I have always loved my Melos SHA GOLD Linestage, but it was such a design nightmare! The servo tied to heater, photoresistor volume control, wonky balance control, remote from hell, three mute circuits, crazy wiring, etc.
Also never was a Headphone guy, even tho this unit apparently was one of the best headphone amps on the planet 30 years ago...
But as a single ended input - balanced out linestage, well, it was just awesome sounding - the 6922 tube, lowish Zout, and true balanced out...
So I finally managed to trace out the linestage - with voltages...comments before I build??
Also never was a Headphone guy, even tho this unit apparently was one of the best headphone amps on the planet 30 years ago...
But as a single ended input - balanced out linestage, well, it was just awesome sounding - the 6922 tube, lowish Zout, and true balanced out...
So I finally managed to trace out the linestage - with voltages...comments before I build??
Attachments
Hi.
It seems the simplest way to achieve the gosl - and the one i love.
Look at the best possible matching of irfs, and even the tubes.
It's not good to go balanced without the best possible simmetry.
It seems the simplest way to achieve the gosl - and the one i love.
Look at the best possible matching of irfs, and even the tubes.
It's not good to go balanced without the best possible simmetry.
It's nice to see this project. Here is my thoughts about a dream preamp:
1. Multiple stages power supply regulation
2. Relay+resistor network for volume attenuation instead of mechanical pot or active volume control
3. Relay for inputs switch
4. Symmetrical left and right channel structure
1. Multiple stages power supply regulation
2. Relay+resistor network for volume attenuation instead of mechanical pot or active volume control
3. Relay for inputs switch
4. Symmetrical left and right channel structure
Yes, I saw it.Albert have you see my schematic? It's the same as the one you found but with voltages
I received your Melos preamp today, will try tomorrow.
Yes, correct - that is not a single diode (otherwise .7v would be correct not 5v). According to the horrible schematic, shows a diode giving the 5v (has to be a string of seven).
I am just thinking of putting a string of three red LED here for the 5v (1.7 x 3 = 5.1v)
I am just thinking of putting a string of three red LED here for the 5v (1.7 x 3 = 5.1v)
Attachments
Last edited:
Looks very nice if one is not prejudiced against mosfet followers, and in particular resistor loaded followers .
What is the deal with the switch? Is it for changing gain?
Adding a negative PS will allow getting rid of the input caps.
What is the deal with the switch? Is it for changing gain?
Adding a negative PS will allow getting rid of the input caps.
So, it is not a switch then. I was referring to the 100k resistor in the right triode plate load.
In this case i don't get the purpose of that resistor. Why should one side be loaded differently?
From what I remember reading regarding the Schmidt Phase Splitter, the one side of PS should have 2 - 4% different plate loading (more or less) than the other to better balance the output...but the 100k makes one side 27k vs 36k plate load difference, not exactly 2 - 4%, more like 33%...
Donno, I just read.
Edit, ya know, tha 100k might jut be in parallel with the first 18k, not both, then resulting in a 8% difference between plate load resistors
Donno, I just read.
Edit, ya know, tha 100k might jut be in parallel with the first 18k, not both, then resulting in a 8% difference between plate load resistors
Last edited:
Found it
The amplitudes can be balanced in practice by making R5 some 2% to 4% greater than R4.
Understanding Audio
The amplitudes can be balanced in practice by making R5 some 2% to 4% greater than R4.
Understanding Audio
I have opened my Melos SHA Gold line stage tried to learn this machine. She offers nice sound that made me happy but her control was not appreciated like John said. I would like to upgrade this line stage to improve her sound even better. However, there was not good reference schematic available. It was hard to identify critical parts or even circuit to be upgraded. She was not great engineering product to be honest. Many jump wires direct solder to top/bottom pads or pins on the board. I have to take many pictures then cut wires to get boards out. I wonder why Melos didn't just make a respin to ease the assembly/service efforts?
From simple look, it was not easy to observe the signal path. So I final decide to do reverse engineering for these boards to create accurate schematic. It takes long time and I am still working on it. I think it may be good to start sharing my progress because some work are making sense now.
Gentlemen may review these schematic and share opinions here. You may answer my questions. We can work together to help the each other especially SHA GOLD owners.
From simple look, it was not easy to observe the signal path. So I final decide to do reverse engineering for these boards to create accurate schematic. It takes long time and I am still working on it. I think it may be good to start sharing my progress because some work are making sense now.
Gentlemen may review these schematic and share opinions here. You may answer my questions. We can work together to help the each other especially SHA GOLD owners.
In the process of back engineering, I tried my best to read or measure the parts value, and get the ref des as much as I can. Because it was very old product, there is no R/C/IC/V etc. for ref des, but all numbers. It hasn't been done yet, but I would comment something here:
1. The INPUT- actually connects to GND
2. There is no feedback for line stage
3. There is Opamp in headphone circuit, I am not sure if it was feedback? Mega ohms resistors in the circuit with 100 amplification.
4. Why the heater be powered by headphone power?
5. I think the input caps 19 and 28 are not needed. Anyone can comment?
1. The INPUT- actually connects to GND
2. There is no feedback for line stage
3. There is Opamp in headphone circuit, I am not sure if it was feedback? Mega ohms resistors in the circuit with 100 amplification.
4. Why the heater be powered by headphone power?
5. I think the input caps 19 and 28 are not needed. Anyone can comment?
Albert - I believe you have it correct - the 475K grid leak resistor should be on the other (downstream) side of the 141 2.1k grid resistor to match the other side- I think someone had flagged this issue on a previous thread on another site.
There are a couple things that are not same as what I see on internet, but who knows which one is correct......
There are a couple things that are not same as what I see on internet, but who knows which one is correct......
In the process of back engineering, I tried my best to read or measure the parts value, and get the ref des as much as I can. Because it was very old product, there is no R/C/IC/V etc. for ref des, but all numbers. It hasn't been done yet, but I would comment something here:
1. The INPUT- actually connects to GND
2. There is no feedback for line stage
3. There is Opamp in headphone circuit, I am not sure if it was feedback? Mega ohms resistors in the circuit with 100 amplification.
4. Why the heater be powered by headphone power?
5. I think the input caps 19 and 28 are not needed. Anyone can comment?
- The opamp is there as a servo to zero the headphones output.
- The heater is used as a resistive load for the headphones output's darlington.
- Home
- Amplifiers
- Tubes / Valves
- Melos SHA GOLD Line Stage