The Stax Thread III

[JimL11]

I thought you might be being sarcastic, but I wasn't sure.

 

[AManAnd88Keys]

So my ears weren't playing trick on me? It is a paradox this amp, little is known outside of Germany it seems, and the designer is eccentric for sure, but obviously knows his subject. If only it had a cheaper variant for either planars or stats only, not both. And forget the amp function. If it was 8k for example, and with that sound, it would be a game changer for the finished product market and high end headphone amps. At 14k it is over priced. Yes the Woo Audio WA33 is about the same price, but at least that looks the part, and in that case has less tubes. I am not so comfortable with 22 tubes TBH.

It would be interesting indeed to pitch it against a DIY T2. I am super curious how much the T2 is different to the BHSE or Carbon. This Mal Valve was so different in it's presentation, it wasn't subtle. Not saying one is correct or even accurate, only the Mal Valve impressed me a lot, as it did to the 3 hifi dealers with me and 2 audio designers as well.

Just goes to show, there are gems around still in this hobby.

It absolute is a high-end amplifier, so Mal Valve at least placed it at an appropiately named show. My biggest issue with it was the heat, it made touching the amp unpleasant. Mr. Mallach mentioned that he could bias the tubes slightly lower without noticeable changes to the sound quality, which would get rid of a little bit of the temperature, but the sheer number of tubes - more than 20, all placed on the top - simply means there's going to be a lot of heat anyways.
The amplifier is basically a integrated amplifier based on the KT120 in push-pull configuration, with a differential driver stage for each channel and a matching input stage. If I'm not mistaken, its fully balanced from input right up to the output transformers. Or to phrase it in a different way: Mallach combined the design of his smaller power amplifiers and that of his preamps in one single device and added an output for dynamic and electrostatic headphones. It's more of a high-end integrated amp for speakers that just happens to also drive headphones very well, than a headphone amp with additional outputs for speakers. From that perspective the design is easier to understand.

Mal Valve has been around for many years, but the company never seemed to really bother about markets outside of Germany, or the EU respectively. Personally, I cannot tell why.
In german speaking countries they've got a very high reputation - their top models are considered absolute references and hardly ever show up for resale. But as you mentioned, prices are steep, and as far as I know they also ask quite some money for technical checkups and repairs.

On this years High-End in Munich I had an opportunity to listen to the new SR-009S driven by the SRM-T8000. From memory, the SR-009 combined with the big Mal Valve amplifier played in a higher league.

 

[JimL11]

The amplifier is basically a integrated amplifier based on the KT120 in push-pull configuration, with a differential driver stage for each channel and a matching input stage. If I'm not mistaken, its fully balanced from input right up to the output transformers. Or to phrase it in a different way: Mallach combined the design of his smaller power amplifiers and that of his preamps in one single device and added an output for dynamic and electrostatic headphones. It's more of a high-end integrated amp for speakers that just happens to also drive headphones very well, than a headphone amp with additional outputs for speakers. From that perspective the design is easier to understand.

That's very interesting. It seems that it would really be overkill as a headphone amplifier, since KT120 tubes run best at standing currents that are significantly higher than the EL34 tubes used in top electrostatic headphone amps, and the latter already have ample standing current. As to its design, it would seem to use differential amps from input to output. Other electrostatic tube headphone amps which share this topology include (from the simplest to most complex), the TubeCAD, SRX-Plus and Megatron, although both of the latter use some solid state in support roles, in the form of current sources. The Megatron with its eight EL34 tubes has been jokingly suggested to double as a space heater, so I suppose the MalValve would double as a how furnace. Probably not the best thing to run in the summer.

 

[wuwhere]

I wonder if the KT120s are swappable with KT150s.

 

[JimL11]

I wonder if the KT120s are swappable with KT150s.

Swapping small signal tubes such as 12A_7s, 6DJ8s, etc. generally only incurs the risk of a malfunctioning circuit due to the swapped tube having inappropriate electrical parameters for the specific circuit. Swapping power tubes, OTOH, carries the additional risk of serious damage to the amplifier if the bias voltage needed by the new tube is too different from the original tube - for example, in the BHSE, inserting a KT77 in place of an EL34 will ZAP (technical term) the amp according to its designer, and is not recommended. OTOH, other amps are designed to allow swapping output tubes - generally these will say so in their advertising. The safe thing to do is consult the designer, who should know. Absent that, if the specs do not specifically state it is OK to swap tubes, I assume the answer is no. The possibility of marginally better sound balanced against the possibility of a blown amp is not worth it - but that's just me.

 

[AManAnd88Keys]

That's very interesting. It seems that it would really be overkill as a headphone amplifier, since KT120 tubes run best at standing currents that are significantly higher than the EL34 tubes used in top electrostatic headphone amps, and the latter already have ample standing current. As to its design, it would seem to use differential amps from input to output. Other electrostatic tube headphone amps which share this topology include (from the simplest to most complex), the TubeCAD, SRX-Plus and Megatron, although both of the latter use some solid state in support roles, in the form of current sources. The Megatron with its eight EL34 tubes has been jokingly suggested to double as a space heater, so I suppose the MalValve would double as a how furnace. Probably not the best thing to run in the summer.

Furnace comes closer than you might think. The attenuator knob at the front was so hot, if I had kept my fingers on it for maybe 20-30 seconds I would have gotten first degree burns. I am NOT exaggerating, not even a bit. That was the reason in the first place why I asked Mr. Mallach if anything could be done about all the heat.

Soundwise that amplifier is pretty much faultless, but for more than 10k I would expect a design that deals better with the temperature. Especially since other manufacturers know how to do it for less money. Not that I don't like the straight forward industrial design, but when two dozen tubes or more are being used in close proximity you should come up with something.

 

[Whitigir]

Nothing good quality can generate less heat, a T2 has a lot of heat, and even if you are willing to throw 10k...you won’t find it LOL

A Carbon also generate a lot of heat

So that amp with all the tubes ? 22 ? Is probably a fire place by it filaments, let alone if it is powerful and good design. May want to have fairy dust coolant for the chassis

 

[AManAnd88Keys]

Swapping small signal tubes such as 12A_7s, 6DJ8s, etc. generally only incurs the risk of a malfunctioning circuit due to the swapped tube having inappropriate electrical parameters for the specific circuit. Swapping power tubes, OTOH, carries the additional risk of serious damage to the amplifier if the bias voltage needed by the new tube is too different from the original tube - for example, in the BHSE, inserting a KT77 in place of an EL34 will ZAP (technical term) the amp according to its designer, and is not recommended. OTOH, other amps are designed to allow swapping output tubes - generally these will say so in their advertising. The safe thing to do is consult the designer, who should know. Absent that, if the specs do not specifically state it is OK to swap tubes, I assume the answer is no. The possibility of marginally better sound balanced against the possibility of a blown amp is not worth it - but that's just me.

I agree. As someone who has build his own tube amplifiers and knows something about how classical tube circuits work, this is how I would approach the issue:

1) ask the designer which kind of tube works in which position of the amp.
2) if there is no designer that you can talk to, because it's an old design or some other reason makes it impossible to find out what the inventor recommends, you need to study the circuit. Sometimes schematics are available, sometimes you need to draw your own. If the circuit is commonplace, you can deduct most of it from other schematics. At times the differences are small, maybe a different tube in one of the stages, other operating points or voltages...
3) If you need to take a closer look at the circuit to find out what works and what doesn't, you probably need to study datasheets too. Otherwise chances are high that you put in a different tube, and you might even set it at a good bias point, but the overall perfomance is worse than it was before. A tube needs to be chosen because it is capable of doing certain things (which also means you need to implement it in the correct way), not because it's a famous or rare tube or because it works so well in another circuit. Yes, some tubes are great designs and simply wonderful examples of engineering, but sometimes they're just the wrong tube for the job.
A simple example: putting in a rare and rather costly early production Siemens ECC83 (12AX7) instead of a some-other-brand ECC82 (12AU7) could work in some circuits, but regardless of sound quality you'd end up with a much higher amplification factor. In some cases that's okay, in others it could mean trouble. If you understand at least some of the main propertiers of the tube you are working with beforehand, your decision will be much better and troubleshooting will also be easier. With sufficient understanding and the right tube(s) at hand, some commercial designs can certainly be improved upon.

 

[JimL11]

Furnace comes closer than you might think. The attenuator knob at the front was so hot, if I had kept my fingers on it for maybe 20-30 seconds I would have gotten first degree burns. I am NOT exaggerating, not even a bit. That was the reason in the first place why I asked Mr. Mallach if anything could be done about all the heat.

Soundwise that amplifier is pretty much faultless, but for more than 10k I would expect a design that deals better with the temperature. Especially since other manufacturers know how to do it for less money. Not that I don't like the straight forward industrial design, but when two dozen tubes or more are being used in close proximity you should come up with something.

While tubes have to run hot - after all, they work by thermionic emission - having the tubes too close together and hence exposed to excessive heat can indeed be a problem. As Toner points out in his book "Getting the Most out of Vacuum Tubes, "the most effective way of reducing all kinds of tube failures is to lower the temperature of their envelopes." I can well imagine that with 22 tubes packed into a relatively small space, even in the open air, that could be an issue.

I agree. As someone who has build his own tube amplifiers and knows something about how classical tube circuits work, this is how I would approach the issue:

1) ask the designer which kind of tube works in which position of the amp.
2) if there is no designer that you can talk to, because it's an old design or some other reason makes it impossible to find out what the inventor recommends, you need to study the circuit. Sometimes schematics are available, sometimes you need to draw your own. If the circuit is commonplace, you can deduct most of it from other schematics. At times the differences are small, maybe a different tube in one of the stages, other operating points or voltages...
3) If you need to take a closer look at the circuit to find out what works and what doesn't, you probably need to study datasheets too. Otherwise chances are high that you put in a different tube, and you might even set it at a good bias point, but the overall perfomance is worse than it was before. A tube needs to be chosen because it is capable of doing certain things (which also means you need to implement it in the correct way), not because it's a famous or rare tube or because it works so well in another circuit. Yes, some tubes are great designs and simply wonderful examples of engineering, but sometimes they're just the wrong tube for the job.
A simple example: putting in a rare and rather costly early production Siemens ECC83 (12AX7) instead of a some-other-brand ECC82 (12AU7) could work in some circuits, but regardless of sound quality you'd end up with a much higher amplification factor. In some cases that's okay, in others it could mean trouble. If you understand at least some of the main propertiers of the tube you are working with beforehand, your decision will be much better and troubleshooting will also be easier. With sufficient understanding and the right tube(s) at hand, some commercial designs can certainly be improved upon.

Having built a couple of tube electronics, I concur with your experiences.

BTW, I see by the moniker that you play the piano? Also, I have a very good friend in Pisa.

 

[astrostar59]

JimL, do you think that Mal Valve amp is fully tube for the gain stags, i.e. no solid state devices in the gain stage? If all tube, mate that is why or part the reason it sounded so good, aside from the design of it. BTW it did feel hot to me as well, as did the 6K 'smaller' headphone amp version, also which power Stats and Planars, and has 'only' 10 tubes LOL.

 

[AManAnd88Keys]

While tubes have to run hot - after all, they work by thermionic emission - having the tubes too close together and hence exposed to excessive heat can indeed be a problem. As Toner points out in his book "Getting the Most out of Vacuum Tubes, "the most effective way of reducing all kinds of tube failures is to lower the temperature of their envelopes." I can well imagine that with 22 tubes packed into a relatively small space, even in the open air, that could be an issue.



Having built a couple of tube electronics, I concur with your experiences.

BTW, I see by the moniker that you play the piano? Also, I have a very good friend in Pisa.

Yes I do, albeit not that much currently. I studied jazz piano for a couple of semesters in Vienna and went for a different career (and a different city) after that period. I still play regularly though, I start to feel uneasy when I haven't played in a while.

Pisa is quite a trip from where I live. I went to Tuscany last March, but didn't visit Pisa unfortunately.

 

[AManAnd88Keys]

JimL, do you think that Mal Valve amp is fully tube for the gain stags, i.e. no solid state devices in the gain stage? If all tube, mate that is why or part the reason it sounded so good, aside from the design of it. BTW it did feel hot to me as well, as did the 6K 'smaller' headphone amp version, also which power Stats and Planars, and has 'only' 10 tubes LOL.

I know that your question isn't directed at me, but given that I know a few things about Mal Valve it's maybe okay if I say something too. As far as I know all Mal Valve designs are 100% full tube, no exceptions. Even Mallachs latest invention, a tube crossover, uses only tubes in the signal path. It's one of his hallmarks.

 

[JimL11]

JimL, do you think that Mal Valve amp is fully tube for the gain stags, i.e. no solid state devices in the gain stage? If all tube, mate that is why or part the reason it sounded so good, aside from the design of it. BTW it did feel hot to me as well, as did the 6K 'smaller' headphone amp version, also which power Stats and Planars, and has 'only' 10 tubes LOL.

With that many tubes, I would think there would be ample gain without resorting to solid state. In the SRX-Plus and Megatron, solid state current sources are used to improve diff amp balance (both), maximize gain (Megatron), improve isolation from the negative power supply (both), and maximize the efficiency of the output stage in delivering current to the headphones (SRX Plus). In the MalValve, which uses KT120 outputs with transformers, the amount of standing current is overkill for driving electrostatic headphones. However, it is possible that some of the tubes are used as tail current sources for the differential stages, as a three, or even four stage diff amp topology, which should supply sufficient gain, wouldn't need 11 tubes per side.

As has been pointed out, this amp is primarily designed as a speaker amp, and while it can be used as a stat headphone amp, it is grossly inefficient for that task. to the point of being environmentally irresponsible. The smaller Mal Valve Three headphone amp uses 2 ECC81/12AT7 input tubes, 4 ECL184 pentode driver tubes and 4 6V6 output tubes. As it has both RCA and balanced inputs, it most likely uses diff amps throughout, with planar and dynamic headphones driven via transformer outputs, and electrostatic outputs either directly or with capacitor coupling.

Now the 6V6 outputs have a maximum rated plate voltage of 315 volts, compared to 450 volts for 6SN7GTB outputs as in the SRX Plus, and 600-800 volts for triode connected EL34s, such as the BHSE uses, so the Mal Valve will have lower voltage output limits compared to most electrostatic headphone amps, although there is probably more standing current available compared to the 6SN7GTB. However, as I have shown elsewhere, the use of constant current source loading of a relatively puny tube such as a 6SN7GTB can actually allow it to deliver more signal current to an electrostatic headphone than an EL34 with a plate resistor load. Assuming the Mal Valve takes the electrostatic drive off the 6V6 outputs with an unloaded transformer in parallel, the question would be, how much signal current can actually go to electrostatic headphones versus how much is burned up in the transformer. It may well turn out that the Mal Valve is significantly less powerful in current as well as voltage limits for driving electrostatic headphones compared to even the SRX Plus. This is the compromise the Mal Valve makes as a universal headphone amp.

 

[astrostar59]

JimL, he told me the amp can do 2400v for stats. I am most curious in this amp as it was so very different to other stat amps in sound that I have heard to date. It played super loud if required with no sign of compression, and the timbre and voicing was superb. Really it made the 009 sound like a totally different HP, I was surprised. I am not talking coloration, rather timbre, accuracy and tonally realistic, as opposed to detail forward and cold sounding. It made me want to buy the amp and the 009s TBH, but I can see the impracticality of the amp (and it's cost). The LCD4 on the same amp was on par IMO, different, but I could love with either. So in 'theory' the LCD4 powered by a top all tube amp would be my path if I was looking at another amp or HP system. There are lots of tube based amps that can drive the Planars, and something I may explore at some point.

This path I may take depends on what I do with my power amp which is solid state. If I go for an Aries Cerat integrated which is all tube output, it would have a headphone out as well, so may 'upgrade' both HP and speakers.

 

[JimL11]

JimL, he told me the amp can do 2400v for stats.

Which amp can do 2400V? I'm assuming the 22 tube amp? As a general rule, for a push-pull amp, the maximum peak-to-peak output is on the order of 4 times the B+ supply, so for the Mal Valve headphone amp, which uses 6V6 outputs, the maximum plate voltage on the 6V6 is specified as 315 volts, which means its max output is on the order of 1200V. With KT120 outputs, the maximum plate voltage is around 600 volts, so if that is the B+ supply voltage it could do close to 2400 volts peak-to-peak.

For perspective, note that the BHSE has a nominal PS voltage of +400/-400 so can put out close to 1600 volts peak-to-peak, and the DIY T2 has a PS of +500/-500 so can put out close to 2000 volts peak-to-peak. However, since subjective loudness is logarithmic, the difference in ultimate loudness between a 400 volt supply and a 500 volt supply is 2 dB, and 400 volt vs a 600 volt supply will produce an additional 3.5 dB loudness, where 10 dB is required to double the subjective loudness in the midrange. Since most Stax headphones other than the SR002/3 are rated around 100 dB for 100VRMS = 283 volts peak-to-peak, any amp with a +/- 300 volt supply can produce a sound pressure level of 112 dB. This is subjectively twice as loud as a jackhammer at 1 meter, and loud enough to cause hearing damage within one minute. An amp that produces 2400 volts peak-to-peak will sound 6 dB louder, at maximum volume, and at that level can cause permanent hearing damage in less than 30 seconds, and may also destroy the headphones.