Since the Melquiades MF channel by design has become a universal 4V DHT amp I wonder anyone heard the PX4 amps?

It looks like the Milq output stage will swallow the PX4 easily:

http://www.mif.pg.gda.pl/homepages/frank/sheets/027/p/PX4.pdf

http://www.mif.pg.gda.pl/homepages/frank/sheets/060/p/PX4.pdf

http://www.mif.pg.gda.pl/homepages/frank/sheets/126/p/PX4.pdf

Here is a good site that described the history and use of PX4:

http://www.px4.org/

I do not want to sound as I am just brainlessly fish for options but until the Milq MF channel is not finished in a way how it shall be finished and until it properly integrated into the system I can’t make my own critical judgment about the sound and about sonic differences between the various tubes. The PX4 it appears looks like a viable candidate….

Which 2A3s and other output tubes have you tried so far, and how do they sound and compare in your amp/system?

Also, remember that you may have to tweak operating points a little to get the most out of any of the tubes you try.

I think I mentioned that I actually liked the Sovtek 2A3 because it could effect a great range of "attitudes", including glorious, placid, bland, dirty, rude and aggressive, as each was called for, and I do not mean by adjusting operating points bit according to the stories of the music, while fixed at the "proper" operating points.  It was never as "pristine" as the old RCA, but it was not limited like that tube, either.

Again, I last heard the Sovtek a couple of years ago, so I don't know if current production is the same.

Best,
Paul S

Here is the picture and you might see in front sitting the Dave Slagle’s MF filtering choke, It is not shield but is hangs along relatively far from anything AC-annoying.  I did not measure the filer yet but subjectively it sounds fine, somewhere where it shall be.

The front end is very much is done permanently and might be closed up. I was playing today the MF channels, trying to evaluate if the non- shield choke would pick up some noise. I am not sure if it does as generally the amp I would call it “noisy”. It is not the I can hear “noise” at listening position but if I stick my heard into the horn then it certainly has some noise and the most important is that I do not really like the harmonics of THAT noise.

Well, I am not alerted yet as the amp if not close up, the output stage is sitting outside and it might suck garbage from anywhere. Still I ma so accustomed for absolute silence from my horn that whatever minor the DHT does is sound different. At this point I am not considering the DC filaments, let see how it will behave when I close it up. Funny is that with this topology it is hard to figure out where the noise is coming from – the front or back end….

Floobydust,

excellent points as usually. I am not too concerned about the AC noise on my filament, at least at this point. I want to get the right “check and balance” operation of the whole amp (all channels together) and then will see how the MF noise will manifest itself. The opinions of my technical advisers are sharply divided: some of then strongly advocate DC and some of them AC. I personally have a strong bias to AC, despite that I have a lot of theoretical justifications that DC shell be more reasonable. Still, I have an open mind about it and it is not difficult to try DC. One of the reasons why I do not want to do it so far is that I do not want to have another variable of sound – to think how the filaments PS would sound.

Your signal-to-noise points are well take and it is how I felt but I have something else in my mind – modulations. The 50Hz hum shall do it a lot into HF. You are also was correct that 50Hz ham on a scope is not the sinewave but rather some multi-wave crap. Another thing that keeps me wondering with my 110dB sensitively and compression driver is the fact that I have very light and very susceptive diaphragm that being stressed with 50Hz shall toss a lot of Doppler distortions. Well, the list might continue and we all agree that there are a LOT of theoretical reasons to DC with high sensitivity – let see how it goes practically. I still, if I will be able to go away with it would prefer to stay with AC.

There is another interesting direction to go – to run filaments from high frequency AC. I think it is what David Berning does. If someone would know any ready to be used 50 kHz -250kHz PS that can do 1-2A then let me know, I would like to try it. I think that subject is so obvious that it shell be some companies out there that shall have some the off the shelf solutions. It truly very simple a 150kHz oscillator and very small class D amp without even any filters in output. I think I need to call to PurePower people and to proposed them to do a new product… I hope someone has already had it.

I was tested the amp in “integrated” mode and even I do have very low noise from the DHT tube but with my sensitivity and with my distances I do hear some ugly noise. I did not realized that the nose is moderate when I lose 3dB of my MF attention to integrate my MF with the rest of channels but when I run it wide open then it kind of unpleasant.

Well, with my entire hate of the DC idea I think I might be forced to go there. It is very pity…

 So, if the dive, or at least trey of DC on filaments would be in order I wonder is any specific DC topology for filaments would be advisable? I might go foe my regular SS Schotcky rectification with following LCRC. I have in the amp a 12.6V and 6.3V, coils, both for very high current, that drives 6C33C that and from there my 2.5 and 4V of LC filtration. But since it will be DC I would like to have an option for multiple voltages…

I have 5A regulated PS up to 33V that I used for my filed coil project; I will try it for temporary used. Still, the filaments in DH are the cathodes and I am VERY suspicion that the quality of the filaments shall matter a lot.

http://www.diyhifisupply.com/files/tenlabs-application-notes.pdf

http://www.diyhifisupply.com/files/tentlabs-dht-heating-methods.pdf

http://www.diyhifisupply.com/files/tentlabs-supply-chart.pdf

Well, I have to admit that I did not listen the whole sound of AC vs. DC filaments yet.  I will bring home today my DC lab supply that I use for my electromagnet coil and will give the try. I still at this point am not concerned about the “whole sound” but I rather a bit pissed about some aspect of “targeted listing” . Thos aspects might be not important at all but it is what I am accustomed and so far with AC those aspects are fucked up.

Listening the AC on DHT vs. the IDHT single-stage I do not feel a lot of advantage of DHT or IDHT. They giving or taking are compatible if I mentally extrapolate myself out of the differences that are intentionally implanted into the sound of those amps. However, when I discard the “whole sound” and look at the “little sound” then the IDHT single-stage amps destroys my DHT amp and at this point I might presume that it is because of AC.

The “little sound” is my own meshuginah that might not be important at all but they are those “some aspect of targeted listing”.  I developed those techniques in 2001-2002 what I was listening a LOT of compression drivers and was trying to see how the drivers’ cones react to different minimal currants from my Lamms ML2.0. Different diaphragms with different suspensions did differently and the best diaphragms did very interesting.

I place me head at the mouth of my MF horn and playing music loud I try filter out in my mind all sound and concentrate my attention only at the sound that are just above auditable level. At this point you can almost hear the curvature of the cone. It hard to explain what to listen but since one get what and how to listen then it is a totally new world – the micro-world , what you deal not with Sound itself but with some kind of molecular sound. Those little artifacts of molecular sound that juts pup up from the diaphragm surface are kind of funny.  You never hear then even at 3feet from horn. You need to dive into the horn, to close your year and spend then foe a while unit you hearing will be able to tolerate the “overload” and to concentrate itself on the “molecular sound”.  Then the “new word” would open up…

I have to admit that as many drivers I heard in the molecular mode no one did as “interesting” as Vitavox S2 with a specific version of magnet. The S2 present the molecular sound as some kind of fantastically sophisticated texture cut with super fine scalpel. Most drivers do not have even a reaction of the levels where S2 already bulb up a whole labyrinth of micro sounds… The key in here is to observe who the “molecular sound” relates to the “whole sound”…

So, my IDHT single-stage is not juts “clean” but it has stunning “molecular sound” quality. My DHT loosing the “molecular sound” butte with HUGHE margin but it is not the worst thing. The worst thing is that “molecular sound” of my DHT is absolutely independent from the rest of the sound. It feels like the DHT’s molecular sound lives with its own live and I very much do not like it.

A friend of my advised that it is not only due to the AC modulations but also the non-literally of the DHT triode anode/grid curves. The curves are not straight and the cathode to anode and cathode to grid voltage of the tube varies for a part of the tube by +/- ½ the peak/peak value of the heater voltage.  So, presumably the DC on heaters would minimize this effect. I think the tonight run of filaments from DC lab supply would show off where I stand on the DC subject. If the DC will be able to make the “molecular sound” of my DHT close to what I have from a single-stage amp then it will be a big plus to the DC idea.

be wrote:

Hey. I have not seen it mentionned here, have you tried, or even considered regulated dc-current supplies for the heaters? Should have the sound quality of ac-voltage supplies combined with the lack of noise of the dc-voltage supplies.?? http://www.tentlabs.com/Products/Tubeamp/Tubefilament/index.html

Romy the Cat wrote:

Well, I have to admit that I did not listen the whole sound of AC vs. DC filaments yet.  I will bring home today my DC lab supply that I use for my electromagnet coil and will give the try. I still at this point am not concerned about the “whole sound” but I rather a bit pissed about some aspect of “targeted listing” . Thos aspects might be not important at all but it is what I am accustomed and so far with AC those aspects are fucked up. Listening the AC on DHT vs. the IDHT single-stage I do not feel a lot of advantage of DHT or IDHT. They giving or taking are compatible if I mentally extrapolate myself out of the differences that are intentionally implanted into the sound of those amps. However, when I discard the “whole sound” and look at the “little sound” then the IDHT single-stage amps destroys my DHT amp and at this point I might presume that it is because of AC. The “little sound” is my own meshuginah that might not be important at all but they are those “some aspect of targeted listing”.  I developed those techniques in 2001-2002 what I was listening a LOT of compression drivers and was trying to see how the drivers’ cones react to different minimal currants from my Lamms ML2.0. Different diaphragms with different suspensions did differently and the best diaphragms did very interesting. I place me head at the mouth of my MF horn and playing music loud I try filter out in my mind all sound and concentrate my attention only at the sound that are just above auditable level. At this point you can almost hear the curvature of the cone. It hard to explain what to listen but since one get what and how to listen then it is a totally new world – the micro-world , what you deal not with Sound itself but with some kind of molecular sound. Those little artifacts of molecular sound that juts pup up from the diaphragm surface are kind of funny.  You never hear then even at 3feet from horn. You need to dive into the horn, to close your year and spend then foe a while unit you hearing will be able to tolerate the “overload” and to concentrate itself on the “molecular sound”.  Then the “new word” would open up… I have to admit that as many drivers I heard in the molecular mode no one did as “interesting” as Vitavox S2 with a specific version of magnet. The S2 present the molecular sound as some kind of fantastically sophisticated texture cut with super fine scalpel. Most drivers do not have even a reaction of the levels where S2 already bulb up a whole labyrinth of micro sounds… The key in here is to observe who the “molecular sound” relates to the “whole sound”… So, my IDHT single-stage is not juts “clean” but it has stunning “molecular sound” quality. My DHT loosing the “molecular sound” butte with HUGHE margin but it is not the worst thing. The worst thing is that “molecular sound” of my DHT is absolutely independent from the rest of the sound. It feels like the DHT’s molecular sound lives with its own live and I very much do not like it. A friend of my advised that it is not only due to the AC modulations but also the non-literally of the DHT triode anode/grid curves. The curves are not straight and the cathode to anode and cathode to grid voltage of the tube varies for a part of the tube by +/- ½ the peak/peak value of the heater voltage.  So, presumably the DC on heaters would minimize this effect. I think the tonight run of filaments from DC lab supply would show off where I stand on the DC subject. If the DC will be able to make the “molecular sound” of my DHT close to what I have from a single-stage amp then it will be a big plus to the DC idea.

I did a few times AC and DC switch with practically no impact. Then, driving the amp from DC I shut doe the B+ with speaker connected the noise is gone instantly. I was looking at my connect Lab Supply and it keep powering the DHT filament. I was looking at the miliampermeter of the DHT current and it she 35mA – the caps still hold the charge.  So, the output stage was fully operation but there was no noise. Ok, it came from the first stage I concluded…

Looking at the first stage I decided that the only element that might be responsible for noise was not shielded filtering choke. I hanged it from anything and who know might be is suck-in some dirt. I disconnected the choke from driver’s grid, and turned the B+ back (the DC filaments kept running on DHT). This time the amp just did not work at all. I was sitting in floor, trying to figure out what I did wrong and then something made me touch to touch anode of driver with a screwdriver. The speaker responded with a very soft click. Holly, shit, it was operational I figured. I tuned “play” in my CD played and the amp was functioning perfectly fine. The scope showed virtually the same level of noise but now it was with zero harmonics – juts a fat thin line.

OK, the problem was found, how about the AC? I disconnected the DC supply and put the AC back. To my huge surprise it was not bad at all. It has some very minor nose if I literally stick my head into the horn. The texture of this noise is WAY more pleasant, it just a soft FM hiss, in a way similar to FM noise, only a bit less uniformed. It is absolutely not auditable from 1-2 feet. Bingo!!! That is exactly what I need!!! Now, with this result like this I can perfectly stay with AC.

Now, I have two problems.

Problem # 1: the MF channel with no filter at all sounded juts spectacular. I play very softly muss with no LF content and with no filter and it was VERY convincing. It was so convincing that I even thought to get a full-range transformer and a speaker-level filter….

 Problem # 2: it is self-evident that my older filtering coil in my custom made shield was good to sitting behind the 12K resistor but it was not good enough with the coil sitting directly on the grid.  (before above  and the after below)

It looks like the choke sitting on the grid acts like an antenna. The new non-shielded coil acts even worse, as in addition it has no resistor between grid… I did not figure out if it is the choke that sucks in dirt or it is the wire from grid to the choke… Nevertheless, the dealing with the filtering choke will be the last bottle-neck that I can envision in this amp. Any ideas are welcomed.

DHT4ME wrote:

So I am aasuming that you are referring the very low level capability of your driver or amps?   The subtle nuances?What other drivers did you find did well beside the Vitavox. I always felt that  most comperssion drivers are a little tightand need to be played at some volume to get them breathing.

You sound like an Altec user.. :-)

DHT4ME wrote:

On a side note have you experienced any problems with startup noises with those drivers directly connected with no crossovers. Sometimes I get some hum in my woofers when I start my amps. I have been considering a multi amped setup with some spuds but fear damaging my drivers.

I did not experience any problems with startup noises or in any danger of damaging drivers. There are many factors involved into and it was discussed somewhere before. in short: with start proper delay and sequencing and with bleeders on caps that drain them out it is not damage of any kind that I invasion.  In past even in my plate-coupled and no filter “Water Dropped” tweeter I was VERY warring in order do not damage it. In practice it turns out to be absolutely not a subject to worry. Nowadays if I wish I just connect B+ to the tube “in fly” with tweeter attached – never had any problem since I practice it.

Last night I built an large and ugly mumetal sarcophagus around the Dave Slagle’s choke and located it a bit aside from the location it was before. The efforts killed perhaps 98% of noise. I did not have good mumetal cans but I have .004 sheet of NETIC® and CO-NETIC®

http://magnetic-shield.com/products/alloys.html

So, I kind of made an ugly container with 15 layers of those sheets. The effect was very positive.  Now I have extremely low noise, something that might be discarded from my point of view. Still, if I move my sarcophagus with choke away from the amp or put around a heavy steal shield then I have ALL residual noise gone, I mean gone absolutely.  Unfortunately the choke is too large to have more shielding on it, the fact that Dave Slagle’s “does not believe in shielding” also does not help. 

So, there are 3 areas where I am looking now to end these problems.

1)      To find more effective shield for my current coke – I want an absolute silence from my MF channel.

2)      Putting a larger grid stopper then I usually use, let say 1-2K. Yes, it will be more impact to the low-pass filter against the Miller capacitance but it would also decouple the choke from the grid and might substantially refuse the effect of the coil in the grid act as a microphone.

3)      To find another choke, a very-very small toroidal choke, designed foe ultra low current that would be in heavy duty sealed mumetal can or even double can. The choke would be so small that it would literally sit on the cable with no leads. I am talking with Brian Sowter about the prospect of such a choke build in case if my above efforts would not succeed.

Still, I have seen the Promised Land and if I have my currant large nickel-core Dave Slagle’s choke in her sarcophagus outsider of the amp then the MF channel is so dead quiet that  it is imposable to say if the channel even on. And it is with AC on filaments!!!

I made today a bunch of experiments with shielding of the MF filtering choke and I have very good progress. I was talking with different people and had a number of good advises (Come one, people the ward “advise” must have a plural form!)  The interesting impression I had after a conversation with Dave Slagle. David dived into, in my view irrelevant for my case, advocating the “evil of shielding for phonostage step-up transformers” and his “disapproval of shielding”. Right he or wrong is irrelevant in practical life, and  while I consider the whole subject is “beside the point” for my case (as my filter is “subtraction” non-series devise) I asked myself: what is the difference between the concept of phonostage step-up transformers and my RL filter? My driver stage has 17 times gain and the direct-coupled output stage has 3. I have 20 times gain (before transformer) or the similar gain to some step-up do, and my filter sits right on the driver’ grid. So, if my Expressive Technologies step-up transformer has 5 separate shields (isolated with air) then why I shall not treat my filtering choke with the same respect, not to mention that my filtering choke sits inside of 6 channels amp with a LOT of sources of all imaginable dirt?

Well, it was what I did last night and today. I wrapped 3 layers of .002’ Conetic sheet around the David’s choke. Then I conversed with a shielding engineer that explained to me that all MuMetal sheets loose permeability at the bend locations. So, I bought the Supper Shield Nickel spray and very richly sprayed the whole Conetic sarcophagus with 3 coating of that Nickel spray substance.

http://cableorganizer.com/mg-chemicals/super-shield-coating.html#features

 Then I have an isolation layer and 2 sheets of .008’ Netic. Then an isolation layer and one layer of think .01’ aluminum sheet. Then another isolation layer and one sheet of .01’ cold-roll steal. David sent me emails suggesting using copper foil as well. I do not know what it might be for but my Expressive transformer does use a thick layer of copper shield as well, so why shell I not? I have ordered the copper foil from MacMaster and will have it tomorrow.

So, for a time being I connected today the choke without the last 2 final layers: it will be a layer of double-wrap copper and the last externals layer of double-wrap of A276 foil)

(Cat. # A276-10 at http://www.lessemf.com/mag-shld.html )

The effect of issue the choke as it is today was insultingly wonderful. There are no problems anymore with noises auditable through the MF driver. Even I stick my head into the horn I need to listen it for a few seconds to recognize that there is “something” in there. The residual noise is so low that it has become a subject of ego and obsession instand of the subject of practicality. I still will put the last two layers tomorrow, juts to be on the save side and it will be it. The problem with the choke noise is gone…

A bit more about noise. I was playing with different grounding schemas, not listenining it but just playing with noise views on the scope. Grounding of the secondary of the Slagle’s Nickel OP transformer looks like drive noise even lowers, interesting that grounding of one or another side of secondary produces absolutely different effect. In terms of number the best I was able to do was 9mV, it is not RMS but peak-to-peak.  The best that I was able to get with single stage Milq was 0.9mV peak-to-peak; the DH channel has a few dB more gain however.  As you understand that 9mV peak-to-peak is way beyond where I need to worry. I have to admit that those 9mV peak-to-peak do have some harmonics (on scope). I put DC on filaments and the nose drooped to 3mV peak-to-peak with virtually no harmonics. Since the noise is not problem anymore with ether PS  on heaters it would might be interesting to evaluate how the types of heater voltage might affect Sound. I however did not do those experiments this time and I would prefer to stay with AC, thankfully now I can. I understand that all problems of modulation that I was named above still there but who knows if the cure more deadly than the illness, so I will stick with George Westinghouse and Nikola Tesla. In my own version of the “War of Currents”

 I think that when I put the last shields and close the amps I might lose another 2-3mV. In additions the LPAD on secondary will kill ~3dB, so I will be way into deep silence…

I put the final two layers on the choke and temperately closed the amp up as it will be finally done, laying all cables right and so on. The loaded MF channel now has 0.8mV RMS. So, forget about any noise.

Playing the amp today a bit I have formed the following area of my dissatisfaction. They are not the dissatisfactions but rather the things that I would like to improve before I end up this project.

1)      The filament voltages. The fix resistors that I use are not good as different brands suck different currents and I have a few tens of volt fluctuation with different tubes. It might be not a big deal but I do not like it. So, why I do not have the two 50W rheostat, like in tube testers, of the step attenuators where I will be able to drive my voltage up and down. If someone knows what I where buy such a thing please let me know.

2)      The output transformer. The 2.5V tubes and 4V tubes work perfectly with the same operation point but it is obvious to me now that all of my 2.5V tubes sound less loaded then all of my 4V tube. I have a transformer 20:1 that is fine but it is clear that the 2.5V tubes with 800R on plate and 4V tubes with 1250R on plate would need 5:1 difference in loading ratio.  I do not know which one is right at this point but it is obvious that in order the 2.5V tubes and 4V tubes sound the “same” this load must be normalized. So, it is highly likely that another counf for my OPT will be called upon…

Otherwise it looks like the amps are at the final lane…

I closed up the amps and moved it to the default operation position, right along with it all Dominus cables and so on.  I played just mono music and was very interesting where I will find myself with this amp Sound.  The right channel still was serving with single stage, I of course EQed their levels.  The spare 3dB on DHT side will do good when I put the rest channels in play.

If you expect that Sound of DHT channel blew out of water my single-stage amp then it would be not corrects. They sound in a way simile with exception that DHT channel was auditable less loaded with all expected consequences. I did not detect the deficiency in Microbe Sound; the single-stage amp still was a bit above but at practically negligible level. Still, benefits from the dynamic and contrast that I got from more idled DHT was hard to discard and it was one of the features that I was going after.

I put a few tubes in play different 2.5V and different 4V tubes. The problem that I see is that all my 2.5V tubes are truly uncomparable to my single-stage amp as they loaded ageist too high impedance. The 2A3 are 800R, I drive then very softly – 38mA, so I presume the true impedance somewhere around 1.5K. The 6K of my current loading does not appear as too high but sonically it feels like it is “at the edge”. The single-stage amp sounds like swimming in a pool filled with oil – graceful, lucrative but essentially too slow (that was my objective to eliminate to begin with and I was not as the single-stage has no more gain). The 4V tubes do much better match to work with single-stage amp as they loaded harder. My 4V tubes have 1200R on plate and at 38mA it probably goes up to 2.5-3K. So, the 6K load is kind of hard load and the 4V tubes sound juts a bit more contrasty but analogous.

I put the Russian YO186, Klangfilm KL71403 and Telefunken RE604. All of them run at absolutly identical operational parameter with 38.6mA – kind of spooky. All of them measured better then new with full emission, none of them broken-in. They all were nice in own way, different and nice, (I will be taking about it what I but all of them up) even though the most proper balance I feel I got from 2A3 made by RCA during WW2.  So, why I named my post “the unexpected insanity”? Well, it is because the sound of the YO186…

It was truly unexpected insanity. Nope the YO186 did not sound “better” then Klangfilm, Telefunken of single-stage amp. It is, as any other tubes, has own balance and own qualities but it has something that I would call “insane” – the bass. OK, you need to understand that I am listing JUST MF channel that has 1000Hz roll off. So, all my talks about bass have to be viewed from this above 1000Hz perspective. Still, the “bass” that YO186 demonstrated was juts stunning, not the quality of it of course but some amassing inflection in lower notes that made notes and lower harmonics absolutely hypnotizing. If you know the coloration of voice that native Americans have then it was in a way similar but with a very a very gentile inflection of very fine velvet texture and very-very interestingly mixed with the rest MF. It was so beautiful that I actually found myself that I would like to drive with this signature my Upperbass Channels and my Fundamentals Channels not the MF Channels …  Of course I have no power in this tube and no inductance in output transformer to do so…

Sure, I need to burn all of this in and listen more of it but the first impression was positive and educational. The lower octave harmonics of YO186 is truly a discovery from me. In a way it reminds me the bass of Tannoy Red but without the tacky softness. The YO186 is not overly extended at HF, the Telefunken and Klangfilm beat YO186 but what it does with voice is a quite eyebrow raising… I think it will be a lot of fun to play it more…