If it is fun that is ok, but then don't complain about it being a money pit ;-)Not at all an issue. All part of the fun
If it is fun that is ok, but then don't complain about it being a money pit ;-)Not at all an issue. All part of the fun
I do not know the difference between all those models. There are simply too many Tango models and Tamura models etc. There is no general Tango sound or Tamura sound as some might think. Tango made some of the best transformers but also some quite poor ones. Lundahl makes more budget priced transformers but for certain applications they make the best ones. Audiophiles are too much about brand names. And tube sounds.
I wouldn’t say that. I know customers of your product who moved on, and not at all for sonic reasons. Changing gear is part of the hobby. Even in your products, I am sure the 45s were great, yet GB is now on 300b. Let’s watch this space two years later.
To be clear I am not talking about you or any specific individual personally. Of course it is fun to find and discover new things, it is part of the learning experience. I do not question that whatsoever.Not at all an issue. All part of the fun
Nice!I don't know if this an acceptable apples vs. apples comparison.
![]()
I built this pair of Class A PP2A3 mono block amps with a Mullard style driver stage sans global feedback + Peerless 16309 OPTs in the early 90s.
![]()
Encouraged by reading Sound Practices, I built this no NFB cascaded 6SN7 > SE2A3 > Tango U808 OPTs. It's essentially half of the PP2A3 amp in terms of topology and power output. I remember the 10 kHz square wave looking similar in both amps. That's as far as I got in terms of testing since I'm not an engineer.
Being a classical musician, my ears chose the SE2A3 because it reminded me more of what I hear in a concert hall. Soon my pair of Spendor LS3/5As were replaced by more efficient Altec 755Cs. I also unloaded the PP2A3 mono blocks to finance higher quality Tamura and Tango SE OPTs.
Twenty five years later, I'm still listening to SE amps ranging from 720mW > 8 watts per channel.
Nice antiques! I think its been a good 45 years since I saw one of those variacs. Longer for the hp...So now, I sit back and enjoy the music!
Except when something breaks down or I want to try a new tube or circuit. Then off to the shop/test bench.
![]()
Thanks! I just made sure that a 1kHz sine wave was as close to symmetrical as I could get before clipping. Hence, the slightly different plate load values in the long-tail driver stage. I could have inserted an AC balance pot but it sounded fine as-is. The PP circuit (with a lower gain 1st stage), was derived from my favorite amp at that time, an Eico HF87 which also didn't have AC balance adjustment. I connected the EL34s in triode mode which was ample power to drive LS3/5As in my small apartment. I started with a PP2A3 amp because OPTs were affordable and easy to find then. I had to save for SE OPTs.Nice!
How did you verify that you had the phase splitter in the PP amp correct? I found that to be a tricky bit, since the older DHT tubes are harder to drive than the kinds of tubes that showed up when PP took over decades ago.
Nice antiques! I think its been a good 45 years since I saw one of those variacs. Longer for the hp...
I bought a Leben CS-300Xs integrated after reading a glowing review of their CS-300X. They compared the sound from the CS-300X (which was an upgraded parts version of the CS-300) with Shindo so I was sold. The upgraded parts, I assume, were matched to the NOS Mullard tubes (12AX7A and EL84)? Problem for me, the availability of NOS Mullard valves was coming to an end so Taku Hyodo had to switch to Sovtek valves, adding the small "s" (for Sovtek) to the designation (CS-300Xs). The design (with upgraded parts) remained the same.Doesn't have to be. It only becomes a money pit if you can't suppress that urge to roll tubes and parts etc. Most of that rolling is turning in circles anyways. A good circuit topology comes first, that trumps all, without that no need to spend much money on tubes or parts. next is decent parts quality that is within the budget. And the last is tubes (I should not say that as a tube manufacturer). Then just enjoy and stop thinking about improvements. There are still plenty dirt cheap tubes out there which can get you top sound quality. Of course if you want the best in the world then don't complain about the cost. When I started my business I had a pair of single ended mono blocks in my offering for 2000 Euros the pair. Way too good sounding for the price, but nobody wanted it.
My question for you, when you say that amplifier design is the most important part of great sound, is that design made around the performance of all valves of the required designation, or for a specific make with whose performance you are familiar with?
I think I understand. The difference in sound between the Sovtek and the Amperex in my Leben was rather plain amplification with Sovtek, and more air, sweetness and relaxing with the Amperex.When I say amplifier design is the most important factor to get best sound I am not only referring to tube brands but also to tube types. Second is quality of transformers used (output, power transformer, interstage and input transfers (if used) chokes. And capacitors.
I prefer an amp with a great circuit design and good quality output transformers using cheap indirectly heated tubes (for example a TV tube like the 6CB5A triode connected) over a poorly implemented 300B amp.
Of course there is a difference in sound quality between different tubes. But IMHO tube selection should not alter the basic sound character of the amp and change the sound a lot. If it does I would try to understand why and remedy that root cause.
An example. Different tubes have different quality vacuum which will result in different levels of grid current. A well designed driver stage will not be bothered with some grid current in the output tube. But many amplifier designs use rather high impedance driver tubes (to get max amplification) and thus use the highest possible grid to ground resistor on the output tube. When the output tube draws grid current (which is signal dependant) this will alter the op point and cause all kinds of effects. A strong driver stage ideally transformer coupled will not be bothered by grid current. And besides insensitivity to the grid current will generally sound better IME.
Also small signal tubes do draw various amounts of grid current which typically starts at a bias voltage of less than -1V. If the circuit and op points are chosen such that there es plenty of headroom, no tube will enter this region. In Preamps and phono stages with tubes like 12AX7 this can be a major cause of different sound between tubes. Again careful circuit design can minimise or eliminate such influences on the sound altogether.
Another point is general choice of op point. I am seeing many designs with marginal op points which means that tubes are operated near the 'knee' of the plate curves (higher distortion area). This is especially often the case when tubes like 300B are used in preamps or DAC output stages. Each tube (even from the same make) will have the knee at a different point. if the op point is now in that region changing the tubes can result in a very different sound. In such cases also a change of supply voltage will alter the sound drastically. This can be caused by the use of different rectifier tubes or by variations in the mains voltage. Moving the op point into a region with more headroom will not only generally sound better but make the amp less sensitive to changes in tube characteristics.
These are just some examples and I explained it in rather simple terms. And this is all IMHO and IME.
If you optimized for symmetrical clipping, the amp would express a greater amount of the 3rd harmonic. Of course being PP you would expect this. But driving 2A3s is a bit harder than driving EL34s even if the latter are triode connected. Did you make any accommodation for them? In my case I found that driving the type 45s was the hardest part of getting my PP design right; most of the phase splitter circuits you see are meant for driving tetrodes or pentodes.Thanks! I just made sure that a 1kHz sine wave was as close to symmetrical as I could get before clipping. Hence, the slightly different plate load values in the long-tail driver stage. I could have inserted an AC balance pot but it sounded fine as-is. The PP circuit (with a lower gain 1st stage), was derived from my favorite amp at that time, an Eico HF87 which also didn't have AC balance adjustment. I connected the EL34s in triode mode which was ample power to drive LS3/5As in my small apartment. I started with a PP2A3 amp because OPTs were affordable and easy to find then. I had to save for SE OPTs.
IME when a tube like the 12AX7 has grid current, its an indication that its operating point is poorly selected. If properly selected the grid current will be in micro amperes; so small (unless clipping) as to be inconsequential. I don't know how well this is known, but the 'Marshall sound' of Marshall's famous 'Plexi' 50 Watt guitar amplifier of the 1960s comes from the input stage being improperly biased in exactly this fashion: the cathode resistor is simply too low a value.When I say amplifier design is the most important factor to get best sound I am not only referring to tube brands but also to tube types. Second is quality of transformers used (output, power transformer, interstage and input transfers (if used) chokes. And capacitors.
I prefer an amp with a great circuit design and good quality output transformers using cheap indirectly heated tubes (for example a TV tube like the 6CB5A triode connected) over a poorly implemented 300B amp.
Of course there is a difference in sound quality between different tubes. But IMHO tube selection should not alter the basic sound character of the amp and change the sound a lot. If it does I would try to understand why and remedy that root cause.
An example. Different tubes have different quality vacuum which will result in different levels of grid current. A well designed driver stage will not be bothered with some grid current in the output tube. But many amplifier designs use rather high impedance driver tubes (to get max amplification) and thus use the highest possible grid to ground resistor on the output tube. When the output tube draws grid current (which is signal dependant) this will alter the op point and cause all kinds of effects. A strong driver stage ideally transformer coupled will not be bothered by grid current. And besides insensitivity to the grid current will generally sound better IME.
Also small signal tubes do draw various amounts of grid current which typically starts at a bias voltage of less than -1V. If the circuit and op points are chosen such that there es plenty of headroom, no tube will enter this region. In Preamps and phono stages with tubes like 12AX7 this can be a major cause of different sound between tubes. Again careful circuit design can minimise or eliminate such influences on the sound altogether.
Another point is general choice of op point. I am seeing many designs with marginal op points which means that tubes are operated near the 'knee' of the plate curves (higher distortion area). This is especially often the case when tubes like 300B are used in preamps or DAC output stages. Each tube (even from the same make) will have the knee at a different point. if the op point is now in that region changing the tubes can result in a very different sound. In such cases also a change of supply voltage will alter the sound drastically. This can be caused by the use of different rectifier tubes or by variations in the mains voltage. Moving the op point into a region with more headroom will not only generally sound better but make the amp less sensitive to changes in tube characteristics.
These are just some examples and I explained it in rather simple terms. And this is all IMHO and IME.
To the best of my knowledge, I did. I maximized the B+ supply to the driver stage and the plate resistors were almost half the value of those found in the Eico HF87. But that's as far I took it as an amateur builder.If you optimized for symmetrical clipping, the amp would express a greater amount of the 3rd harmonic. Of course being PP you would expect this. But driving 2A3s is a bit harder than driving EL34s even if the latter are triode connected. Did you make any accommodation for them? In my case I found that driving the type 45s was the hardest part of getting my PP design right; most of the phase splitter circuits you see are meant for driving tetrodes or pentodes.
If you ever see fit to do something like that again, its very helpful to replace the 6.2KOhm cathode resistor in the phase splitter with a constant current source. That will decrease distortion, increase bandwidth and gain all at the same time. We also have a nice output transformer for PP 2A3s as we used to make a PP 2A3 amp that made about 16 Watts. The circuit we used for that was very similar to yours.To the best of my knowledge, I did. I maximized the B+ supply to the driver stage and the plate resistors were almost half the value of those found in the Eico HF87. But that's as far I took it as an amateur builder.
More like nano Amperes. A micro amp would correspond to 2V across a 2M grid to ground resistor. It can be in the micro amp region for output tubesIIf properly selected the grid current will be in micro amperes;
Thank you for the tips!If you ever see fit to do something like that again, its very helpful to replace the 6.2KOhm cathode resistor in the phase splitter with a constant current source. That will decrease distortion, increase bandwidth and gain all at the same time. We also have a nice output transformer for PP 2A3s as we used to make a PP 2A3 amp that made about 16 Watts. The circuit we used for that was very similar to yours.
I’m really impressed by the sound of the Emission Labs 10th anniversary 45 tubes, which I’ve been listening to for the past few days. They are easily the most revealing 45 tubes I’ve heard. There’s a nice description of these tubes here:
Heres what they say in the documentation:
“This is a special tube, because the Anode (Anode) system is not attached to the glass bulb at any position. Neither at the bottom, nor at the top. So the only physical connection to the outside world is via two small bars, that are also the electrical connection. These are the two small bars you can see at an angle of 45° coming out through the glass. The three other (mid position) bars you can see coming through the glass are not connected directly to the Anode. These are for the grid and the filament, and are attached with flexible foil. The foil will greatly reduce mechanical coupling to the grid and filament. So indeed the whole tube system is solidly connected only at two points.
The good part of this is, the minimal acoustical from the outside world into the tube. There is always some small 50 (60) Hz mechanical hum present on the amplifier chassis, coming from the mains transformer. Also acoustical coupling from the air (=loudspeakers) via the glass is reduced. Last but not least the globe tubes simply look nice
Please note, very good care needs to be used when shipping these, since the classical way to connect the Anodes to the glass is also a shipment protection. So when ordering those, please choose insured shipment. In case you do have a shipment issue, it will be the Anode system being 4...8 degrees off axis. However this can be easily corrected by gently tapping on the front or the back of the tube. Actually this is the same procedure as with NOS Globe tubes.”
Sonically, the tubes are starkly revealing. When I hear familiar recordings, it’s like veils of obfuscation are removed. Everything stands out in remarkable clarity. This is worlds removed from the rich gauzy sound of the 300B tubes, which tend to sugar coat the sound, making everything sound luxuriant. That is most certainly not what these 10th anniversary EML 45s do. They lay bare every recording. You hear every rustle, every squeak, every whisper. Highly recommended to those who want a highly transparent sound. It may not be as flattering a sound as a typical 45 tube produces. But I like it!
View attachment 101591
I'm always thrown off when I see 'EML' since I play an EML (Electronic Music Labs) model 101 synthesizer in my band.@godofwealth
EML makes very high quality tubes. I’m not surprised that they are working out well for you. They sound terrific and are exceptionally durable/reliable.
Charles