Is there an Inherent Leanness to OTL Designs? Does the Speaker have to be Warm for OTLs to Work their Magic?

caesar

Well-Known Member
May 31, 2010
3,514
248
498
#1
OTLs when paired with the right speakers are some of the most incredible combinations in audio. Soundlab and Atmasphere is an incredible combo, one of the best in all of high end , regardless of price. But Soundlab seems much warmer than its distant cousin, the Martin Logan CLX.

Also, Stenheim with Einstein is outstanding, but Stenheim seems like a warm speaker. Ditto for Einstein with Acapella.

So, is there a leanness to OTL designs or is it just an unexplained, amp-speaker synergy thing / coincidence?
 

Alrainbow

Well-Known Member
Dec 12, 2013
1,854
518
370
NYC , USA
#2
any given amp speaker relationship has a cause and effect. perhaps OTL is a bit less effective in making pairing easy.
The real issue is most buy a speaker they love and then make poor choices on amps or vice versa.
 
Likes: tima

Mike Lavigne

Member Sponsor & WBF Founding Member
Apr 25, 2010
9,095
3,091
820
#3
if you listen to a number of OTL tube amps compared to tube amps with big transformers the common theme is linearity in the bass verses a relatively warm/fat bass. but each type amp can be voiced to be either way, or tubed to be either way. and when i say 'warm/fat' bass, i'm mostly referring to SET bass.

i fell in love with my 75 watt Tenor OTL's back in 2001 and that linearity on the the top and bottom was what pulled me in (along with some fire and sparkle in the mids). and other OTL's do those same things. yet; there are lots of variables involved as well as speaker matching considering it's less tolerant of low impedance.

with most tube amps it's the transformer that is dominant (and most expensive single part) in the sonic signature. so it's logical that lacking a transformer is also a dominant characteristic.

i don't enjoy all OTL amps equally. my two favorite i've heard are the Tenor and Berning 211/845. both i would call linear. i would not use the term 'lean-ness'; i would say natural (like my darTZeel). read Martin Colloms darTZeel review. he relates their sound to a linear tube sound. exactly.
 
Last edited:

Duke LeJeune

[Industry Expert]/Member Sponsor
Jul 22, 2013
584
663
340
Princeton, Texas
#4
So, is there a leanness to OTL designs or is it just an unexplained, amp-speaker synergy thing / coincidence?
Imo it's an EXPLAINED amp/speaker synergy thing, but not many people are aware of the explanation.

Most speakers are voiced on solid state amps which behave like a constant-voltage source, which means they put out MORE power into impedance DIPS, and LESS power into impedance PEAKS. (Some of Nelson Pass's designs are exceptions, and there are probably others.)

OTL amps tend to be constant-power sources, which means they put out approximately the SAME power into the speaker's impedance curve pretty much REGARDLESS of the peaks and dips, as long as they aren't too big. Most transformer-coupled tube amps also exhibit similar behavior, though the more global negative feedback they have, generally the more they lean towards behaving sort of like a constant-voltage (solid state) amp.

Some OTL amps (and some Nelson Pass amps) behave very much like a constant-current source, putting out significantly LESS power into impedance DIPS, and significantly MORE power into impedance PEAKS. The Aries Cerat Collatio comes to mind.

So if we have an 8-ohm speaker designed for use with solid state amps, it will be "voiced" with the expectation of the halving its power into the 16-ohm bass impedance peak, doubling its power into the broad 4-ohm dip in the upper bass region, and sending 1/3 of its power into the 24-ohm peak in the crossover region. (I'm using numbers that make the math easy.)

If we then drive this speaker with an OTL amp, which puts out approximately the same power across the spectrum, the low bass will sound loose and underdamped (twice as much power as expected is going into the 16-ohm bass peak); weak and anemic in the broad upper-bass region (half as much power as expected is going into that 4 ohm dip); and forward/bright/harsh in the crossover region (THREE times as much power as expected is going into the 24-ohm crossover peak).

So if you had NO IDEA that this sort of thing happens, and swap out your solid state amp for an OTL amp, your immediate and logical conclusion would be that OTL amps sound thin and harsh and have flabby bass. But as you can see, the real story is much more complicated than that.

EVERY SINGLE TIME I have been able to find out the impedance curve of a speaker that sounded bad with a good OTL amp, the explanation was obvious.

(Imo one way to design a speaker which works well with both solid-state and OTL amps is this: Keep the impedance curve fairly high and as constant as is reasonably possible, and if it's a ported box, provide for variable port tuning frequencies. Ime there is often 1/4 to 1/3 octave more bass extension available as virtually a "free lunch" when going from solid state to OTL, if the port tuning can be adjusted to take advantage of the OTL's behavior.)

This is my perception from the point of view of a speaker designer. For the view from an amplifier designer's perspective, I suggest this short article:

http://www.atma-sphere.com/en/resources-paradigms-in-amplifier-design.html
 
Last edited:

Ron Resnick

Site Co-Owner, Administrator
Jan 25, 2015
8,208
2,633
730
Beverly Hills, CA
#5
I love the crystalline clarity and transparency of the Einstein OTL Silver Bullet amplifier. I suspect this "see-through" clarity and transparency is common to OTL amplifiers. Unfortunately for me, I hear this as a kind of "leanness."

When I reported to Ralph Karsten that I appear to like the slight warmth of transformer-coupled tube amplifiers he said that I am enjoying their warm-sounding transformer haze (or words to that effect). As usual, Ralph is correct.

Intellectually and theoretically I like the crystalline transparency of OTL, but practically I like the sound of VTL. I wish I didn't. (Oh, and how I would love to commission Ralph to build me a pair of MA-3s!)
 

marmota

Well-Known Member
Feb 3, 2016
89
99
105
#7
Imo it's an EXPLAINED amp/speaker synergy thing, but not many people are aware of the explanation.

Most speakers are voiced on solid state amps which behave like a constant-voltage source, which means they put out MORE power into impedance DIPS, and LESS power into impedance PEAKS. (Some of Nelson Pass's designs are exceptions, and there are probably others.)

OTL amps tend to be constant-power sources, which means they put out approximately the SAME power into the speaker's impedance curve pretty much REGARDLESS of the peaks and dips, as long as they aren't too big. Most transformer-coupled tube amps also exhibit similar behavior, though the more global negative feedback they have, generally the more they lean towards behaving sort of like a constant-voltage (solid state) amp.

Some OTL amps (and some Nelson Pass amps) behave very much like a constant-current source, putting out significantly LESS power into impedance DIPS, and significantly MORE power into impedance PEAKS. The Aries Cerat Collatio comes to mind.

So if we have an 8-ohm speaker designed for use with solid state amps, it will be "voiced" with the expectation of the halving its power into the 16-ohm bass impedance peak, doubling its power into the broad 4-ohm dip in the upper bass region, and sending 1/3 of its power into the 24-ohm peak in the crossover region. (I'm using numbers that make the math easy.)

If we then drive this speaker with an OTL amp, which puts out approximately the same power across the spectrum, the low bass will sound loose and underdamped (twice as much power as expected is going into the 16-ohm bass peak); weak and anemic in the broad upper-bass region (half as much power as expected is going into that 4 ohm dip); and forward/bright/harsh in the crossover region (THREE times as much power as expected is going into the 24-ohm crossover peak).

So if you had NO IDEA that this sort of thing happens, and swap out your solid state amp for an OTL amp, your immediate and logical conclusion would be that OTL amps sound thin and harsh and have flabby bass. But as you can see, the real story is much more complicated than that.

EVERY SINGLE TIME I have been able to find out the impedance curve of a speaker that sounded bad with a good OTL amp, the explanation was obvious.

(Imo one way to design a speaker which works well with both solid-state and OTL amps is this: Keep the impedance curve fairly high and as constant at is reasonably possible, and if it's a ported box, provide for variable port tuning frequencies. Ime there is often 1/4 to 1/3 octave more bass extension available as virtually a "free lunch" when going from solid state to OTL, if the port tuning can be adjusted to take advantage of the OTL's behavior.)

This is my perception from the point of view of a speaker designer. For the view from an amplifier designer's perspective, I suggest this short article:

http://www.atma-sphere.com/en/resources-paradigms-in-amplifier-design.html
Spectacular response, a pleasure to read!
 
Likes: lordcloud

Alrainbow

Well-Known Member
Dec 12, 2013
1,854
518
370
NYC , USA
#8
Imo it's an EXPLAINED amp/speaker synergy thing, but not many people are aware of the explanation.

Most speakers are voiced on solid state amps which behave like a constant-voltage source, which means they put out MORE power into impedance DIPS, and LESS power into impedance PEAKS. (Some of Nelson Pass's designs are exceptions, and there are probably others.)

OTL amps tend to be constant-power sources, which means they put out approximately the SAME power into the speaker's impedance curve pretty much REGARDLESS of the peaks and dips, as long as they aren't too big. Most transformer-coupled tube amps also exhibit similar behavior, though the more global negative feedback they have, generally the more they lean towards behaving sort of like a constant-voltage (solid state) amp.

Some OTL amps (and some Nelson Pass amps) behave very much like a constant-current source, putting out significantly LESS power into impedance DIPS, and significantly MORE power into impedance PEAKS. The Aries Cerat Collatio comes to mind.

So if we have an 8-ohm speaker designed for use with solid state amps, it will be "voiced" with the expectation of the halving its power into the 16-ohm bass impedance peak, doubling its power into the broad 4-ohm dip in the upper bass region, and sending 1/3 of its power into the 24-ohm peak in the crossover region. (I'm using numbers that make the math easy.)

If we then drive this speaker with an OTL amp, which puts out approximately the same power across the spectrum, the low bass will sound loose and underdamped (twice as much power as expected is going into the 16-ohm bass peak); weak and anemic in the broad upper-bass region (half as much power as expected is going into that 4 ohm dip); and forward/bright/harsh in the crossover region (THREE times as much power as expected is going into the 24-ohm crossover peak).

So if you had NO IDEA that this sort of thing happens, and swap out your solid state amp for an OTL amp, your immediate and logical conclusion would be that OTL amps sound thin and harsh and have flabby bass. But as you can see, the real story is much more complicated than that.

EVERY SINGLE TIME I have been able to find out the impedance curve of a speaker that sounded bad with a good OTL amp, the explanation was obvious.

(Imo one way to design a speaker which works well with both solid-state and OTL amps is this: Keep the impedance curve fairly high and as constant at is reasonably possible, and if it's a ported box, provide for variable port tuning frequencies. Ime there is often 1/4 to 1/3 octave more bass extension available as virtually a "free lunch" when going from solid state to OTL, if the port tuning can be adjusted to take advantage of the OTL's behavior.)

This is my perception from the point of view of a speaker designer. For the view from an amplifier designer's perspective, I suggest this short article:

http://www.atma-sphere.com/en/resources-paradigms-in-amplifier-design.html
When you refer to power I assume you watts ? Am I correct ?
 

Alrainbow

Well-Known Member
Dec 12, 2013
1,854
518
370
NYC , USA
#9
Imo it's an EXPLAINED amp/speaker synergy thing, but not many people are aware of the explanation.

Most speakers are voiced on solid state amps which behave like a constant-voltage source, which means they put out MORE power into impedance DIPS, and LESS power into impedance PEAKS. (Some of Nelson Pass's designs are exceptions, and there are probably others.)

OTL amps tend to be constant-power sources, which means they put out approximately the SAME power into the speaker's impedance curve pretty much REGARDLESS of the peaks and dips, as long as they aren't too big. Most transformer-coupled tube amps also exhibit similar behavior, though the more global negative feedback they have, generally the more they lean towards behaving sort of like a constant-voltage (solid state) amp.

Some OTL amps (and some Nelson Pass amps) behave very much like a constant-current source, putting out significantly LESS power into impedance DIPS, and significantly MORE power into impedance PEAKS. The Aries Cerat Collatio comes to mind.

So if we have an 8-ohm speaker designed for use with solid state amps, it will be "voiced" with the expectation of the halving its power into the 16-ohm bass impedance peak, doubling its power into the broad 4-ohm dip in the upper bass region, and sending 1/3 of its power into the 24-ohm peak in the crossover region. (I'm using numbers that make the math easy.)

If we then drive this speaker with an OTL amp, which puts out approximately the same power across the spectrum, the low bass will sound loose and underdamped (twice as much power as expected is going into the 16-ohm bass peak); weak and anemic in the broad upper-bass region (half as much power as expected is going into that 4 ohm dip); and forward/bright/harsh in the crossover region (THREE times as much power as expected is going into the 24-ohm crossover peak).

So if you had NO IDEA that this sort of thing happens, and swap out your solid state amp for an OTL amp, your immediate and logical conclusion would be that OTL amps sound thin and harsh and have flabby bass. But as you can see, the real story is much more complicated than that.

EVERY SINGLE TIME I have been able to find out the impedance curve of a speaker that sounded bad with a good OTL amp, the explanation was obvious.

(Imo one way to design a speaker which works well with both solid-state and OTL amps is this: Keep the impedance curve fairly high and as constant at is reasonably possible, and if it's a ported box, provide for variable port tuning frequencies. Ime there is often 1/4 to 1/3 octave more bass extension available as virtually a "free lunch" when going from solid state to OTL, if the port tuning can be adjusted to take advantage of the OTL's behavior.)

This is my perception from the point of view of a speaker designer. For the view from an amplifier designer's perspective, I suggest this short article:

http://www.atma-sphere.com/en/resources-paradigms-in-amplifier-design.html
One more question is the imp curve you post of in part due to an OTL relative high imp output ? I do feel your level imp curve and can add an above 8 ohm relative is better. but I think it’s not just due imp curve of speaker but also the inherent low amp high voltage output of tubes. any thoughts ?
 
May 30, 2010
16,977
1,744
720
Portugal
#10
I love the crystalline clarity and transparency of the Einstein OTL Silver Bullet amplifier. I suspect this "see-through" clarity and transparency is common to OTL amplifiers. Unfortunately for me, I hear this as a kind of "leanness."

When I reported to Ralph Karsten that I appear to like the slight warmth of transformer-coupled tube amplifiers he said that I am enjoying their warm-sounding transformer haze (or words to that effect). As usual, Ralph is correct.

Intellectually and theoretically I like the crystalline transparency of OTL, but practically I like the sound of VTL. I wish I didn't. (Oh, and how I would love to commission Ralph to build me a pair of MA-3s!)
No experience with the Einstein OTL Silver Bullet , I have owned Atmasphere MA50, MA2 and and MA2mk3. I think Ralph was addressing the general tube amplifiers, not particularly the VTL Siegfried II - I do not notice warm-sounding transformer haze in this amplifier and I do not consider the Siegfried's as warm! The VTLs are however more powerful in the bass than the MA2, that seems "faster". IMHO the difference you refer can be associated more to the preamplfiers than to the power amplifiers.

If it was not for the mechanical humming problems of the MA2 power amplifiers used with 50 Hz mains at that time and the heat I would still have considered the Atmasphere option. Unfortunately many excellent US made amplifiers and equipment suffer from this problem and I can't endure with it - some local people are more forgiving than me in this aspect.

The Graph OTLs with adequate speakers were not lean at all - I remember that they were warmer than the MA50.
 

spiritofmusic

Well-Known Member
Jun 13, 2013
11,887
2,661
653
E. England
#11
I love the crystalline clarity and transparency of the Einstein OTL Silver Bullet amplifier. I suspect this "see-through" clarity and transparency is common to OTL amplifiers. Unfortunately for me, I hear this as a kind of "leanness."

When I reported to Ralph Karsten that I appear to like the slight warmth of transformer-coupled tube amplifiers he said that I am enjoying their warm-sounding transformer haze (or words to that effect). As usual, Ralph is correct.

Intellectually and theoretically I like the crystalline transparency of OTL, but practically I like the sound of VTL. I wish I didn't. (Oh, and how I would love to commission Ralph to build me a pair of MA-3s!)
Is the Alieno that Emile of Taiko Audio and Audioquattr both use, an exception to this description?
 

Ron Resnick

Site Co-Owner, Administrator
Jan 25, 2015
8,208
2,633
730
Beverly Hills, CA
#12
Is the Alieno that Emile of Taiko Audio and Audioquattr both use, an exception to this description?
That is a good question. I don't know. It might be.

That amp uses transistors for current gain, if I remember correctly. I think it's probably inapposite for this OTL analysis.
 

christoph

Well-Known Member
Dec 12, 2015
2,405
1,610
290
Principality of Liechtenstein
#13
Keep the impedance curve fairly high and as constant at is reasonably possible,
This is why the Apogee Studio Grand/Studio Ribbon Array is so very easy to drive:
6 Ohm impedance and ruler flat over the whole frequency range :cool:
 

morricab

Well-Known Member
Apr 25, 2014
4,657
1,211
388
Switzerland
#14
No experience with the Einstein OTL Silver Bullet , I have owned Atmasphere MA50, MA2 and and MA2mk3. I think Ralph was addressing the general tube amplifiers, not particularly the VTL Siegfried II - I do not notice warm-sounding transformer haze in this amplifier and I do not consider the Siegfried's as warm! The VTLs are however more powerful in the bass than the MA2, that seems "faster". IMHO the difference you refer can be associated more to the preamplfiers than to the power amplifiers.

If it was not for the mechanical humming problems of the MA2 power amplifiers used with 50 Hz mains at that time and the heat I would still have considered the Atmasphere option. Unfortunately many excellent US made amplifiers and equipment suffer from this problem and I can't endure with it - some local people are more forgiving than me in this aspect.

The Graph OTLs with adequate speakers were not lean at all - I remember that they were warmer than the MA50.
The Graaf GM20 OTLs sounded fantastic with Reference 3A speakers. I also had a pair of Silvaweld OTL Reference monos (100 watt with 6C33C output tubes) that were pretty amazing sounding on ribbons, my Acoustats and some JBL C50 Olympus speakers. Only the heat drove me to get rid of them (8 x 6C33C can make really a lot of heat). I also had a pair of Transcendent Sound Beast monos that sound really awesome with normal speakers but with my electrostats they became unstable and would start oscillating, so I had to sell them.
 

morricab

Well-Known Member
Apr 25, 2014
4,657
1,211
388
Switzerland
#15
One of the best sounds I heard with Odeon speakers was with the Einstein OTLs...shockingly transparent... a real eye opener... Those same OTLs also sounded really great with the Living Voice Avatar OBXR2 speakers.
 

Atmasphere

[Industry Expert]
May 4, 2010
1,014
225
485
St. Paul, MN
www.atma-sphere.com
#16
So, is there a leanness to OTL designs or is it just an unexplained, amp-speaker synergy thing / coincidence?
Its not OTLs. It is the match between the amp and speaker like it always is :)
. but each type amp can be voiced to be either way,
IME anyone that 'voices' their amplifier is making a mistake and I don't know anyone that actually does that. It means your product is married to a synergy. This usually means that it will not be as transparent and low distortion as it could be.
Unfortunately for me, I hear this as a kind of "leanness."

When I reported to Ralph Karsten that I appear to like the slight warmth of transformer-coupled tube amplifiers he said that I am enjoying their warm-sounding transformer haze (or words to that effect)
The actual words were "lower ordered harmonics"; not so much the transformer. OTLs FWIW do not have to sound lean, but they are load sensitive and so its a good idea to audition them against other amps on a speaker that does not have an adverse impedance at low frequencies. The impedance curve does not have to be flat. These days its very common to have speakers with dual woofers which are 4 ohms in the bass and then nominally 8 ohms in the mids and highs. OTLs are going to sound lean on such a speaker, unless they have so much feedback applied that they can act as a voltage source. This will allow them to be flat on the speaker, but seriously, you are putting any amplifier made at a disadvantage on such a speaker!! The reason is simple- tubes don't do well driving lower impedances (with output transformers, you can easily lose as much as an octave of bass bandwidth in the output transformer going from the 8 ohm tap to the 4 ohm tap, and the output transformer will also run warmer). But solid state doesn't do well either as its distortion goes up into lower impedances.

Now think about that last bit carefully. If your distortion is higher with your solid state amp at 100 Hz, at 3000 or 7000 Hz you will be hearing higher ordered harmonics of the bass notes, and since the ear interprets higher ordered harmonics as brightness and harshness this will cause the system to sound less musical. There's no good argument for 4 ohms, and there are plenty of arguments against.

The VTLs are however more powerful in the bass than the MA2, that seems "faster".
It is, and its easy to hear and measure. The risetime of the output section of the MA-2 is about 600V/microsecond. Most transformer coupled tube amps are more like 15V, so well over an order of magnitude faster.
If it was not for the mechanical humming problems of the MA2 power amplifiers used with 50 Hz mains at that time and the heat I would still have considered the Atmasphere option. Unfortunately many excellent US made amplifiers and equipment suffer from this problem and I can't endure with it - some local people are more forgiving than me in this aspect.

The Graph OTLs with adequate speakers were not lean at all - I remember that they were warmer than the MA50.
We solved the 50Hz noise thing in the MA-2 about ten-twelve years ago. The original EI core transformers made noise even on 60Hz (actually it was only the B+ transformer). We went to toroids and no more issues, plus the amp is lighter :) The M-50 had less power than the Graff, so its likely to be drier, but the M-50 was discontinued about 1993 (27 years ago for those keeping track) and replaced by the M-60, which was in turn replaced by the M-60 MkII, the M-60 MkII.1, .2 and .3, then the M-60 MkIII and now we're on the M-60 Mk3.3, if you get my drift. They have more authority and and distortion of all types is much lower; we've actually made improvements in the last few decades.
How cool would that be! Well ... you will need to modify your room to include open stacks or chimneys above them to give the heat somewhere to go.

500 watts and 54 tubes ... per channel
FWIW with that many tubes the output section is so efficient that each power tube runs quite a bit cooler than in any of our smaller amps- so it makes barely any more heat than a set of MA-2s! Plus if you don't need the power the output tubes can be shut off entirely in banks- there are three banks (as you can see in the photo) so 1/3rd power or 2/3 power is there to help you manage heat. But seriously, if you can afford an amp like that you can also afford to vent the ceiling for a few hundred dollars with some inexpensive duct work and bring the heat outside, so you don't have to overload your air conditioning. A customer of ours with MA-3s did exactly that. He lives in Guatemala City and you can be comfortable in his listening room all day and all night, even without air conditioning. I know- I've been there.
 
May 30, 2010
16,977
1,744
720
Portugal
#17
(...) We solved the 50Hz noise thing in the MA-2 about ten-twelve years ago. The original EI core transformers made noise even on 60Hz (actually it was only the B+ transformer). We went to toroids and no more issues, plus the amp is lighter :) (...)
Great to know that the problem has been solved. If I manage to sell the ML3's I will probably consider a MA2 again - I still have several large boxes filled with 6AS7G's and my next listening room will have air conditioning!
 

olivier

VIP/Donor
Nov 27, 2016
42
30
225
Munich
www.olivier-schmidt.com
#19
No experience with the Einstein OTL Silver Bullet , I have owned Atmasphere MA50, MA2 and and MA2mk3. I think Ralph was addressing the general tube amplifiers, not particularly the VTL Siegfried II - I do not notice warm-sounding transformer haze in this amplifier and I do not consider the Siegfried's as warm! The VTLs are however more powerful in the bass than the MA2, that seems "faster". IMHO the difference you refer can be associated more to the preamplfiers than to the power amplifiers.

If it was not for the mechanical humming problems of the MA2 power amplifiers used with 50 Hz mains at that time and the heat I would still have considered the Atmasphere option. Unfortunately many excellent US made amplifiers and equipment suffer from this problem and I can't endure with it - some local people are more forgiving than me in this aspect.

The Graph OTLs with adequate speakers were not lean at all - I remember that they were warmer than the MA50.
A bit late but i read your answer just a few seconds ago....
I had a 50hz problem with my CAT JL5 and solved it by using my PS Audio P10. My dealer told me to try this. First i didn´t want to use it cause i
didn´t want to loose dynamics but there is no dynamic lack with it
Perhaps this helps people having this 50hz problem
 

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. A place where audiophiles and audio companies discuss existing and new audio products, music servers, music streamers and computer audio, digital to audio converters, turntables, phono stages, cartridges, reel to reel, speakers, headphones, tube amplifiers 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