6C33 abs. max ratings for unmatched tubes and a operational life of 750hr
Without cathode resistor:
Tubes in paralell / Plate dissipation / Plate current / Peak
1 60,0W 0,600A 1,200A
2 42,5W 0,425A 0,850A
3 36,4W 0,364A 0,728A
4 33,3W 0,338A 0,676A
With 10 ohm cathode resistor:
Tubes in paralell / Plate dissipation / Plate current / Peak
1 60,0W 0,600A 1,200A
2 47,2W 0,473A 0,946A
3 42,8W 0,428A 0,856A
4 40,8W 0,410A 0,820A
As anyone can see you would need at least 3 tubes in paralell each with a 10ohm cathoderesistor (preferable 4 when now resistors are used )to get a healthy 750h lifespan.
My advice, run the 2 tubes in paralell OTL with a more healthy load of 16ohm or use a autotransformer if you have to use a 4 ohm load.
With 8 ohm a transformer will still pay off in the long run, alltough the tubes will survive the only now and then occuring beating.
Without cathode resistor:
Tubes in paralell / Plate dissipation / Plate current / Peak
1 60,0W 0,600A 1,200A
2 42,5W 0,425A 0,850A
3 36,4W 0,364A 0,728A
4 33,3W 0,338A 0,676A
With 10 ohm cathode resistor:
Tubes in paralell / Plate dissipation / Plate current / Peak
1 60,0W 0,600A 1,200A
2 47,2W 0,473A 0,946A
3 42,8W 0,428A 0,856A
4 40,8W 0,410A 0,820A
As anyone can see you would need at least 3 tubes in paralell each with a 10ohm cathoderesistor (preferable 4 when now resistors are used )to get a healthy 750h lifespan.
My advice, run the 2 tubes in paralell OTL with a more healthy load of 16ohm or use a autotransformer if you have to use a 4 ohm load.
With 8 ohm a transformer will still pay off in the long run, alltough the tubes will survive the only now and then occuring beating.
Last edited:
Probably a lot depends on what one means by "operational life." These tubes were intended for use in MIG fighter jets and the like, where they had to be able to sustain high G forces and not be the cause of a failure that led to the trashing of the jet. Rather more exacting than someone's home stereo system. They probably replaced them early rather than risk the whole jet plus pilot for the sake of a cheap tube.
Personally, I've been running OTLs using 6C33C output tubes for many years, and never had a tube failure. Others on this forum have probably pushed them harder even than I have. I read a posting by tubetvr where he left a 2-tube OTL running at 25W continuously for 24 hours (which would mean the 107W tube power dissipation I mentioned) and it came out OK afterwards. And as I mentioned, actual music, as opposed to a steady full-power sine wave, will have a much lower average power level.
Unless you have specific experience of 6C33C failures in OTL applications I would be inclined to go with the body of accumulated evidence that they are tough tubes that can take a fair bit of abuse.
Last edited:
You forgot something, if you want 100W/8ohm with 2 paralell tubes you have to increase the Ub by 20V otherwise those 2,5A just wont happen, and your losses go up to 125W, not down to 94W as you stated.
Furthermore, paralelling unmatched 6C33 means you have to derate!
Last edited:
No I didn't. If you get 25W into 8 ohms then it means 2.5 amps peak. I agree that means there is, at the peak, only 130V from anode to cathode, because of 20V dropped across the 8 ohm load. (In fact that was taken into account in the integral I did in order to calculate the 107W average power dissipation. i.e. P = 1/(2 Pi) integral_0^Pi (150 - 8 * 2.5 * Sin t ) 2.5 Sin t dt .)
In my experience one can achieve 25W into 8 ohms with a pair of 6C33C tubes. It is also what is reported by others. The question with paralleled tubes is what is the necessary anode to cathode voltage to be able to get 2.5 amps per tube; how much lower than 130V can you go. It's not obvious what the minimum voltage is.
It is true that with four output tubes you won't actually get quadruple the power of two tubes, which is why I said "roughly quadruple" the power. I agree that the extra voltage drop across the load is the reason why the calculation changes a bit for the quadruple tube case.
But these are relatively small effects. The main point I wanted to make was that stocktrader was way off when he spoke of "100s of watts of dissipation" in the tubes. 94W, 107W, 125W, it doesn't matter too much. We are talking roughly double the nominal 60W rating, give or take something. Experience suggests that is not a big deal for these tubes, and in any case, crucially, in realistic conditions listening to music, the average power dissipation will be much less than this.
Last edited:
Do you have documentation for this (i.e. that the given rating is not applicable to the high-G usages that the tube was designed for)? Of course, most people's stereo systems will presumably in any case just be sitting in the living room, not flying on a jet fighter!
But more crucially, do you have experience with using 6C33C tubes in OTL applications where you have encountered these kinds of lifetime limitations you are speaking of?
Last edited:
As I said, these are relatively small effects, and one cannot in practice expect to get exactly four times the power output by doubling the number of tubes, at least while keeping the HT voltages fixed. The main point is that one is talking of about 110 +/- a bit watts of tube dissipation at full continuous output power, which is really not that big a deal. Principally because music signals, even if listened to at a level where they are just reaching clipping at maximum power, will correspond to an average power level that is way below the steady power of a sine wave driven to clipping.
sure, with speakers of 8 or above ohms and decent efficiance, it will work. Not the best possible way but it will work, even a rozenblit design...
My point is, with a autotransformer and tertiary coil derivered nfb it would work even better in every other respect (except BW outside the range that is needed for audio)
My point is, with a autotransformer and tertiary coil derivered nfb it would work even better in every other respect (except BW outside the range that is needed for audio)
As for documentation what this tube was designed for you can find it russian data and tranlations on the internet from wich you can see that it was designed for use in serial regulated powersupply, Uhk 300V points to that also.
If used under extreme conditions of high acceleration (MIG-figthers) the following applies:
Environmental Conditions :
In the table, environmental conditions are specified which shall be survived by any sample valve taken from a production batch without suffering permanent degradation.
They do not, however, represent sensible operating conditions for maximum valve performance and service life.
Vibration (10-300 Hz) 6 g
Multiple impacts (duration ≤ 50ms) 150 g
Single impact (duration ≤ 50ms) 500 g
Constant acceleration 100 g
Operating temperatures (ambient) - 60° C .. +100°C
Relative humidity (at ambient θ= + 40°C) 98 %
In case you are interested in other "MIG-figther" tubes that can withstand multiple impacts of 500g look at for example the E130L (or many other rugged built tubes).
If used under extreme conditions of high acceleration (MIG-figthers) the following applies:
Environmental Conditions :
In the table, environmental conditions are specified which shall be survived by any sample valve taken from a production batch without suffering permanent degradation.
They do not, however, represent sensible operating conditions for maximum valve performance and service life.
Vibration (10-300 Hz) 6 g
Multiple impacts (duration ≤ 50ms) 150 g
Single impact (duration ≤ 50ms) 500 g
Constant acceleration 100 g
Operating temperatures (ambient) - 60° C .. +100°C
Relative humidity (at ambient θ= + 40°C) 98 %
In case you are interested in other "MIG-figther" tubes that can withstand multiple impacts of 500g look at for example the E130L (or many other rugged built tubes).
Last edited:
My limited experience or yours or others with only the experience of a rather limited range of samples does not count because non has the experience the manufacturer of this tubes offers.
Therefore I go with information published by the manufacturer.
Offcourse, almost any tube can take severe beatings.
If this is made a basis of a design it comes at a cost.
There are better ways
Last edited:
The manufacturers meaning of operational life can be found in the datasheet.
Typical operating data:
Filament current serial heater conn. –filament voltage 12,6 V 3,3 ± 0,3 A
Parallel heater connection – filament voltage 6,3 V 6,6 ± 0,6 A
Plate voltage 120V
Cathode resistor for cathode bias 35 Ohms
Warm-up time to steady state ≥600 s
Plate current 550 ± 80 mA
Rated Service Life under above operating conditions: ≥750h
Criteria for End of Service Life:
Reverse grid current≥15 uA
Plate current≤340 mA
Reduction of emission compared to new valve ≥30 %
Last edited:
They were used in power supplies with forced air cooling, but still servicing required routine replacement of their sockets.
If going for OTL, I would rather used a bunch of 6S19P tubes.
I agree that anecdotes alone don't provide a solid basis for judging the potential reliability of a design, but taken in combination with actual calculations and measurements, they can form a useful component in forming a general picture.
If we take the example of the 2-tube 25W OTL, then the calculations show that even if one runs the thing at full power on a sine wave, the average power dissipation of the tube is only 107 watts, which is less than twice the nominal maximum continuous power dissipation of 60 watts. By the way, the average plate current is about 0.8 amps for the case of the full-power sine wave.
But crucially, actual music is not like listening to a full-power sine wave, and once the crest factor that gives a measure of the average audio level versus the level at the loudest peaks is taken into account, one finds that the corresponding average power dissipation in the output tubes will be considerably less than 60 watts.
So in practice, listening to music, even loud music, the power dissipation in the output tubes will be well within bounds. And the empirical evidence from those who have been running 6C33C OTLs for years is that there is no particular problem with tube failures.
FYI...
I am running my current set of tubes since 2015.
Spec sheet also shows two scenarios for tube life. 750 hrs if driven at absolute max level of 60w (both triodes with plate current <250V. Rosenblitz T4 runs at 170V.) At 45W it is rated for 3000 hrs.
Also, given so much concern about my 110W claim I went back to see what the claims actually were. The patent mentions 110W with 4 tubes. Does not mention speaker assumption so likely assumed 16ohms to make it sound big. FWIW, a review at the time quoted 80w into 8ohms and 50W into 4ohms.
BTW, you can pull 2 tubes and run it on the remaining 2 dividing all these numbers by 4. The amp sounds fine with 2 tubes but with 4 it just has a more convincing grunt across the spectrum. I happily use 4 despite 2x tube cost.
For an OTL with low output impedance, (as for a SS amplifier) the output voltage level will be almost constant regardless of load impedance, i.e. power will increase in lower impedances as long as the maximum available power for each impedance is not exceeded.
An OTL as in the above example that can give 110W in 16 ohm as above will for a certain fixed input level that gives say 5W in 4 ohms give 2.5W in 8ohm and 1.25W in 16 ohm, (give or take some margin for the fact that output impedance is not 0 ohm).
An OTL as in the above example that can give 110W in 16 ohm as above will for a certain fixed input level that gives say 5W in 4 ohms give 2.5W in 8ohm and 1.25W in 16 ohm, (give or take some margin for the fact that output impedance is not 0 ohm).
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.