| Re: New Production Tung Sol 6550, Penta KT88s [message #97845 is a reply to message #97825] |
Sun, 07 July 2024 20:26   |
positron
Messages: 113
Registered: May 2020
|
Viscount |
|
|
gofar99 wrote on Wed, 03 July 2024 11:09Hi, I suspect that there are a number of other factors involved as well. Class of operation (A,AB,U/L etc) actual load impedance and any possible output load side anomalies, stability of the B+ from no load to full load. Regardless there is considerable room for all of us to find tubes that sound and perform well in our gear. Good discussion.
Yes, there are some other possibilities. However, my designs are superior
designs, and well within parameters. For instance, my line voltage only
varies 1 volt, 2 volts maximum during all conditions.
Concerning Class of operation, for newbies.
Class A has maximum plate dissipation at idle (some add a few ma. at Max signal output). How close to maximum plate dissipation will affect gas
release by the heated plate. As the output wattage increases (louder
volume), that output wattage goes to the speaker, not in the form of heat
on the plate. Thus the output power is subtracted from the maximum plate
dissipation.
UL operation is similar to triode, unless the screen dissipation is exceeded, then possible catastrophic consequences.
Class AB normally has higher peak cathode current vs idle current, thus higher peak plate dissipation vs idle dissipation. The average cathode current, and plate dissipation is a consideration.
I would think the main worries are space charge depletion at peak cathode current and a higher average cathode current, which shortens life in general. The peak, and thus average is often determined by the output transformer. A 2k ohm to secondary X output transformer will have a higher peak and average cathode current vs a 5k ohm to secondary X transformer.
A poorly designed amplifier design could have oscillation problems,
low frequency and/or high frequency. This can cause high peak and average cathode currents, which would shorten tube life vs lighter running of the tube(s).
An oscillation often causes a rectangular waveform, thus extended high
cathode currents. Of course rectangular waveforms have very high distortion characteristics. Guitar amps often use distortion in expression of their music. But it can occur in high fidelity amplifiers if a designer is not careful.
Stability of the high voltage usually results from two areas, line voltage
variations and inferior DC filter section designs.
Anyway, I shall see if I can determine, via experiments, what is the main
limitation of the JJ E88CC small signal tubes, and the new Tung Sol output
Tubes under excellent operating conditions.
One last consideration is the lack of plate dissipation. It takes a quite hot temperature for getters to work. I have noticed over the decades that the top getters are not as efficient as side getters, which are near the hotter plate structure. That tells me that tubes without side getters could easily have more internal gas than tubes with side getters, thus faster poisoning of the cathode oxides.
Cheers
pos
|
|
|
|
|
|
| Re: New Production Tung Sol 6550, Penta KT88s [message #98039 is a reply to message #97973] |
Sun, 06 October 2024 22:36 |
positron
Messages: 113
Registered: May 2020
|
Viscount |
|
|
I performed a check between a used pair of new Tung Sol 6550 with
hundreds of hours on them vs a brand new pair of TS 6550s. All were
part of a matched octet.
I ran each output tube at 30 watt plate dissipation and
6.0 vac, no regulation. The digital measured line voltage has
always been 121/122 vac.
Comparing idle cathode current, the used tubes averaged 68 ma while the
brand new tubes averaged 69.3 ma. Pretty amazing since the side getters
were absorbing internal gas, ~1/3 milky.
Lately, I have dropped the filament voltage to ~5.75 vac to see the
results. So far, working nicely. Will keep informed.
cheers
pos
|
|
|
|
Posting Rules
Members
Search
Help
Register
Login
Get a T-Shirt!
