I have seen 6L6 datasheets with class AB2 design center values and maximum rating specifications but not for 6V6, the datasheets I have for 6V6 is Tung-Sol and General Electric.
Can the 6V6 be operated in either A2 or AB2 without any issues? The reason I ask is because I have plenty of 6V6GT and 6L6GB bottles but I only have the iron for 6V6. I have a 330-0-330 power transformer w/ 5V and 6.3V taps and push pull output transformers 8k/8ohm.
Can the 6V6 be operated in either A2 or AB2 without any issues? The reason I ask is because I have plenty of 6V6GT and 6L6GB bottles but I only have the iron for 6V6. I have a 330-0-330 power transformer w/ 5V and 6.3V taps and push pull output transformers 8k/8ohm.
The 6V6 data sheet does not explicitly indicate positive g1 current service. Therefore, don't try Class "x2" operating conditions.
I am of the other opinion. I will try A2 or AB2 on any tube. Some tubes are non responders, AB2 doesn't make any more power than AB1. Many sweep tubes are like this since they are already saturated before the grid goes positive. Some 45's particularly old well used tubes don't respond to AB2 even though AB2 IS specified in the data sheet because the filament is emission limited.
I have never blown a tube from A2 or AB2 operation, so try it.
Depending on the power rating on your OPT's and the current capability of your power transformer's 6.3 volt winding, you could try the 6L6GB's. They will work into 8K, they just won't deliver 50 watts. Triode connection for the 6L6GB may be another possibility.
I have never blown a tube from A2 or AB2 operation, so try it.
Depending on the power rating on your OPT's and the current capability of your power transformer's 6.3 volt winding, you could try the 6L6GB's. They will work into 8K, they just won't deliver 50 watts. Triode connection for the 6L6GB may be another possibility.
Hahaha it's like the devil on one shoulder saying "go ahead, try it" and the voice of reasoning on the other saying "don't try it" 🙂
I figured the 8k isn't optimal for the 6L6GB's but as tubelab points out I am concerned about the almost doubled current draw on the 6.3V tap, it's a pretty wimpy power transformer and gets pretty hot when used with 6V6's. I am not trying to get extra power by going with x2 operation, merely trying to avoid "blocking" and crossover distortion because the amp will be driven hard. I have some of Pete's A2 boards I wanted to try.
I figured the 8k isn't optimal for the 6L6GB's but as tubelab points out I am concerned about the almost doubled current draw on the 6.3V tap, it's a pretty wimpy power transformer and gets pretty hot when used with 6V6's. I am not trying to get extra power by going with x2 operation, merely trying to avoid "blocking" and crossover distortion because the amp will be driven hard. I have some of Pete's A2 boards I wanted to try.
Last edited:
You can use those A2 boards to prevent blocking distortion and not expose the control grids to damage. 😉 Instead of "routine" grid stoppers, put something substantial, say 47000 Ω, in those positions. I2R heating is what causes damage and BIG stoppers hold the g1 current down to something safe.
Member
Joined 2009
Paid Member
6V6's are common enough to burn a couple - why worry about a large grid stopper - put the pedal to the metal and see what it does for you !
The devils advocate 😉 I like it!!
I have some nice VT107A's from 1945, RCA and Sylvania. I don't want to damage them, even though they are more rugged than current production they are getting rarer and worth more $$$. I might just buy some JJ's and see how long hey last........if they hold up then the VT107A's should too.
All of JJ's recent Octal production is garbage. In current production, the EH variant is OK and the "reissue" TungSol is excellent. Poindexter of Musical Machine fame prefers the current production TungSol labeled tubes to NOS. 😉
OK, how about the voice of good engineering proposing some middle ground.....
I don't like huge grid stoppers for several reasons, and a large stopper is not needed here.
I am assuming that Pete's A2 board is a mosfet follower. I am assuming that you can connect the drain of the mosfet to it's own supply. I tend to use 100 to 150 volts for this supply.
Even though AB2 or even AB anything is not mentioned in the 1962 RCA HB3 data sheet for the 6V6GTA, grid current curves are supplied! It looks like driving the grid to +15 volts will draw about 10 mA of grid current for all plate voltages above 50 volts. Simply choose the stopper value to limit grid current to 10 mA and your grid dissipation will be 150 milliwatts. This is 15K for a 150 volt supply.
Eli,
Thanks for giving me the heads up on the JJ's not being any good. I will try the Tung-Sols out.
Tubelab,
Thank you for pointing out the grid current curves on the datasheets, looks like I won't need too large a grid stopper after all.
Thanks for giving me the heads up on the JJ's not being any good. I will try the Tung-Sols out.
Tubelab,
Thank you for pointing out the grid current curves on the datasheets, looks like I won't need too large a grid stopper after all.
!
I have a box full of old mostly used 6V6's and the breadboard from the 6L6GC in AB2 thread. I would have no problem finding a few crusty samples to "test", but the tubes and all my equipment is all packed up and waiting for my new house to be built. It will be mid next year before I can do any testing at least.
Note:
I vaguely remember wiring a 6V6GT into a TSE which uses PowerDrive to do A2 SE back when I ran all the 6AV5GA experiments. I don't remember the details but I did find a picture on my computer of a grey glass RCA tube wired into the TSE with clip leads. Chances are that was a 6V6GT. I think the only grey RCA's I have are 6V6GT's and 6SN7's.
If you are running the 6V6s as pentodes, you have to have a loadline that crosses above the knee of the Vg = 0 line in order to have any benefit from class *2 operation. If you don't, you get no benefit and you will just be heating up your screen and control grids for no reason since you are driving the tube to saturation.
If you run them as triodes, you will get increased power regardless of your load since positive grid voltage will always comes before saturation.
As long as you don't abuse the control grid (say by driving them deep into saturation with a continuous sine wave) I doubt you would hurt the 6V6s.
I'll post some additional info when I get time.
If you run them as triodes, you will get increased power regardless of your load since positive grid voltage will always comes before saturation.
As long as you don't abuse the control grid (say by driving them deep into saturation with a continuous sine wave) I doubt you would hurt the 6V6s.
I'll post some additional info when I get time.
Several years ago, when I knew considerably less about tubes, I did and AB2 experiment with triode-connected KT88s(EH new production). I drove them directly with mosfet source followers.
I decided to drive them with a sine wave input and see how low I could get the plates to swing on the signal peaks. (At this point, more knowledgeable readers have alarm bells going off in their heads.) Anyway, I noticed that as I cranked up the voltage I got a compressed sine wave. As I cranked it more, I got what looked like a square wave with little rounded domes on top.
Suddenly, my experiment was cut short by a voltage regulator going up in smoke. It was the regulator that supplied power to the mosfet drains. That was unexpected, I thought, since I had given a lot of thermal margin for that regulator's heatsink. I would later learn that driving the tubes the way I did, deep into saturation, was putting a lot of current into the control grids of those tubes. Later I would fire those tubes up again to find out that a couple of them (Yes, I did this to four brand-new tubes) could not be completely cut off, even with -150V or so on the grids. They would always let a few mA of plate current flow. Obviously, the grids sustained some damage.
I would go on to hit the books and learn more about what I had done, but when I reflect back on it I am surprised that I was able to drive them as long as I did without much worse damage to the tubes. Playing regular music, since it has a high peak to average ratio, would be much easier on the tubes.
From what I have learned, KT88 have a pretty fragile grid, made of fine wire and placed very close to the cathode to give the tube high gm. 6L6 and 6V6, being lower gm tubes should have a more robust grid, all other things being equal. (I would love to do some tube dissections to confirm this.)
So I wouldn't worry about making a class *2 amp with 6V6s but I would recommend caution in how you test it. Any grid stopper resistors will cut down on grid dissipation but will add distortion.
I decided to drive them with a sine wave input and see how low I could get the plates to swing on the signal peaks. (At this point, more knowledgeable readers have alarm bells going off in their heads.) Anyway, I noticed that as I cranked up the voltage I got a compressed sine wave. As I cranked it more, I got what looked like a square wave with little rounded domes on top.
Suddenly, my experiment was cut short by a voltage regulator going up in smoke. It was the regulator that supplied power to the mosfet drains. That was unexpected, I thought, since I had given a lot of thermal margin for that regulator's heatsink. I would later learn that driving the tubes the way I did, deep into saturation, was putting a lot of current into the control grids of those tubes. Later I would fire those tubes up again to find out that a couple of them (Yes, I did this to four brand-new tubes) could not be completely cut off, even with -150V or so on the grids. They would always let a few mA of plate current flow. Obviously, the grids sustained some damage.
I would go on to hit the books and learn more about what I had done, but when I reflect back on it I am surprised that I was able to drive them as long as I did without much worse damage to the tubes. Playing regular music, since it has a high peak to average ratio, would be much easier on the tubes.
From what I have learned, KT88 have a pretty fragile grid, made of fine wire and placed very close to the cathode to give the tube high gm. 6L6 and 6V6, being lower gm tubes should have a more robust grid, all other things being equal. (I would love to do some tube dissections to confirm this.)
So I wouldn't worry about making a class *2 amp with 6V6s but I would recommend caution in how you test it. Any grid stopper resistors will cut down on grid dissipation but will add distortion.
Also, in my reading, I learned that when control grids are made to dissipate lots of power, they are typically made with copper supports, non-gold plated wire, thicker wire, and radiator fins on the supports. Just something to look for if you have 6V6s from different manufacturers.
I read that gold plated wire is problematic because when the wire gets too hot, the gold will vaporize, contaminate the vacuum, and also contaminate the cathode coating, reducing emission.
I read that gold plated wire is problematic because when the wire gets too hot, the gold will vaporize, contaminate the vacuum, and also contaminate the cathode coating, reducing emission.
Thanks guys for all the input, I am confident about the 6V6's now being operated in x2 mode. But after finding the specifications for my power transformer, it can handle the extra current draw from the 6L6's. Now I just have to draw some load lines for the 8k load and the 6L6GB's and see what I can come up with. I have both RCA and GE black plate versions😎
There is an example in the data sheet with a 9K load, cathode bias, and a 360 volt supply. Start there.
Want an excellent, but somewhat complicated amp, look here. I had 6L6GA's in it, smoked glass RCA's from the 40's too, but I used a 6600 ohm OPT.
http://www.diyaudio.com/forums/tubes-valves/133034-6l6gc-ab2-amp.html?highlight=6l6gc+ab2
Last edited:
Using inspiration from George's 6L6GC AB2 thread and some new boards from Pete Millett, I started a thread a couple of weeks ago here:
http://www.diyaudio.com/forums/tube...riode-push-pull-driver-pcb-6l6gc-ab2-amp.html
No comments yet, but if anyone wants to weigh-in that will motivate me to continue the thread. There are many ways to implement a driver stage, I just happened to want to experiment with Compactrons. The little AB2 Driver Boards are sure convenient and could be used for alot of different scenarios.
http://www.diyaudio.com/forums/tube...riode-push-pull-driver-pcb-6l6gc-ab2-amp.html
No comments yet, but if anyone wants to weigh-in that will motivate me to continue the thread. There are many ways to implement a driver stage, I just happened to want to experiment with Compactrons. The little AB2 Driver Boards are sure convenient and could be used for alot of different scenarios.
Provided one uses new tubes will full emission, 6V6 will just work fine in A2 or AB2. I have all sorts of 6V6 (namely: Sylvania VT-107B coke bottle with grey coated glass, US Navy RCA's 6v6 GTY with coated glass and true micanol brown base, 6v6GTA Fivre with black coated bigger glass and Reflector 6V6GT with black coated glass) I haven't found any issue for HiFi use. No need of protecting the grid and the usual DC coupled CF (or MOSFET source follower is you prefer) without any limiting resistor worked best. Actually I think that A2 or AB2 are the best solutions to get some efficiency in triode connection. With a plate voltage of 300-320V one can achieve 2.5-3W in SE, twice as that in PP A2 (with much lower THD) and easily up to 10W in AB2.
Last edited:
- Status
- Not open for further replies.