Hi,
I have changed the 5881 valves for a matched quad set of 6L6GC Russian NOS valves.
Since the change as well as the improvement in the sound, the output transformers are running a lot cooler. Why would that be?
I am also puzzled that despite having 5881 valves from new, the official Yaqin schematic shows 6L6GC valves which although pin compatible have very different power capabilities. Maybe the Chinese don't make 6L6GC valves and 5881 valves are cheaper?
I have changed the 5881 valves for a matched quad set of 6L6GC Russian NOS valves.
Since the change as well as the improvement in the sound, the output transformers are running a lot cooler. Why would that be?
I am also puzzled that despite having 5881 valves from new, the official Yaqin schematic shows 6L6GC valves which although pin compatible have very different power capabilities. Maybe the Chinese don't make 6L6GC valves and 5881 valves are cheaper?
on my mc5881A,the best result is the 6P3S-e quad. when I bought it (broken down) it was equipped with 6p3s. after repairing it and making some modifications, I tried several types of power tubes and input tube and phase shifter. best result with ECC85 / 6N1P e-v voskhod / 6p3S-e. there is a long thread on diyaudio on the corrections to bring to this amp. in my case, the main ones are a REAL adjustment of the individual cathode bias. a heating delay. a passage in pseudo triode the addition of an additional filtration stage the change of all the chemical as well as all the capacitors of connection by IOPs. I liked it very much and wanted to keep it until a friend made me an offer that I could not refuse and that I sold to him.
There is no such thing as a NOS Russian 6L6GC in my book. There are many different tube types and versions from both Russian and Chinese production. The problem is that you'll find these labeled as whatever Western customers know and desire, even when they are something else. For instance, there is a Russian tube that I have seen labeled as 6L6GC and with another brand as 6L6GB. In the '90s Groove Tubes sold them as uprated 6V6's as they didn't hold up to Western 6L6 power and voltage specs. This is why on this forum you often see the original designation: 6P3S.
This is supposed to be a copy of a 6L6GB. It has a rather large black base. The 6P3S-E is a different design, a lot tougher. Easily recognized by it's wafer (thin) base. Which one do you have? It surprises me the outputs transformers are running cooler now. They should not get warm at all. Are they very close to the valves and warmed up by their heat? It sounds like the new set runs at less current, thus dissipation. How are they biased? Fixed, adjustable, cathode resistor?
This is supposed to be a copy of a 6L6GB. It has a rather large black base. The 6P3S-E is a different design, a lot tougher. Easily recognized by it's wafer (thin) base. Which one do you have? It surprises me the outputs transformers are running cooler now. They should not get warm at all. Are they very close to the valves and warmed up by their heat? It sounds like the new set runs at less current, thus dissipation. How are they biased? Fixed, adjustable, cathode resistor?
The amp is cathode biased. The valves are just marked 6L6GC. This is a picture of the valves. The output transformers are at 90 degrees to, and either side of the mains transformer, and quite close to the output valves.
An externally hosted image should be here but it was not working when we last tested it.
Last edited:
I've never noticed any of my tube amps with the output transformers running hot. The power transformer will vary in heat output depending on the rectifier tube. Hot being related to how long I can leave the palm of my hand on the power transformer.
BTW- I have the same Ruskie tubes and NOS American made 5881 tubes. The 5881s sound louder. I can set the bias on each tube, no excessive heat noted, but I don't use a laser thermometer or FLIR, so I can't tell you how hot the power transformer is.
BTW- I have the same Ruskie tubes and NOS American made 5881 tubes. The 5881s sound louder. I can set the bias on each tube, no excessive heat noted, but I don't use a laser thermometer or FLIR, so I can't tell you how hot the power transformer is.
The rectifier is solid state on this amp.
An externally hosted image should be here but it was not working when we last tested it.
It looks like they are picking up heat from the output valves and the mains transformer, as they are all in close proximity.
An externally hosted image should be here but it was not working when we last tested it.
1. OK. Thermals are one thing. So are the possible differences when you operate those tubes in UL. 6L6GC (with special screens that are different than every other 6L6 type) and 5881 (same or similar screen as all the non 'GC' screens) may not respond the same with screen voltages higher than the plate voltages. That is what UL does, put more voltage on the screen than the plate. Pentode mode with screen dropping resistors often puts less voltage on the screen than on the plate. And then there are Pentode mode with regulated screen voltage which often puts the screen voltage much lower than the plate voltage.
Another question is, what are the voltages across the parallel 500 Ohm self bias resistors (250 Ohms)?
6L6GC? _______
VDC 5881? _______
VDC More current = more power = more heat.
2. As a completely different experiment: Why don't you get a second 100uF 100V, and individually self bias each output tube? You will be able to easily check the quiescent current balance of the 2 tubes. When you tie the cathodes together, you will never know if the current is balanced, unless you put a current sense resistor in each individual cathode. With a common cathode resistor, even "matched" tubes often draw different currents. Push Pull transformers do not work at their best with unbalanced currents.
Another question is, what are the voltages across the parallel 500 Ohm self bias resistors (250 Ohms)?
6L6GC? _______
VDC 5881? _______
VDC More current = more power = more heat.
2. As a completely different experiment: Why don't you get a second 100uF 100V, and individually self bias each output tube? You will be able to easily check the quiescent current balance of the 2 tubes. When you tie the cathodes together, you will never know if the current is balanced, unless you put a current sense resistor in each individual cathode. With a common cathode resistor, even "matched" tubes often draw different currents. Push Pull transformers do not work at their best with unbalanced currents.
Last edited:
You've found the issue. Have you considered a PC fan? I use one on a Chinese built Audio Nirvana PP amp.
Like this: Amazon.com: AC Infinity MULTIFAN S5, Quiet Dual 80mm USB Fan for Receiver DVR Playstation Xbox Computer Cabinet Cooling: Computers & Accessories
I place it right next to the amp near the output tubes so that the air is drawn away from the tubes, plug it into the wall wart. It's quiet and you can find them all over the inter webs.
BTW- the Audio Nirvana amp mentioned above has been used daily since I bought it in 2016. Same tubes that came with the amp.
Like this: Amazon.com: AC Infinity MULTIFAN S5, Quiet Dual 80mm USB Fan for Receiver DVR Playstation Xbox Computer Cabinet Cooling: Computers & Accessories
I place it right next to the amp near the output tubes so that the air is drawn away from the tubes, plug it into the wall wart. It's quiet and you can find them all over the inter webs.
BTW- the Audio Nirvana amp mentioned above has been used daily since I bought it in 2016. Same tubes that came with the amp.
Duke 58,
Louder.
Define louder:
If the Audio Nirvana has a fair amount of global feedback, then changing the output tube should not make the amp have any more Gain.
But different tubes could play louder before Clipping. Is that what you meant? When you set the bias for the different tubes, did you set for a specific quiescent current? With more bias voltage, the amp will swing further before clipping.
Louder.
Define louder:
If the Audio Nirvana has a fair amount of global feedback, then changing the output tube should not make the amp have any more Gain.
But different tubes could play louder before Clipping. Is that what you meant? When you set the bias for the different tubes, did you set for a specific quiescent current? With more bias voltage, the amp will swing further before clipping.
Last edited:
Once there was 6L6, it was good, popular.
Often used HARD. They broke.
Tung-Sol built a tougher "6L6" called 5881, for sockets where 6L6 were failing.
TV makers wanted "a 6L6" with higher ratings for sweep-work. They got a top-cap model for that, and a no-cap model of the same guts was offered to audio and servo designers. It got to be cheaper (because of TV production) than 6L6(mAB) or 5881.
The newer types were fudged to replace the older types without circuit changes.
Several "copies" of all these tubes were widely made, from the excellent KT66 to the low-price stuff for Russian civilian PA systems.
The US and Euro tube factories closed. Some of their tooling went to other countries, who already had similar tooling. Sources of known metallurgy dried up. There's been economic upset, and exploitation of the never-ending guitar amp tube market. I'm not sure WHAT is inside any specific bottle these days, except it "works" in classic tube (guitar) amps with those type-numbers on the sockets.
Which to use on any given day may depend what prices are offered by the various factories.
I'm sure any "5881" not made in/near Newark NJ before 1970 is a rebadged 6L6 frame, probably a 6L6GC or enlarged KT66 frame to give the robustness without Tung-Sol's excellent processing.
There always was much variation in 6L6/6L6GC, it is not a precision tube. Cathode bias reduces variability but not a lot.
I too would be curious to know the current. However you say it sounds better and runs cooler, so just enjoy.
Often used HARD. They broke.
Tung-Sol built a tougher "6L6" called 5881, for sockets where 6L6 were failing.
TV makers wanted "a 6L6" with higher ratings for sweep-work. They got a top-cap model for that, and a no-cap model of the same guts was offered to audio and servo designers. It got to be cheaper (because of TV production) than 6L6(mAB) or 5881.
The newer types were fudged to replace the older types without circuit changes.
Several "copies" of all these tubes were widely made, from the excellent KT66 to the low-price stuff for Russian civilian PA systems.
The US and Euro tube factories closed. Some of their tooling went to other countries, who already had similar tooling. Sources of known metallurgy dried up. There's been economic upset, and exploitation of the never-ending guitar amp tube market. I'm not sure WHAT is inside any specific bottle these days, except it "works" in classic tube (guitar) amps with those type-numbers on the sockets.
Which to use on any given day may depend what prices are offered by the various factories.
I'm sure any "5881" not made in/near Newark NJ before 1970 is a rebadged 6L6 frame, probably a 6L6GC or enlarged KT66 frame to give the robustness without Tung-Sol's excellent processing.
There always was much variation in 6L6/6L6GC, it is not a precision tube. Cathode bias reduces variability but not a lot.
I too would be curious to know the current. However you say it sounds better and runs cooler, so just enjoy.
5881 and 6L6 are not exactly the same. They have a bit different curves. With the same bias voltage as the result 5881 runs a bit cooler.
You can compare them side by side:
'6P3S (6L6GT) and 6P3S-E (5881)
You can compare them side by side:
'6P3S (6L6GT) and 6P3S-E (5881)
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.