can SE bass be as good as PP ?

[Bigun]

I have read that PP has more bass and can have just a good a mid range and treble as SE.

This isn't a SE vs PP thread - although I expect it'll get caught up in that a bit.

What I'd like to learn is whether there are key design approaches that you have experience of, which give SE the bass performance that is often credited only to PP amps.

I don't mean frequency extension, I mean the subjective performance, comments I read use terms like 'tight bass' 'well defined' or on the other hand 'warm' , 'fat'

I'm not sure if the whole story is tied up with the predominant 2nd harmonic distortion of SE or if there is another factor here. Such as PSU design etc.

 

[mike567]

Good bass response can be had from any amp. Better bass response from a PP amp would be that it generally has much power.
SE amps over 10 watts get very pricey very quickly!

 

[sreten]

Hi,

Objectively it depends on the SE design and the speaker its connected to.
Subjectively is a minefield of different opinions and budget. For a given
budget, an amount of iron, PP will go louder, e.g. 10W SE versus 30W
PP renders all comparisons academic if the SE doesn't go loud enough.

Whilst 30W is only 4.7dB louder in the midrange, it sounds like
more in the bass, due to the bunched F&M curves in the bass.

rgds, sreten.

 

[kevinkr]

It's really all down to design target, within their power ratings an SE can be designed to have very good bass performance, this does require larger output transformers and somewhat lower impedance power supplies than is typically implemented.

I use a pair of diy 20W SE GM70 amps to drive a pair of 11 cu ft Onkens with Iconic 165-8G woofers; the speaker system has an F3 of ~35Hz and a pair of push pull Luxman MB-3045 provided rather similar performance, not better. The Luxman's additional power reserve was not tapped at any tolerable volume level.

The amps were designed for these speakers, other than that nothing special about them.

 

[Globulator]

I have read that PP has more bass and can have just a good a mid range and treble as SE.

-------------

What I'd like to learn is whether there are key design approaches that you have experience of, which give SE the bass performance that is often credited only to PP amps.

I get powerful bass from my SE amplifier. I achieved this by removing the OPT from the GNFB loop, and implementing the feedback around driver and power tubes. This lowers the impedance that drives the OPT and improves feedback quality so the bandwidth of the transformer increases and you get less high order harmonics. My amp is now very difficult to clip despite only being 8W or so.

 

[MrCurwen]

I've found that this type of output section has as powerful a bass as any amp, PP or SS included. Excellent definition also down to the lowest frequencies.

SE will of course have a bit of 2H (amount depending on the tubes curves) that the balanced wouldn't have. Excellent performance otherwise.

 

[Bigun]

Interesting, not only low impedance drive all the way through, but paraded output. I wonder, to what extent, we can say they are 'responsible' for the good bass performance as neither of these qualities are necessarily present in a PP amplifier (where bass is reported, in general, to be good).

 

[DF96]

It all depends on what you mean by "good bass".

SE will, other things being equal, have more second-order distortion than PP. This means that you could get the perception of more bass from a speaker which doesn't actually do much bass, via harmonics effectively 'synthesizing' the missing bass in the brain of the listener. Useful for small speakers.

The increased distortion of SE means that it may cope with feedback less well than PP, so less feedback may be used (and SE fans sometimes dislike feedback anyway), so the output impedance may be higher than PP. This leads to poorly damped bass, which some may prefer.

I suspect that 'tightly controlled bass' and 'warm fat bass' are opposites. Typically, SE is more likely to lead to the latter but as always it depends on the details of the circuit and the quality of the OPT. Good SE could be better than random PP.

 

[Phats]

Forgive my breaking in as a newby, I have read so many interesting posts of you guys along the way, but my two-cents opinion is that the core of output transformers of SE amps get saturated quite easily. Perhaps I am saying this the wrong way, I am not really versed in (especially not in the dark art of transformer winding) but as I understood a large part of the quality of an SE amp came down to the quality of the output'iron' (when all else stays the same, good design and such).

Cheers, Stef.

 

[haddar]

I do not know where are you are from but I see Canadian flag beside your nick. Therefore, if you are anywhere close to Toronto in two weeks (Oct31-Nov2) there will be hi-fi show TAVES. Audio Note and Audio Note Kits rooms will have SET amps playing with AN/E speakers and many of the other rooms have PP amps.

From my DIY experience SET bass can be from soft/boomy to tight and clean. Using same OPTs, same speakers and zero NFB but depending on by witch schematic amp was built and components used.

 

[planet10]

Objectively it depends on the SE design and the speaker its connected to

To take this important point further. And let's forget the generalization made wrt PP & SE designs.

Loudspeaker, amp (and cable connecring) are a system.

Most commercial loudspeakers are designed assumming that the amplifier has close to zero output impedance. This gives higher damping factor amplifiers an edge in terms of keeping the speakers bass appropriately damped.

There are counter examples to that and diyers can choose to build their loudspeaker with differing assumption.

If you have a speaker that prefers a high output impedance (ie most SE amps) then they achieve better bass with the high output impedance amp.

dave

 

[multi]

planet10;
You are dead right; i have a horn system using Altec 515, no boom even with no feedback SET or PP. Recently been working on a 18 Cubic ft cabinets, some of the speakers I tried had lots of boom even with feedback. Ended up using a Chinese made Neo speaker, about the right amount of boom..
Must admit I am just getting used to T&S parameters.
Phil

 

[MrCurwen]

Interesting, not only low impedance drive all the way through, but paraded output. I wonder, to what extent, we can say they are 'responsible' for the good bass performance as neither of these qualities are necessarily present in a PP amplifier (where bass is reported, in general, to be good).

I'd venture a guess that the low impedance drive provided with the output section presented is 98% responsible for the increased performance. Parafeed allows one to use a cheap trafo, possibly a PP one, but if you have a big, good quality OT, that part of the equasion probably is not very important. But, for the output tubes linearity it matters a great deal of course, and for the builders wallet.

 

[Shoog]

What effect does the asymmetrical drive current play in the perceived difference between PP and SE. I would have thought that the low drive current on once side of the waveform in SE would have caused lower speaker control and required a high damping speaker driver to work.

Am I missing something here.

Shoog

 

[DF96]

What effect does the asymmetrical drive current play in the perceived difference between PP and SE. I would have thought that the low drive current on once side of the waveform in SE would have caused lower speaker control and required a high damping speaker driver to work.

I think you may be confusing two different effects:
- SE produces more second-order distortion
- SE often has higher output impedance
It is not the drive current which provides speaker damping, but the impedance from which that current comes.

However, there is a germ of truth in your idea, as an SE output may have a non-linear output impedance too which varies across the audio cycle more markedly than PP.

 

[Bigun]

I would have thought that the low drive current on once side of the waveform in SE would have caused lower speaker control and required a high damping speaker driver to work.

Interesting thought. If I follow what you mean here, you are saying that the AC current flowing through the OT primary has a different impedance depending on which direction it's flowing?

 

[MrCurwen]

Add a CCS from the tube anode to ground, to solve that problem completely. Or rather the plate load source and ground. That way, both the sourcing and sinking is low impedance at all signal phases.

 

[Bigun]

Isn't a CCS a high impedance ?

 

[Shoog]

Since the current through the output triode in SE varies with signal amplitude the current delivered to the transformer must be variable over the waveform. If this is not what is delivered to the secondary then the energy storage of the transformer must be smoothing the current variability from the positive going signal to the negative going signal. This seems an illogical idea since the current is carrying the signal and so if the transformer is altering it in this way it is distorting the signal grossly.

Think of what I am trying to say here as the positive going signal approaches saturation (2x standing current) at the same time the the negative going signal is approaching cutoff (zero current). Both these conditions exist at the same instant and represent a difference in power at the positive and negative peaks of the waveform.

The fundamental difference between this scenario and a PP output stage is that current delivery is equal on both the positive and negative going peaks in PP because it is handled by two triodes with a summing of their opposite signals in the transformer.

Again I ask, am I missing something fundamental here ? This is a genuine point which logically doesn't make sense to me.

Shoog

 

[DF96]

Since the current through the output triode in SE varies with signal amplitude the current delivered to the transformer must be variable over the waveform.

I think this is the point where you get confused. You mention "current" and "signal" as though they are two separate things. The current is the signal! The varying current is the waveform.

In a triode SE output the anode impedance varies with current - this is part of the reason why you get second-order distortion. More current means smaller impedance. With no feedback this gets reflected through the OPT straight to the speaker, so that sees a driving impedance which varies through the waveform. This does not happen with a pentode SE, and happens to a smaller extent with UL SE.