Tube preamp & Class D Power Amp. Is it best of both worlds?

captrajesh

Moderator
Staff member
Joined
Oct 18, 2009
Messages
5,230
Points
113
Location
Mumbai/ Hyderabad
Sometime back, (Somewhere between March and May of this year) there was a discussion on types of preamps and power amps, synergies between them etc. and I had read about a post by our Moderator Kanwar stating that having a tube preamp and a Class D power amp would offer best of both the worlds.

He had given a wonderful explanation as to why it is so. I've been searching for the post for over a month but can't seem to find it. So pinged Kanwar but he couldn't locate it either. As advised by him, I'm starting this new thread to discuss the issue. This would not only be an education for me as I'm thinking of pairing a DHT preamp with Class D Power amps but to others as well.

My interest of starting the thread is, as many of the regular FMs may be aware, I've made my vintage 15" Tannoy Monitor Gold DC Drivers in Rectangular GRF cabinets. I've not got them up and ready. Unfortunately, the active crossover box hasn't been completed fully and this is giving trouble in playing the system with issues of excessive hum and thumps. I intend to fix it ASAP. I want to also get a clear direction on which way to go because I already have a Pass DIY B1 Buffer preamp, Class D Audio CDA254 & Topping Power amps.

I'm inviting Kanwar (and also other knowledgeable FMs) for doling out his expert advice.
 
Rajesh

I dont know if one can be categorical about anything in audio-"Best of Both worlds"? I have both valve pre and ss pre and Class D and ss and valve power amps. To my mind each combo is a different flavour. Having just come back from a trip to Calcutta and sampled Bong sandesh at Balaram Mullick and Bhojohari Manna, I loved it. Is it the best? I dont know. Tres Luches at Mexico City is an outstanding dessert, as is gelao in Italy and so is Moong dal ka halwa and on and on. At the end of the day I think the valve pre and SS or Class D power is a good combo, but one of many good combos.
 
Rajesh

I dont know if one can be categorical about anything in audio-"Best of Both worlds"? I have both valve pre and ss pre and Class D and ss and valve power amps. To my mind each combo is a different flavour. Having just come back from a trip to Calcutta and sampled Bong sandesh at Balaram Mullick and Bhojohari Manna, I loved it. Is it the best? I dont know. Tres Luches at Mexico City is an outstanding dessert, as is gelao in Italy and so is Moong dal ka halwa and on and on. At the end of the day I think the valve pre and SS or Class D power is a good combo, but one of many good combos.

Being Bong, and loving Rosogolla and other sweets and any food that is sweet, I couldn't stop appreciating above statement.

Can quality of a music be measured? NO. Let me know if it really can.

Sound quality can be measured but not with a music lover's ears.

Bob Dylan's song, a Baul singing in a village through an old mike may sound wonderful to me but could sound as noise to someone habituated to listening live programme of silk voice Jagjit Singh only (no pun to anyone but pure personal opinion).

We should do what best we can and enjoy the result.

I have an SET amp for mid and high and a tiny Lepai for low frequency drivers of my OB. It always surprises me. My expectation is not low (be reminded):).
But that same SET played with a 2.5 way FS speaker alone sound little boomy and sloppy in bass section.

Go ahead, make one. We all will be somehow benefitted along with you.
 
@GeorgeO, fully in agreement with you as far as flavours are concerned.

What Kanwar described in that post was more with regard to technical merits of the combination, making use of the goodness of tubes and that of the Class D amps from technical perspective.

I wanted to brush up my knowledge regarding that to get clarity on which way to go.
 
Captn, in these past few years i have heard tube amps sounding Hard with Slam, Transistor amps sounding soft and Mushy and completely digital setups sounding as organic as it gets !
So talking in terms of a tube Preamp and Class D amp would be getting caught in terminologies rather than the sound.

Since you have your speaker and your room defined the choice is in finding a Amp which can drive it well ie able to supply all the current needed by the speaker at the instant it needs it....and since all amps are not the same presenting it in the way your ears want to hear it

Since you have gone for a Vintage Driver, it would be ideal to go around to check what are the rest of the folks driving it with to get a good idea of your target.

Unless we have folks here who know what a Tannoy gold+ GRF is all about , i dont suppose the info is going to be as good as what existing owners can give. From what my reading was on this when i was looking at GRF/RHR speakers, it needs a lively amp unlike the Altecs which need mellow ones. While the ideal pairing I have heard about are old Macintosh SS or Tubed Amps or a QUAD amp , they are older amps.

I would think a good DHT Pre and a 50-100 w SS or Tube amp should go very well. Class D is not something I have been exposed to and am sure as long as it is done well, it will sound great. I believe the older drivers were far more forgiving and so were the amps that used to run them in those days. putting in components of today which are a lot more detailed could go either way ;)

Do check this gentlemans journey on audiogon with HPD drivers AudiogoN Forums: The Summit
 
Last edited:
if you are looking at high power designs, class D might make sense, but otherwise, given the low wattages most class T amps run, and the power requirements for High sensitivity drivers, you're better off implementing a class A stage, preferably with FET output drivers, like the hybrid zen stages. FET's have very tube-ish output characteristics, which go well with a low negative feedback design
 
Thank you all for the replies guys. :)
Go ahead, make one.
I think your post crossed with my second post where I had clarified as to why I had started the thread.

We all will be somehow benefitted along with you.
That's the whole idea. :)

Captn, in these past few years i have heard tube amps sounding Hard with Slam, Transistor amps sounding soft and Mushy and completely digital setups sounding as organic as it gets !
So talking in terms of a tube Preamp and Class D amp would be getting caught in terminologies rather than the sound.
if you are looking at high power designs, class D might make sense
While I agree with you, I just wanted some elaboration on Kanwar's post that I had referred in the first post which talks about the technical merits of going in for tube preamplifier and Class D Power amplifier. Moreover, I already have a power amplifier from Class D Audio and it sounds quite organic to me.

Unless we have folks here who know what a Tannoy gold+ GRF is all about , i dont suppose the info is going to be as good as what existing owners can give. From what my reading was on this when i was looking at GRF/RHR speakers, it needs a lively amp unlike the Altecs which need mellow ones. While the ideal pairing I have heard about are old Macintosh SS or Tubed Amps or a QUAD amp, they are older amps.
Since I'm going with an active crossover, I require four channels of power amplification. I've had some brief sessions mating the B1 with a DIY Class AB power amp (A 6 channel power amp given away to me by Hydrovac for my HT setup) and my Class D Audio Power amp driving single speaker at a time. Both have very distinctive sound signatures.

I would think a good DHT Pre and a 50-100 w SS or Tube amp should go very well. Class D is not something I have been exposed to and am sure as long as it is done well, it will sound great.
I've loaned a Topping amp from FM Ravi_d to drive the tweeters and intend to use the Class D Power amp to drive the woofer. But the problem I have is, the output impedance of the Lyrita DHT preamp is way higher for the 7 KOhm input impedance of the Class D power amp. Not sure of the input impedance of the Topping though.

Do check this gentlemans journey on audiogon with HPD drivers AudiogoN Forums: The Summit
Thanks for sharing the link Arj, will go through it.

given the low wattages most class T amps run, and the power requirements for High sensitivity drivers, you're better off implementing a class A stage, preferably with FET output drivers, like the hybrid zen stages. FET's have very tube-ish output characteristics, which go well with a low negative feedback design
Yeah but the Class T amp would drive only the tweeter. The Woofer is driven by CDA254 power amp from Class D audio which churns out 125 WRMS @ 8 ohms. Since the woofers are 15 Ohms, it would still provide about 60 to 65 WRMS.
 
IMHO the OT wanted technical discussion but some of our FM's including Mod as always derail such technical threads with subjectivism :D Sarson da Saag bhi mangwalo :licklips:

it wasnt meant to derail it and I dont think Rajesh or others felt it was derailing it.

Though my post was to discuss 'Technical merits' of going in for an implementation of Tube pre and Class D Power, I certainly wouldn't mind subjective discussion regarding the traits of various amplifier topologies.

Kuch Chat Pata ho jaye. ;) (Let's get something spicy;))
 
Kuch Chat Pata ho jaye. ;) (Let's get something spicy;))

I thought only Keith has the knack of stirring controversies. :D


Ok Coming back to the Topic...........

Tubes are inherently linear in voltage amplification than semiconductors without the need of feedback. This trait of tubes is very useful in extracting best out of them which is using them purely for voltage amplification.

Whereas Class-D amplifiers are highly versatile when delivering reactive current which are needed by loudspeakers in order to drive them. No other linear amplifier class comes closer to the reactive current drive feature of Class-D amps. They effortlessly give reactive current SINK-SOURCE and again recycle them back to the power supply.

The distortion mechanism completely changes when the loudspeaker is consuming the current which is not in synced phase with voltage. Linear amplifiers suffer from many problems here while delivering reactive currents. Though they can happily drive linear loads with ultra low distortion profiles but when it comes to phase lead/lag between voltage and current waveforms the distortion kicks in and byproducts are heard in sonics.

Class-D hence can be used for producing the current required to drive the hardest of loudspeakers, in form of buffers with unity gain.

In simple terms use the Tubes for voltage amplification and Class-D for current amplification. Here the combo will provide the tube dynamics with ample reactive current drive from the class-D buffer amp. Large signal dynamics are preserved in greater extent in such a way.

There is alot more to this..... to be continued.
Cheers,
Kanwar
 
Last edited:
........ a tube preamp and a Class D power amp would offer best of both the worlds.........

This isn't a pro and con type reply.

A very good tube preamp should be transparent and so it shouldn't have a 'sound of it's own' like 'warm' or 'fuzzy' or 'musical' etc. So the sound shouldn't change in or out of circuit before the power amp.

However a tube preamp can come in handy because they hardly load the source due to ( usually) high input impedance and often that can make a difference to the sound.
If they also have low output impedance ( like a Cathode follower) or other configuration they will not alter the sound going into the low zin power amp.

I am currently using a two stage tube preamp ( no nfb) with a SMSL classD power amp. Sounds superb. Better than my EL34 PP tube power amp ...to my ears. The SMSL seems to load the source differently on it's own and so having the tube preamp does help a bit. But the sound is good both ways, with or without the tube preamp even though there is a slight difference. The difference cannot be called a 'tube sound' !

But as always, your system is different from mine and can behave differently. Do try what doesn't cost an arm and a leg. I might add that done carefully there is no 'hum' problem with a tube front end. With a very low Zin of your classD the tube stage should make a very big difference. NOT by making it sound 'tubey' ! ;)

Edit: Just remembered before going to bed.

You can easily get significant second harmonic distortion to make the sound ' warm' ! Just remove any buffer output stage. That will place a larger load on the amplifying stage and cause higher distortion, mostly second harmonic for a triode. However, not a good idea I think !
 
Last edited by a moderator:
FYI, Rogue Sphinx is one such amp. It is an integrated amp that has a tube preamp stage and a Class D (Hypex module) power amp and a linear power supply. I am sure there are others but this one came to mind. Also happened to recently read its review on stereophile so I thought I would share.
 
Captain beware of 'HUMMING' in some Tube DHT preamps, Class-D amp are very sensitive to such issues and with high sensitivity speakers you ears will get irritated. I hope you got what i have meant :D
 
I came across a couple of posts from Jason Stoddard the other day, about why Schiit does not make D-class amps. I don't know if it is ok with Head-Fi.org to cut&paste complete posts, so please click the following to read

To answer the questions about our prejudice against Class D, we'd have to go into both practical and philosophical domains.... (Click to read)

[He means ultrasonic noise ...] Nope, I mean RF radiated noise..... (Click to read)

Kanwar can probably relate from the manufacturer's point of view, and suggest what a DIYer may need to do to stop RF radiation from the amp affecting the flavour of food in the microwave oven :eek:!
 
I thought only Keith has the knack of stirring controversies. :D
No controversies at all. All that I'm saying is, all types of discussions are welcome. :)

Tubes are inherently linear in voltage amplification than semiconductors without the need of feedback. This trait of tubes is very useful in extracting best out of them which is using them purely for voltage amplification.

Whereas Class-D amplifiers are highly versatile when delivering reactive current which are needed by loudspeakers in order to drive them. No other linear amplifier class comes closer to the reactive current drive feature of Class-D amps. They effortlessly give reactive current SINK-SOURCE and again recycle them back to the power supply.

The distortion mechanism completely changes when the loudspeaker is consuming the current which is not in synced phase with voltage. Linear amplifiers suffer from many problems here while delivering reactive currents. Though they can happily drive linear loads with ultra low distortion profiles but when it comes to phase lead/lag between voltage and current waveforms the distortion kicks in and byproducts are heard in sonics.

Class-D hence can be used for producing the current required to drive the hardest of loudspeakers, in form of buffers with unity gain.

In simple terms use the Tubes for voltage amplification and Class-D for current amplification. Here the combo will provide the tube dynamics with ample reactive current drive from the class-D buffer amp. Large signal dynamics are preserved in greater extent in such a way.

There is alot more to this..... to be continued.

Hmm good read Kanwar. Now if the virtues of tubes as great voltage amplifiers (for line stage) and Class D as great current drive amplifiers (for power amplification) are known, I'm wondering as to why hasn't this become a industry standard (yet!)

However a tube preamp can come in handy because they hardly load the source due to ( usually) high input impedance and often that can make a difference to the sound.
If they also have low output impedance ( like a Cathode follower) or other configuration they will not alter the sound going into the low zin power amp.

Since it is an established norm that the input impedance of the power amp must be atleast 10 times the output impedance of the preamp, how does one get over the problem where the power amplifier has unusualy low input impedance?

I am currently using a two stage tube preamp ( no nfb) with a SMSL classD power amp. Sounds superb. Better than my EL34 PP tube power amp ...to my ears. The SMSL seems to load the source differently on it's own and so having the tube preamp does help a bit.

Great to know. Which tube preamp is this and what is it's output impedance? Also, what is the input impedance of the SMSL Power amp?

With a very low Zin of your classD the tube stage should make a very big difference.
Didn't get you here. Are you referring to the low input impedance of the Class D power amp?
 
Last edited:
I came across a couple of posts from Jason Stoddard the other day, about why Schiit does not make D-class amps. I don't know if it is ok with Head-Fi.org to cut&paste complete posts, so please click the following to read

To answer the questions about our prejudice against Class D, we'd have to go into both practical and philosophical domains.... (Click to read)

[He means ultrasonic noise ...] Nope, I mean RF radiated noise..... (Click to read)

Kanwar can probably relate from the manufacturer's point of view, and suggest what a DIYer may need to do to stop RF radiation from the amp affecting the flavour of food in the microwave oven :eek:!

Thad,

After reading those posts i can only say one thing : Schiitt Audio could be good at designing analog audio but playing with PWM/Class-D RF power electronics is not their game. If he is unable to design a proper pcb which can help in cutting the radiated RF emissions, its his problem only.:)

That guy is saying about non approval from FCC , let me tell you how come manufactures such as QSC are getting the FCC [Class-A] approvals for their Class-D + SMPS amplifiers.

PLD4.2 Power Amplifier with DSP - QSC

Click specs and scroll down for agency approvals.:D

Someone ask that guy how come PCs and laptop chargers meet FCC [Class-B] Approvals?

FCC Class-A are for commercial and Class-B are for consumer.

There are ways and there are people who know how to do it, isn't it?




Hmm good read Kanwar. Now if the virtues of tubes as great voltage amplifiers (for line stage) and Class D as great current drive amplifiers (for power amplification) are known, I'm wondering as to why this hasn't become a industry standard!

I am yet to see any direct-coupled DHT Tube preamp on this planet, Please show me if any?? Designers are still using Capacitive & Transformer coupling in traditional manner. No innovation at all.

Class-D is new phenomenon and I can only see few people who can deliver the performance in that arena. Hence it will take some more time for it to become an industry standard, if that is how you perceive it.


Cheers,
Kanwar
 
Last edited:
Since it is an established norm that the input impedance of the power amp must be atleast 10 times the output impedance of the preamp...
...... Which tube preamp is this and what is it's output impedance? Also, what is the input impedance of the SMSL Power amp?
....Didn't get you here. Are you referring to the low input impedance of the Class D power amp?

That 1:10 is only a guideline. It can be 1:1 if you know what you are doing etc.... It was suggested just to indicate that most of the signal will appear on the load if the following input impedance is significantly larger than the output impedance. This also does not mean that if the preamp output impedance is 100 ohms you can use a 1 K ohm load ! You could over load the preamp severely.

My tube preamp is a two stage preamp. First stage amplifies and the second stage is a buffer , a cathode follower ( gain just under x1) for a low output impedance. Uses a 12AU7 and a 6922 . Output impedance hasn't been measured. Simulation shows it to be 110 ohms.
The SMSL-50 has an input impedance of 50K ohms when measured in the off state with a DMM. ( minimum was 41K ohms with wide open volume ) . However it might be lower than this when the unit is powered up. With the circuit switched on the impedance of the active circuit will come in parallel with the volume control. I have found that I usually do not advance the volume control beyond about 1/3 rd of a turn and so the effective input impedance will be quite high.

I meant that using a low Zout tube preamp with your low Zin classD amp should work out well. The tube stage will provide the high input impedance required for the system.

There's nothing exotic about tube circuits ! We are all using what was designed in the 40's or so ! You just need to know what value of components to use to suit your purpose. As preamps they are often quite forgiving with component value changes. If you spend a few days on the Net you will get to know everything about them ! You can then talk like an 'expert' !:)
Isn't the Net great ?;)

PS. You will find a lot of people on the Net disliking the 12AU7 in the amplifying stage. It's less linear than many other tubes but in my application it 'sounds' fine. I have a 6CG7 ( 9 pin version of 6SN7 ) which is far more linear but I don't think I'll get an audible improvement though there will certainly be a measured improvement. Main thing is that I can't hear any degradation with the current tube stage in or out of circuit. That's enough for me.
 
Last edited:
This also does not mean that if the preamp output impedance is 100 ohms you can use a 1 K ohm load ! You could over load the preamp severely.

Mr. Fantastic

Kindly explain under what conditions 1 kilo-Ohm load can severly overload a preamp with output impedance of 100 Ohms.... ??:)
 
Last edited:
Mr. Fantastic

Kindly explain under what conditions 1 kilo-Ohm load can severly overload a preamp with output impedance of 100 Ohms.... ??:)

Please build a suitable tube preamp stage and test it all all acceptable signal levels.
I've been doing it for over 30 years. Maybe I know a thing or two !
If you are saying a 100mV signal will not overload , please, stop being simplistic ! This isn't what the statement meant !
 
Please build a suitable tube preamp stage and test it all all acceptable signal levels.
I've been doing it for over 30 years. Maybe I know a thing or two !
There's nothing exotic about tube circuits ! We are all using what was designed in the 40's or so ! You just need to know what value of components to use to suit your purpose. As preamps they are often quite forgiving with component value changes. If you spend a few days on the Net you will get to know everything about them ! You can then talk like an 'expert' !:)
Isn't the Net great ?;)

Sir,
I assume you have spent numerous days on internet also and can talk like an 'expert' !:)
You might be knowing 100 more things, but if you are doing this for 30yrs, then it means you must be having the knowledge & experience to tell under what certain conditions the 100 Ohm impedance would get overloaded by 1k. I have asked you about the certain set of conditions for overloading which you have failed to answer me. Please don't tell me to build tube preamps as i know what i am talking about, been there done that.

If you are saying a 100mV signal will not overload , please, stop being simplistic !This isn't what the statement meant !

^^^^ Where did i said that, Please don't try to put your own words in my mouth but let me clear the air here.... :)

For example, a certain tube preamp which has voltage/current drive swing limitation at high levels due to inability of buffer or tube to drive the load at that much voltage/current swing, then the quoted output impedance of 100 Ohms is nothing but cheating on the part of designer, therefore he/she must specify upto which level the Z out of 100 ohms is applicable. It is always stated in specs usually [Max voltage swing 30V into 600 ohms<<< like this] but a cheat brand can make as many claims.
 
Last edited:
".... under what certain conditions the 100 Ohm impedance would get overloaded by 1k. ..."

QED. ;)

"...I assume you have spent numerous days on internet also and can talk like an 'expert'.."

Could be ! :)
 
Last edited:
A beautiful, well-constructed speaker with class-leading soundstage, imaging and bass that is fast, deep, and precise.
Back
Top