Rerurn to Romy the Cat's Site

Melquiades Amplifier
Topic: What Might Have Been...

Page 1 of 2 (21 items) 1 2 »


Posted by Romy the Cat on 12-18-2007

Was talking with various audio people about the 6C33C I was amazed that people are so clueless about this tube. The people are so filled with some foolish prejudges and faulty notions about 6C33C that I deseeded to put together a brief “6C33C user manual”. Pretty much what I will say here I have said before in various posts in my site and somewhere else but I think it would be a good idea to put all my thoughts in one place calling is “6C33C Survival Guide”.

I am very much not an electronic engineer authority or an experienced DIYer by I am a person who since 1996 went through number of 6C33C-based power amplifiers and burned over 60-70 of those tubes. That makes me kind of familiar with 6C33C as a user. Also, I use in my audio endeavors my ears and common sense that in order to make audio recommendations makes me much overqualified among the rest of audio crowd. It would be worth to mention that I have no problems of complains with 6C33C, nether with it’s sound nor with exploiting of this tubes – so something I do right…

All my comments about 6C33C will be related to use this tube in single-ended applications with output transformer. Those comments of mine might be applicable to OTL and PP application but sine I personally never built or DIY-experimented with OTL or PP I would offer to others to think how applicable my comments would be in respect to  those topologies.

So, here is the datasheet file on 6C33C:  http://www.goodsoundclub.com/PDF/6C33C.pdf

The 6C33C has only one equivalent: 6C18C. The 6C18C is an early version of 6C33C, might be used as drop-in replacement. The 6C18C is very good tube, in some instances is better sounding then 6C33C but in may experience the 6C18C has a large percentage of faulty tubes. Among fifteen new 6C18C that I used 5 of them were discarded immediately due to the pins shortage, 2 of the “good” tubes failed a few hours after they were heated and one of the surviving failed in a few month when it was subjected to shaking. In contrary the 6C33C is very reliable; among 60-70 tubes that I have burned I had only one that was defective.

So, here is goes my “6C33C Survival Guide”.

1) A new 6C33C has pins cover with something that I call “white shit”. I have no idea what it is, perhaps some solidified oil or paraffin. Anyhow it should be scrubbed out before you use 6C33C for a first time. If you do not clean up the “white shit” from the pins then it will short the lifespan of your 6C33C tube socket

2) There are six 6C33C tube sockets (ordered from worst to the best)

         a) Chinese Ceramic
         b) Russian Plastic
         c) Russian Ceramic
         d) Russian Exotic (with side handles) 
         e) American Johnson
         f) Japanese Exotic

Each of them works fine but the best sockets will last longer. The Chinese sockets will die after 1-1.5 year  of heavy use and generally it is a good idea to replace a new 6C33C right along with replace of the tube socket. The American sockets might last for years and the Japanese socket where pins hold by a full cylinder might last for a near a ife-time

3) The 6C33C tube sockets should be mounted at well ventilated location with nothing closer to the tube then ~ 2-3 inches. Out of chassis mount is highly advisable, but if the in-chassis mount use then punch necessary holes to crated natural air-convention or use forced cooling.

4) The Filaments should use AC. The filaments wires should be 16ga or larger. The wires to the socket’s leads should be soldered with high temperature solder.

5) If forces air use then do not cool down tube but cool down tube socket.

6) 6C33C should be use with fixed bios only. If you feel that in your amplifier 6C33C sounds better with automatic of semi automatic bias then ether you are in a strong need to perform a lobotomy  over your reference point of your amplifier is a piece of crap to begin with.

7) Biasing resistor that goes to the 6C33C grid should be the best possible quality and has value less than 100K.

8) The multi-turn precession attenuator that sets up you 6C33C’s bios should be also very high quality.

9) The 6C33C bias supply lines are very critical for 6C33C’s sound; even the proximity to the ground of the negative bias wire is auditable. Use transmission line techniques to set up that line.

10)  Although the 6C33C marked for 60W plate dissipation and some of them will care even more with no problem still made for yours an absolute rule: the maximum plate dissipation of 6C33C is 50W, with recommended plate dissipation 40-45W for a full tube and 30-35W of a half tube.

11)  The half 6C33C sound better then a whole 6C33C, the half 6C33C sound 30000 times better than one 6C41C.

12)  Set up proper operational parameter for your 6C33C. The proper operational point will be the properly of your plate load, type of acoustic system you use and a few other factors. Still, with all things considered your 6C33C should experience power clipping by currant and by voltage at the SAME time.

13) Generally look for 180-210V on 6C33C’s plate and 180-230mA of plate current for a full 6C33C. still, the paragraph above (#12) should be denominating rule

14) The 6C33C has HUGE inconsistency of parameters when it is new, so huge that it is impossible to measure the tube in order to determine how much of cathode emission left or what mutual conductance would be is. The only possibility with such inconsistency of new parameters is to monitor how the parameters were changed for a given tube.

15) Make bias supply available from minus 50V to minus 120V

16) Generally the signs of the 6C33C dying are (in case the tube  still operates properly):

      a) Appearance from time to time mechanically sounding loud “ping” through loudspeaker.
      b) The tube’s getter loosing it’s gloss
      c) Change of bias voltage more than 50%
      d) Tube is used over one year in amplifier that works dally.
      e) Too small amplitude of “start up gap” (more about it later)

Since the tube is dirt cheap ($5 in Russia) it is advisable to replace the 6C33C each yeas or what the above mentioned effects show up.

17)  All 6C33C have different gain.  Gain of 6C33C is the only parameters that should be matched. If you have 3 of those tubes on a row with the same gain then you are incredibly lucky and you should stop play with your amplifier and should become a stock broker.

18)  When a well worked 6C33C starts in your amp the filaments should be pre-burn for =>2 minutes.

19) Do not apply a signal that would exceed more than ¼ of class A1 to a “cold” 6C33C. The 6C33C stops become “cold” ONLY AFTER 30-40 minutes of burning by its cruise plate currant (I call it “Hot Mode”). Practically it means that you should not play your 6C33C amps too loud for the first half hour, particularly if you have low sensitively loudspeakers of too large listening room.

20) When 6C33C start with none-regulated power supply then it has the “start up gap”. The voltage is higher than cruise plate voltage and the currant is lower. With time, when the plate current hits up the tube the voltage drops and the current raises to it references cruise values. Usually a worked 6C33C reaches its 95% of cruise currant within 10 minutes.

21) Use the 6C33C “start up gap” to monitor the life cycle of 6C33C. If your cruise current is 220mA then a worn 6C33C might start at 200mA and then within 10 minutes reach 220mA. A newer tube will start at 150mA and then within 10 minutes reach 220mA.

22) When you take as brand new tube, then heat it’s filaments up for 2 minutes and start the amp with 20mA on plate. In 10-15 minutes the plate current will rise to 150mA – let it be this way.  Burn the tune for another 3-6 hours with 100mA-150mA and then consider it ready to be use at full power.

23) Violation of proper pre-heating rules and application excessive grid current to “cold” 6C33C lead to overheating of anode. Once anode was overheated it will have tendency to be more susceptible to overheating in this tube. A properly pre-burned tube, running “Hot Mode”, with fix bias and in class A1 is absolutely indestructible and perfectly thermo-stabilized.

24) It is advisable once a 6 moths to inspect the 6C33C’ filaments pins for the sights of corona, partially if your use not American trapezoid tube sockets buy Russian and Chaise triangular sockets. It worth also, if you have nothing else to do, once as 6 month to listed the balloon of cold 6C33C with juts applied filaments voltage with a stethoscope. A present of corona on the filament’ pins will, sound like a very light buzzing.

25)  Keep, power supply impedance that drives 6C33C as low and you can.

That is all that I have atop of my mind. Properly used 6C33C is truly a problem-free tube with opportunity to have excellent sound if it is appropriately used in amplifier. So, I was disappointed with moronic myths that audio people invent about this tube and to address that stupid audio-prejudicial folklore I have put together that “6C33C Survival Guide”.

Rgs, Romy the caT

Posted by Romy the Cat on 11-10-2008
fiogf49gjkf0d

Srajan Ebaen made a write up about a Polish amp from Ancient Audio

http://www.ancient.com.pl/e_single_six.htm

http://www.6moons.com/audioreviews/ancient3/single6.html

It is good that Srajan do not write about sound anymore, whenever he did it appeared to me as it was delirious blabbering, so it is better Srajan do not even touch sound and do his “reviews” as extended new reports. His is much better for this.

So, the polish 6C33C amp with the Srajan’s 6-mooning and repent the manufacture comments. I do not agree with everything that manufacture said, some if his comment even sounds even idiotic but generally it a not complitly wasted educational read about 6C33C, if you care about the tube.

The caT

Posted by guy sergeant on 11-10-2008
fiogf49gjkf0d
Is it true that they were only made at one factory and are not made any longer?

I realise that not many people are using them (either manufacturers or DIY people) but presumably there are large stocks lying somewhere?

Posted by Romy the Cat on 11-10-2008
fiogf49gjkf0d

I do not know if Russians still make them. I do not think they do as they hardly manufacture anything nowadays. I presume that there a large stock of them in Russia as Victor Khomenko (from BAT) told that during the peak of the 6C33C production Russians made 100.000 tubes per month.  I think a few years back Victor told that they are still being made…

The 6C33C were not made at one factory. Before 1970 there were produces by Svetlana in Leningrad, the Winged. Then for a short time the production reportedly was moved to Poljaron in Lvov, it is Western Ukraine. Then from beginning of 70s the 6C33C were made by Ulyanovsk plant. The Svetlana’s tubes from 60s usually are more pricy and “rare” but I do not find it justifiable. I never had any Lvov-made tubes and most of tubes that you will find today are Ulyanovsk-made from 70s and 80s. I did see tubes made in 91-92 but I never have seen anything later then 1992. I use 6C33C since ~ 1995-96.

I do not worry about the long-term availability  anymore.  I kind of had preoccupation in past with stashing of those damn tubes. When I discovered 7788-7721 I then stashed a few dozen of them of all imaginary manufactures – I have quite unique collection of them. When Dima pitched to me his 5E5P/6E6P and I learned how good they were I got from Russia a near 1000 of them, all imaginable vintages. Who cared how many if they were 11c-19c each and the cost of shipping was the most pricy part of the transaction. The same was with 6C33C. Before I gone for the 12-chennal Melquiades DSET I got from Russia 200 of 6C33C somewhere $4.85 per tube + $200 of a bribe to Russky custom service. The tubes are sitting in boxes in my storage… However, last year I it suddenly hit me how Moronic it all was. We need or two a tube per year and we waste our time, money and attention to create a mental deficit in our own minds. If I do not have those tubes in my stash then what would be a big deal to go to the open market and to buy the tube WHEN I NEED IT? Well, it will be slightly more expensive – so what, big deal!!  Those tubes cost the same as a large provolone sandwidge in downtown Boston - we do not count money when we by a sandwidge during lunch, so why we so preoccupied with availability (that is juts only money) when we pay the same price for a tube twice a year? I just recognize that it ridicules behavior is and I do not do the tubes or any other parts stuffing any longer. Anyhow, the long term availability of tubes or anything else does not bother me anymore.

The Cat

Posted by Romy the Cat on 12-07-2008
fiogf49gjkf0d

Another memeber of the 6c33c family

http://www.audioart.no/satiamp.php

Have no ide what it is but looks very sexy.

The Cat

Posted by Romy the Cat on 01-16-2009
fiogf49gjkf0d

If you read my post at:

http://www.romythecat.com/GetPost.aspx?PostID=4235

Then you are generally informed about the history of the 6C33C but here is another wonderful and uncommon illustration. It is reportedly a guidance system part of Russian cruise missile from 60s. You can see the 6C33C employed at her full glory.

6C33C_military_1.jpg

6C33C_military_2.jpg

6C33C_military_3.jpg

Romy the Cat

Posted by Paul S on 01-16-2009
fiogf49gjkf0d
Seeing this guidance system, I can't help but wonder if they used the same stupid steel tube sockets that Lamm used.  Had "we" known then that "they" did, we could have slept peacefully at night, maybe spent more on arts, education and universal health care.

Best regards,
Paul S

Posted by Romy the Cat on 03-15-2009
fiogf49gjkf0d

I already wrote a few times at my site about 6C18C. Search if you are interested.

 http://www.GoodSoundClub.com/TreeItem.aspx?PostID=2292

If you use the 6C33C for amplification or regulation then you would hardly avoid attention to 6C18C. It was the ancestor of 6C33C and it has larger anode mass – a very good sign. I got a couple years back a party of 6C18C from 1964, 1968 and 1969 and I run them at Milq bass channel. I do like them BUT… The 6C33C was introduced and “more reliable version of 6C18C” according to BAT’s Victor Khomenko. Leaving for 2 year with them I understand what he meant – the dam tube is just too fragile and to mechanically unstable. A few of them failed catastrophically and a few of them I did not risk to put in my amp – it was a bunch of garbage riddling in them. One of them was perfect – in fact very perfect. The rest were good but… did not live for a long time.

The tubes are 40 year old and then are in perfect shape, with dense and glossy getter. However, as soon those tubes begin to work then they very rapidly losing getter. The getter is necessary to remove the gas that made up by cathode bubbling from vacuumed tube. I do not know if the cathode in 6C18C produce too much gas or the tube heat up and is loosing it’s hermetic status but I do know that one 3-4 weeks of use and the tube has no dark silver deposit atop: 

Naked_6C18C.jpg

So, it is not good. I was trying many different things that I was advised to stop this process but it did not help. An old man told me that in order to preserve the getter I need to dance naked each full moon around the tube driver by 1.666V. I tried and it did not help. Now, I have a small army of the tubes like above and I wonder what to do with them – to use them or not? The funny part is that I do not hear any worsening of sound from those tubes. The next step would be to pump the marihuana smoke into the 6C18C balloon and to play exclusively Luigi Russolo’s music…

The Cat

Posted by Romy the Cat on 08-21-2009
fiogf49gjkf0d

Hm, a German company Audio Valve clams that this new Assistant 100 amps has

“110 w sinus wave - 1khz / 1% dist. / 8 ohm load, per channel”

http://www.audiovalve.info/nav/assi100_en.html

That is bold statement even if you drive the thing very hard, not that I see a reason to do it. As many people before they drive the 6C33C at near or over 60W on play and stabilize current by applying fixed bias dymickly.

I do know what sonic advantage the 6C33C has by running it very hot – it might but the question I have to ask – of you need 110W for your speakers then …why you do not blame your speakers?

Anyhow, I do think that Assistant 100 amp is an interesting idea BUT I would like to see the Audio Valve to do a 2X50W amp with a single 6C33C per channel. THAT might be interesting to drive bass channels, of course replacing the Audio Valve’s originals output transformers. 

BTW, it would be interesting to try putting the Audio Valve bias circuit on the Milq LF channel.  It would not be hard to do – just measure current and slide the bias. The Milq LF has 450mA gap in output transformer, 5000mA power choke and 1A power transformer.  I with 400mA and 200V on plate I will have 80W on anode, interesting what would be the plate impedance in this mode. I presume it will be sub 50R. It might be fun to try this mode another day. I wonder what happen with sound of 6C33C in this mode with low level signals…

The Cat

Posted by Romy the Cat on 08-21-2009
fiogf49gjkf0d
I understand that AudioValve would like to keep their generic chasses for all their models but I think what they did with Assistant 100 amp is very wrong in term of chasses architecture. They claim the “special familiarity with 6C33C” but it is not what I see and sine it is a new amp and not one has it I would like to warn people about the faulty (in my view) design decision the AudioValve took.

Take a look the 4 6C33C clustered VERY close together on the circuit board. The meters are grades up to 500mV, that I am sure are mA of B+. So, they do drive the 6C33C very hot. The 4 sitting together 6C33C even at 60W on plate produce tremendous amount of heat. 240W on plate + 12.6x3.5x4 = 176W. So, we have 420W only from output stage that is concentrated in very narrow spade – almost a chimney.  I have seen that a single 6C33C in enclosed space burns 120 degree capacitors from 3 inches. My Melquiades has 3 6C33C, two of them run with half-heaters; it has 4 time larger chasses then AudioValve, no power supplies and the tubes outside. Still, if I shut down the forced cooling then after 3-4 hours the amp gets uncomfortably hot.  The AudioValve has 4 6C33C sitting INSIDER and no signed of forced air as there are no holes drilled through the circuit board.  Those parts that are between the tubes will be burn soon on later; I put all money in my pocket to bet on it.

The only justification might be that they run the amp in some kind of tricky B or C class and PP where in idle mode they dissipate not a lot of power on plate. Well, if so then I do not think it shall sound right … I hope it is a parallel SET and if so then they must advertise is as a …toaster.

The Cat

Posted by aische on 01-19-2010
fiogf49gjkf0d
hi all - I found this "toaster" note: http://www.dlaudio.com.tw/front/bin/ptdetail.phtml?Part=new15&Category=326765

Paul

Posted by Romy the Cat on 01-19-2010
fiogf49gjkf0d

Thanks, Paul.

My Googleised Chinese allowed me to understand that they use some kind of forced cooling but there was not mentioning of it. Upon further inspection I come across to this image:

6C33C_Cooling_1.jpg

So, the amp has two fans aside to cool this thing. I still very much do not like the design from cooling perspective. 4 of 6C33C within one inch from each is very insane amount of heat. It would be very high if they were the stand-alone 4 tubes but that are on the PC board surrounded with other components – that all will be roosted very soon.

The amp has 2 fans that I am sure take work as output fans. I think it is not only not enough but located at very wrong position. Take a look at the second image:

6C33C_Cooling_2.jpg

The hot air coming from the 4 x 6C33C chimneys is taking across the bank of power caps to the output fans. Even if they are high temperature caps then no caps will be able to a long time to serve in this condition. One 6C33C running at 50W heat up to 280 degrees Celsius, or 536 degrees Fahrenheit. 4 tubes in high proximity and confined space would do 300-330 degrees Celsius. So, the hot air blowing across those caps are in vicinity of 250 degrees Celsius – you might very much do not use those caps as in time they are not effective anymore.

In addition the side cooling in close proximity is a problem. When cooling air hit a tube from one side and another side is super-heated by another tube then it create not even cooling of glass that is very bad for tube. Doe this “bad for tube” has any practical impact? I do not know. I do not know if it affects sound. I am sure it affects the longetivety of the tube but who cares about it – in one year you need to trash 6C33C no matter what…. Still, I think this sided cooling is not too effective.

I understand that they were trying to make a compact and elegant design but it the elegance contradicts functionality then it is not elegance. I would like in this amp instead of moronic decorative gas to see a very powerful outtalking fan atop of the amp. The side, the top and any other vents shell be closed and the air shall be taking from the bottom of the amp. The outtalking fan I would make 18-20cm and would run it at relatively high 700-800rpm with thermostat. The circuit board I would redesign making a LOT of holes in it let air to pass across the circuit board.  I would even to make the amp 2-3 inches wider, letting the pairs of 6C33C to stay a bit further from each other. Perhaps it would be a good idea to list up the circuit board for 3 inches up and put the heat-sensitive elements under the bottom of the board…

The Cat

Posted by aische on 01-19-2010
fiogf49gjkf0d
I don`t understand chinese language, can someone translate for more details ?

Thx, Paul

Posted by aische on 01-22-2010
fiogf49gjkf0d
Hi guys, it looks as if translated the review itself ... here




Posted by Romy the Cat on 01-22-2010
fiogf49gjkf0d

I see, what they did – the built a perfect sealed chimney with two fans to suck cold air in and two fans to suck hot air out. Well, it might work, even though I do like this idea. I absolutely do not see why such a hot and large tube as 6C33C shall work INSIDE of a chassis; in this case it will be 4 of 6C33C!!!! There is absolutely nothing that would prevent then to put 4 sockets atop of the chassis and mount the tubes there. But, hey this is not my amp….

The Auto Bias for 6C33C, well it is obviously not my cap or tea. The PP operation on such wildly different tube tubes as 6C33C also is not very much useful. Look at the Lamm ML1 – a lot of power but very little séance. Who else does the PP around 6C33C? Balanced Audio Technology famously did one, I think it was 60W and they also used automated bias. I do not know, I do not like this idea.

The Five-Element Diagram for the Stereo in the review is “different”. I do not know how and if it relates to sound in practical terms.

The Cat

Posted by Romy the Cat on 09-28-2011
fiogf49gjkf0d
The mean KOTriode in his thread “A different breed of 6C33C amplifier“ jinxed me with his 6C33C vulnerability and today my output tube at right midbass channel went down. The characteristic pings, which via the midbass horn sound like house room is collapsing. The tube is about 2 year old, the last year it ran at 55W, so it was time… BTW, for the all time I use the half of 6C33C (about 6 years in infection and upperbass channels at 150mA) I never had any of them die on my.  The 6C33C runs at 355F but a half of 6C33C at 275F, I think this is the reason why.

6C33C_Cooking_Getter_Activation.JPG

Anyhow, the new party of 6C33C and 6C18P went today for getter activation. I run it for 6 hours at 275F in dry oven and then for 2 hours slow roast it in Georgian Pirosmani wine with Italian peppers and Uzbekistanian herbs.  It gives very tender Sound and goes along very nice with Parmesan Cheese. 

Rgs, Romy the Cat

Posted by son on 04-26-2012
fiogf49gjkf0d
Hallo Romy, i'm interested with half triode at 6C33C , but im confuse how to set it,... 
Full Triode : pin 1&2 give 6,3v , pin 6&7 give 6,3v.
Half Triode : series pin 2&6 1&7 give 12,6v or just use pin 1&2 give 6,3 v
thanks before...

Posted by Romy the Cat on 04-27-2012
fiogf49gjkf0d
 son wrote:
Hallo Romy, i'm interested with half triode at 6C33C , but im confuse how to set it,... 
Full Triode : pin 1&2 give 6,3v , pin 6&7 give 6,3v.
Half Triode : series pin 2&6 1&7 give 12,6v or just use pin 1&2 give 6,3 v
thanks before...
It is truly very simple. Use ether pins 1&2 or pins 6&7, dive it with 6,3V. I mean you need to use only one set of heaters and this will bring your tune to half-anode mode. Drop your plate current to ~65% of what you use to have. Be advise the you will lose ~3dB power (in fact less than 3 dB and will rave twice higher pate impedance. If you would like to maintain the same loading of your pate (I would) then you will lose 3dB of gain by transformer ratio. If you have only 12,6V PS for your amp and stereo configuration chassis then you can connect your half triode 6C33C in series. It is a good question to ask: which side to disconnect: left or right. I was not able to detect which one is better. I would advise to use the side that has better cooling and farther from any cooling obstacles. Another rational would be to sue the side that is easier to wire from under hood and that further from your signal paths.

Posted by son on 04-27-2012
fiogf49gjkf0d
Its very clear right know,so i decide to build with 12,6v in series,... Thank's Romy.. : )

Posted by Romy the Cat on 04-27-2012
fiogf49gjkf0d
 son wrote:
Its very clear right know,so i decide to build with 12,6v in series….

Son, when I proposed the series configuration I thought that you have already have an amp and would like to convert it to single plate configuration. However, if you “decide to build” then I very much recommend you to go for 6.3V and just to provision extra current in your filament transformer of the transformer secondary.

You do not want to lose an action to go full power with your 6C33C, so extra current will help of you run both filaments in parallel. Parallel configuration has, in my view, benefits that it might be shielded. I presume you run AC on filaments and if you use series connection then there is no way to twist the wires and if you do run the twisted pairs with series configuration then it will be longer wire and not elegant. With parallel configuration the AC supply to filament wires are much short and much more accurate...

The Cat

Page 1 of 2 (21 items) 1 2 »