Konnichiwa,
As I always say, that's 300B territory. If you don't want 300B's you can get away using both sections of a 6AS7 in parallel, but linearity is pretty poor.
Or you could use an all EL84 Amplifier, 4pcs trioded EL84 in the output stage, loaded by 2K5 (or even 1K2) and another EL84 as driver. Then there are EL34 & KT88/6550 which can be triode connected for something in the general vicinity of 8 Watt.
Other options might be a Trpple PE 10Y Amplifier, using a 5K Primary Output Transformer. If you want to you can use a 845 at low current and get 8 Watt out of that.
Finally there are more excotic Triodes out there with High Mu and Anode Power which can be used as massive "spud" (Spud for single tube, single stage Amplifiers), there are a few around in russia as DHT transimtter types and in the US IDHT Compactron Beam Triodes like the 6HS5. I suspect even a 572 could be coaxed into a 8WPC "Spud" Amp....
So, there are no hard and fast answers, cheapest are probably triode wired generic tetrodes/pendoes, but they will likely sound worst of all options and given the cost for decent output transformers you might as well shell out for some inexpensive 300B's. The more esotheric stuff, well that is another story.
Sayonara
GeirW said:
As I always say, that's 300B territory. If you don't want 300B's you can get away using both sections of a 6AS7 in parallel, but linearity is pretty poor.
Or you could use an all EL84 Amplifier, 4pcs trioded EL84 in the output stage, loaded by 2K5 (or even 1K2) and another EL84 as driver. Then there are EL34 & KT88/6550 which can be triode connected for something in the general vicinity of 8 Watt.
Other options might be a Trpple PE 10Y Amplifier, using a 5K Primary Output Transformer. If you want to you can use a 845 at low current and get 8 Watt out of that.
Finally there are more excotic Triodes out there with High Mu and Anode Power which can be used as massive "spud" (Spud for single tube, single stage Amplifiers), there are a few around in russia as DHT transimtter types and in the US IDHT Compactron Beam Triodes like the 6HS5. I suspect even a 572 could be coaxed into a 8WPC "Spud" Amp....
So, there are no hard and fast answers, cheapest are probably triode wired generic tetrodes/pendoes, but they will likely sound worst of all options and given the cost for decent output transformers you might as well shell out for some inexpensive 300B's. The more esotheric stuff, well that is another story.
Sayonara
Gday.
TnX. for the swift reply.
Well. for two 300B cheapo, it will cost me (local prices) 330$.
(i have no intension to spend that on tubes. "yet")
The 845 option looks promising, but still i reckond
this would atleast require 550-600v. (maybe less,
have just had a quick look, beside.. math is not my strongest
side.)
At the look of it, KT66 would can probably give me 8w around 400v.
I guess between theese tubes mentioned, 6550 or EL34
would be it. 6550 is rated 12w. at 250v, and EL34 is 8w. at
250/65. But distortion is high though.
6550 have 7%. and 34 have 10%, beside that, i think
6550 would perform better than 34.
As EL84 goes (i love them), well. i got the imression from various
threads, that SEP is secondary amp`s? (qualitywise)
TnX. for the swift reply.
Well. for two 300B cheapo, it will cost me (local prices) 330$.
(i have no intension to spend that on tubes. "yet")
The 845 option looks promising, but still i reckond
this would atleast require 550-600v. (maybe less,
have just had a quick look, beside.. math is not my strongest
side.)
At the look of it, KT66 would can probably give me 8w around 400v.
I guess between theese tubes mentioned, 6550 or EL34
would be it. 6550 is rated 12w. at 250v, and EL34 is 8w. at
250/65. But distortion is high though.
6550 have 7%. and 34 have 10%, beside that, i think
6550 would perform better than 34.
As EL84 goes (i love them), well. i got the imression from various
threads, that SEP is secondary amp`s? (qualitywise)
Hi Geir,
Do you need 8W in particular, say, over 4 or 5W, which would get you (just barely) into 2A3 territory.
I think you'll find that most members with an opinion on the subject will tell you SEP is less desireable than single-tube operation. However, once you go beyond two tubes in parallel, to three to ten tubes, the table perhaps turns, at least for some (think conrad johnson ART preamp). If you look at the distribution chart of tube characteristics and the statistical spread of deviation from ideal, you'll find that a group of tubes (say, 5-10) stands a better probability of being more linear than any given tube. If you're adventurous, you might wire together a number of, say, smaller triodes---even smaller DHTs---to see what happens on the result.
Cheers.
Do you need 8W in particular, say, over 4 or 5W, which would get you (just barely) into 2A3 territory.
I think you'll find that most members with an opinion on the subject will tell you SEP is less desireable than single-tube operation. However, once you go beyond two tubes in parallel, to three to ten tubes, the table perhaps turns, at least for some (think conrad johnson ART preamp). If you look at the distribution chart of tube characteristics and the statistical spread of deviation from ideal, you'll find that a group of tubes (say, 5-10) stands a better probability of being more linear than any given tube. If you're adventurous, you might wire together a number of, say, smaller triodes---even smaller DHTs---to see what happens on the result.
Cheers.
Attachments
Gday.
Interesting stuff serengetiplains.
i certainly will look into it, sooner or later.
2x5 or 2x8 wouldn`t be such a big stepp. (i think)
I certainly wount be crossing 10w, for that am planning a PPT
amp.
Interesting stuff serengetiplains.
i certainly will look into it, sooner or later.
With this spesific amp. am would like to stay as simple as possible, triode, medium voltage and 2x10w. as a limit.
2x5 or 2x8 wouldn`t be such a big stepp. (i think)
I certainly wount be crossing 10w, for that am planning a PPT
amp.
Konnichiwa,
Considering russia is just around the corner from you, why not buy direct?
That said, if you build the Amp with for example a set of James universal Outputs like the 6123 and a suitable Mains Transformer with enough +B and Output Heater taps you could start with a EL34/KT66/KT88/6550 or whatever you have and later change the Valve sockets and try 2A3's & 300B's at your lesure.
Parallel connection looses you some, but the EL84 is readily available, sounds very good and using a truckload is actuall;y quite feasible. I mean you could use 7 in parallel and the 8th as Driver with an Output transformer around 1K or lower, something like those made for SE 6S33.
In any case, the "massive parallel" amplifiers can be quite sucessfull.
Sayonara
GeirW said:
Considering russia is just around the corner from you, why not buy direct?
That said, if you build the Amp with for example a set of James universal Outputs like the 6123 and a suitable Mains Transformer with enough +B and Output Heater taps you could start with a EL34/KT66/KT88/6550 or whatever you have and later change the Valve sockets and try 2A3's & 300B's at your lesure.
GeirW said:
Parallel connection looses you some, but the EL84 is readily available, sounds very good and using a truckload is actuall;y quite feasible. I mean you could use 7 in parallel and the 8th as Driver with an Output transformer around 1K or lower, something like those made for SE 6S33.
In any case, the "massive parallel" amplifiers can be quite sucessfull.
Sayonara
Konnichiwa,
Before I forget, there is a qute nicely researched project by Claus Byrith which uses Standard tetrodes/pentodes in a local feedback arrangement that is very similar to what Audiopax calls "LM3" operation.
8W Singe End, EL34 amplifier designed by Claus Byrith, Royal Academy of Music in Aarhus, Denmark.
Sayonara
Before I forget, there is a qute nicely researched project by Claus Byrith which uses Standard tetrodes/pentodes in a local feedback arrangement that is very similar to what Audiopax calls "LM3" operation.
8W Singe End, EL34 amplifier designed by Claus Byrith, Royal Academy of Music in Aarhus, Denmark.
Sayonara
Take a peek at this
Here's one for you: 1 and 2. Note the coupling method, which biases the output tube by means of the voltage difference appearing across the anode resistor of the driver stage. The benefits of this topology are: 1) no cathode resistor (or resistor+capacitor) in the output tube (= better energy transfer to the speaker + better damping + less distortion etc), 2) no coupling capacitor or transformer between driver and output (= reduced grunge + phase shift etc), and 3) driver supply grunge typically amplified by the output tube of a single-ended circuit (ie, as fed to the output tube grid only) is partly cancelled by being coupled in-phase to the output grid *and* cathode.
Here's my take on benefit 3. Of the few advantages push-pull circuits hold over single ended, one of more significant is the inherent cancellation of power supply noise when a push pull stage is fed by one power supply. Assume, on such stage, the power supply voltage goes down slightly (ie, the supply responds to a momentary high current draw and supply voltage drops on supply impedance). That momentary PSU voltage dip appears on the plates of both tubes in the stage in question (not so bad on the relative scale of dip voltage as a percent of overall supply voltage) and is coupled to the grid of the next stage (bad on the relative scale of dip voltage as a percent signal voltage). But because the voltage dip appears in the same phase on both tubes, which are amplifying the signal 180 degrees out of phase, the dip effectively cancels as a source of noise.
Noise cancels as such, and completely, only when both tubes of a push-pull circuit are precisely balanced. No two tubes achieve perfect balance, statically or dynamically, leading to imperfect noise cancellation the degree of which bounces around like a bucking horse: don't no one know where it will go next. I consider this important because I suspect the ear is acutely sensitive to changes in distortion relative to frequency, phase and voltage level. Also, note the cost of attaining even the imperfect degree of noise cancellation a push-pull circuit attains: a significant increase in circuit complexity.
Compare noise cancellation on a push-pull circuit with noise cancellation on the single-ended circuit I posted above. In this latter, PSU noise cancels only partly---PSU noise voltage is reduced by passing through the anode/bias resistor and thus appears at differing voltages on the cathode and grid of the next stage---but the degree of cancellation does not bounce around, and is inherently more linear in phase, frequency and voltage level. Now, compare the complexity of a typical push-pull circuit with that of the circuit I posted. I can't imagine a simpler working circuit than the latter.
GeirW said:
Here's one for you: 1 and 2. Note the coupling method, which biases the output tube by means of the voltage difference appearing across the anode resistor of the driver stage. The benefits of this topology are: 1) no cathode resistor (or resistor+capacitor) in the output tube (= better energy transfer to the speaker + better damping + less distortion etc), 2) no coupling capacitor or transformer between driver and output (= reduced grunge + phase shift etc), and 3) driver supply grunge typically amplified by the output tube of a single-ended circuit (ie, as fed to the output tube grid only) is partly cancelled by being coupled in-phase to the output grid *and* cathode.
Here's my take on benefit 3. Of the few advantages push-pull circuits hold over single ended, one of more significant is the inherent cancellation of power supply noise when a push pull stage is fed by one power supply. Assume, on such stage, the power supply voltage goes down slightly (ie, the supply responds to a momentary high current draw and supply voltage drops on supply impedance). That momentary PSU voltage dip appears on the plates of both tubes in the stage in question (not so bad on the relative scale of dip voltage as a percent of overall supply voltage) and is coupled to the grid of the next stage (bad on the relative scale of dip voltage as a percent signal voltage). But because the voltage dip appears in the same phase on both tubes, which are amplifying the signal 180 degrees out of phase, the dip effectively cancels as a source of noise.
Noise cancels as such, and completely, only when both tubes of a push-pull circuit are precisely balanced. No two tubes achieve perfect balance, statically or dynamically, leading to imperfect noise cancellation the degree of which bounces around like a bucking horse: don't no one know where it will go next. I consider this important because I suspect the ear is acutely sensitive to changes in distortion relative to frequency, phase and voltage level. Also, note the cost of attaining even the imperfect degree of noise cancellation a push-pull circuit attains: a significant increase in circuit complexity.
Compare noise cancellation on a push-pull circuit with noise cancellation on the single-ended circuit I posted above. In this latter, PSU noise cancels only partly---PSU noise voltage is reduced by passing through the anode/bias resistor and thus appears at differing voltages on the cathode and grid of the next stage---but the degree of cancellation does not bounce around, and is inherently more linear in phase, frequency and voltage level. Now, compare the complexity of a typical push-pull circuit with that of the circuit I posted. I can't imagine a simpler working circuit than the latter.
,
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.