|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
68.75.162.241
In Reply to: 6550 closest to GE. Current production only posted by gnugear on April 8, 2007 at 18:22:26:
I currently have the Penta/Shuguang KT88SC from Doug's Tubes (no affiliation) and they sound great. I am running Dynaco Mark 3 in triode mode and I have 500 volts for my B+. They look and feel like better quality when compared to my JJ KT88. The sound is open and airy without being harsh up top. I have also used the Svetlana KT88 (from year 2001 so now they are probably known as SED winged C) and they sounded good but I can not compare as I cooked them before I got the KT88SC. I happen to have a quad of the GE 6550A but I have not tried them as I was not wanted to use them up just yet. They came with my amp so I am not sure how much life they have left. Plus since I am running 500 volts, that is kind of hard on a 6550 tube but I have been told that they can take it so maybe I will give it a try and report back to you.
Follow Ups:
"Plus since I am running 500 volts, that is kind of hard on a 6550 tube..."I'm not sure where that idea came from. GE lists a plate maximum of 660 volts (design maximum), the RCA books lists 600 volts (design center), the Tung-Sol data sheet (old Tung-Sol, not reissue) is the same as the GE.
They do list lower screen voltage limits (unless you run ultralinear), but 500 volts is no problem for a decent 6550. And as I have written before, voltage (in and of iself) does not wear out a tube - current and dissipation are the culprits.
The funny thing is the GE's seem to be easier to come by than the SED's. And the GE's sound better overall. But that's all relative.
Good point. I forgot to say that the Mark 3 standard bias setting runs each KT88 with 70mA to get me to 35 Watts at idle per tube. From page 2 of the data sheet of the GE6550A, it lists the triode connection max plate voltage of 500V and does not list the screen voltage. Sometimes my plate voltage is over 500 since my line voltage varies from 120 to 124.xx Vac.
I was just afraid of the combination of all that being hard on the 6550 and actually the KT88 as well. But I do not know much about tubes so let me know if I am worrying for nothing.
"From page 2 of the data sheet of the GE6550A, it lists the triode connection max plate voltage of 500V and does not list the screen voltage."You need to look at the rating system used. The "design maximum" rating system is the middle of the three, they rank like this (from most conservative to least conservative):
1. Design-center
2. Design-maximum
3. Absolute-maximumDid you read what I said about voltage? It appears not... so let me write it one more time - voltage of and by itself is not hard on a tube! Unless the voltage is so high it causes interelectrode breakdown/leakage (or a big fat arc!) it is a non-factor. It is only a concern when combined with current to create heat the tube must dissipate. Remember, even at 500 volts if the current is zero the tube dissipates ZERO watts.
With the design-maximum at 500 volts you have no problem with your amp. And when triode connected the plate and screen are at identical voltages.
First off, 70mA is not zero mA. I run 35 Watts at idle, not zero.Where I got my data sheet does not list the absolute maximum just the design maximum, so I kind of lumped design max with absolute max.
That was the source of my problem. Now I understand.Do you have a better data sheet link? I got mine from triode electronics http://www.triodeel.com/tubedata.htm
All my data sheet knowledge has been in semiconductors. So my predisposition is that maximum is maximum. They always have one section very clear with absolute maximums.
So the fact that the design max and absolute max are different make sense but it just confused me since I did not have all three ratings in front of me.
Jim is just suggesting that you not get hung up on the specs. Those do not apply to most of today's offerings anyway.
I would go with the SED 6550C. I like its sound because it has the "crunch" I like on Robben Ford's recordings and some rock. You'll pay about $110 for a matched quad. I also have the Ruby/China (not sure of the maker) 6550ASTR which I have yet to audition. It looks very well made and the bottle seems tough. The Ruby brand is expensive too as they match for plate current and transconductance and offer a six month warranty on individual tubes.
Good luck in your hunt.
"First off, 70mA is not zero mA. I run 35 Watts at idle, not zero."I didn't say your amp ran at zero ma. I used that example to illustrate the point about voltage (with the caveats I listed) not being an issue.
I assume from the tone that you found that illustration offensive or condescending somehow. That was absolutely not the intent, but you are free to interpret it as you wish.
Maybe someone less offensive or condescending can help you from here. I'm done.
I really am not offended. I just don't like for you to keep quoting me out of context. The other other half that you leave out totally messes up the meaning. This started in my 2nd response and 3rd response.One thing that puzzled me was this exchange: in my 2nd post I also said:
"Good point. I forgot to say that the Mark 3 standard bias setting runs each KT88 with 70mA to get me to 35 Watts at idle per tube."
You still came back with:
"Did you read what I said about voltage? It appears not... so let me write it one more time - voltage of and by itself is not hard on a tube!"So I really started to think it was you that wasn't reading my post fully. No big deal, I just get kind of confused why you take it down that way.
I want to thank you for clearing up my misunderstanding about the design max versus absolute max. As I said, I did not even have the absolute max on my data sheet just the design max. That was the source of my problem. Not the understanding of the voltage versus power issue. So I guess I need to get tube data some books.
I spent much time at my last job using various high voltage discharge units to get parts to fail from arcing. Sometimes high voltage section of the circuit would arc to ground and reset the microcontroller. However...
I also spent equal amounts of time connecting just one more unit up to see of the master unit could handle just a few more mA of current before a resistor, circuit trace, transistor or diode would fail from overheating. From time to time components had to be made more robust for each case. So I have some understanding of the differences at least in the solid state world.
You could consider the design max as the standard operating point to use when maximum output power is desired.Solid state devices have very hard limits on voltages. Tubes are way different. Consider your 6550's in a push pull amp with an output transformer. Normal bias point is 500 volts on the plates and 75 milliamps which equals 35 watts. Now that 35 watts is the total current as it is what goes through the cathode. The design center for allowable wattage is 35 for the plate PLUS 6 for the screen so your operating point isn't all that tough (about 85% which is hot for a class AB amp but the plate voltage is supposed to be 475 on your amp and is high due to higher wall voltages today).
Now we get to the real difference in tubes. We have outlined the bias condition above but what happens when the music comes along? On a positive going waveform one tube will be driven into higher conduction, draw more current, and its plate voltage will fall. But at the same time its partner tube will draw less current and its plate voltage will rise. If the input signal is large enough it might drive that tube near cuttoff and that tube might now see a plate voltage near TWICE the inital 500VDC. The tube specs take that into account for output tubes intended to work into a transformer. But even some small signal tubes are designed to sit near cuttoff with plate voltages many times the tube's design center values.
Now try that with them semi-conductors:) It is also worth noting that in many cases amps were designed that ran tubes way over their rated voltage specs. Fender amps with 6V6's come to mind.
Bottom line, your GE 6550's will be fine in your Dynaco (but your heater voltage is likely high also and that is more of a problem). In fact they would normally last a shorter period of time with a lower plate voltage. That is because a lower plate voltage almost always means higher current and sometimes class A operation, both of which decrease life compared to class AB operation as used in your amp. This is so because the cathode coating (where the electrons come from) is only so thick and is depleted as a function of current not wattage (high wattage is a heat problem and can cause the internal parts of the tube to glow red and sag which could create internal shorts).
Perhaps you now better understand Jim's comments about current being the problem and not plate voltage.
Russ
P.S. Some of my examples above relate to pentode operation only and your amp is an ultra-linear amp. But still the concepts are the same. Google "TDSL", for tube data sheet locator.
Thanks. I will use my GE 6550A without a worry.Actually, I had already read some of Jim's posts in the past about current being the issue. I was still worried about the 500 Volt limit for triode mode was in the spec sheet, just did not know what I was reading or enough about the inner workings of the tubes. I was tripped up by the design max versus absolute max and the difference since my data sheet did not even list ab max numbers. So I am not worried about it any more. I do appreciate your explanation of how it works in the push pull amp.
Thanks for the tube data sheet suggestion too. I have a lot of reading ahead of me.
I came full circle with tubes when I got to try my dad's W-5M about 7 years ago. My first receiver, about 29 years ago, was a freebie Fisher 500C that my brother's friend did not want any more. Then a few years later I "upgraded" (not really an upgrade) to a Hafler DH-220, sounded like mono compared to the Fisher, but I was told that it sounded better. Being a youngster, I stopped using the Fisher, did not know any better. Leap forward to 7 years ago as my dad's W-5M and my newly acquired Mark 3 sounded more lifelike than my $2400 SS amp. So I sold the SS amp. Would like to try the HK Citation if I ever get the money.
This post is made possible by the generous support of people like you and our sponsors: