply voltage.
---
### Conclusion:
Your original formula and basic process were correct, but for Class A single-ended operation, the voltage swing is typically smaller, which results in slightly lower power output than you originally calculated. With these adjustments, you can now confidently estimate the actual power output based on your design!
Let me know if you need further clarification or have any additional questions!
I am convinced that we are talking to some rather low end AI from these posts. I do not feel it is necessary to be involved any longer in this thread.
Your power supply schematics is incorrect, it will not work and short the transformer secondaries. I think AI has not evolved enough to come with something useful. It sure does not know how to connect anything. This project will cost you an arm and a leg to get working. I am sorry but AI needs to learn to think, not just spew out things gathered in the ether. Why does it even bother with schematics, it should produce a populated PCB for the user just to build and be done with it. Who needs engineers or technicians if we have AI to stuff everything up.
What is wrong with the schematic? I am human, and I will work until it is correct. I am here for guidance. Please be specific, and I will make changes.
From the AI:
I’m sorry to hear about the feedback, and I appreciate that you’ve taken the time to share it. Let’s go over what might have gone wrong with the power supply section and correct it step by step.
### Key Issues in the Power Supply Section
The most common mistake when designing power supplies—especially with multiple rectifiers and transformer secondaries—is incorrectly connecting the AC inputs of the rectifiers, which can cause short circuits in the transformer secondaries.
### Let’s Address the Core Problem:
Here’s what we need to do:
### Step 1: Transformer Winding Isolation
If the two 150V windings of your transformer are isolated secondaries (meaning they are independent of each other), you should:
1. Ensure each 150V secondary winding has its own rectifier.
2. Make sure there are no direct connections between the secondaries unless the datasheet specifies they are meant to be connected in a particular way (such as in parallel or series).
### Step 2: Rectifier Connections
### Updated Instructions for the Power Supply Section:
#### 1. Transformer Setup:
#### 2. Full-Wave Bridge Rectifiers:
- Each rectifier should have:
- Two AC inputs: These will connect to the two ends of each 150V winding.
- + DC output: This will connect to the filter capacitor (100µF, 400V) for smoothing the output voltage.
- − DC output: This will connect to ground (or common ground after filtering).
#### Example of Correct Rectifier Connections:
Important: Each rectifier’s AC inputs must only connect to its own 150V secondary winding. Do not mix the connections between windings or rectifiers.
---
### Next Steps:
1. Review your schematic to ensure each rectifier is connected to its own independent 150V winding from the transformer.
2. Check for any connections between the secondary windings, which could be causing shorts.
3. Once you’ve corrected this, let me know, and we can double-check everything together.
If you’re unsure about the connections, feel free to take a picture of your schematic or share the key parts of it, and I’ll help you troubleshoot the setup.
Again, I appreciate your patience, and I’m here to make sure we get this right!
--------------------
From me:
Yeah, these are independent 150V windings, and I'm pretty sure I drew them correctly. I sure could use some help. I don't know what I did wrong.
I probably crossed some wires in my drawing. The purpose of the drawing is just so that I understand the connections well enough to put the schematic into kicad. But any wire crossings will not be duplicated in kicad. Sorry if I drew it too messily. I was just trying to get something on paper, to make some progress.
I'm not going to be discouraged. I will build this amp with or without your help.
I’m sorry to hear about the feedback, and I appreciate that you’ve taken the time to share it. Let’s go over what might have gone wrong with the power supply section and correct it step by step.
### Key Issues in the Power Supply Section
The most common mistake when designing power supplies—especially with multiple rectifiers and transformer secondaries—is incorrectly connecting the AC inputs of the rectifiers, which can cause short circuits in the transformer secondaries.
### Let’s Address the Core Problem:
- Bridge rectifiers need to be connected carefully to the transformer secondary windings to avoid creating shorts.
- You must isolate the secondaries properly if you are using two separate rectifiers for two different supplies (like the tube B+ supply and the MOSFET supply).
Here’s what we need to do:
### Step 1: Transformer Winding Isolation
If the two 150V windings of your transformer are isolated secondaries (meaning they are independent of each other), you should:
1. Ensure each 150V secondary winding has its own rectifier.
2. Make sure there are no direct connections between the secondaries unless the datasheet specifies they are meant to be connected in a particular way (such as in parallel or series).
### Step 2: Rectifier Connections
- For each rectifier, you will connect the two AC inputs directly to the two ends of the secondary winding (one secondary for each rectifier).
- The DC outputs of the rectifiers (positive and ground) should remain separate unless you're combining them at a common ground after regulation or filtering.
### Updated Instructions for the Power Supply Section:
#### 1. Transformer Setup:
- Ensure you have two completely separate 150V windings from the transformer, and each winding will power its own rectifier.
- There should be no connection between the windings unless the transformer datasheet specifies otherwise.
#### 2. Full-Wave Bridge Rectifiers:
- Each rectifier should have:
- Two AC inputs: These will connect to the two ends of each 150V winding.
- + DC output: This will connect to the filter capacitor (100µF, 400V) for smoothing the output voltage.
- − DC output: This will connect to ground (or common ground after filtering).
#### Example of Correct Rectifier Connections:
Code:
For Rectifier 1 (Tube B+ Supply):
150V Secondary (AC)
|
+-------->|----+ <-- + DC output (to filter capacitor)
| |
--+ +-- <-- - DC output (to ground)
| |
+-------->|----+
|
150V Secondary (AC)Important: Each rectifier’s AC inputs must only connect to its own 150V secondary winding. Do not mix the connections between windings or rectifiers.
---
### Next Steps:
1. Review your schematic to ensure each rectifier is connected to its own independent 150V winding from the transformer.
2. Check for any connections between the secondary windings, which could be causing shorts.
3. Once you’ve corrected this, let me know, and we can double-check everything together.
If you’re unsure about the connections, feel free to take a picture of your schematic or share the key parts of it, and I’ll help you troubleshoot the setup.
Again, I appreciate your patience, and I’m here to make sure we get this right!
--------------------
From me:
Yeah, these are independent 150V windings, and I'm pretty sure I drew them correctly. I sure could use some help. I don't know what I did wrong.
I probably crossed some wires in my drawing. The purpose of the drawing is just so that I understand the connections well enough to put the schematic into kicad. But any wire crossings will not be duplicated in kicad. Sorry if I drew it too messily. I was just trying to get something on paper, to make some progress.
I'm not going to be discouraged. I will build this amp with or without your help.
Last edited:
Each 150V winding has its own rectifier and capacitors. Then the one for the tube goes off to the tube, and the one for the mosfets goes through a voltage regulator and its resistors and caps before heading over to that section.
The schematic is incomplete. The section going to B+ is not grounded. The wire going on to B+ is not labelled. I will finish that now, but I won't have time today to keep working on this, unfortunately. I will begin the tube section tomorrow, unless you have specific things I need to change about my power section, other than completing the connections to ground and the tube section, which I should have included in the schematic. I didn't think leaving that off would be such a sin that people would no longer have any interest in helping me. I am sorry.
current follows in the direction of the arrow. your bridge seems like causing a short circuit of the transformer.
Last edited:
It's quite possible that I didn't draw the bridge correctly, with the arrows in the right direction. However, there should be a symbol already in the kicad libraries for a full-wave bridge rectifier, so I'll be sure I find one that has the arrows in the right direction. I think I will stop uploading my hand-drawn sketches, since my ineptitude with drawing them seems to be an issue. I will do the sketches first, then put them into kicad, and then upload the kicad schematics for each section as I complete their rough drafts.
And please do bear in mind that this is a rough first attempt. I will learn and get better at this. I appreciate your patience and instruction.
And please do bear in mind that this is a rough first attempt. I will learn and get better at this. I appreciate your patience and instruction.
Me thinks, AI is not as smart as you think it is. Seriously though, every aspect of this design you have presented so far will either not work at all or will not work as you had hoped. This forum is an immense database of knowledge with some extremely intelligent and helpful members... if one is willing to put in the "leg work" and learn. If you continue down this path of relying on AI you're going to be in for a whole lot of frustration, disappointment and wasted time and money. Trying to go from not knowing what a diode or capacitor is to designing a working circuit is a huge undertaking. Why not start with building a known working design. A few hybrid head amps that come to mind are Borbely's hybrid amp, Pete Millet's Starving Student and its variants, Pete Millet's Nutube amp.
Electronic schematics are not pictures of things, each have an exact meaning and purpose. Drawing it up-side-down means something totally different. You read a schematic like you read a text book, having to understand each part and its function exactly. Getting a "picture" of something is not the object, knowing how it fits into the grand scheme of things are. Besides it technical constraints, it takes engineers more than a weekend to learn what it is al about and a lot of mathematics to support what they intend.
Have you ever read Wireless World, Practical Electronics, or similar magazines and tried to understand the designers objectives?
I am starting to think that simulators are evil things that turns ordinary handymen into design engineers.
You first need your concept, then design and the simulator is used to refine your design without having to physically build it and tinker with it. If you have no idea how to design, simulators are a total waste of time.
Have you ever read Wireless World, Practical Electronics, or similar magazines and tried to understand the designers objectives?
I am starting to think that simulators are evil things that turns ordinary handymen into design engineers.
You first need your concept, then design and the simulator is used to refine your design without having to physically build it and tinker with it. If you have no idea how to design, simulators are a total waste of time.
Last edited:
I just want the particular amp that I want, and I'm willing to do what it takes to get it by the most direct route possible. I learn by doing. I'm pig-headed about building this amp in the immediate future. I am not at all pig-headed about how to do that. Verifying and correcting AI's suggested design is why I am here, hoping some of you will volunteer your time and expertise to going with me on this journey. If you have specific areas of the design that you would like to point out as incorrect, I would be very glad to learn about that.
My understanding as of right now, from AI and from looking at the datasheet, is that the 150V transformer white goes into the ac + input of the bridge rectifier, the yellow of that same winding goes into the ac - of that rectifier. I will arrange the diodes as directed. I will verify that arrangement with people here and what I can find online. I will do that so that I can be sure I find the right schematic symbol, and so that I will build it correctly. From the rectifier, the dc - out goes to the negative end of the 100microFarad, 100V capacitor, and on to ground. The dc+ out goes to the positive end of the cap, and on to power B+.
Is that incorrect?
I cannot recall in the 50+ years of engineering design anything, that I started out with a bill of materials, down to the nut and bolt. Good heavens, design must have surpassed and left me in the dark a long, long time ago.
My personal advice is to fire AI, he/she is useless. A raw engineering student on his first day of work would be a pleasure in comparison. We are at post 76 and we are still only at the recitfier, how long are you willing to wade through the masses of knowledge required to achieve what you set out to do, you will be better of getting you engineering degree and then start to wonder how to proceed understanding things. You need first principles or grass root knowledge first. I propose the same as @[U]thirdicomplex[/U] where to start and progress from there.
Last edited:
Well, I am going to proceed with the specifics of this plan until someone suggests specific changes to make to it. But I approach this as a consumer, not as an engineer. I want the specific amp that I want, and it does not exist right now. I could get something just as good or better, but it would cost thousands. If I'm going to spend thousands on an amp anyway, I might as well learn what I can and end up with one that is truly my own. I have the time, and am willing to learn. But I must proceed with this project.
Ok, I gotta take care of less fun stuff today. I'll check in tonight, Lord willing. Thank you for your kind attention and well-meaning and sincere advice. I will take all your specific suggestions with gratitude.
I would hate my doctor tell me "please do bear in mind that this is a rough first attempt. I will learn and get better at this. I appreciate your patience" when they push me into the theater 
Well, Nico Ras, that would seem like an occasion for mortal terror. I can assure you, this is not such an occasion. No one will plug this amp into a wall until a certified electrician has decided it's ready.
I'll be back in a day or two, however long it takes me to get the power section into kicad some kinda way. You won't like it, I know, but I'm still going to post it here when I have it. Hopefully this weekend, but we'll see how it goes.
I'll be back in a day or two, however long it takes me to get the power section into kicad some kinda way. You won't like it, I know, but I'm still going to post it here when I have it. Hopefully this weekend, but we'll see how it goes.