No, probably blew a transistor and trying to sort it out.
Without using any math.
Without using any math.
Well, the rough-and-tumble discussion may be seen as unkind, rather than dancing gracefully, as per Alan Watts' cosmic dance.
Running single transistors may not be everyone's cup of tea, but there are lots of things that can be learned and incorporated. I, along with many people, jumped into 2 and 3-way speakers before gaining a better appreciation of 1-way. E.g. how to reduce IMD and extend a smooth frequency response without "kicking the can" by adding more drivers and writing-off the crux of the problem as unsolvable?
Going the "optimized class-B route" before mastering some minimalist class-A amplifiers seems similar. It's a bold statement of understanding: "I'm going to get better results and save electricity while I'm at it!"
As an aside, I accidentally blew $100s on electric heating last month, because of a mystery hot water tap that had been left in the wrong position. In that context, 'excess' power consumption for purposeful listening enjoyment is nothing.
Listening to individual transistor devices surely has value, and I'm tempted to do some of my own experiments, if only to get a better feel for my design decisions, which are otherwise blind in a sense.
You would think that he’d have had enough time to master the single transistor amplifier and move on by now. This thread isn’t his first rodeo. I built my first 1 transistor amplifier when I was 8. With about the same level of understanding. Some of Cummb’s other proposals are downright preposterous, and one really does start to wonder if it’s just to **** engineers off.
1-transistor amplifiers can be refined to actually sound “good”, but so far I’ve seen no real attempt to do so. I’ve seen some attempts to make charcoal. Add some sulfur, and we can make gunpowder.
1-transistor amplifiers can be refined to actually sound “good”, but so far I’ve seen no real attempt to do so. I’ve seen some attempts to make charcoal. Add some sulfur, and we can make gunpowder.
Continued...
Specifically, I know that a resistor-loaded BC850 buffer has particular grainy-textured distortion, and is a failure for high quality sound. I thought 20V rails would be sufficient to prevent audible modulation with a mere ~0.5V headphone level signal, but I was wrong. But I never followed up on that discovery with any methodical "what if" tests...
-- CCS load instead of resistor? How perfect is it? Is there some kind of bipolar or asymptotal behaviour? E.g does the quality of the distortion seem to get worse and "fight against" the muting effect of the (theoretically) improved circuit? Adding a CCS is kind-of all or nothing, so that's not so easy, but I could compare some different CCS variations.
--Would I be better off with a resistor load on a MOSFET or JFET? Would a CCS be a worthwhile improvement or a waste of time?
--Cascodes? They look promising in theory and I could use the impedance conversion in a couple of places, but is it the same story with modulation squared? Looking at the M2x amplifier, a small transformer was used to achieve a modest voltage gain. It seems to me that a transistor cascode could be used in similar places, but are we looking at a tone-generating musical instrument amplifier by another name?
So those are just a few of my unanswered Qs at the moment for playing with transistor sound.
Specifically, I know that a resistor-loaded BC850 buffer has particular grainy-textured distortion, and is a failure for high quality sound. I thought 20V rails would be sufficient to prevent audible modulation with a mere ~0.5V headphone level signal, but I was wrong. But I never followed up on that discovery with any methodical "what if" tests...
-- CCS load instead of resistor? How perfect is it? Is there some kind of bipolar or asymptotal behaviour? E.g does the quality of the distortion seem to get worse and "fight against" the muting effect of the (theoretically) improved circuit? Adding a CCS is kind-of all or nothing, so that's not so easy, but I could compare some different CCS variations.
--Would I be better off with a resistor load on a MOSFET or JFET? Would a CCS be a worthwhile improvement or a waste of time?
--Cascodes? They look promising in theory and I could use the impedance conversion in a couple of places, but is it the same story with modulation squared? Looking at the M2x amplifier, a small transformer was used to achieve a modest voltage gain. It seems to me that a transistor cascode could be used in similar places, but are we looking at a tone-generating musical instrument amplifier by another name?
So those are just a few of my unanswered Qs at the moment for playing with transistor sound.
My father built me a variable power supply out of an old car battery, by driving brass screws through the case to tap off 6, 8, 10V besides the 12. Mine used some old Delco part on the car radio heatsink I pulled with it, and was driven by a crystal radio, with line antenna strung out to a tree.
I wasnt very smart even then and couldnt understand why the speaker "sucked in" when used in the circuit. I didnt understand capacitive coupling... However I knew some car radios had a transformer thingy between the transistor and the speaker; transformer thingy being an autotransformer. Almost a joke there.
"Cumbb" connects his B200-6 to the very latest MP3 Class A SE Amplifer and describes today's sound result as the latest thing.
;-) ;-) ;-)
Full of joy, the tinkerer forgets that the amplification factor is determined by the complex load, the loudspeaker - a broadband /full range chassis. Among other thing!
Particularly noteworthy is always the number of " ;-) ". Of course, building such primitive amplifiers is fun, but it will never be hi-fi or high-end.
Children,
better build the ZENDUCTOR number 2 from Nelson Pass.
Bye,
HBt.
Particularly noteworthy is always the number of " ;-) ". Of course, building such primitive amplifiers is fun, but it will never be hi-fi or high-end.
Children,
better build the ZENDUCTOR number 2 from Nelson Pass.
Bye,
HBt.
One can get a more pronounced version of this with a pentode tube amplifier or TDA2003 strapped into transconductance mode. Built-in smiley face EQ. That’s why old table radios with a 4 or 6 inch speaker we’re actually listenable back in the day.
Use a two way with a 12dB LR textbook crossover, and you get a nice fat peak in the midrange. An “in your face” sound - enough to send you screaming from the room and claim all multi way speakers sound bad.
Use a two way with a 12dB LR textbook crossover, and you get a nice fat peak in the midrange. An “in your face” sound - enough to send you screaming from the room and claim all multi way speakers sound bad.
cumbb is the guy who tells everyone that mica sheets sound bad and that the transistor package matters more than the overall circuit.
And he gets into threads with confused newbies who are trying desperately to wrap their heads around it all. With half of what is stated going directly against the laws of physics, the result reminds me of the Star Trek episode “I, Mudd”.
Nah, just issue a general warning, similar in principle to those on high voltage safety. He is entitled to his opinions, and experimenting is not without merit. Just everybody around needs to be able to make an informed decision on whether to take the red pill or the blue pill.
Well, many in this world tune their dials to maximize controversy. The dead give away? Try to assist along the lines of their idea. The dead giveaway is the "dial tone" response. That's because you're not challenging their position, you're going along with it in an assistive sort of way. Not what's wanted, apparently.
Interacting with someone who only values the controversy they can stir up is a waste of breath. IMHO, you cant ban someone who's like that; best you can do is not let yourself get sucked into that vortex a second time. Or - if time on your hands - point out the nature of their participation.
This practice is ancient, BTW, going back to before the internet was even a thing. I remember one guy dropped he could hear the difference regarding radio shack 90 deg RCA adapters in line or not. What a massacre... All on ASCII terminal display. He was "well fed", shall we say, over that one.
Dear audio friends, I am signalling a different problem which even cumbb is overlooking.
TO220 or any other casing (TO247 and the like) use non-directional copper leads to transfer the signal in and out the transistor.
First about the copper. Is it oxygen-free, mono-crystalline, 99.99% pure? Is copper a suitable material for audio conductors anyway?
But then, most importantly, what about the directivity of the leads?
Is this directionally correct copper anyway. And if it is, how is it directed?
For the base lead it should be clear, the power is directed into the transistor.
But the collector and emitter leads are a different story.
If the transistor is used in single ended CE configuration, the power is directed into the collector, and out of the emitter.
In a CFP class AB amplifier it is the other way around, the power goes into the emitter and out of the collector.
It is impossible that a transistor designed for EF configuration might sound good in CFB configuration if the leads have the improper directivity.
I hope cumbb is able to shed some light onto this.
TO220 or any other casing (TO247 and the like) use non-directional copper leads to transfer the signal in and out the transistor.
First about the copper. Is it oxygen-free, mono-crystalline, 99.99% pure? Is copper a suitable material for audio conductors anyway?
But then, most importantly, what about the directivity of the leads?
Is this directionally correct copper anyway. And if it is, how is it directed?
For the base lead it should be clear, the power is directed into the transistor.
But the collector and emitter leads are a different story.
If the transistor is used in single ended CE configuration, the power is directed into the collector, and out of the emitter.
In a CFP class AB amplifier it is the other way around, the power goes into the emitter and out of the collector.
It is impossible that a transistor designed for EF configuration might sound good in CFB configuration if the leads have the improper directivity.
I hope cumbb is able to shed some light onto this.
By that same logic an NPN transistor will sink to the bottom of the ocean because the base has holes in it.
The current flows in the same direction regardless of whether it’s EF or CFP. The only difference is who is on top and who is on the bottom. One can also design using same sex transistors, and whether to use NPN or PNP is simply a design decision. You can use any criteria you want, even what a Ouija board tells you to use. But don’t expect anyone who has scientific measurement equipment to agree with it.
The current flows in the same direction regardless of whether it’s EF or CFP. The only difference is who is on top and who is on the bottom. One can also design using same sex transistors, and whether to use NPN or PNP is simply a design decision. You can use any criteria you want, even what a Ouija board tells you to use. But don’t expect anyone who has scientific measurement equipment to agree with it.
- Status
- Not open for further replies.