I am now using Transcendar 3k opt's with new production 6l6wxt+. Using 680 ohm cathode resistor and B+ of 440v on the nose. Sounds fantastic in triode or ul, but might like to try old tubes. 6l6 has so many designations and I don't know what is safe to use. Can anyone help? I would like to not change the resistor all the time either.
While I'm at it does anyone know how to wire the Transcendar opt's for negative feedback?
While I'm at it does anyone know how to wire the Transcendar opt's for negative feedback?
Check out duncan amps TDSL data for 6L6 types.
TDSL Tube search [6L6]
The 6L6GC types can handle lots more power than the older, non -GC types, so you would need to adjust your bias current and/or B+ to run the older designations at less output power. The B+ voltage and bias adjustments may need to be reduced significantly......
To calculate power dissipation, multiply your B+ voltage by your bias current....then look at the max dissipation values for the various 6L6 types. For UL and pentode mode, the above gets you close, but neglects screen grid current.
TDSL Tube search [6L6]
The 6L6GC types can handle lots more power than the older, non -GC types, so you would need to adjust your bias current and/or B+ to run the older designations at less output power. The B+ voltage and bias adjustments may need to be reduced significantly......
To calculate power dissipation, multiply your B+ voltage by your bias current....then look at the max dissipation values for the various 6L6 types. For UL and pentode mode, the above gets you close, but neglects screen grid current.
440 minus the cathode voltage, to get the voltage across the tube - will probably be more like 425 -420 volts.
That kind of voltage should not be a problem for any true 6L6GC, 7581A, or KT66. The 6L6GC/7581A plate voltage rating is 500.
The lesser variants are rated at 360 and can take more. A 5Y3 or 5W4 might shed enough voltage to let him use those.
Win W5JAG
That kind of voltage should not be a problem for any true 6L6GC, 7581A, or KT66. The 6L6GC/7581A plate voltage rating is 500.
The lesser variants are rated at 360 and can take more. A 5Y3 or 5W4 might shed enough voltage to let him use those.
Win W5JAG
For what they are worth, these are my rules of thumb:
A 5W4 will drop about 55 more volts than a 5AR4;
A 5Y3 will drop about 50 more volts than a 5AR4;
A 5R4 will drop about 35-40 more volts than a 5AR4;
A GZ-37 will drop about 25-27 more volts than a 5AR4;
A 5T4 or 5U4 will drop about 23-25 more volts than a 5AR4;
A 5V4 will drop about 15-16 more volts than a 5AR4.
Each of these tubes has reduced current ratings compared to a 5AR4, and the amount of current the rectifier has to provide will affect the current drop, so YMMV.
Also, if you have not done so, the cathode voltage is subtracted from the voltage at the tube plate (anode) to determine the actual voltage across the tube.
Win W5JAG
edit: also, the Sovtek 5Y3GT is not a 5Y3GT. It's more like a reduced current 5V4. I haven't tried JJ's new 5Y3, but it looks like a real 5Y3.
A 5W4 will drop about 55 more volts than a 5AR4;
A 5Y3 will drop about 50 more volts than a 5AR4;
A 5R4 will drop about 35-40 more volts than a 5AR4;
A GZ-37 will drop about 25-27 more volts than a 5AR4;
A 5T4 or 5U4 will drop about 23-25 more volts than a 5AR4;
A 5V4 will drop about 15-16 more volts than a 5AR4.
Each of these tubes has reduced current ratings compared to a 5AR4, and the amount of current the rectifier has to provide will affect the current drop, so YMMV.
Also, if you have not done so, the cathode voltage is subtracted from the voltage at the tube plate (anode) to determine the actual voltage across the tube.
Win W5JAG
edit: also, the Sovtek 5Y3GT is not a 5Y3GT. It's more like a reduced current 5V4. I haven't tried JJ's new 5Y3, but it looks like a real 5Y3.
Last edited:
So by changing the rectifier tube I can use regular 6l6's and g's. And not be stuck with gc's only?
Can anyone tell me roughly where I stand for voltage across the tube with 5ar4 , B+ at 440v, using a 680
Cathode resistor in the standard sse layout? If not, I'll pull it out of the cabinet and take measurements.
Can anyone tell me roughly where I stand for voltage across the tube with 5ar4 , B+ at 440v, using a 680
Cathode resistor in the standard sse layout? If not, I'll pull it out of the cabinet and take measurements.
Pull it out, take a few measurements, then model what you have in PSUDII, it's very straightforward. There are plenty of folks here to help you. Once you get the model close to reality, then change the rectifier type and current draw in the PSUD model to your heart's desire and see what B+ you end up with.........
If you need help with the PSUD model just ask.......
If you know your power transformer specs (or at least close) and the current draw of your power tubes (measure across the cathode R and use ohm's law), you're half done with your model.
Georges tables on the Tubelab "tubes and transformers" page can give you lots of direction...study that and keep in mind that the power out values there are for triode mode.
If you need help with the PSUD model just ask.......
If you know your power transformer specs (or at least close) and the current draw of your power tubes (measure across the cathode R and use ohm's law), you're half done with your model.
Georges tables on the Tubelab "tubes and transformers" page can give you lots of direction...study that and keep in mind that the power out values there are for triode mode.
OK, but you might not like what I have to say. My SSE has the same Allied 6K7VG. The B+ is 430 volts with KT88's biased up to 100 mA each! It goes over 440 volts when I drop the current back to 65 mA each. Granted my wall outlet never goes below 120 volts and has been seen over 126 volts, but this is too hot for even some 6L6GC's.
The inrush surge with this transformer is tough on rectifier tubes. Some 5Y3's and 5R4's will spark out. Some poorly built new production 5AR4's will spark out on initial power up. Putting a 1N4007 in series with each red transformer lead will help prevent this. It will be implemented in the next PC board revision. Tube purists can always install a jumper or a resistor.
To run 6L6GC's in my amp I need a cathode resistor in the 620 to 680 ohm range. Even with a 5Y3 rectifier and a 750 ohm cathode resistor the voltage across the tube will be over 360 volts which is the screen grid maximum for a 6L6GA. Did that stop me from stuffing them into the amp? NO! Did any blow up? YES! Were they already bad? Maybe.
I found a box full of used and dirty 6L6GA's in my warehouse. I plugged in a 5Y3 and ran several of these tubes through the amp. Some worked, some didn't and two sparked out. the second sparking 6L6GA took out the 5Y3, which was also rather old and crusty looking. That was two weeks ago and I haven't even replaced the blown fuse yet. If I have some time tomorrow I might get over to the warehouse to grab all the tubes. I plan to test them and sell some of the good ones on Ebay. Most are smokey glass RCA's.
I believe that the dissipation was too high since I could see a faint glow through the smoked galss on some tubes. Maybe an 820 ohm cathode resistor would be needed. I have 750 ohms in my amp and I have a collection of resistors with small alligator clips that I use for boosting the current by clipping them across the 750 ohm. Some users have a switch.
A 3K ohm load is a bit low for a 6L6GC running at over 400 volts. It will make more power but the distortion will be higher and the damping factor will be lower. I have Transcendar 3K OPT's in that amp. It all depends on your speakers, but I get better sound with my "8 ohm" speakers plugged into the 4 ohm jacks. This puts a 6K load on the tubes. It's not quite as loud, but the bass is cleaner and the vocals are much more detailed. The only time I use the 8 ohm taps is with KT88's in UL with cathode feedback, the 15 inch speakers and something loud (Pink Floyd...) in the player.
How did you wire the transcendar's for nfb? Straight or reverse like on your site? I'm getting good bass out of 6l6gc's in triode, 3k, 4ohm tap, and 4ohm rated 15inch efficient woofers. Tight and clean. Loud as heck. I can't listen past 1/3 of volume knob with cd's. Too loud. With my low gain phono stage I can get it to clip, but I am way up on the volume. I am VERY happy as is, but I constantly tube role that's why I asked about the other 6l6 types. Did I make a huge mistake using the 3k opt's? I think I might have limited myself a bit, but not a huge mistake.
I have the inrush limiters in place and watched the thing power up slow and steady with a voltmeter. I even switch the 5ar4 to a 5u4 and no inrush problems as long as I start when cool.
I have the inrush limiters in place and watched the thing power up slow and steady with a voltmeter. I even switch the 5ar4 to a 5u4 and no inrush problems as long as I start when cool.
Possibly, but, as George has noted, the plate and screen rating for all of the lesser 6L6 variants is 360 volts, and you will wind up around 385-375 volts, maybe a bit more, across the tube with a 5Y3 or 5W4. It will also be right at, or over, the output current rating for the rectifier tube. The cathode resistance for the power tubes will likely need to be increased.
The late production 6L6WGB's should easily handle these voltages, they have the guts of a late production 6L6GC in them.
Personally, I have run RCA 6L6G's, ( real ) Tung Sol 5881's, older 6L6WGB's, 5932's, and even 807's at 375-380 volts across the tube and have not lost one, but I accept the risk that it can happen and that expensive amplifier damage can also occur when going outside the ratings.
I have some real nice 6L6GA's and 6L6GAY's, but I have not run them past the ratings because I am unwilling to risk frying them.
There are also a few threads on using a choke input power supply. IIRC, Ty Bower used an LCLC setup to run 6L6GA's in his SSE. That requires a destructive modification to the pcb, however.
Win W5JAG
The mod to the circuit board wasn't so bad. It only involves cutting one trace, which can be jumpered back together without too much trouble. Here's my post where I showed the results:
http://www.diyaudio.com/forums/tubelab/125439-another-simple-se-builder-3.html#post2377321
6l6
I like the lclc idea. I'm told that makes a better supply. For now I'll try messing around with 6l6gc's and tubes that are close like 350b. BTW I took measurements today. 440v B+, 435v plate voltage, 33v across cathode resistor. Measurements were taken with 5ar4, 6201, and two el34 after warm-up. 680 ohm cathode resistor. Don't forget 3k opt. Based on these measurements, I think I can drop the resistor to 560 ohm.
I like the lclc idea. I'm told that makes a better supply. For now I'll try messing around with 6l6gc's and tubes that are close like 350b. BTW I took measurements today. 440v B+, 435v plate voltage, 33v across cathode resistor. Measurements were taken with 5ar4, 6201, and two el34 after warm-up. 680 ohm cathode resistor. Don't forget 3k opt. Based on these measurements, I think I can drop the resistor to 560 ohm.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.