6C33 Circlotron

[pat49hk]

Dear All,

I am going to build a 6C33 Circlotron OTL as schematic attached. I hope can get help to :

1) evaluate and give comment to modify the circuit;
2) a effective speaker protection circuit to apply;
3) a soft-start HT circuit to avoid pop noise with arround 2 minute start up run-in.

Thanks in advance.

Patrick

 

[tubetvr]

1) evaluate and give comment to modify the circuit;

I assume you are used to design SS based amplifiers? It looks un-necessary complicated to me and I am sure that you would reach a good result quicker with a less complicated design, e.g http://members.aol.com/aria3/index.html if you want a Circlotron or something like this if you can accept a non circlotron design http://www.tubetvr.com/otl.html

I dont like the Circlotron myself due to higher complexity, higher output impedance and DC level on both outputs without any advantages.

a effective speaker protection circuit to apply;

Not really necessary but if you want you can do a window comparator that measure the DC level on the output and switchoff the B+ if DC offset is too high

a soft-start HT circuit to avoid pop noise with arround 2 minute start up run-in.

Not necessary either if you switch on power stage heater + drive stage heater + drive stage B+ first, wait one minute and then switch on B+ to output stage.

Regards Hans

 

[pat49hk]

6C33 OTL

Dear Hans

Thanks for your response and kindly advise.

Yes, I am agree that the circuit is complicate a little bit and look like SS design as it is one of my goal to try and get the result for applying those concepts in a tube amplifier. (Somebody told me that it was good but another said that it sounds "dead-hard" even the measurement data are excellent. Please let me know if anybody has experiences about this.)

Your links enclosed are very useful. I had build a Futterman OTL with 7591 in triode connection long times ago and its ¡§sweet¡¨ sound still in my mind.

Can you explain in which view of point you are thinking that the speaker protection is not really necessary? It is so different with my own thinking.

The start-up procedure is right but I just want the auto proceed to make my wife comfortable.

Hope to hearing from you soon.

Regards,

Patrick

 

[tubetvr]

Hi,

My firm belief is that one of many reasons why tube amplifiers can give better sound than SS designs is that it is possible to make a design with fewer components. In early days of solid state designers tried to make amplifiers copying old tube circuits, it didn't work well as design criterias are very different for tubes and solid state and for the same reasons I don't believe in applying ss design to tubes.

Some examples: in your design the first diff stage is connected to a 2nd diff stage or what tube designers called a CF input GG output stage, what is the reason of this? Is it to not put un-necessary load on the diff stage? It is not necessary with tubes as the grid has all but infinite input impedance. I also note that you put constant current sources on all drive stages, is it because you want to maximise gain? The gain of a triode doesn't differ that much if the load is changed so you dont get so much by this, maybe 3dB, with transistors it is different as they behave more like a penthode where the gain is ~GM * Rl but this is not true for triodes. I would remove all CCS and get a much simpler design.

Re speaker protection, the most common failure modes of 6C33 is reportedly grid cathode short with a large increase in cathode current and what will happen is either that a fuse in the B+ circuit will blow or that the cathode link in the tube will blow. Either of these, (the fuse or the cathode link) will burn so quickly that a speaker doesn't have the time to get damaged so therefore you probably don't need a speaker protection. I have built several OTL amplifiers for my frends and some of them strongly requested a speaker protection so I made a simple one with a relay that switch off the B+ if the DC offset on the output exceed a certain value, it works very well when I test it but it has never been activated in practise.

The start up procedure I described cab be handle by a simple relay timer with turn on delay that switch on the B+ after a certain time, these can be bought ready made.

Regards Hans

 

[pat49hk]

Dear Hans,

First of all, I need to clarify that I am simple circuit pursuer also. But in another hand, I believing "what you pay and what you get". The only question is - It worth to pay or not? (Sometimes trouble is one sort of "get", right?)

Actually I had preparing two sets of driver board (one with-out CCS) for this experiment. I agreed the differences between SS and tube circuit, but I just want to know the possibility to gear-up the strength and weak of them. If you have any ideas, please let me know.

I still trend to insert speaker protection due to uncertainty. Your practices are good reference but never been activate does not means never happen.

Ready made seems a bit off-track of DIY; I wish to have my own design.

Regards,

Patrick

 

[tubetvr]

The only question is - It worth to pay or not? (Sometimes trouble is one sort of "get", right?)

Agreed but it usually some kind of "diminishing return on investment" and often complexity doesn't pay off at all.

I don't know how you came up with your design but if you think it is good then try it, when I am doing a new design I first sit down and think about what I want to achieve and then I design the circuit to fulfil those goals. This is how I work in my day to day work, (I am working with radio circuit design) and this is the way I have found is the most efficient. Of course it is so that I do a lot of experiments but I always get a good understanding on how something is supposed to work before I build it. My personal goal is to predict how something is going to work and then confirming it by building it, this I can achieve not always but often.

I will not give more comments to your design as we probably have completely different thinking and/or philosophy regarding eventual benefits of using solid state circuitry in a tube circuit.

I still trend to insert speaker protection due to uncertainty. Your practices are good reference but never been activate does not means never happen.

I understand that you are worried about this issue and I have given you 2 ideas, either use a quick blowing fuse in the B+ lines of the Power stage or design a circuit as I did using a window comparator checking DC on the output, both methods work well.

Ready made seems a bit off-track of DIY; I wish to have my own design.

I assume you mean the timer for start up? I used a ready made turn on delay relay in my first OTL but I have used an electronic timer for the others I built, It uses the same circuit board both for the speaker protection and for the turn on timer.

Regards Hans

 

[wiligruen]

Hi,
have you ever thought of using/adapting this design for 6C33?
http://www.vacuumstate.com/various/dpa300B_brochure_lo-rez.pdf

 

[tubetvr]

have you ever thought of using/adapting this design for 6C33?

I don't know if you directed the question to me or Patrick but the answer from me is no, I have stopped building transformer coupled tube amps and am concentrating on OTL's only. I have never found a transformer coupled amp that is as good as an OTL and then I have auditioned probably more than 50 different amps including SET, push-pull, hybrid and OTL.

Regards Hans

 

[kambule]

where you get the 6c33B ??

Patrick,

I live in HK as well. Where you get the 6c33B and the reliable sockets ??


kb

 

[pat49hk]

Dear Hans

I am glad to sharing your work experience and it is precious for my daily job too. (I was AM/FM Multi-band PCB designer but turned to Cordless Phone PCB design now.)

As I said that this is an experiment; the main goals is to comparing the differences and try to seek the ways to gear it up. I think this can only be finding out from practices and not concerning about any prediction and/or philosophy (May be the only prediction is to prove). I wish somebody could tell me which performance/specification need to analyze thoughtfully and how.

Yes, I am worrying of this issue. Can I have your schematic of what you told - window comparator - for reference? My own idea attached hereafter but I am considering to modifying it.

I think we got the same idea and please comment my preliminary board design.

Regards

Dear Wiligruen,

One of my friends had built a 6C33 class-AB amp with output transformer, it sounds good. But there are wide range of tube selection for transformer output amp not only 6C33 - a hard handling tube.

Regards

Dear Kambule,

My friends in Mainland bought it for me. I don't know the socket is reliable or not now, until I proceed the experiment.

Regards

 

[pat49hk]

Continuous

 

[kambule]

loudspeaker protection

Dear Patrick


Loudspeaker protection and soft starter, try Rod Elliot's page :

http://sound.westhost.com/project33.htm

I have copies of other designs but they are basically the same.


I have read a lot about the reliability of 6c33B and its socket on this forum. Be extremely careful.

kb

 

[tubetvr]

I have read a lot about the reliability of 6c33B and its socket on this forum. Be extremely careful.

Reliability is not a problem as long as the anode dissipation is 40W or lower, real Russian sockets are to be prefered over chinese as the latter are made of too soft materials and doesn't give enough contact pressure, it helps to clean the tube pins also before they are used first time as they are often badly oxidised.

I use real Russian sockets in the OTLs I use myself and the ones I have built for others, I never had a problem with reliability or any socket problems, My own OTL has been operating now for more then 5 years without adjustments or tube changes.

The speaker protection described on Rod Eliot's page is similar to my own but I use a LM393 comparator instead of a discrete design also I don't use a relay in the speaker line but instead switch off the power feed to the output tubes.

Regards Hans

 

[pat49hk]

Dear Hans,

I don't know why I just can put one attachment, so I combine the spk protection and board layout in a pdf file to send.

The sockets I bought seems strong enough, however, I will pay attention in this point.

Regards

Dear Kambule,

Thanks for your advice.

Regards

 

[kambule]

speaker protection cct

Patrick

Just look at the input of the cct (connection to output of amp), the NPN only detects excessive +ve, how about excessive -ve dc voltage ?? Unless the amplifier use only +ve supply, it needs another PNP transistor to detect -ve .

Study the Rod's cct. It is one of the best.

Hans has a very good suggestion : cutting the power supply. This way it eliminates a contact resistance to the speaker.


kb

 

[pat49hk]

Dear Kambule,

Sorry for late response.

The bridge could invert -ve to +ve and I just worrying about the sensitivity. Too high or too low and how to control it? I am considering to use OP as Han's suggestion. But I need to built another dual voltage supply.

Actually there is pwr-cut control to cut-off the HT which synchronise with the spk-cut action.

Regards

Patrick

 

[kevinkr]

I had problems with arc over in 6C33 during warm up in the circlotron otl I designed and built about 8 yrs ago. Its really important to delay B+ to the output tubes for a couple of minutes until the filaments are fully warmed up. I would recommend switching the primary side of the supply transformer to avoid issues with relays (contact resistance, need for extra relays, etc.) I would put one fuse in the output connections to the speaker and one in a supply lead - in the event that a tube fails dc current in the voice coil will blow that fuse and sustained fault current will blow the supply fuse as well. In the event the supply fuse fails the speaker fuse will open, hopefully before the voice coil is toast..

Incidentally I ruined an expensive woofer when an output tube failed. The internal connections in a 6C33 can take several 100 milliseconds to open under fault conditions and this can be enough to ruin some voice coils. In another case just the arc over was enough to destroy the voice coil in the woofer of a spica tc-50.

I would keep the design very simple, and remember that the voltage drive requirements for decent output power levels are surprisingly large. I found I needed more than 10dB of headroom in the driver in order to get the output power I wanted. Unfortunately I don't remember the details. Take a look at my website for the saga: http://www.kta-hifi.net/projects/amp_page/6c33amp/6c33.html who knows you might find it amusing.

Mine sounded pretty good without negative feedback incidentally which was a surprise to me at the time as I was firmly in that camp.

 

[Mark A. Gulbrandsen]

I have never found a transformer coupled amp that is as good as an OTL

Tubetvr,

Have you tried a BAT VK-60 or the new version of same? I had one once and a pair of M-60 MK2 otl's at the same time and the M-60's didn't even come close to the performance of the BAT. I dumped the M-60's and kept the BAT for well over two years. There is just something really good about their toroid OP transformers.... The BAT even gave my KSA-80B a run for the money in the dynamics department too. Its a pretty amazing amp that really lets the 6C33's shine through.

Mark

 

[tubetvr]

Have you tried a BAT VK-60 or the new version of same?

I have tried the old version but not the new one.

However when you mention M-60 I want to modify my statement slightly to: "I have never found a transformer coupled amp that is as good as a good OTL"

I don't consider the M-60 or any other of Atmasphere amplifiers as the best of OTL's, for one thing I think the output impedance is too high. What kind of speakers did you try the M-60 with?

The best Circlotron type OTL I heard is the Graaf GM-20 and the best OTL ever is a Japanese limited production unit made by San-Ei, (who where the first ones to use 6C33 in an OTL).

Regards Hans

 

[Mark A. Gulbrandsen]

I definately agree on the Atmospheres!! They need to add more output tubes to get the output impedence lower, also many mods to get them sounding their best and then they are still not really all that good. The only OTL I owned that was worth a darn and could give other tube or solid state ampe a run for their money was the Fourier Triomphe which I also owned for quite a while. The Fourier was very prone to eating tubes and popping fuses though but sounded great when it was working. Have you heard any of the Trancendent OTL's? I was driving Dyanudio Audience 9's at that time. Both the BAT and the Fourier drove them to decent levels.... The Atmospheres got lost in the speaker someplace..... The BAT VK-60 is also very easy on the tubez.

Mark