SETs and Horns

DaveC

Industry Expert
Nov 16, 2014
3,899
2,141
495
In many/most situations I agree.

Which also sort of points to the gain structure of the system, which is probably much more important than a lot of people realise. Amplifying the small source signal more than it needs then attenuating it more than a little, or even more than once is a recipe for degrading sound quickly.

Two things in my playback I have been very careful to optimise: gain structure; noise. Very loud listening with all my sources is with about 4dB-10dB of total attenuation in my case via a Slagle silver autoformer at the output to the preamplifer (which is technically perfect in this situation - not necessarily so for all applications). I also have zero audible noise even with my head in the horns, so micro-dynamics are unhindered.

Some SET amps are lower gain and some need more input voltage for maximum output than most 'traditional' SS and hybrid amplifiers and depending on the remainder of the system a preamp with different gain may be required. This is not exclusive to SET amps, but I prefer to build the the best sounding amplifier I can and then feed it whatever voltages are required via the preamp. My sources all are low output circa 1V-1.25Vrms and my amps are high input, then accounting for well recorded music with high dynamic range (an extra say 6dB) I found in the end that 8x gain (18dB) for the preamp is what suited my system and set about finding the best high gain, low output impedance, high current (also specific requirements for my playback - not necessarily for others) preamp to perform this task.

Why is this important? Well, if you are attenuating significant amounts, particularly with some kind of resistive device (eg. potentiometer, ladder or series resistors, chip) or even digital in software, odds are that you can improve sound just by reducing gain somewhere in the system. I listen at 11 o'clock on the volume knob...listening at 9 o'clock would be worse...6 o'clock I do not even want to think about. Listening at 12 o'clock and needing to go to 13 o'clock would be a travesty. This is a long-winded way of saying that swapping out amplifiers often changes the gain structure of the system and that in this situation the changes heard are not neccesarily entirely attributable to that amplifier, but perhaps may be due to the attenuator, perhaps, a kind of system synergy.


I'm super jealous of your silver slagleformers... :) I've been looking at them for years now. I'll definitely get a set eventually but digital solutions seem to be getting better so it'll be interesting to compare silver vs digital vs a good ladder-type resistive attenuator. I'm currently using a tortuga ldr vc/source switch in my own 6SN7 buffer preamp. Ideally, I'd like to convert the preamps in the link below to a remote controlled slagleformer with LCD readouts.


I agree on gain. My voltage amp used to use 6SN7s, but after putting together my current horn speakers the amp is now used >400 Hz and I have more SPL capability, so I needed to switch to 6SL7s. This gives me about 26 dB gain from source to amp output. I use a Broski Aikido gain stage, it's almost the same as a White cathode follower but with a PS noise cancellation scheme. I use the same scheme with my EL34s.

I also forgot to mention another advantage of SET is the ability to use an all, or primarily, film-cap power supply. My EL34 amp is PtP wired w/ all Clarity Cap TC Series PS caps. They sound so much better vs electrolytic, the amp is both super-fast, grainless and effortless sounding with the film cap PS. Can't do that with SS without a dishwasher-sized box full of caps, lol...
 

Atmasphere

Industry Expert
May 4, 2010
2,336
1,837
1,760
St. Paul, MN
www.atma-sphere.com
All this “theoretical “ is nice but I have owned balanced gear in the past and sorry those advantages don’t manifest much in practice. The most important part, the sound, was also not better.
As to noise, my amp is dead quiet with 100db speakers...no need for power supply rejection when it is done right.

Any resemblance of class d to tubes is purely superficial...a remotely in depth listen reveals quickly otherwise.
I forgot to mention something called AES48 which is the balanced line standard. Its pretty important if you want all the benefit of balanced operation. That Ultrapath preamp does not support it. Not picking on it in particular- most high end audio balanced products don't support the standard. The most egregious area this happens is how the ground is handled. In a balanced line, ground is ignored. This is done to prevent ground loops from being amplified, and also to help prevent the interconnect cable from affecting the sound. As soon as ground is referenced and used as a signal return path, this ability goes out the window. So its no surprise to me that you experienced what you did. Balanced line should not be controversial- you'd think audiophiles would embrace something that prevents cables from causing colorations; allowing you to extract more musical information. But the waters were sullied by high end audio's general refusal to embrace the standard (although I suspect many manufacturers don't know it even exists).

In a nutshell, the standard is
1) pin 1 ground (chassis), pins 2 and 3 are signal
2) the system is low impedance (this is part of why LOMC phono cartridges work so well balanced)
3) Pin 2 signal is referenced to pin 3 and vice versa- ground is shield only and no audio on it at all

When playing SET amps with high-efficiency, the sound - in terms of clarity and detail, actually sounds like solid state, without the horrifical analytical sound artifacts SS brings to the table. On the other hand, hooking up push pull tube gear to big box speakers (Wilson, Magico, Sonus, etc.) , I am hearing a soft, warm, syrupy, colored sound....

So I find SET amps, actually sounding more transparent than both regular push - pull tube amps, even when possibly distorting on a 95 efficient or higher speaker, and more musical than SS?
You're not talking apples to apples here; most modern loudspeakers are not all that tube friendly being both adverse impedances and low efficiency. This puts even push-pull amps at a disadvantage. Plus there is the issue that not all push-pull amps are the same in terms of bandwidth and distortion- that warm syrupy sound is actually distortion.

I'm of the opinion, and in my experience if you want the best out of any amplifier dollar investment, its a good idea to put that amp on a speaker of moderate to higher impedance, and one with higher efficiency. You can measure how much lower the distortion is of any amplifier when driving a higher impedance and you can hear it too- since a lot of that extra distortion driving lower impedances is going to be higher ordered harmonics.

This IMO is one of the advantages of horns since if they are designed correctly they are as transparent as the best ESLs but not in need of tons of power.


I also forgot to mention another advantage of SET is the ability to use an all, or primarily, film-cap power supply. My EL34 amp is PtP wired w/ all Clarity Cap TC Series PS caps.
Its arguable that in many amplifiers, the power supply capacitors can be seen as coupling caps. This isn't true of all designs but in SETs it certainly is. Our amps use a Circlotron output, and even with no filter caps at all it sounds pretty good as the waveforms in the power supply can cancel. So in our case, the caps can't be construed somehow as coupling caps- this makes the amp less sensitive to what kind of filter capacitance is used. If the driving circuit is also differential, you have a similar power supply immunity (Cross Mode Rejection). But to be clear, any push-pull tube amp can employ film capacitors in the power supply. Its simply a matter of cost and space. But if you have a push-pull transformer coupled output, again its not going to affect the sound all that much- so long as the output section is properly balanced. Rather than use bias controls to do that (which doesn't work) IMO/IME its better to use matched power tubes and then use a constant current source in the cathode circuit of the power tubes to force them to operate differentially. That's when you get the best performance out of the output section.

If we're talking about a tube amplifier I'm a big fan of point to point wiring (or stake boards, like you see in the HK Citation 2) as its so much easier to insure reliability and serviceability down the road. If you do your layout right, you can also get better bandwidth.

Now its well-known that if you really want to make a tube circuit perform, its often a good idea to use cathode bypass caps in the circuit. Traditionally this task is handled by electrolytic capacitors. You could put film there of course, but the device would be enormous in most cases and could 'lead' to layout problems.

But a funny thing happens with electrolytics. If you bias the capacitor with a DC voltage at least 20% higher than the amplitude of the audio waveform that has to pass through it, its distortion can be quite low- with performance similar to film caps. That's why electrolytics can sound just fine as cathode bypass caps, but not as output coupling caps. This is also a problem with non-polar electrolytics used in crossovers, since they don't have a DC bias at all they can generate a bit of distortion. Its because of this sort of mis-use of electrolytics that they've gotten a bad rap with audiophiles.
 

morricab

Well-Known Member
Apr 25, 2014
9,391
4,988
978
Switzerland
I forgot to mention something called AES48 which is the balanced line standard. Its pretty important if you want all the benefit of balanced operation. That Ultrapath preamp does not support it. Not picking on it in particular- most high end audio balanced products don't support the standard. The most egregious area this happens is how the ground is handled. In a balanced line, ground is ignored. This is done to prevent ground loops from being amplified, and also to help prevent the interconnect cable from affecting the sound. As soon as ground is referenced and used as a signal return path, this ability goes out the window. So its no surprise to me that you experienced what you did. Balanced line should not be controversial- you'd think audiophiles would embrace something that prevents cables from causing colorations; allowing you to extract more musical information. But the waters were sullied by high end audio's general refusal to embrace the standard (although I suspect many manufacturers don't know it even exists).

In a nutshell, the standard is
1) pin 1 ground (chassis), pins 2 and 3 are signal
2) the system is low impedance (this is part of why LOMC phono cartridges work so well balanced)
3) Pin 2 signal is referenced to pin 3 and vice versa- ground is shield only and no audio on it at all


You're not talking apples to apples here; most modern loudspeakers are not all that tube friendly being both adverse impedances and low efficiency. This puts even push-pull amps at a disadvantage. Plus there is the issue that not all push-pull amps are the same in terms of bandwidth and distortion- that warm syrupy sound is actually distortion.

I'm of the opinion, and in my experience if you want the best out of any amplifier dollar investment, its a good idea to put that amp on a speaker of moderate to higher impedance, and one with higher efficiency. You can measure how much lower the distortion is of any amplifier when driving a higher impedance and you can hear it too- since a lot of that extra distortion driving lower impedances is going to be higher ordered harmonics.

This IMO is one of the advantages of horns since if they are designed correctly they are as transparent as the best ESLs but not in need of tons of power.



Its arguable that in many amplifiers, the power supply capacitors can be seen as coupling caps. This isn't true of all designs but in SETs it certainly is. Our amps use a Circlotron output, and even with no filter caps at all it sounds pretty good as the waveforms in the power supply can cancel. So in our case, the caps can't be construed somehow as coupling caps- this makes the amp less sensitive to what kind of filter capacitance is used. If the driving circuit is also differential, you have a similar power supply immunity (Cross Mode Rejection). But to be clear, any push-pull tube amp can employ film capacitors in the power supply. Its simply a matter of cost and space. But if you have a push-pull transformer coupled output, again its not going to affect the sound all that much- so long as the output section is properly balanced. Rather than use bias controls to do that (which doesn't work) IMO/IME its better to use matched power tubes and then use a constant current source in the cathode circuit of the power tubes to force them to operate differentially. That's when you get the best performance out of the output section.

If we're talking about a tube amplifier I'm a big fan of point to point wiring (or stake boards, like you see in the HK Citation 2) as its so much easier to insure reliability and serviceability down the road. If you do your layout right, you can also get better bandwidth.

Now its well-known that if you really want to make a tube circuit perform, its often a good idea to use cathode bypass caps in the circuit. Traditionally this task is handled by electrolytic capacitors. You could put film there of course, but the device would be enormous in most cases and could 'lead' to layout problems.

But a funny thing happens with electrolytics. If you bias the capacitor with a DC voltage at least 20% higher than the amplitude of the audio waveform that has to pass through it, its distortion can be quite low- with performance similar to film caps. That's why electrolytics can sound just fine as cathode bypass caps, but not as output coupling caps. This is also a problem with non-polar electrolytics used in crossovers, since they don't have a DC bias at all they can generate a bit of distortion. Its because of this sort of mis-use of electrolytics that they've gotten a bad rap with audiophiles.
Aries Cerat uses film based super caps for the output stage. No big electrolytics.
 

Solypsa

Well-Known Member
Jun 7, 2017
1,811
1,399
275
Seattle
www.solypsa.com
...
.I use a Broski Aikido gain stage, it's almost the same as a White cathode follower but with a PS noise cancellation scheme. I use the same scheme with my EL34s.
...
I built for myself a diy phono using Broskie's Aikido but altered to use LCR for the RIAA. Pretty good I think. As with all things implementation is a lot of the outcome and I need to move the PS outboard to ameliorate some hum. Projects projects projects :)
 

acg

Well-Known Member
Dec 25, 2013
75
84
323
I built for myself a diy phono using Broskie's Aikido but altered to use LCR for the RIAA. Pretty good I think. As with all things implementation is a lot of the outcome and I need to move the PS outboard to ameliorate some hum. Projects projects projects :)

I've started a nothing-left-on-the-table build of a two stage tube based passive LCR Phonostage...21 pieces of iron to help convert 0.2mV MC signal to circa 1V output for the preamplifier. Everything ounce of noise tries to get into those little inductors so lots of mumetal and steel shielding in the build.
 

DasguteOhr

Well-Known Member
Sep 26, 2013
2,354
2,514
645
Germany
a two stage design with ecc 808 tubes telefunken, active riaa. only 12 components for phono. perfect tube for step up transformer at the entrance very low miller capacity. is copied by Klein & Hummel (Neumann) from 1965 and slightly modified.
 

DasguteOhr

Well-Known Member
Sep 26, 2013
2,354
2,514
645
Germany
I've started a nothing-left-on-the-table build of a two stage tube based passive LCR Phonostage...21 pieces of iron to help convert 0.2mV MC signal to circa 1V output for the preamplifier. Everything ounce of noise tries to get into those little inductors so lots of mumetal and steel shielding in the build.
Love LCR Phono , i listen to a Allnic Phono sounds great
 

Solypsa

Well-Known Member
Jun 7, 2017
1,811
1,399
275
Seattle
www.solypsa.com
. Everything ounce of noise tries to get into those little inductors so lots of mumetal and steel shielding in the build.
Yes...in my build I used the Silvercore LCR modules and am quite certain they are picking up the PS hum. Oh boy need to get another chassis and move all the PS.
 

Solypsa

Well-Known Member
Jun 7, 2017
1,811
1,399
275
Seattle
www.solypsa.com
a two stage design with ecc 808 tubes telefunken, active riaa. only 12 components for phono. perfect tube for step up transformer at the entrance very low miller capacity. is copied by Klein & Hummel (Neumann) from 1965 and slightly modified.
Which original module is the design referenced from if I may ask?
 

DasguteOhr

Well-Known Member
Sep 26, 2013
2,354
2,514
645
Germany
klein & hummel has built studio amplifiers and studio speakers, and for music lovers they have set up a joint venture with Saba telewatt. the original is made of telewatt vs 110 amp.
20210217_094710.jpg
 
Last edited:
  • Like
Reactions: Solypsa

Kingrex

Well-Known Member
Feb 3, 2019
2,803
2,312
350
For a SET and Horn thread, this really became a SET thread. I find that interesting for the following. Of the 3 amps I have at my house, a class D. A class AB push pull KT88 and a 845 SET, the amps all sound surprisingly similar. Of the 3 sets of speakers, PAP trio 15 horn, PAP trio 10 Voxative and Sonus Faber Liuto, the speakers are dramatically different in their voicing.

My personal belief is a speaker has much more impact on playback performance than any amp. Of course, you need to have the correct pairing of each other.

I also very much disagree a SET should power the mid/high horn and a SS power the woofer. I researched pretty hard biampimg and in the end gave it up as everyone who did it successfully said they used the exact same amp on the woofer as the top end. As in, the exact same amp. Not the same type of amp. Never did blending of different types of amps lead to cohesion of drivers. They always stood appart. Or so I was lead to believe.
 
  • Like
Reactions: morricab

Al M.

VIP/Donor
Sep 10, 2013
8,683
4,473
963
Greater Boston
I also very much disagree a SET should power the mid/high horn and a SS power the woofer. I researched pretty hard biampimg and in the end gave it up as everyone who did it successfully said they used the exact same amp on the woofer as the top end. As in, the exact same amp. Not the same type of amp. Never did blending of different types of amps lead to cohesion of drivers. They always stood appart. Or so I was lead to believe.

In my monitor/subwoofer system the monitors are powered by tube amp, the subs by internal, high-power (1800 W) class D amps. Both the tube amp and the subs are fed by tube preamp. Blending of subs with monitors is excellent.
 

Kingrex

Well-Known Member
Feb 3, 2019
2,803
2,312
350
Well, Ok. If were going to make subs truly subs, then blending can probably be done pretty well. But when I think of horn as truly an all horn setup, I believe the woofers can climb as high as 1200 hertz. That's probably where mine cross as my Behma mid/high drivers is only rated to 900 herts and up. Then again I am not all horn.
I have though of bringing JL subs into my system for better bottom end since I roll off around 38 herts or so.

But still, this thread was about Horns and SET and to me its seems an inordinate amount of attention is going to the item that has the lesser impact on the overall presentation of playback. In my view. I would think the horn speaker you select is going to have a significant impact on what you hear. The Amp, provided it is not totally failing in capability is going to make up less of the total picture.
 
Jan 18, 2012
2,323
2,409
1,475
Drobak Norway
my dual JBL frontloaded midbasshorns operate from 75-550 where TAD 4003 takes over up to 6K and TAD 2002
all above 75Hz is driven by an all out 300B
tried Nagra VPA but my 300Bs had better midbass too...probably due to properly executed PSU as opposed to Nagra peanut supply
 
  • Like
Reactions: matakana

Parsons

Well-Known Member
Mar 31, 2018
101
80
113
Ohio
I'm (now) a full-range high-efficiency driver guy...whether horned or open baffled (or perhaps even boxed), I think augmenting the true bass with solid state amplification, at least below 120 (some full range can effectively cover a bit lower), is required for the lowest punch and a very-defined set of bass instrumentation, drum strikes, and even piano strikes and effective decay. If you are running multiple drivers (i.e. not "full range") that number obviously might change upwards depending on the driver coverage. I think tubed SET in the 8-15 wpc range can effectively cover the midbass. Just my experience on 4 sets of high-efficiency speakers--two sets of classic horns, and 2 sets of modern open baffles. This was also the approach my 3 sets of full-range box speakers took. YMMV of course.
 

Parsons

Well-Known Member
Mar 31, 2018
101
80
113
Ohio
For a SET and Horn thread, this really became a SET thread. I find that interesting for the following. Of the 3 amps I have at my house, a class D. A class AB push pull KT88 and a 845 SET, the amps all sound surprisingly similar. Of the 3 sets of speakers, PAP trio 15 horn, PAP trio 10 Voxative and Sonus Faber Liuto, the speakers are dramatically different in their voicing.

My personal belief is a speaker has much more impact on playback performance than any amp. Of course, you need to have the correct pairing of each other.

I also very much disagree a SET should power the mid/high horn and a SS power the woofer. I researched pretty hard biampimg and in the end gave it up as everyone who did it successfully said they used the exact same amp on the woofer as the top end. As in, the exact same amp. Not the same type of amp. Never did blending of different types of amps lead to cohesion of drivers. They always stood appart. Or so I was lead to believe.
I agree with your first couple of comments to a very large degree. However, SET amps vs. some of the topologies you mentioned (which I think coincidentally do tend to sound a lot alike--owning all of them you mention) can, however, sound very different--even when all are good "matches" for the same speaker. But yes, changing speakers will change the sound a lot more than changing amps, in general.

Additionally, I can promise folks that properly configuring SET on top and SS on bottom can VERY MUCH pay off. With proper crossover points and methods (for the bass) and attenuation (maybe for both) in the equation, you can be shocked by how effective it can be. I don't think it's a coincidence that a fair number of modern speakers have moved to that model in a lot of their higher-end speakers--by including powered sub drivers integrated into the speaker, and encouraging the usage of lower-powered tubes (including by not limited to SET) on top. I will admit that my success here is mostly on cross-over-less full-range drivers on top, with augmented bass below...if you are trying to get 4-way or 5-way driver arrays all crossed-over properly, then the complications on effective driver integration is obviously made even more complicated by multiple amp types, power-levels and impedance of the different amps.

I had a harder time getting bi-amping with different amps to work as well on two sets of classic Altec horns, where I even had custom crossovers made. So it's NOT a slam dunk--but it's possible.

Just my 2 cents.
 

Solypsa

Well-Known Member
Jun 7, 2017
1,811
1,399
275
Seattle
www.solypsa.com
. ..
My personal belief is a speaker has much more impact on playback performance than any amp. Of course, you need to have the correct pairing of each other.
Never did blending of different types of amps lead to cohesion of drivers. They always stood appart. Or so I was lead to believe.
What is super interesting to me is the apparent contradiction in these statements. Not that I am contesting them. So 'really good amps' of different topologies are more similar than not, yet mixing them yields unsatisfactory results. Thats pretty interesting imo...
 

Kingrex

Well-Known Member
Feb 3, 2019
2,803
2,312
350
What is super interesting to me is the apparent contradiction in these statements. Not that I am contesting them. So 'really good amps' of different topologies are more similar than not, yet mixing them yields unsatisfactory results. Thats pretty interesting imo...
That's a good point. I'm chuckling inside.

What I was hearing from a few people was speaker cohesion is hard to obtain when the amps are different designs. If it was slam dunk easy, I think there would be less of variation in sound between speaker types. I say that as just the makeup of individual drivers varies so much. Not to mention the box surrounding it, or not. So now you have a paper woofer on a baffle, in a box or say in a horn feed with a powerful SS amp. Then you try and blend it with a light aluminum diaphragm, or silk dome horn fed by a quick SET tube. Speaker manufacturer with all their computer modeling and big brain engineers struggle to make it all work. And here I, or some other member on a forum are going to think we can achieve great results by playing around with a mini DSP before sending the settings to some shop to make an active analog setup for us.

I also knocked my head into issues in that the analog active crossover made for my speaker had a 100 ohm input. My preamp want to see 1000 ohms. So right there I may be running into issues. Or I go all digital. But that rubs me wrong when I'm running a very nice STST Motus II TT you set me up with and an open reel 15 ips tape machine I just spent good money on having a custom preamp made to direct wire the play head out. I don't want to turn all that analog into digital.

I truly wish it was easy as I do agree, I am probably taxing my amps and driving them to distortion when I turn it up. Ralph said I have maybe 7 useful watts with my Audion 845. I also don't doubt they make some bloom that is not in the recording. But its pleasant at the volume levels I play it at. And there in, is what it all comes down to. Do you like how it sounds when you play music. At the moment I'm pretty happy. But I get the itch like many other.

Erik, we need to talk phono preamps. I'm curious where it goes beyond my Allnic H1201. Maybe we can chat next week. This week is busy.
Rex
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing