Paul S wrote:

Romy, you would probably know this:  I would be interested to know if the ML2's 6AK5 tube is in any way related to feedback in the ML2, because I would guess that it is.  If not, then, I'm wrong (again).

Paul,

the 6AK5 has nothing to do with ML2’s gain stage feedback. Lamm uses 6AK5 as an amplifier in a classic voltage regulator circuit, where the V1 tube (6C33C) used as a voltage facet.

You might go to Steve Bench article about the tube regulators and familiarize yours how it works:
 
http://members.aol.com/sbench/reg1.html

(Go for higher pages)

Thanks for the link, Romy.  I'll check it out.  Too bad there isn't more information about the ML2 available.  A schemo with some values would be nice.  Do you know if anyone has sorted it out yet?

Steve Bench happens to live near me.  I was set to meet him but wound up having to cancel, then had to move.  I am now reminded of him, an interesting person.

Whether it is actually physically related or not, the ML2 bias took off soon after I changed out the 6AK5s.

I mention this for anyone else who may swap out that tube.

Best regards,
Paul S

I have noticed the same kind of control and "density" of sound from KR audio amps, which are zero feedback.  Have you tried this brand of amp?  The one thing that I can tell conclusively about the two different amps is that neither one suffers from bass distortion due to output transformer saturation.  I think this is where the level of control comes from that you are thinking comes from the negative feedback.  There are MANY tube amps with negative feedback and poor bass control and I bet if you look at their distortion vs. frequency plots you will see rising distortion into the bass. 

The one thing that the Lamm ML2 does differently than nearly all tube amps is regulation of the output stage.  They do this with the 2nd 6C33C that is on the chassis.  Even the KR amps don't do this and I think it is having a serious effect on the sound of the amp. 
Cheers,
Brad

morricab wrote:

The one thing that the Lamm ML2 does differently than nearly all tube amps is regulation of the output stage.  They do this with the 2nd 6C33C that is on the chassis.  Even the KR amps don't do this and I think it is having a serious effect on the sound of the amp.

This is an interesting subject. I spend quite some time thinking and experimenting with the subject of 6C33C regulation and although I have some observations I would hardly feel that my findings are final.

Yes, the output stage on ML2 is regulated, it isn’t not so seldom future in SETs (thought I never have see anyone on West did it commercially with 6C33C before Vladimir). There are 3 aspects in this subject:

• Benefits of the output stage regulation
• Benefits of 6C33C regulation
• Benefits of specific type of regulation

Benefits of the output stage regulation

What I have seen then with regulation generally and with 6C33C in particularly is that it allowed having more natural harmonics and better bass. A regulated out stage was able to play slow (in my book it is a VERY positive sign). The regulation (at least as I tried) also eat general dynamics (I used very fast pentodes in my regulators and feed the voltage reference tube not from regulated side but from raw voltage side)

Benefits of 6C33C regulation

That is a big subject. Russian do a LOT of amp around 6C33C and they are very accustomed to argue the subject if the 6C33C needs of regulation for better sound. Within any Russian audio community you will see arguments for it and against it. There is no consensus among them and there is no consensus in my mind. Generally I feel that 6C33C is too rough and too violent tube to be used with fixed base “as is” –this tube does need some kind of civilizing jacket. So, it is possible that regulation might be beneficial for 6C33C. When I was experimenting with 6C33C in context of Milq I did not reach the point where I had needs for regulation since I had discovered the sonic effect of gas-tube-dumping within my driver stage. You see, in Milq 6E5P bias is coming from gas tube and that gas tube (in context of the given bias model) has very interesting contribution. I have many tests trying it with and without the gas version and then concluded that Milq must be gassed – if not then the whole amp will be not good.  In 2000 John Atwood wrote in TubeCad:

“Gas regulators do not have very good dynamic response. Due to the time lag of the heavy ions that make up the discharge, they have time lags in response on the order of tens of microseconds. Redistribution of the discharge in the tube can make them even more sluggish. For this reason, they should not be relied on to stabilize strong audio-frequency variations.”

Mr. Atwood is absolutely correct BUT in context of the bias voltage the slowness and sluggishness of the gas tubes are God-sent. The gassed bias in addition ‘comforts” the currents floating from the 6E5P’s grid back to PS, creating all together VERY different sound. If to add to this soup the 6E5P’s own dynamic idiosyncrasy (it is not linearly-dynamic tube) then the result becose very, I would say -  “idiosyncratic”.

It is interesting that all dynamic compensations and harmonic comforting in my version of “how to use 6C33C” took place at the driver tube not in the output tube.  After I got what I got from Milq I had no needs to experiment with regulation of output tube anymore as the 6C33C did not sound to me as it needed anything else – the complete sound from the whole “packaged” was what I would call “dynamically-harmonically correct”.  Since I consider that a properly used ML2.0 is the absolute and incontestable king in the world of harmonic authority I used it as intellectual references in regards of harmonic education of myself. Sure, since I stopped to use ML2.0 and went for Melquiades it would reasonable for me to admit that Milq was better in portraying harmonics but I would not do it. Milq and ML2.0 do harmonics similar but in a different way. For anyone who played with C633C’s Sound it would be obvious that the 6C33C is subdued somehow in ML2 and in both amp the “subduing” has “different feeling”. It was VERY interesting to put both amps next to each other and to learn what they both do in reference to each other. I am sure some folks out there, among those who will build Milq, will do this experiment and it will be educational to hear from them. From my point of view Milq was free from all shortcomings that I implicated ML2.0 and it was “enough” for me.

Did this tail portray anything about the subject of 6C33C regulation? I think it does. The regulation itself is not the self-abscessed goal and it should be views on context of the “total Sound”. It is possible that Lamm has accomplished good things regulating his 6C33C but it also possible that he did not as the core of “ML2 Sound” might come from a different source.  We do not know how the ML2 made and what make it to sound as it is. It is possible that the regulation is just an externals “visible” thing that we are able to recognize but we do not see more and deeper…

Benefits of specific type of regulation

It is possible that 6C33C might need a specific type of regulation. During my experiments with regulating the 6C33C I drove it with 3 paralleled 807 tubes, I played also with paralleled 6C19P.  The 6C33C itself is a great regulator, Lamm knows this tube and he might know how to regulate it “properly” in terms of sound. From another side Lamm is not in a business of making Sound (he used to be) but in a busses of profit maximization of his transactions. I have seen before in his thinking and in his actions that he implemented costly solutions, absolutely not necessary (or in some case even evil) in terms or results, but he went for it juts in order to make the equipment “larger” and more expensive. I certainly do not say that the 6C33C’s regulation in his ML2 was one of those “features” but you never know with Vladimir. Agree, if you see a manufacture does it once then the manufacture is losing trust credentials forever…. So, you never know with Lamm if the regulation was a well warranted design decision or it was just his attitude-fart via electronics….

Returning back to the subject of regulation. Do not forget that 6C33C in Milq case is choke-regulated.  The voltage regulation of course is not tight as in case of ML2’ active regulation but in transient ability the choke-regulation is much faster than tube-regulation. Also, the choke-regulation does very good current buffering that might also be useful.

So, go figure the “truth” about the subject or regulation. Still, I would not feel that juts a fact that an output stage use regulation is automatically set one amp above others. I have seen some low-current amps with regulated output stages that did not sound well….

Paul, there is a number of inaccuracies in your post.

Paul S wrote:

Lots has been written already about tube replacement, but I want to mention my recent experience when replacing the 6C33C output tubes (V2).  I had the previous tubes in for only about 1 year; in this case I estimate about 300 hours, although I do not keep a log.

One year is a normal life expectancy for 6C33C, those 300 hours sounds a little too short.  I would say that a normal 6C33C driven at sub 50W should serve 800-1000 hours and then it might be replaced even if it works fine. Sure, being a Russian tube and partially as low QA as 6C33C it might do whatever she wants…

Paul S wrote:

I got the previous tubes from Lamm because I could not find anyone else at that time who could/would match them for gain.

It is extremely simple to do yourself – it will tale a few minutes.

Paul S wrote:

Lamm now wants 192.00 USD for a pair, so I  shopped harder.  My new ones are from Ken Chait.  We'll see what happens.

I do not believe ANYBODY do tube matching, particularly the 6C33C. They all BS-sayer and 99% of them know nothing about 6C33C. How that Ken Chait match them? I would love to mock his assurances and his warranties of “matching”. Lamm does test and match his tubes, those his price of $192 sounds to grotesque, particularly knowing that he buys then in Novosibirsk for $5 each. Now, the sad truth is: the 6C33C for ML2 should not be matched at all. Well, they should be matched by gain that could be done by 8 year old child. It is SET amplifier not PP amp, what the hell we are talking about matching? Even the bias selection is not necessary in ML2 as the V2’s bias resistor is wide enough to accommodate wide offset of 6C33C’s bias.

Paul S wrote:

So, who knows how much additional "service life" the previous V2 tubes had/have on them?  It's hard for me to believe that the V2s only last 300 hrs, although I might only be getting 100 hrs from the stupid 6N6Ps.  Anyway, the bass started to go, it sounded like V2, and new output tubes got it back.  I will continue to use the "old" output tubes as voltage regulators (V1), as Romy suggested, since it has worked fine so far.

The 100 hrs from the 6N6P sound too strange and not credible. Also, when the 6C33C doing south it does not loose any bass. Paul you need to revise your views about the dying tubes…

Paul S wrote:

But the main thing I wanted to pass on here is to remind other ML2 users to be careful about bias settings when installing new V2.  My old tubes had just checked at 185V/28mA, and I was only reminded to check the new tubes when I could smell them.  At that point they were 64mA!  No harm done that I can tell, but I'm glad I stuck around rather than leaving the room while the amps "warm up" for 20 minutes, like I usually do!

Most likely it was not 28mA and 64mA but 280mA and 640mA.  I have written many times that a new 6C33C should be properly broken in. Read about the “start up gap” in the “The short "6C33C Survival Guide":

http://www.romythecat.com/Forums/ShowPost.aspx?PostID=6140

When you installs a new time in ML2 you need to turn the plate current regulator all the way to the left. Install the tube (it will be 10mA-30mA), let this tine to burn like this for 30 min. Then set the plate current 150mA and let the tube to experience 3-4 times a full cool down and then > 30 min. After then the 6C33C will loose most of it “start up gap” and might set to it’s cruse current after it was buried more then 30-40 minuets.

It is shame that Lamm never explained in his manual to his customers how to properly prepare the new 6C33Cfor use. It is obviously that he prefers to charge $192 from his naive customers in order to kill the “start up gap” in his house.  Interestingly that this strange Lamm’s “caring” hurts him very aggressively and I have no idea why he never learns. I know/knew ton of ML2 owners who avoid buy tubes from Lamm just because the ridicules price. I am sure if Vladimir open his mouth and explain to people the satiation then people might not feel that he juts fucking them up. Well he does but not in the moment that people know.  The very same took place with Lamm L2 preamp. Over the years I say to people to replace regulators and the rectifier but the freaking Lamm kept saying “good for over 5 years”. The result? Tone of people have sold the L2 preamp because it did not performed well due to the dead tubes in PS. In fact, in 2004 the avalanche of the people who were getting rid of L2 was so high that Lamm was forced to make a public warning that the regulators should be replaced after one years (in realty 2-3 months, though I have see that the sound of L2 went down after 2-3 weeks because of the 6C19P). Funny is that Lamm deal with 6C19P for 30 years and well knows what it going on – he juts do not give a shit. Well, so did I and other people when we get rid of our L2 preamps…

I have no idea why he behaves as he does, well at this point I do not care as I have my answers.

I have stated several times that the ML2s are less sensitive to bad electricity than my phono stage, and this is true.  However, this is not to say that the amps are immune to bad electricity.  In fact, it is the effects of bad electricity on the ML2s that got me thinking more about audible effects from the amps themselves, because these effects are mostly more noticible when the amps "struggle" with bad electricity, including voltage drop.

 One thing I think about the ML2s is that they seem to want/need LOTS of good electricity to do their best.  Nothing unique in this, really, but I have found that the ML2s normal "effects" on the sound are far less "transparent" when the electricity is bad, meaning that bad electricity has actually given me some insight into how the ML2s go about their business.  Unfortunately, now that I've become more aware of the processes on bad electricity days I am also more aware of these processes when the electricity is good.

For one thing, the "rules" of SETs still apply to the ML2s, meaning these amps do their very best when operating well within their comfort zone.  And that comfort zone is a lot smaller when electricity is bad, as it often is.  Because the ML2s are so "powerful" I can see where someone might get the idea that they can be pushed, since they will deliver the SPLs with difficult loads and bad power, and they are still lots better than other amps when so stressed.  OTOH, once you start to figure out what really makes these amps special, then you are also bound to notice when they are not doing their considerable best.

So, how much good power do these things need coming in to put out their best?  I suppose its 120 clean, well-shaped VAC.

If I say it is clear now that the corrective elements at work in the ML2 are a "factor" at all times, then the Morons will immediately "understand" that this is a "flaw" that therefore "devalues" the ML2.  So perhaps I should say that the wonderful things the ML2 does come - like everything - at a price.  And perhaps we pay some sort of price like this for any plug-and-play solution.

At this point I can hear the ML2s "failings", but I still can't imagine "actual" amps that might do better in the context of my present system.

I liked Romy's idea for the motor driving the generator, so I looked into it a little deeper and found out that the more "serious" generators require up to 20 hp to drive them!

I can actually see the generating station from my listening room.  Maybe a dedicated line...

Paul S

Paul S wrote:

I have stated several times that the ML2s are less sensitive to bad electricity than my phono stage, and this is true. However, this is not to say that the amps are immune to bad electricity. In fact, it is the effects of bad electricity on the ML2s that got me thinking more about audible effects from the amps themselves, because these effects are mostly more noticible when the amps "struggle" with bad electricity, including voltage drop.