I'm new to tube HiFi and trying to learn a bit before building something. I know that a lot of people like the sound of SET vs. P-P but I don't know why. What are the sonic differences? When I look at charts, I notice that SET's put out a lot more 2nd order harmonic content. Is this what people are hearing that they like or is it something else? What is your personal preference?
For my application, I will be using high efficiency full range drivers and I don't rattle the windows, so I don't need the extra power from P-P or even parallel. So I'm looking at SET. What do you think?
For my application, I will be using high efficiency full range drivers and I don't rattle the windows, so I don't need the extra power from P-P or even parallel. So I'm looking at SET. What do you think?
Hi Sprinter,
If taken in isolation then, yes, the harmonic content would be a significant factor in the sound of an amplifier. However, with SET amplifiers the high output impedance comes into play, and this aspect must be considered when choosing loudspeakers. If you were building your own loudspeakers then the calculated value for qts would ideally take this into account. Basically, the output from a high impedance amplifier, such as a SET, will follow the impedance curve of the connected loudspeaker. This behaviour produces a small but audible boost in the bass and the treble where the loudspeaker impedance is greatest and, rightly or wrongly, it's probably this characteristic which accounts for what many like about SET amplifiers. Now, if you intend to use full-range, high-efficiency drivers (which are usually highly damped low qts drivers that lack in the frequency extremes), it can be quite a welcomed addition. With other speakers it may not be the case.
I use an SET amplifier myself, and this has been used with both Lowther horns and a two-way reflex cabinet. With something like a 'conventional' two-way reflex, it certainly helps to keep the impedance as flat and as near to the correct loading for the amplifier as possible, and for the cabinet design to take into consideration the higher output impedance.
If you already have the full-range high-efficiency loudspeakers that you mention, then try to borrow a good SET and a solid-state. The difference will be quite audible. As to which you prefer.....
Steve
If taken in isolation then, yes, the harmonic content would be a significant factor in the sound of an amplifier. However, with SET amplifiers the high output impedance comes into play, and this aspect must be considered when choosing loudspeakers. If you were building your own loudspeakers then the calculated value for qts would ideally take this into account. Basically, the output from a high impedance amplifier, such as a SET, will follow the impedance curve of the connected loudspeaker. This behaviour produces a small but audible boost in the bass and the treble where the loudspeaker impedance is greatest and, rightly or wrongly, it's probably this characteristic which accounts for what many like about SET amplifiers. Now, if you intend to use full-range, high-efficiency drivers (which are usually highly damped low qts drivers that lack in the frequency extremes), it can be quite a welcomed addition. With other speakers it may not be the case.
I use an SET amplifier myself, and this has been used with both Lowther horns and a two-way reflex cabinet. With something like a 'conventional' two-way reflex, it certainly helps to keep the impedance as flat and as near to the correct loading for the amplifier as possible, and for the cabinet design to take into consideration the higher output impedance.
If you already have the full-range high-efficiency loudspeakers that you mention, then try to borrow a good SET and a solid-state. The difference will be quite audible. As to which you prefer.....
Steve
Thanks for that info. What I have in mind is to use the SET to drive the full-range drivers in a bi-amped active-xover setup. The speakers would be a 2-way open baffle arrangement involving a full-ranger like a Fostex FX166 or FE126 for example, and a separate 15" woofer crossed at about 200 Hz. The augies would be driven by a cheaper SS amp good to 20 Hz.
These full-rangers are typically around 93 dB and Qts around .3 to .5 . They are supposed to be OB friendly. Do you think they (or Audio Nirvana Super 6.5 which is Qts .2) would be good choices for SET's?
The big newbie question is, What is the significance of Qts and how does it affect amplifier/speaker/enclosure choice?
These full-rangers are typically around 93 dB and Qts around .3 to .5 . They are supposed to be OB friendly. Do you think they (or Audio Nirvana Super 6.5 which is Qts .2) would be good choices for SET's?
The big newbie question is, What is the significance of Qts and how does it affect amplifier/speaker/enclosure choice?
Ever since my dad put his Audio Nirvanas BLHs and 2A3 amp together, he says he has been 'wallowing' in their sound. I've heard that 2A3 on my OB Audio Nirvanas, and it sounded nothing short of phenomenal.
In short, yes, it will be an excellent match with your intented application. On an OB though, I must warn you, the SET might run out of steam.
In short, yes, it will be an excellent match with your intented application. On an OB though, I must warn you, the SET might run out of steam.
They are very OB friendly, and in that application I wouldn't
worry too much about the Qts. Probably you want to pay
more attention to the Qts of the OB woofer, which I prefer
to be around .7 or so. You may find as I have, that a SET
can do a good job driving both upper and lower with a
simple passive crossover.
worry too much about the Qts. Probably you want to pay
more attention to the Qts of the OB woofer, which I prefer
to be around .7 or so. You may find as I have, that a SET
can do a good job driving both upper and lower with a
simple passive crossover.
Yeah, that's what I want to do, wallow. Thanks for encouragement.
BTW, what I have in mind for the amp is to build a Tubelab SimpleSE with 6L6's because I already have some 5881's. If I need more power, I can go to EL34. I even have some 6146's to experiment with. I know it's a little iffy with OB's but the drivers are really efficient and I don't rock the house usually.
About Qts vs output impedance and the like...
(disclaimer - this is the product of a thought experiment, based around different things people use to alter Qts of speakers...)
A small resistor in series (with a solid state amp) is often used to raise the Qts of drivers of drivers, usually on an open baffle. The amplifier loses some of it's grip on the speakers, electrical damping goes down and Qts goes up. Bass performance improves on an open baffle due to the bass lift a high Q brings.
So, transfer this over to valve amplifiers. They tend to have high output impedances. This would act roughly (exactly? I don't know enough about electronics to say) equivalent to a resistor in series with the output. This, too, would act to raise speaker Qts.
As a listening test, we took some 300B SETs, and a 1w/ch class A amplifier. The speakers were co-axial 12" units, suited to a ported cabinet (indicates a fairly low Qts). Bass was fine with the class A amplifier, but something was amiss with the valve amplifiers. The rest of the frequency range, they each had their differences, but both were good.
Now, the amiss bass with the valve amplifiers - I think it was down to the drivers' Qts being raised, so they're no longer so suited to a ported cabinet. There was an audible difference in the lower regions, particularly where "punch" was concerned.
It was down to a mis-match between amplifiers and drivers, so beware they might not always work exactly as you might think.
This might help. If not, feel free to ignore it. As I said, it's the result of some dangerous thinking. If someone with more experience in this area tells you otherwise, believe them.
Chris
(disclaimer - this is the product of a thought experiment, based around different things people use to alter Qts of speakers...)
A small resistor in series (with a solid state amp) is often used to raise the Qts of drivers of drivers, usually on an open baffle. The amplifier loses some of it's grip on the speakers, electrical damping goes down and Qts goes up. Bass performance improves on an open baffle due to the bass lift a high Q brings.
So, transfer this over to valve amplifiers. They tend to have high output impedances. This would act roughly (exactly? I don't know enough about electronics to say) equivalent to a resistor in series with the output. This, too, would act to raise speaker Qts.
As a listening test, we took some 300B SETs, and a 1w/ch class A amplifier. The speakers were co-axial 12" units, suited to a ported cabinet (indicates a fairly low Qts). Bass was fine with the class A amplifier, but something was amiss with the valve amplifiers. The rest of the frequency range, they each had their differences, but both were good.
Now, the amiss bass with the valve amplifiers - I think it was down to the drivers' Qts being raised, so they're no longer so suited to a ported cabinet. There was an audible difference in the lower regions, particularly where "punch" was concerned.
It was down to a mis-match between amplifiers and drivers, so beware they might not always work exactly as you might think.
This might help. If not, feel free to ignore it. As I said, it's the result of some dangerous thinking. If someone with more experience in this area tells you otherwise, believe them.
Chris
Thanks, Mr. Pass. What I have in mind for woofer is an Emminence Alpha 15a or Goldwood gs1858, both of which have Qts >1.0. If you recognize this configuration as patterned after MJK's OB system, you'd be right...They are very OB friendly, and in that application I wouldn't
worry too much about the Qts. Probably you want to pay
more attention to the Qts of the OB woofer, which I prefer
to be around .7 or so. You may find as I have, that a SET
can do a good job driving both upper and lower with a
simple passive crossover.
I just prefer to have the additional control over xover with the active approach. Also, this way, I can build the amp with less expensive output xfmrs since they won't have to work under about 200 Hz or so.
I still don't understand how Qts is used to design a system. You said you prefer around .7 for a woofer. Is greater than .7 okay too? I do understand that in OB you need the fullranger to have greater efficiency to compensate for edge losses in the woofer, but I figure that is really only important with a single amp and passive xovers. Would that be true? My system would be bi-amped so I would have individual control.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.