Does anyone know how the metal back of a TO3-PL, 2SA1943 and 2SC5200 transistors, is bonded to the plastic package?
Metal and plastic do not bond together which means a mechanical bond should be in use. I had to enlarge the mounting holes of these transistors or buy a new heatsink; I chose to enlarge the mounting holes. My concern driving me to ask this question, is whether a 4mm stainless steel screw will make contact with the metal back of the transistors. Direct contact can be easily inspected with a lense.
I want to make sure there is enough isolation between the heatsink and transistors. The mounting screws are in electrical contact with the heatsink.
With the help of a megger this problem would have been solved without too much hassle, but I do not have such an instrument.
Metal and plastic do not bond together which means a mechanical bond should be in use. I had to enlarge the mounting holes of these transistors or buy a new heatsink; I chose to enlarge the mounting holes. My concern driving me to ask this question, is whether a 4mm stainless steel screw will make contact with the metal back of the transistors. Direct contact can be easily inspected with a lense.
I want to make sure there is enough isolation between the heatsink and transistors. The mounting screws are in electrical contact with the heatsink.
With the help of a megger this problem would have been solved without too much hassle, but I do not have such an instrument.
I tried to drill 2.5mm mounting holes in another heatsink but ended up dissatisfied. Holding an electric drill with the hands is not good for decent looking holes that are at right angles to the surface of the heatsink. The holes I drilled looked more like made by a drunkard after drinking 2 bottles of Whisky! So, again, I had to revert to wider holes to get more acceptable work.
The power transistors were originally fixed with pressure clips, two transistors per clip: I do not want that.
The power transistors were originally fixed with pressure clips, two transistors per clip: I do not want that.
You need a centre punch a drill press and stand, some machine oil to lubricate your hole so the drill bit does not over heat and a metal tap to create the right size thread. If you don't have these tools and the skills to use these then it may be better to get a metal work shop to do the job for you.
If you don't use a centre punch to make the initial indent the drill bit will skate around on the metal with the rotational force applied by a power tool freehand more -so depending on the speed. Tapping a thread requires patience and has to be done slowly clearing swarf regularly by working back and forth as you go. You can start with a small drill bit and work up in diameter until you get to that which leaves enough for the spiral cut to be made by the cutting edges of the tap.
Aluminium loses its strength very quickly as temperature is increased. At about 150C, a temperature easily reached in a household oven or over a kitchen burner, its strength drops by a considerable amount. This may enable me to make the holes more perpendicular to the surface.
Ladies and gentlement, this is the hardest and most cumbersome of the entire project. Sincerily, I cannot boast with my failures...
My aim is to revive that amplifier with a grounding scheme that makes more sense. However, I lack the tools, and to ask other people who have the tools, it would mean, having to wait for several months only for this. No one will lose money doing something like this if other orders are less cumbersome to do and earn one more money.
Ladies and gentlement, this is the hardest and most cumbersome of the entire project. Sincerily, I cannot boast with my failures...
My aim is to revive that amplifier with a grounding scheme that makes more sense. However, I lack the tools, and to ask other people who have the tools, it would mean, having to wait for several months only for this. No one will lose money doing something like this if other orders are less cumbersome to do and earn one more money.
Some Questions:
a) Is a simple pre-driver amplifier circuit possible to have a very low distortion and high stability? [ distortion < 0.01%; thermal quiescent current stability, no self oscillations at all output levels ]
b) I suspect my never-ending quest to devise a pre-driver circuit with the above characteristics, and yet having a very low component count, is an impossible feat. Am I right?
c) Is it possible to use a very simple circuit to drive the drivers of a large power amplifier?
I tried the above until I couldn't stand more my aching tired head.
If I am running after the wind, please, reply explaining why, but please do it civilly.
a) Is a simple pre-driver amplifier circuit possible to have a very low distortion and high stability? [ distortion < 0.01%; thermal quiescent current stability, no self oscillations at all output levels ]
b) I suspect my never-ending quest to devise a pre-driver circuit with the above characteristics, and yet having a very low component count, is an impossible feat. Am I right?
c) Is it possible to use a very simple circuit to drive the drivers of a large power amplifier?
I tried the above until I couldn't stand more my aching tired head.
If I am running after the wind, please, reply explaining why, but please do it civilly.
I thought you were happy enough to build a prototype of your design a few days ago but then you keep moving the design goals.
High power high voltage amps are never straightforward and I know you were trying to make your design retain an existing output stage configuration which again is another limiting factor.
Sometimes you can be better to start again with a clean sheet and explore different topologies (and their problems) and hopefully their solutions.
Success always stands on the shoulders of failure (unfortunately).
but then you keep moving the design goals.High power high voltage amps are never straightforward and I know you were trying to make your design retain an existing output stage configuration which again is another limiting factor.
Sometimes you can be better to start again with a clean sheet and explore different topologies (and their problems
) and hopefully their solutions.Success always stands on the shoulders of failure (unfortunately).
That means expecting a very simple pre-driver circuit is contrary to amplifier design experience.Mooly said:
I saved all circuits: those that performed well in LTSpice are saved. I can refer to those.
I am modifying the original amplifier circuits so that the output transistors are isolated from ground. This will allow me to use the common grounding scheme that uses the 0V rail as ground.
The MJE15033 and MJE15032 are rated at Ic_max=8A and Vce_max=250V. A maximum current of 8A means the die area is too big to be used for a VAS as the parasitic collector-base capacitance is dependent on the area of the die. While I will keep the best performing circuit unchanged I will use the 2S3600 and 2SA1406 which are rated at Vce_max=200V and Ic_max=100mA. This is a reduction of current handling by a factor of 80 times. With this change I expect the VAS's parasitics should be greatly reduced.
I am about to move the output Inductor-Resistor filter from the 0V rail to the hot output terminal where it is usually found. This will need a small separate PCB which will be bolted to the main PCB. I will keep the connecting wires the shortest possible.
In the original circuit a series Resistor-Capacitor filter was placed after the Inductor-Resistor filter. Since readers of this thread including Mooly, the moderator, insisted the Inductor-Resistor filter should be connected AFTER the Resistor-Capacitor filter, I would like a last comment before I make the actual physical circuit changes. This is not modifying a schematic to be simulated by LTSpice or equivalent, but the actual circuit.
I am attaching the .asc schematic. Is this where the mentioned filters are usually connected?
In the original circuit a series Resistor-Capacitor filter was placed after the Inductor-Resistor filter. Since readers of this thread including Mooly, the moderator, insisted the Inductor-Resistor filter should be connected AFTER the Resistor-Capacitor filter, I would like a last comment before I make the actual physical circuit changes. This is not modifying a schematic to be simulated by LTSpice or equivalent, but the actual circuit.
I am attaching the .asc schematic. Is this where the mentioned filters are usually connected?
I am trying to buy the two 0.1 Ohm resistors. I looked for non-inductive wirewound resistors in a local shop but couldn't find any such resistors. What I found were heatsink mounted high power resistors that I seriously doubt are suitable.
Can anybody please post some existing wirewound resistor types, that can be used as shown in the last schematic posted to this thread?
I found these:
10 Watt Resistors Datasheet
Are these suitable?
Can anybody please post some existing wirewound resistor types, that can be used as shown in the last schematic posted to this thread?
I found these:
10 Watt Resistors Datasheet
Are these suitable?
I decided in favour of using an array of 10 metal film resistors in parallel for each 0.1 Ohm resistor. These will be soldered to a separate PCB that will be soldered to the main PCB. It will have three contacts two of which will connect with the output transistors' collectors and the middle point to the output terminal.
When modifying the original PCB is ready, I will turn to testing the resultant circuit for leakages. I will start with a very weak +/-12V DC power supply that can hardly supply 0.5A, and move up the power ladder as leakage tests are passed. I will do most of my testing on +/-42V DC which I can obtain using a centre tapped mains autotransformer that will be used to feed power to the amplifier. The autotransformer is rated at 1500W, 240V.
The +/-85V DC will NOT be used before the circuit passes all tests using lower voltages. Notwithstanding of all the precautions, I will make sure the high voltages will have to feed current through a voltage dropper in the event of malfunction.
Something that I would like to ask is whether a transistor based circuit can be used to avoid the power-on surge that is heard as a load 'plop'. I would like to avoid having to use a relay.
When modifying the original PCB is ready, I will turn to testing the resultant circuit for leakages. I will start with a very weak +/-12V DC power supply that can hardly supply 0.5A, and move up the power ladder as leakage tests are passed. I will do most of my testing on +/-42V DC which I can obtain using a centre tapped mains autotransformer that will be used to feed power to the amplifier. The autotransformer is rated at 1500W, 240V.
The +/-85V DC will NOT be used before the circuit passes all tests using lower voltages. Notwithstanding of all the precautions, I will make sure the high voltages will have to feed current through a voltage dropper in the event of malfunction.
Something that I would like to ask is whether a transistor based circuit can be used to avoid the power-on surge that is heard as a load 'plop'. I would like to avoid having to use a relay.