6c33 - ready, SET, GO

[rvrazvan]

About two months ago i just finishing my first amp. An el84 SET. A very nice amp, small, but because of this baby i'm now starting the build for the second one. 6c33 SET based on Borbely's design).



You can find the schematic and more in this article from Glass Audio
Borbely 6c33 SE



Yesterday i've recived my custom made OPT's.




The dimensions of the problem ). the small tube is an el84(equivalent)


I'm gonna post more photos once the build will advance .

[azazello]

Óñèëèòåëü íà 6ñ33ñ
This 2 stages schematic /I made SE ampl. 2 time/ sounds great and amazing!
/Instead 6E5P, You can use 6P15P or 6P9/

[rvrazvan]

hmmm. i might try this schematic this weekend. Borbely's schematic has a special feature callled "distortion cancelation". the distortion in the second stage will cancel with those in the final tube(180 phase shift).
I'm not convinced about mixing pentodes in the schematic. i will do some research and maybe i'll change my mind.

[the_manta]

Hi,
I wouldn't go for the Borbely design. Always use a cathode resistor when dealing with 6C33C or at least a mixed topology with fixed and auto bias (At leat more than 1R as in the schematic !!). 6C33C drifts nicely if no bias compensation is used.

Furthermore I prefer a better driver. Why SRPP as first stage ? SRPP gets you huge voltage swings (when connected to a suitable high Ub) to a unique fixed load impedance. Nothing of that is needed in the first stage.

You will get best results with the 6Э5П, of course connected as Pentode. Use max Ub which the 6Э5П is designed for. Results are even better when you use an anode choke. Feedback in old fashioned way to the 6Э5П's cathode (parallel derived, series applied)
Have fun !

Edit: Can you tell me where you got those OPT's ?

[rvrazvan]

I have some reservation regarding the use of a pentode in the circuit. I'm gonna try something this weekend maybe. The ears will know better.
About the bias....I know the problems...but for the best results i'm go this way. In the bias supply i'm gonna use only mkp(motor caps). The amp will have an amp meter for each channel and the bias pot's will pe mounted on the case.

Tube will be burned in for at least 24h(filaments only). This from what other people state will stabilise the tubes...i hope.

I can provide you all the contact information if you need them. I recomand the source. Cheap and well build.
Prost! )

[knoberik]

Originally Posted by the_manta

Always use a cathode resistor when dealing with 6C33C or at least a mixed topology with fixed and auto bias (At leat more than 1R as in the schematic !!). 6C33C drifts nicely if no bias compensation is used.

This opinion is over emphasized. 6C33C will not drift, if you are not driving it with too high current. Fixed bias is the only right way in output stages, try the sound differences and you will understand.

And using OPT is just wasting in case of using 6C33C. You can get those 15 sine watts easily without OPT. The best 15W you can imagine.
Every transformer in signal path is wrong. If this got your attention, I can add some experience to share.

[Blitzen]

Originally Posted by knoberik

And using OPT is just wasting in case of using 6C33C. You can get those 15 sine watts easily without OPT. The best 15W you can imagine.
Every transformer in signal path is wrong. If this got your attention, I can add some experience to share.

Let's hear it!

[the_manta]

Originally Posted by knoberik

This opinion is over emphasized. 6C33C will not drift, if you are not driving it with too high current. Fixed bias is the only right way in output stages, try the sound differences and you will understand.

Well, lets put it this way - You have to know what you're doing here. My 6C33C batch was rather inconsistent. One drifted more, one less. The burn in time didn't change it that much. I don't know the currents anymore, but thats an interesting point you're referring here.

Originally Posted by knoberik

And using OPT is just wasting in case of using 6C33C. You can get those 15 sine watts easily without OPT. The best 15W you can imagine.
Every transformer in signal path is wrong. If this got your attention, I can add some experience to share.

OTL's are a different issue. But saying that every transformer in Signal path is wrong is a bit tough. Bandwith limiting, stray inductance, etc we all know that but you can build excellent sounding amplifiers with output transformers. Designing a stable OTL is another story. Especially with 6C33C. Moreover you will need more than one tube as in SE Design. Mostly circlotrons or series push pull with feedback (Futterman and all the realted designs as Rosenblit, Murray, Gomes, Sandman, Taylor). But I'm looking forward to seeing your design. Still have those 7242 lying around

[rvrazvan]

Well. I won't give up on the project now or change it to much...but if you have a nice schematic for an otl amp...bring it on!
Maybe the next one will be otl.

[knoberik]

Originally Posted by the_manta

Well, lets put it this way - You have to know what you're doing here. My 6C33C batch was rather inconsistent. One drifted more, one less.

I started to use 6C33C tubes also quite lately, for now I tested about 20+ tubes. I built a special schematic for burn in procedures and studying the tube's nature, using anode current source with current limiter. I also first read and was concearned about the current drift warnings etc. But in all those cases I did not see the big bad current drifting some people are warning. Yes, the 6C33C has a bit special nature, but if you know it, there is no major issues about using the tube. I'd say there is a need to check all the 6C33C before starting to use them. In the other hand I do not see the need to run them 24h+ some people recommend for brake in. 1-2h with heater only and 1h more with anode current will be enough.
There were some tubes bad as born, inner shorts mainly, one has air inside I suspect. But not a single good tube went in current drift. 6C33C has a slight current rise after switching on, you have to concearn about it, but the current stabilizes after 15 minutes warmup. So called warmup-gap. For example, if you set anode current lets say 170 mA (after 2-3 min heating without anode supply), then after 15 minutes current will be 200-230 mA.

OTL's are a different issue. But saying that every transformer in Signal path is wrong is a bit tough. Bandwith limiting, stray inductance, etc we all know that

You listed the main issues to be avoided by eliminating OPT

Designing a stable OTL is another story. Especially with 6C33C.

Glory for the next goes all to user tubetvr, as he is the creator of this wonderful schematic, based on my 6C33C OTL amp is created.
Since I built a stable OTL amp with 6C33C tubes and did not experience serious difficulties, I can surely say it's not worth to bother with OPT, if you have some of those tubes. Result will amaze you as much as me. And yes, this schematic is rock stable (at least in my interpretation).

Since then I built also several amps with OPT and I was also thinking it is the only way nowadays to get enough current to low ohm output. I was also in constant searching of better OPT. Then just for fun tried the schematic mentioned before and I was and still am amazed every time listening my amp, how good it sounds.
It is a serious issue about amplifier you use. One bring out every detail and finesse, another is giving you just some sort of a mumbling. Speaking about tube amps only of course.

Moreover you will need more than one tube as in SE Design. Mostly circlotrons or series push pull with feedback (Futterman and all the realted designs as Rosenblit, Murray, Gomes, Sandman, Taylor). But I'm looking forward to seeing your design. Still have those 7242 lying around

I see no problem using more than one tube in output. I dislike circlotron as idea. I have no experience of circlotron on tubes, but I once built a 25W circlotron on FET's, basically it's close enough to get the idea.

Hans Bejiner's 6C33C OTL amp I can really suggest. It has the sound I was searching. My construction is absolute linear inbetween 10Hz-40kHz (Hans said he got up to 200kHz, he uses different tubes and details). I'm using it with 90dB Audes Gloria 037 3,5 band speakers. 15W power per channel is suprisingly enough even with 90dB multi band speakers. The amp gives nice fast and precise basses, which OPT amp lacks of. And the basses are powerful and rich at same time. Very detailed but not tireing mids. I just discover my records again. The amp shows all mistakes of recording engineering, and in other hand it plays all the brilliance of sound engineer has put on record. Very crisp and clean high end. Very much air in whole sound, all instruments/interpreters locate very precicely on stereo panorame.
The amp needs a warm up time 15-20 minutes, because of the 6C33C tubes nature I described earlier. After this it opens it's capabilities fully and sounds so good I quit about even thinking of OPT amps. Let those be for guitar players.
You can see some images of my creation here.