• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

can SE bass be as good as PP ?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Isn't a CCS a high impedance ?

In that position the CCS will sink current much much more efficiently than the tube. The top MOSFET on the other hand will source current very efficiently.

So, with the source and sink, OT is well controlled. Well damped, and all the parasitics are better behaved. Don't have to worry about OT quality as much; your wallet will be grateful.
 
MrCurwen said:
In that position the CCS will sink current much much more efficiently than the tube. The top MOSFET on the other hand will source current very efficiently.

So, with the source and sink, OT is well controlled. Well damped, and all the parasitics are better behaved. Don't have to worry about OT quality as much; your wallet will be grateful.
Either I don't understand what you are saying, or you don't understand what you are saying.

A CCS does not make anything "well damped". It presents a high impedance, ideally infinite. In the right place it reduces damping; in the wrong place it does nothing useful at all.
 
Member
Joined 2003
Paid Member
I've read these posts from the start..

It's interesting.. DF96, I have a 5w SE Pentode(EL84), The bass has more control, and sounds better, than some of PP amps I have.. Volume is the limitation..

I have about 10 different tube amps, and heard at least ten others over last 2 years,and SE bass can be better than PP.. But, all things aren't equal, various design levels, integrated vs separates etc..

Good read, keep it up folks:)
 
Joel, With all that listening experience, can you draw any correlations between bass performance and the topology/implementation of the amplifier ?

I suspect that the power supply has a role to play here - almost as if in the case of SE the power supply fills the role of the 'other triode' in a PP design.
 
In Florida I had an SE amp / speaker system that could rattle the windows in the house across the street with deep solid bass. The amp was an SSE with KT88's in UL mode with a few db of cathode feedback and high idle tube current (100 mA) to reduce the amp's output impedance. Big OPT's and as mentioned speaker drivers DESIGNED for being driven with high output impedance SE tube amps. The drivers were Hawthorne Silver Iris 15 inch coaxials with 96db efficiency mounted in old console radios for OB cabinets.

The amp, speaker, listening room parameters, and music choices all interplay to affect the perception of bass content. The typical SE amp is usually made with a triode output tube and zero or minimal feedback, while the typical push pull amp is often made with pentodes and considerable feedback. These parameters individually affect how the amp interplays with the speaker system and generalizations of SE VS P-P may not always be valid.

The typical SE tube amp has a high output impedance resulting in a damping factor in the range of 1 to 5. Connecting this amp up to a typical modern speaker system designed for a solid state amp with a typical damping factor in the range of 50 to 200, CAN produce the illusion of good bass, but an FFT analysis with a good mic will reveal some fundamental bass notes, but excessive second harmonic distortion. It can also result in flabby poorly controlled bass. Tweaking the amp's bias current and other parameters can affect the bass by changing how the amp interplays with the woofer's resonant impedance peak.

It is true that the slew rate and the DYNAMIC output impedance of an SE triode amp can be asymmetrical, especially in a zero feedback design. This generally results in higher levels of second harmonic distortion, which can affect the perception of bass content. A triode push pull amp will not have the second harmonic content, but will have a higher damping factor. If all other conditions are constant, the two amps will sound different in the bass region, but the determination of which will have "better bass" may be up to the user, their speakers, and their choice of music. The push pull amp will generally sound punchier and more dynamic.

Pentodes typically have a much higher output impedance that is more constant under dynamic conditions. This is usually lowered by application of negative feedback, which also alters the dynamics and the harmonic spectrum of the system. The amount and method of feedback application will affect the perception of bass content. Again the speaker system and music choice will have a huge influence of how the system sounds in the bass region.
 
It is true that the slew rate and the DYNAMIC output impedance of an SE triode amp can be asymmetrical, especially in a zero feedback design. This generally results in higher levels of second harmonic distortion, which can affect the perception of bass content. A triode push pull amp will not have the second harmonic content, but will have a higher damping factor. If all other conditions are constant, the two amps will sound different in the bass region, but the determination of which will have "better bass" may be up to the user, their speakers, and their choice of music. The push pull amp will generally sound punchier and more dynamic.

This is what I was discussing.
The SE amp may have a fuller bass, but it will lack the definition of a good PP amplifier. I would say that this will carry forward even to zero feedback PP amps of similar damping factor to the SE, since the PP will have more consistent dynamic impedance.

For me this difference is important since it will be noticeable on light jazz with double bass and also electro dance music where much of the critical music information is in the bass range. Rock and pop less so.

Shoog
 
Last edited:
A wealth of learning from Tubelab !!

Is there nothing that can be done to improve the SE amp topology, other than feedback, to improve the bass performance - is there a power supply approach that helps ? Are low Rp tubes like the 6AS7 to be preferred or does the winding ratio of the OT make that a moot issue.
 
Member
Joined 2003
Paid Member
Joel, With all that listening experience, can you draw any correlations between bass performance and the topology/implementation of the amplifier ?

I suspect that the power supply has a role to play here - almost as if in the case of SE the power supply fills the role of the 'other triode' in a PP design.

Difficult Question.. They're all unique, especially in the bass.. Probably easier to define the rest of the spectrum by topology/implementation.. I don't built them so need to learn more about the different design configs and why they sound the way they do..

Right now I'm running 2A3 Mono's, Class A, Double PP driver stage, auto bias. Hi level design Wide open, lots of recorded detail, delicate highs, bass slam isn't huge but all other bass characteristics are very good.. The PT is Hammond 302AX and I plan to run it at 240v when I put the circuits in.. I've used SS amps at 240V before, this is one upgrade that is easy and worth doing IME

In general, I've really like all my EL84 amps in the bass region and have heard many of these at various levels of parts selection, and including, SE, UL, PP pentode.. Think I got that right:)
I also have 300B SET mono's, zero feedback, single 12ax7, tube rec.. Most crisp 300B than any I've heard but not like El84 or 2A3..

Next dream is trying 40-50w KT 88, also, when I get to 3way speaker I'm planning, running 2A3's for mids and highs, SS for bass..

6Mb8 4 tube amp is nice on my Tv system and one for bedroom system..
 
Bigun said:
I suspect that the power supply has a role to play here - almost as if in the case of SE the power supply fills the role of the 'other triode' in a PP design.
Yes - No.

For SE the last PSU cap is in series with the output valve and OPT, but the effect of this (like everything else) can be reduced by using feedback. The cap does not play a role anything like the other half of a PP output.

Is there nothing that can be done to improve the SE amp topology, other than feedback, to improve the bass performance - is there a power supply approach that helps ?
Provided the PSU is adequate for the task, then no. Improvements come from PP and feedback. The SE topology is inherently flawed, except for small signal use. Curiously, there are some people who insist on balanced design for small signals (unnecessary) while other people (or even some of the same people?) insist on SE for power outputs.
 
Just a few (subjective) impressions - I've gotten excellent bass performance - with 15" Eminence woofers in UREI 813A cabinets - using single-ended amplification.

One was a EF86 driver / EL156 (and later a 6550) fixed bias output in ultralinear. Built as monoblocks. GNFB was applied. The output transformer was the hefty Hammond 1627SEA.

More recently it was a fixed-bias pentode amplifier using a 5687 driver and any number of the classic octal output tubes - 6550s, EL34s, 6L6GCs - with a James 6123HS output transformer. Once again I used global feedback. The screens of the output tubes were also regulated via VR tubes.

The UL amplifier was a little more loosey-goosey than the pentode - but again, subjective impression. Both, however, could drive the UREI speakers plenty loud though the harder clipping characteristics of the fixed-bias pentode amplifier were more noticeable at (too) loud listening levels.

However, the winner - at least with these speakers - were a pair of Eico HF-60 push-pull monoblocks. I'll attribute this to the better iron... and the circuit.
 
Either I don't understand what you are saying, or you don't understand what you are saying.

A CCS does not make anything "well damped". It presents a high impedance, ideally infinite. In the right place it reduces damping; in the wrong place it does nothing useful at all.

I'm referring to the schematic I posted earlier in this thread.

http://www.diyaudio.com/forums/atta...219d1413729980-can-se-bass-good-pp-27_71a.png

In this schematic, the plate load MOSFET can provide current very well to charge up C2, and the CCS (on the far right) can sink current very well to discharge it. Tube is relieved and operates much more linearly. The OT is better controlled by this kind of drive, and the load (speaker) is better damped. If tube is linear, no NFB needed.

My opinion is, after some comparisons made with my personal ears (yours may disagree), is that this kind of drive provides best bass definition and power (perception of "moving air" even at low volumes, not watts per se) SE has to offer, and as good as PP or SS.
 
Let me qualify my own opinion. My experience is that if you use a SE amplifier well below its max power it will sound just about the same as any PP at similar power levels. In order to test this you really need very sensitive speakers of the order of +100db/W.

Shoog
 
almost as if in the case of SE the power supply fills the role of the 'other triode'

The power supply is in the signal path for nearly all amp topologies including P-P, although some of it's effects are partially cancelled in P-P. The power supply has no capability to act as half of a P-P circuit. It is merely A source of power, and an impedance in series with the OPT in an SE amp. The OPT is usually 3000 to 8000 ohms, so even a 10 ohm power supply will have minimal effects on the output stage alone. It can allow coupling into other stages if they aren't designed for immunity (PSRR).

In all cases the power supply should ideally have a zero output impedance for all frequencies from DC to the frequency where the entire amp's gain drops below 1 (usually several hundred KHz).

This can be accomplished with an active regulator design, but the design is critical especially in the ultrasonic frequencies.

In practice the output impedance can be brought into the milliohm region with the right combination of output caps. I find that a good electrolytic in parallel with a polypropylene motor RUN (not start) cap of at least 50uF will get you 99% there.

Is there nothing that can be done to improve the SE amp topology, other than feedback,
You can improve the damping factor some by using a tube with a low plate resistance and running it at a fairly high current. Use the highest possible OPT impedance consistent with your power needs. In my bass monster SSE I used a triode wired KT88 (switchable to UL) with 100 mA of tube current, and a 5K OPT instead of the usual 3K or 3.6K.

The 6AS7 does have a low Rp but virtually no gain and poor linearity. The 300B Has more gain and excellent linearity, but poorer Rp. Some triode wired pentodes are excellent, but many of them (sweep tubes) have severe screen grid voltage limits. It is possible to run a pentode as a pentode and reduce its Rp to triode levels with local feedback.

The SE topology is inherently flawed

Both SE and P-P have their inherent weaknesses and strengths.

SE does have its damping factor limitations, and requires a big expensive OPT, but that OPT can be operated in the sweet spot in its BH curve.

P-P can have solid state like damping factor but the OPT's magnetic domains must transition through zero magnetization and reverse direction often. This can mask low level detail in the music. Good OPT's can do a great job, but don't use more OPT than you need. Too much iron can cost you detail.

My current opinion:

If you are a fan of single driver loudspeaker systems build a good SE amp for 70 Hz up (or wherever your speaker stops working) and use a solid state or P-P amp for the subwoofer.

If you are into multiple driver systems, build an amp for each driver and use an active crossover. Use SE for the highs, and maybe the midrange, P-P or SS for the woofer, and a 500+ watt class D amp for the thunderbox!
 
Member
Joined 2003
Paid Member
"If you are into multiple driver systems, build an amp for each driver and use an active crossover. Use SE for the highs, and maybe the midrange, P-P or SS for the woofer, and a 500+ watt class D amp for the thunderbox"

This suggestion is like putting a stiff drink in the hands of an Alcoholic!

I guess something I might strive toward for a reference 4way system..Hmm, couple of years maybe:D
 
"If you are into multiple driver systems, build an amp for each driver and use an active crossover. Use SE for the highs, and maybe the midrange, P-P or SS for the woofer, and a 500+ watt class D amp for the thunderbox"

This suggestion is like putting a stiff drink in the hands of an Alcoholic!

I guess something I might strive toward for a reference 4way system..Hmm, couple of years maybe:D

We seem to all end in that place eventually.

Shoog
 
For What it's worth...

Over a few years I have used the same Boston A100 speakers with various types of amps. 10"x1" 2way acoustic suspension, 91db.

Class A SE Mosfet- home brew
Class A/B Transistor- Home brew
Pioneer 9000sx -
Marantz 2226B
Dynakit ST35 (P-P) with Marantz section as preamp
Dynakit ST35 (P-P)with Various home brew preamps (ss and tube)
Currently using a home brew SE 1-2W (6L6 or 6V6...wiring easily changed)

My current SE Home Brew is good enough that I am looking at selling my ST-35 and other amps. Detailed highs, nice mids, solid enough bass, great detail/clarity and most importantly...balanced. Im currently using crap power transformers as OT's...so I can only imagine when I get real OT's.

I prefer the SE over P-P...both SS and Tube. Clarity, detail and when done right...balance. Just as if you had a good P-P design.
So yes...a well designed SE can be at least as good as P-P
 
If you are a fan of single driver loudspeaker........

I might add that I had a system assembled in Florida that used a TSE with either NX-483 tubes (like a 5 volt 45) for 2 WPC or 300B's for 6 WPC. This fed a pair of Planet 10's FH3 horn speakers for a very nice and plenty loud system in my small work room. There was very limited bass with the tiny full range drivers, so I added a solid state powered sub, and this worked out very well. I had that running for about the last 6 months I was in Florida and it was my favorite small system. My favorite big system at that time was a 125 WPC push pull amp fed to those same 96db OB speakers that I mentioned previously. It doesn't take a rocket scientist to figure out that this is "rock concert in your living room" territory.

If you are into multiple driver systems.....

I was working on a system for sound reinforcement / musical instrument / DJ use. The prime criteria was LOUD with good sound. This in contrast to what I was hearing in south Florida, which wad plenty of bottom end BOOM, but muddy mids and no top end. I wanted a portable system with tubes for everything but the sub bass. I had built a pair of very nice sounding speakers for 100 Hz and up with 6 inch pro audio PA drivers and bullet horn tweeters. They were driven with a P-P tube amp of 125 WPC to produce loud and clear electric piano and guitar at loud SPL's. I was working on the bass box at the time I had to leave Florida ahead of schedule, so it isn't finished. A pair of 4 CF boxes with pro audio 15 inch woofers and 300+ watts per box should keep up with the higher freq boxes, but the crossover frequency will be higher.

All that fun stuff is packed away while my new house is being built, and I am currently using a $20 Lepai amp through Yamaha NS-10M studio speakers.

I will have a much bigger work room in the new place, so I will need to find a new "favorite system." I have several amps including SE tube amps in sizes from 2 WPC to 40 WPC, P-P tube amps from 8 WPC to 250 WPC, a 450 WPC tube amp partially breadboarded, a couple of crummy SS amps and class D chip amps from 10 WPC to 300 WPC.... I'm sure I'll figure something out.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.