I've never build any SS amp apart from a kit, only build tube amps but wonder what a low power SE SS amp might be like compared to say SE 6V6, EL84 etc. Something simple, just one mosfet or power transistor. I've came across some Zen amp and few others but really not many. Maybe there is a reason people don't build them much. What do they sound like compared to tube, do they produce 2th order harmonics or just the typical transistor crackling distortion if driven to clipping? I want to give it a go but not sure what to build so I get some results.
Last edited:
SE solid state requires an output coupling capacitor and a correctly setup AB doesn't. It'll still clip like a transistor when driven too hard and it's gonna burn at least 75% of its input energy off as heat while doing it.
Fortunately, it's easy and cheap to breadboard a pure class A SS amp.
To sum, you get all of the drawbacks of SE tube operation with the drawbacks of transistor amplification.
Fortunately, it's easy and cheap to breadboard a pure class A SS amp.
To sum, you get all of the drawbacks of SE tube operation with the drawbacks of transistor amplification.
That is a bit of an oversimplification, and these amplifiers do not need output capacitors by definition.
Not unexpensive (transformer), and appropriate loudspeakers a must.
Here some examples:
http://www.firstwatt.com/pdf/art_sit_nemesis.pdf
https://linearaudio.nl/nemesis-rebuild
Not unexpensive (transformer), and appropriate loudspeakers a must.
Here some examples:
http://www.firstwatt.com/pdf/art_sit_nemesis.pdf
https://linearaudio.nl/nemesis-rebuild
Last edited by a moderator:
Have a look at the ACA amp from Mr Pass, it will give you 6W/ch. Try it with a 24Vdc supply.
I've build many tube amps; low power, lower DF and higher THD do not bother me but I've noticed that people in general will likely be more attracted by high power output amps, to me the more power doesn't mean better sound, I actually truly enjoy my Darling 0.75W SET amp.
BR,
Eric
Not true. See Nelson Pass' Aleph series of amplifiers for higher powered single-ended amps using MOSFETs. They sound very nice, are direct coupled, and enjoy up to 50% efficiency using his modulated current source. Schematics are here and there on the forums. Output power from 10W for mini to 100W+.
Another nice example is Pass' 2015 Burning Amp Festival circuit. It is designed for 50W single-ended (can be scaled down). While the original design is capacitor coupled on a single supply, it can be direct-coupled if using a dual supply.
And single-ended tube amps don't? At least solid state amps don't require extra power just to light a filament...
Last edited:
I'd take the first part but definitely don't want to be listening to SS distortion. But I was also wondered if it has the sound stage like SE amp normally does, that openness and space.
I've only heard an SE SS amp and I'm not good with the audio descriptions but it sounded great.
You won't hear the distortion until it's clipping and at that point it doesn't really matter what amp you're listening to unless it's for guitar effects.
You won't hear the distortion until it's clipping and at that point it doesn't really matter what amp you're listening to unless it's for guitar effects.
Look at old car radios
To try out some SE SS circuits, look up the schematics of the first "Transistor" American car radios, built from about 1957 to 1963. They had tubes in the RF and IF stages, running with a plate voltage of +12V, but had a single germanium power transistor (often a 2N176 or 2N301) as the output stage. The output was choke-loaded, sometimes with a tap on the choke. Look for circuits in the RCA tube and transistor manuals of the time, or dig up some Sams PhotoFacts.
The use of a choke load is very beneficial: there is little resistive voltage drop, there is inherent speaker protection, and the energy-storage ability of an inductor allows the output voltage to swing to twice the Vcc! The low impedances involved mean that only milliHenries are needed and the high frequency response is no problem. Just make sure it is sized large enough for the DC current flow and low frequency response - about the same size as a similarly-powered SE tube amp.
You will probably want to use a modern silicon power transistor such as the 2SA1302, since the old germanium transistors are not very stable, have bad hfe vs Ic distortion, and have really lousy frequency response.
These circuits are typically run as common-emitter, so if run without feedback, will have high output impedance (low damping) just like a pentode or beam power tube. It doesn't take much feedback to tighten up the bass.
I've played around with these circuits, and they sound pretty good, although not as good as a similarly-powered tube amp, IMHO. You might like it, though.
- John Atwood
To try out some SE SS circuits, look up the schematics of the first "Transistor" American car radios, built from about 1957 to 1963. They had tubes in the RF and IF stages, running with a plate voltage of +12V, but had a single germanium power transistor (often a 2N176 or 2N301) as the output stage. The output was choke-loaded, sometimes with a tap on the choke. Look for circuits in the RCA tube and transistor manuals of the time, or dig up some Sams PhotoFacts.
The use of a choke load is very beneficial: there is little resistive voltage drop, there is inherent speaker protection, and the energy-storage ability of an inductor allows the output voltage to swing to twice the Vcc! The low impedances involved mean that only milliHenries are needed and the high frequency response is no problem. Just make sure it is sized large enough for the DC current flow and low frequency response - about the same size as a similarly-powered SE tube amp.
You will probably want to use a modern silicon power transistor such as the 2SA1302, since the old germanium transistors are not very stable, have bad hfe vs Ic distortion, and have really lousy frequency response.
These circuits are typically run as common-emitter, so if run without feedback, will have high output impedance (low damping) just like a pentode or beam power tube. It doesn't take much feedback to tighten up the bass.
I've played around with these circuits, and they sound pretty good, although not as good as a similarly-powered tube amp, IMHO. You might like it, though.
- John Atwood
I have built and listened to both single ended tube and MOSFET amps, and can't really detect any perceptible "SS distortion." If it existed, solid state amps would have much less of a following these forums. If you drive them to clipping, the tube would clip more softly and the solid state amp will clip sharply, so this may be what you mean when you say "SS distortion." But, just don't clip the amp and you don't have the problem. This seems a simple enough task since most solid state amps (even single-ended ones) will put out at least as much power as most single-ended tube amps.
As my username suggests, I came here first to learn about and build tube amps. Solid state - in particular amps derived from Nelson Pass' circuits - have since caught my eye. His First Watt amps all aim for a certain sound quality that many would describe as "tubey". So far, I have been impressed and continue to build similar MOSFET-based circuits. Maybe they don't sound as holographic as some tube amps, but the overall distortion is considerably lower... Although, apparently the First Watt J2 has displaced more than one tube amp in reviewers' systems in its day.
A simple experiment might be to build the Amp Camp Amps as someone suggested above. It is cheap and can be built in a couple hours.
What you actually want to build is a SE amplifier where a transistor or MosFet is driving a SE output transformer.
That will sound quite close to what a Tube SE amp does.
I much respect the SE Hi Fi projects mentioned here where a single transistor drives a resistor or a current source, but they are absolutely different to a transformer coupled one, which to boot is way more efficient but most important, has the inductive kick and very high output impedance the others lack.
The fully active types will clip with boring flat tops, the ones using a transformer will have biting overshoot and ringing.
That will sound quite close to what a Tube SE amp does.
I much respect the SE Hi Fi projects mentioned here where a single transistor drives a resistor or a current source, but they are absolutely different to a transformer coupled one, which to boot is way more efficient but most important, has the inductive kick and very high output impedance the others lack.
The fully active types will clip with boring flat tops, the ones using a transformer will have biting overshoot and ringing.
Yes, I don't think people realize how much an air-gapped output transformer with DC current flowing through its primary contributes to the sound of a SET tube amp. In some cases, it dominates the sound.
Not sure I see how it's "way more efficient." How so exactly?
And what exactly do you mean by "inductive kick"?
se
Good point. Plus, you can use depletion power MOSFETs such as those made by IXYS which bias up just like a tube using some source resistance.
Although I wasn't the person asked, I can answer:
When a SE amp pulls against an active current source (another device), that device has to be dropping as much voltage and the same idle current as the SE amplifying device does. In that case, the source dissipates as much power at idle as the SE device, and the max possible efficiency (assuming no cheats like modulating the current source, which is really just another form of push-pull IME), is 25%.
When an inductor is used as the current sourcing element, it drops very little DC voltage (but still the same idle current as the SE device), so the inductor dissipates very little power. Maximum possible efficiency in that case is 50% (which is the same as the max possible with push-pull).
The inductor source has some other advantages: it inherently filters the supplied current from the power supply, and if the speaker is returned (usually via a capacitor) to the other end of the SE output device, then the power supply filter capacitor(s) aren't even in the output signal path. If a high frequency switch mode power supply is being used, very little ripple gets through the inductor (a lowpass filter) to the output, so that makes a very good marriage (except for anyone who is philosophically averse to anything switching at higher than the 120Hz that normal AC supplies already switch at).
Disclaimer: the "efficiency" I mention above is with the maximum available unclipped sine wave output power, assuming ideal *everything*. There are always other losses, though, and music doesn't run at constant sine wave output. At more typical musical average powers, the actual efficiency is MUCH lower. With class AB, the efficiency at lower output power is much better than any true class A, because the class AB idles at lower current. And most class D runs with ridiculously higher efficiency than either A or AB at average music outputs. If efficiency and heat generation are of concern.
I am using a pair of these at the moment and they are really quite good. Even if you find you need more power at some point, you have paid so little for the ACA it can be moved over to drive the computer speakers.
dave
Take an inductor, tie it to some positive voltage and pull the other end toward ground so that current flows through it. Then, let go (stop pulling current). The end you were pulling on will sail up ABOVE the positive supply voltage of the other end. So, with an inductor feed (gapped, to be practical and handle the idle currents), and the output driven from that point, you can get up to twice the supply voltage swing when biased at the current needed to drive the load impedance. Same current, though, so possible output power is only doubled (rather than quadrupled).
Last edited:
Butt out, pal! 😛
Yes, I understand that.
Yes, but where did the original poster say anything about using an inductor as a current sourcing element? I only saw him mention using an output transformer like tube amps do and it was in that context I asked the questions that I did.
se
See above.
Where does the original poster say anything about using an inductor as a current source element?
se
An output transformer behaves like an inductor -- the current flows through the primary winding, which is on a gapped core and has significant inductance. Say it's a triode, if you scope the voltage on the triode's plate pin, you'll find it can swing up to near double the DC voltage the triode rests at -- inductor stored energy in action.
The only difference is how the power is pulled out. With just an inductor, the output is effectively taken (in an AC sense) from across the inductor. With the transformer, the inductor has a second winding across from which the power is taken from, providing the option of an impedance conversion (needed with the kilohms of output load wanted by a triode to convert it to the 10ish ohm range of a speaker).
The only difference is how the power is pulled out. With just an inductor, the output is effectively taken (in an AC sense) from across the inductor. With the transformer, the inductor has a second winding across from which the power is taken from, providing the option of an impedance conversion (needed with the kilohms of output load wanted by a triode to convert it to the 10ish ohm range of a speaker).
How is it going to behave like an inductor unless you have no load tied to the secondary? Transformers reflect impedances. So let's say for the sake of argument that you tie an 8 ohm resistor across the secondary. Looking into the primary, the load will appear resistive.
se
- Status
- Not open for further replies.