I have no idea why this guy is and what he says, I even have no idea of what language it is but here it is: the 6E5P-6C33C amp in PP configuration.

http://drzlab.com/projekti/TPPP_6c33c.html

Interesting that he uses Russian military Teflon caps that I found to be very bad sounding but this guys open them up and do something with them. I have no idea what as I can’t read it.

http://www.moxtone.com/PTFE_COND.htm

Paul S wrote:

Seems like a good basic concept for a laconic PP amp, though...

The fix bias of  driver and out stages are fine. Why he need to have 1:8 input transformer. The driver stage does 32 time gain, the output stage does 3 time. Is it not enough?  Does he has his speakers with sensitively of 85dB and 2000 cub feet room? I think for the gain he made with 1:8 input transformers I think the amp will not have enough power. However, the biggest thing that I do not like in there is the PP concept itself. In this PP implementation the out tubes shall be match super prissily and it is imposable with 6C33C. Even if you do match them initially then the different 6C33C age with different tempo. The only way to use them in PP is to use automated base that make the 6C33C to sound like wet dog, courtesy to BAT and others. So, I am not so impressed with what this guy does. Still his use of 6E5P makes him deserved to be mentioned. I do not know what result he gets from this amp. Even if I was able to read his text then I would not believe that he say anything about sound. As I underrated he is from DiyAudio.com site and those people know only about soldering and measuring but they are clawless about Sound.

vuki wrote:

Anyway, my plan was to make powerful triode NFBless amplifier for relatively small money. I'm really satisfied with what I did. Maybe I should have done GM70 SE, but what would be my next project then?

vuki wrote:

To tell you the truth I was already very satisfied with sound, but I just wanted to make that powerful triode amplifier - I really don't need it. But it's nice to have it. As you know in this hobby it's not so much about the end of the road but more about getting to it. So there is probably no audiophile who, from time to time, doesn't want to try something new. If audio electronics is your hobby then even more so. Audiophile who is completely satisfied with the sound of his system doesn't have the perfect system - he just gave up searching for it.  That is why I would probably also make GM70 SE amplifier. Just to enjoy making it and then tweaking it's sound.

It is nice to see the de-centresation of audio manufacturing. Formerly US, UK, Germany, France, a few Scandinavians and off the wall Japanese were making high-end audio.  Today the lead positions in interesting high-end audio have moved to Poland, Bulgaria, Korea, Greece, Lithuania and few other countries that 10 years ago were not big on the world hi-fi map. That is very good. I can talk a lot about the reasons but I am sure that not the last of the reasons that English-writing audio publications idiots who have been fucking up public audio intellect for years, did not yet learn nether Bulgaria nor Lithuanian…

Anyhow, here is another interesting product from Greece.  The company Cosmos audio Labs and they have 2-stage SET with “my” 6C33C.

http://www.cosmosaudiolabs.com/products.html

A pleasant design and a present implementation. I have no idea what another 6C33C does in the back of the amp – is it a regulator for PS or what? The description does not say anything about it. The driver is 7721. This is very good tube, I used it in output of my phonocorrector. I am not sure if I like it as a driver for two-stage amp. The 7721 has very little power but the 6C33C would like to be pushed very hard. If someone from Greece then pitch to the Cosmos people my 6E6P, it is the same pins, practically the same gain but it would run 8W as a driver. I think that 15W that they specify are mostly restricted by clipping of the driver stage – I am sure they might use more current on the driver.

Sure it turns me off that they provide Frequency Response not at full power – to me this is an indication of evil-minded stupidity. Still, I would like to hear the amp and to learn more about the design. If it will be distributed in US then let me know.

Dear Romy, thank you so very much for mentioning our amplifier to your excellent forum. Indeed, during the last 5-6 years there are a lot of manufacturers “from non traditional hi-fi manufacturing countries”. In particular, in poor old Greece we are more than 50-60 very small companies building some interesting audio products even with the extremely awkward local financial situation which is leading to extinction a lot of us.

Anyway, as the description on our site on second paragraph says, the third 6C33 located at the back, is used as the pass element of the regulator (controlled by solid state circuitry) providing 220V with 100uV of random noise & Hum at full load with very low and uniform, across the audio band, output impedance.

On the other hand, 7721 was chosen amongst many other driver tubes tested including 6E6P-DR and 6E5 too. To tell you the truth I really liked 6E6 but the gain was lower than needed and after 1 year on the test amp it started to degrade in sound and characteristics. Also 7721 (specified for 10000hours service life which is good for a commercial design) seem to my ears marginally more analytical than 6E6, though 6E6 had marginally better articulated bass. 7721 is not a small tube at all and at 170V(at anode) and 14mA (triode strapped) provides very low distortion even at full power. In any case the circuit can easily accommodate 6E6 with very minor internal adjustments (though no soldering is needed).

The quoted 15W of output power is restricted only by the distortion characteristics of 6C33, if you do not mind higher distortion (say 7-8%) the output power can reach easily 20 or more watts. Moreover, because of the adjustable regulator, someone may increase voltage, from 220VDC specified, to 230 or more and play with the operating point, but the sound is not that correct to my ears, I really prefer 220V @ 210-220mA but my taste is just my taste.

You are correct though that frequency response is not at full power and you can be assured that I am not at all evil minded as I am not stating that the response is at full power. The frequency response stated is measured at 8W@8Ohm done just for ease of measurement as the 8V mark on my millivoltmeter happens to be right above the 0db mark. For your information, I just measured the actual full power bandwidth at 15W and it is -3dB from 9Hz to 122KHz  which is not bad and not very much different from what it is stated at our site and very respectable for a zero feedback SET amplifier. This good bandwidth is the product of good old transformer calculation and construction execution using a total of 10 primaries and 5 secondaries carefully layered. The core is of type E120 with 8cm of laminations providing a total of 10H gapped for 250mA. 

On the other hand, we got no distributor in the US yet, as we launched our products in Munich Hi-End Show (together with Tune Audio Anima speakers) in May and had no contacts or interest from any US distributor.

Sorry for the long post,

Remaining at your disposal for any questions you may have

Cosmos wrote:

Anyway, as the description on our site on second paragraph says, the third 6C33 located at the back, is used as the pass element of the regulator (controlled by solid state circuitry) providing 220V with 100uV of random noise & Hum at full load with very low and uniform, across the audio band, output impedance.

Cosmos wrote:

On the other hand, 7721 was chosen amongst many other driver tubes tested including 6E6P-DR and 6E5 too. To tell you the truth I really liked 6E6 but the gain was lower than needed and after 1 year on the test amp it started to degrade in sound and characteristics. Also 7721 (specified for 10000hours service life which is good for a commercial design) seem to my ears marginally more analytical than 6E6, though 6E6 had marginally better articulated bass. 7721 is not a small tube at all and at 170V(at anode) and 14mA (triode strapped) provides very low distortion even at full power. In any case the circuit can easily accommodate 6E6 with very minor internal adjustments (though no soldering is needed).

Cosmos wrote:

You are correct though that frequency response is not at full power and you can be assured that I am not at all evil minded as I am not stating that the response is at full power. The frequency response stated is measured at 8W@8Ohm done just for ease of measurement as the 8V mark on my millivoltmeter happens to be right above the 0db mark. For your information, I just measured the actual full power bandwidth at 15W and it is -3dB from 9Hz to 122KHz  which is not bad and not very much different from what it is stated at our site and very respectable for a zero feedback SET amplifier. This good bandwidth is the product of good old transformer calculation and construction execution using a total of 10 primaries and 5 secondaries carefully layered. The core is of type E120 with 8cm of laminations providing a total of 10H gapped for 250mA.

Cosmos wrote:

Indeed it took me a very long time to decide upon regulate or not. I ended up with regulation option to save some chassis space and weight, otherwise I would need at least 10 kilos more and 30% bigger space that would make the whole amp impossible to move and redicoulously big for a 15 watter. On the other hand regulation is good for a commercial machine since its easier to live with less adjustments for the final user, moreover the sound of the regulator is very similar with slightly better detail and bass, however chokes provide slightly easier and softer sound to my ears, but there are times you got to make some choices and regulator was mine.

Cosmos wrote:

You are absolutely right on frequency response method of testing and I will correct my site asap, however that was not at all intentionally as expalined in my previous post.

Cosmos wrote:

On the other hand my measurements are correct and really do not understand why you do not believe these, just for your information my equipment is recently calibrated and are the following, tektronix SG502 signal generator, HP3478A, Fluke 8840A, Fluke 8060A, HP400H, tektronix D15 and 561A, these are used for measurement of frequency and Vrms and also EMU0202 as spectrum analyzer. Note also that the load used for these measurements is a 50W 8ohm power resistor. If you would like me to measure something else please be my guest. I really do not get why you are saying that the response you are getting is only 20-22KHz, maybe your output transformers are not up to the task? Maybe I can help you on that.

Dear Romy,

Obviously it is impossible to convince you upon the frequency response of our transformers and in any case that is fine with me since everybody is entitled to its own assumptions and opinions and this is not bad at all, on the contrary, I love different opinions.

However, let me illustrate some very-very basic and obvious knowledge-strategies that we used for these transformers, not at all kinky but sometimes forgotten (some times intentionally).

In low rp tubes and especially as low as 6C33 which is sub 100Ohms, our design is intentionally targeting on decreasing leakage inductance in expense of increased inter-winding capacitance.

Specifically, our output transformers may well have several nano-farads of inter-winding capacitance but the leakage inductance is only 1,5mH.

The above, in conjunction with ultra low 6C33(sub 100ohm) plate resistance, allows a transformer construction with -3db at more than 100KHz. Thing which is completely impossible to accomplish with almost any other tube, even with 300B (with a lowish 800ohm rp), no matter how well the transformer is calculated and executed.

The only viable possibility to construct a transformer for 300B with 100KHz + response is to make the transformer so small that the bass will be almost absent.

Of course your ideal implementation of DSET is capable of producing even more impressive results, but unfortunately this is not a design to be used for general commercial use.