• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Availability of 6C33C tubes? I need some for OTL amps

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi

I have heard some worrying news about the availability of 6C33C tubes, some people say that they are not manufactured anymore. Does anyone know the facts?

I use them in my OTL amplifier and today I am building a couple of these for my friends, it would not be so good if the tubes are unavailable soon if they would be needed for re-tubing.

My OTL is here: http://www2.gol.com/users/tube/otl.html

Regards Hans
 
diyAudio Senior Member
Joined 2002
FROM INSIDE THE INDUSTRY.

Hi,

The 6C33-C has not been in production for the passed 4 years.

However,don't panic...prices have never been lower at current dealers and stock levels are still estimated at 750.000 units worldwide.

The lack of breakthrough for audio use is one factor for the current hold,the war thread in the East however is such that the factories keep all tooling ready in case war comes into play.

Nowadays even the US-Army uses these types in various applications.

Relax...if we can still unearth NOS tubes from the 40's to this day you can be rest assured the 6C33-C will still be around for the next 20 years at least.

Ineteresting your SRPP OTL.:cool:
 
Frank,

Thank you for this valuable information, that is exactly what I wanted to know, if stock is that much I can understand it shoudn't be a problem to source tubes for 2 or 3 amplifiers in the future.

I have used my OTL now for ~ 3 years and during this time it has been very stable and I have not needed any re-adjustment of bias. I have heard that many people experience problems with 6C33C but as I have understood it, it depends either due to bad sockets or that the tubes are not run in properly. I adjusted my sockets so that the connectors is pressing on the pins as much as possible and I also let the tubes run for one week with heaters only before switching on B+.

2 of my friends, one here in Japan and one back in Sweden really like this amplfier so I promised to build 2 more.

Regards Hans
 
diyAudio Senior Member
Joined 2002
ALLOW ME PLS.

Hi,

What kind of sockets have you used?

Icecold,the septar socket is not up to carry the heater currents.

What is the reasoning behind the 1 week burn-in?

This is a low mu,high transconductance valve operated at high current.
It will need weeks to stabilise and change even after that...John you should know your teletubes.

Moreover the septar will have a hard time carrying this kind of heater current,it will eventually quit its' job.

Cheers,;)
 
Hi,

The sockets I use I bought here in Japan from the same place from where I got the tubes. These sockets are not the same as ordinary SEPTAR types as they have an extra U-shaped spring that push the split connectors together, (I have used ordinary Septar sockets without these extra springs when I built RF amplifiers with YL1070 a long time ago). The sockets I use in my OTL I assume come from Russia as the tubes, the tubes where delivered in original boxes, (bad quality card board type).

The reason I did a one week burn in is that was recommended in the build instructions I got together with my 6C33C SE amp kit from SAN-ei, this amp was designed by Mr Dakesue who was the first one who imported 6C33C to Japan, (renamed EC33C)in the seventies.

As Frank says the 6C33C being a high transconductance tube is bound to show more drift than a lower transconductance tube like 300B, the distances between the catode and grid is smaller and therefore any mechanical tolerances or thermal drift, (where by these mechanical dimensions changes) will show up as a comparably larger drift in GM, (GM is inversly dependant on cathode to grid distance)

I measured some drift of bias current when I used my SE amp so therefore I use the burn in method also in my OTL and it seem to have worked well.

The tubes I use in my OTL is from Ulianov production, (they have Ulianov marking). I use 250mA bias current in my OTL so it works in class AB.

Regards Hans
 
diyAudio Senior Member
Joined 2002
6C33-C

Hi Hans,

Glad to meet another OTL fan.:cool:

I actually found your website a couple of weeks ago while I was trying to help another member with his Cuffioli OTL.

Since your amp seems to be inspired by the same topology I wondered if you care to solve something I have a lot of trouble understanding:

On his latest version he shows a single channel and three separate secondaries for the bias supply.
One is common for both channels and naturally if left this way adjusting one channel naturally changes both channels simultaneously.

What bothers me most is that biassupply connection C appears to be common to both channels.
How on earth can this work?

Why did he not use four independent bias supplies?

I would appreciate any clarification on this since Mr.Cuffioli has not yet replied to the e-mail sent by our fellow member.

Thanks,;)
 

Attachments

  • version 8.2_otl.gif
    version 8.2_otl.gif
    22.7 KB · Views: 943
Frank,

Just seen this and having a quick look....

As I read it C would be connected to the -160V rail of just one channel so the bias volts are 'referenced' to this point and each channels bias is tapped off via 68K resistors... I guess we are assuming A1 operation...

I think this will work but it means that the two channel will be biased differently by the amount of the of the difference in their -160V rails...not ideal.

If point C is common to both amps - then the -160V rails are tied together... should also work I think providing the last psu cap is a true ac short (this is a valve that is happy at vhf+ and we have common grid connection if this isn't true!)) ... but there are going to be some interesting circulation currents in the psu...and the failure modes would be doubly spectacular!

What have I missed?

James
 
diyAudio Senior Member
Joined 2002
Re:What have I missed?

Hi James,

This is precisely what baffles me.

I think the best solution would be to use four bias supplies.

If C is really common to both channels it would be degrading stereo separation to such an extend that it would be rather ridiculous,not to mention dangerous.

Connection of C to one channel only would work but the channels would keep on tracking each other,whatever you add to one would be detracted from the other.
Neither one would be biased correctly either.
All OTL amps I know off are all AB1 and from what I see so is this one.

If I were to build such beasts for my own use I would definitely opt for monobloc construction.

Thanks,;)
 
Time for Classes...

Frank said

All OTL amps I know off are all AB1 and from what I see so is this one.

Quite...:eek:

No grid current was what I meant and I got sloppy and didn't think or check it through...

Now a Class A1 OTL would be nice... And 600R speakers to connect it too would be even nicer :D

And I could turn the heating off in winter :) and get a bigger airco unit for summer :(

ciao

James
 
diyAudio Senior Member
Joined 2002
Re:Classes.

Hi,

Now a Class A1 OTL would be nice...

As you may know a Class AB1 OTL amp doesn't suffer from the same crossover distortion a typical OPT amp does.

From that standpoint opting for Class A would only reduce effeciency.

And 600R speakers to connect it too would be even nicer

Philips once had a lovely 800 ohm FR for their EL86 OTL amp,they used it for stereo sets and amps in radios.
Very good sound but reported to be rather fragile.

One of our members is studying this high impedance speaker as a possiblilty for a commercial product.

In the mean time stacked FR (seried) 16 Ohm speakers are the only practical and economical solution I'm afraid.

While I am at it a White CF could be used to make a nice SE OTL amp .
That would require a coupling cap at the output of course but with the available BG caps it certainly opens up perspectives.

Cheers,;)
 
Regarding point C in Andrea Ciuffoli's diagram I believe he intended to connect that to negative supply of only one channel. I tried that method earlier and found that there where problems to adjust bias for the 2 sides equally although it was possible to make it work also that way.

What I did later and still am using is to use one power supply for both channels with only one pair of capacitors. Altough this can seem strange it works very well also in theory given that the capacitors exhibit a low impedance in the audio range. If the positive and negative average supply currents are equal, which is true for normal audio signals, the connection points between the 2 capacitors will act as a virtual ground point and will not move at all and therefore it is possible to connect the 2 sides together without them interacting.

This setup is somewhat similar to what you use in microwave circuitry where it is quite common to work with virtual grounding, I have a background in that field and usually try to see similarities when building audio circuits, (it is only the frequency band that is different, the theory is the same).

In reality it works very well and there are no problem what so ever with stereo separation, I haven't measured it, (probably the only thing I haven't measured) but will do so in the near future, however there are no hearable leakage between the 2 channels and it shouldnt be.

Regarding safety and speaker protection I use ordinary separate fast fuses in the positive and negative supply lines, I am confident that they will blow before I destroy my speakers although I never had a tube failure (yet).

Regards Hans
 
diyAudio Senior Member
Joined 2002
6C33-C

Hi,


Hello, I'm very interesting by this amp but I have a pair of Proac 2.5, so I don't know if there isn' t mistmatch between this amplifier and this speakers.

If the impedance curve of the Proac is reasonably flat that should be not problem.

A pair of 6C33-Cs will give you about 20W into 8 Ohm.

If you need to fill a large room with music it may not be enough.

Have you had any stability problem?

If any problems occur it is usually within the first weeks of use.

Cheers,;)
 
As Frank writes a reasonable output power from a pair of 6C33C tubes in OTL is abot 20W at 8ohm, at 4 ohm it will be a bit less. Some companies claim up to 60W! from a pair of these tubes but I feel it is very unrealistic and I would expect they wouldnt last long at that power.

I get 25W in 8 ohm but as I use Lowther Fidelio horns the amplifier has a very easy life but I have tested with 25W in 8 ohm for something about 24 hours continously without any ill effects.

The proac 2.5 is a very insensitive speaker, (86 dB) and as such it can't play very load with this kind of power but as I understand it the impedance curve is not too bad.

With the amount of feedback I use the ouput impedance is 1.5 ohm which is better than most SE tube amps but less then solid state or PP tube amps.

The output impedance can be lowered by more feedback but then some changes to the circuit is necessary in order to make it stable.

Regards Hans
 
I have recently built this OTL, following closely the Ciuffoli schematic and I get about 22W at 8Ù with the 6C33's biased at ~200mA. My speakers are insensitive at 86db and the amp has a hard time to play loud in my 56m2 room.

I would suggest paralleled output tubes for >50Wrms but in this case a close matching would be necessary and this is not easy with high transconductance tubes such as the 6C33. A monoblock construction would help as, if nothing else, it will make the adjustment of the operating point of the two channels totally independent (by not sharing the same floating ground)

Other than that, the amp plays really beautiful within its limits and I guess Hans will have a great sound with the Lowthers

BTW I would really like to thank Hans and Frank for the valuable help and advice they provided in fine-tuning the amp :nod: :nod: :nod:
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.