No I´m rebuilding my PC atm so it´ll be awhile until I can do that.
I´ve set the 807 up for 6J51P again,but I`m wondering if:
How high should I go?
Running the 6J51P on just above 200v on the anode and 9mA.
I know you shouldn´t judge the sound only after a few records but I feel the 6J51P is better on the upper part and 6P15P are better on the bass part..
Just wish there was a way to comine the 2.
But I aint gonna do any more tweaking until I get my PC done so I can mesaure the frekvensy response....
I know you shouldn´t judge the sound only after a few records but I feel the 6J51P is better on the upper part and 6P15P are better on the bass part..
Just wish there was a way to comine the 2.
But I aint gonna do any more tweaking until I get my PC done so I can mesaure the frekvensy response....
Ake, the 16k would work well around 500V - around 30-35mA bias is optimal for 16K load.
8k around 500V would require twice the current, which well exceeds the plate dissipation, so that's out of the question at such "low" plate voltage. 8K load would work well around 800-870V in tetrode mode (with 300V g2), though... ; ).
Just installed the new Edcor OPT on the weekend into my 807, replacing the Hammond 125ESE (5K pri - 15W - 80mA, about half the Edcors size). The Edcors are 9.5lbs, 25W/200mA core, 10K pri - 6R sec (that is 13K reflected from 8R speakers). First impression was much improved imaging, especially in the depth; much lower extension; and a sense of quiet: much lowered sound floor. Also, more dynamic. Also take into account that I went from the PSE with Hammond to SE with the Edcor, that is my loading changed from 10K to 13K.
I can highly recommend the Edcor, was a huge step forward, but at 670Vp I'd rather recommend 10K load - 13K seems to be overkill, the drop in power is noticeable compared to 10K. Will see how it breaks in, my suspicion about lowering the load might be premature... I had less than 3 hours on it yet. Although at 670Vp 10K load had an optimum around 32-34mA current, and the 13K seems to prefer 22-25mA.
8k around 500V would require twice the current, which well exceeds the plate dissipation, so that's out of the question at such "low" plate voltage. 8K load would work well around 800-870V in tetrode mode (with 300V g2), though... ; ).
Just installed the new Edcor OPT on the weekend into my 807, replacing the Hammond 125ESE (5K pri - 15W - 80mA, about half the Edcors size). The Edcors are 9.5lbs, 25W/200mA core, 10K pri - 6R sec (that is 13K reflected from 8R speakers). First impression was much improved imaging, especially in the depth; much lower extension; and a sense of quiet: much lowered sound floor. Also, more dynamic. Also take into account that I went from the PSE with Hammond to SE with the Edcor, that is my loading changed from 10K to 13K.
I can highly recommend the Edcor, was a huge step forward, but at 670Vp I'd rather recommend 10K load - 13K seems to be overkill, the drop in power is noticeable compared to 10K. Will see how it breaks in, my suspicion about lowering the load might be premature... I had less than 3 hours on it yet. Although at 670Vp 10K load had an optimum around 32-34mA current, and the 13K seems to prefer 22-25mA.
Update on the Edcors (CXSE OPT 10K pri, 6R sec, 670Vp for the 807s, 21-26mA, tetrode mode):
Put well over 100hrs on the Edcors with the 96dB/8R Voigt pipes, and they did show continuous signs of break-in, but the base was very lacking compared to the top end and has not opened up much.
Switched in a pair of Mirage 760 speakers (4R, sensitivity maybe 89-90dB?, 30 years old 2 way bass reflex), which had plenty of base to boot with, but volume output was very low - moderate at best. Rebuilt the crossover with fresh new caps (MKPs instead of the dying old bipolars) and new wires. As the crossover broke in, the amp could drive it to earsplitting volumes without slightest hint of distortion or stress, with great base, transparency and imaging that I did not expect. With the fresh, modern caps in the crossover (and Kimber / silver internal wires), the old Canadian speakers became very serious, I have a hunch that they would give a tough run for modern speakers for their price ; ).
My first idea was that the Edcors match 4R much better than 8R - and I regretted not to have gone lower than 10K for the primary instead of going higher.
However, when I switched back to the pipes this weekend, I noticed that the base was there! With the sensitive speakers the Edcors were never going to break in (as they were never pushed hard). Needed to pump real power through the power hungry lower sensitivity speakers to break in! I guess that's what you get for running 200mA rated iron at 26mA: it just does not break in, unless you push it to the max.
An experience with the 807 idle current:
-At 26mA it sounds richer, warmer. The 3D stage is in front of the speakers, and more in the center. Yet, there's much less headroom.
-At 21mA it sounds leaner, the soundstage behind the speakers and much wider. Much more headroom.
Put well over 100hrs on the Edcors with the 96dB/8R Voigt pipes, and they did show continuous signs of break-in, but the base was very lacking compared to the top end and has not opened up much.
Switched in a pair of Mirage 760 speakers (4R, sensitivity maybe 89-90dB?, 30 years old 2 way bass reflex), which had plenty of base to boot with, but volume output was very low - moderate at best. Rebuilt the crossover with fresh new caps (MKPs instead of the dying old bipolars) and new wires. As the crossover broke in, the amp could drive it to earsplitting volumes without slightest hint of distortion or stress, with great base, transparency and imaging that I did not expect. With the fresh, modern caps in the crossover (and Kimber / silver internal wires), the old Canadian speakers became very serious, I have a hunch that they would give a tough run for modern speakers for their price ; ).
My first idea was that the Edcors match 4R much better than 8R - and I regretted not to have gone lower than 10K for the primary instead of going higher.
However, when I switched back to the pipes this weekend, I noticed that the base was there! With the sensitive speakers the Edcors were never going to break in (as they were never pushed hard). Needed to pump real power through the power hungry lower sensitivity speakers to break in! I guess that's what you get for running 200mA rated iron at 26mA: it just does not break in, unless you push it to the max.
An experience with the 807 idle current:
-At 26mA it sounds richer, warmer. The 3D stage is in front of the speakers, and more in the center. Yet, there's much less headroom.
-At 21mA it sounds leaner, the soundstage behind the speakers and much wider. Much more headroom.
A very good question. Burning-in or breaking-in does not have scientific explanation. However, many (myself included) observed that sound of audio components improves with use. It is some kind of black magic.
Many a musician will tell you that sound of musical instruments mellows with time, and instrument should be played in order for this mellowing to occur.
Isn't amplifier a music instrument?
Although there is no scientific explanation, strong belief exists, that in this case we deal with low level (hardly measurable) added noises, arising due to imperfections at various kinds of materials (wires, conducting films or foils, welded and soldered joints, tube plate and grid materials, etc.).
Burning-in is quite physically understandable effect, resulting from annealing of imperfections. I use burning-in procedure after any resoldering or parts replacement, with dummy resistors at amp output, and the procedure is much faster if output power higher. This is well described by the Annealing Activation Energy term.
Last edited:
Thanks for post your impressions
This lead me to some doubts:
What 807 are you using?
I unknew OPTs needs break-in, what break-in do in the transformer?
I suppose the core dont need any burn-in just the winding.
Thanks
Hi FullRangeMan,
The power transformer for the driver and screen high voltage has about 15 years of burn-in (about ?20K hours), the filament transformers have about 2-3000 hours, and the 700V high voltage transformer also about 2-3000 hours. I'd say they are well broken in, at least, all have well stabilized by now with the exception of the 700V transformer, which has x20 current capability compared to the demand, so will take probably a decade or longer to break in. As the idle current has been stable (thus load on PT has not changed), the changes observed are most probably due to the OPT break-in.
I had a discussion with a friend of mine who has vast experience with operating and designing tube gear and electronics (he held training courses for NASA, knows his stuff well.)
I also thought that it's the wire that needs break-in, but he said that the transformer lams also need break in - the BH loop changes with break in (it is measureable), especially in the lower frequencies. He said the bigger the iron, the bigger the change: do not expect significant base response from a big SE OPT if you are running it under a fraction of the rated idle current until it breaks in.
Burn in is a well studied phenomena at NASA, although I am not aware that much has trickled down to audio designers of that lore. For example, do you recall the NASA burn in CD? Luckily, it was available for purchase in the 90s for a brief period. That contains the signal they use to burn in the satellite transponders with before they start broadcasting. Before the burn-in, it is a fact that the satellites bandwidth is limited. (Don't know if they still do this, but has been common practice around 1990-2000. I remember finding a number of satellites with my dish with the colored lines showing, and this audio signal going on for months.) So, even solid state electronics is plagued by burn in - and of burn out as well.
I regularly run the burn-in CD, every couple months. If you have not done it in years, you will be shocked how much the imaging opens up and how more analogue-like it sounds after placing the burn-in CD on repeat for overnight. It's also great to burn-in amps and speakers, but will drive you mad - it sounds like a jumble of test signals. They are not random, they were designed to demagnetize the circuitry they are running through. Weird stuff, sounds weird, but developed by hard-core engineers driven by a critical market (communication satellites).
Thanks for post your impressions
What 807 are you using?
Thanks
Using RCA NOS; they had about 2-3000hrs of break in in the amp with the old OPTs.
I'm using the amp a lot, it is on for at least 4 hours a day, and often I forget to turn it off at night. Usually it's on for 3-4 days at a stretch, or even a week goes by without turning the amp off. I've been running the 807 for the past year at 60-70% plate dissipation. Before the major rebuild, I used to run them at 400Vp at 85-95% plate dissipation, tubes equipped with copper cooling fins, with a pair of 12V fans blowing on them - and that way the tubes were down to 95% of their initial value after 10 years, about 10K-20K hours or so. I put that pair of tubes to rest for a while, as I replaced them with a different pair... got to do some tube rolling once in a while, shame to have a box of 807s just to collect dust ; ).
I'd say when they have the copper cooling fins, they last forever when run at 80% dissipation... can't imagine how long they will last with 65%
Also, I used to bake them once a year. (There's a special heating procedure to activate the getter to eliminate gas.) Now I use demagnetization instead of baking, I suspect it works better than baking. (Demagnetize the tubes with a tape demagnetizer when the amp is on.)
In brief: keep the tubes cool and keep the vacuum tight, and a pair of 807s will outlast you ; ).
I'd say when they have the copper cooling fins, they last forever when run at 80% dissipation... can't imagine how long they will last with 65%
Also, I used to bake them once a year. (There's a special heating procedure to activate the getter to eliminate gas.) Now I use demagnetization instead of baking, I suspect it works better than baking. (Demagnetize the tubes with a tape demagnetizer when the amp is on.)
In brief: keep the tubes cool and keep the vacuum tight, and a pair of 807s will outlast you ; ).
Don't have any experience with those tubes, but I assume they will last a very long while if you keep plate dissipation at 80% max, and help with cooling - with either sticky aluminum tape or copper tape. (Plus, make sure the screens are kept strictly under 300V.)
I'll try to post a picture with what I do with my power tubes; )
I'll try to post a picture with what I do with my power tubes; )
From Post 107
"As long as plate voltages are kept in the 400V range UL should be OK. The strict 300V screen grid rating applies ONLY for tetrode mode. In triode or UL you can run it higher, provided that the screen is ALWAYS at a lower potential than the plate - which it always is in triode or UL mode."
Actually, I have measured the screen voltage to be higher than the plate voltage in my Ultra linear connected 807. There is a 343V screen and a 337V plate Voltage. This makes sense as the screen tap it closer to the B+. There is no screen stopper resistor. I will have to add one which should get the screen voltage under the plate voltage. Also in triode mode, wouldn't the voltages be the same for the screen and grid without a screen stopper resistor?
It was this thread mostly that convinced me to buy the Old buffalo amp. OldBuffalo 807 fu7 tube amplifier
Although from the only review I was expecting a 480 supply voltage, it is only 337 volts.
The sound is unbelievable clear compared to the mid priced transistor amps I have. Such as the NAD C320BEE. The depth and definition of reverb is astounding. And the bass definition is the best I have ever heard. Led zeppelin II was one of the first albums I ever bought and it is the first time I can hear every bass note super clear. But it feels lacking in dynamics and weight in the lower frequencies. I am sure this amp has plenty of room to tweak it up.
I will update on the progress. Next step is the screen stopper resistor.
"As long as plate voltages are kept in the 400V range UL should be OK. The strict 300V screen grid rating applies ONLY for tetrode mode. In triode or UL you can run it higher, provided that the screen is ALWAYS at a lower potential than the plate - which it always is in triode or UL mode."
Actually, I have measured the screen voltage to be higher than the plate voltage in my Ultra linear connected 807. There is a 343V screen and a 337V plate Voltage. This makes sense as the screen tap it closer to the B+. There is no screen stopper resistor. I will have to add one which should get the screen voltage under the plate voltage. Also in triode mode, wouldn't the voltages be the same for the screen and grid without a screen stopper resistor?
It was this thread mostly that convinced me to buy the Old buffalo amp. OldBuffalo 807 fu7 tube amplifier
Although from the only review I was expecting a 480 supply voltage, it is only 337 volts.
The sound is unbelievable clear compared to the mid priced transistor amps I have. Such as the NAD C320BEE. The depth and definition of reverb is astounding. And the bass definition is the best I have ever heard. Led zeppelin II was one of the first albums I ever bought and it is the first time I can hear every bass note super clear. But it feels lacking in dynamics and weight in the lower frequencies. I am sure this amp has plenty of room to tweak it up.
I will update on the progress. Next step is the screen stopper resistor.
Continuing from post 238.
The screen voltage was still higher than the plate voltage after installing the 100ohm stopper resistor on the screen. I did not feel comfortable with this and with further research found these tubes are not ultra linear friendly, so I converted the amp to triode mode.
It sounds much better now. In ultra linear mode the high frequencies were too emphasised and lacked dynamics. This did make listening to reverbs facinating, but the amp sounds much more relaxed and balanced now with a more present and deeper sound stage. I don't know why anyone could say a bad thing about single ended zero feedback tube amps.
There is not much to mod on this amp. Just some capacitor upgrades really, perhaps a larger choke. That's it. I will let you know when I do these upgrades. But at the moment it sounds very nice. There is no going back to transistors now.
One thing to note is the original Chinese tubes sound a bit fuzzy. I have installed NOS Sylvania and Russian tubes which have better focus and sound stage.
The screen voltage was still higher than the plate voltage after installing the 100ohm stopper resistor on the screen. I did not feel comfortable with this and with further research found these tubes are not ultra linear friendly, so I converted the amp to triode mode.
It sounds much better now. In ultra linear mode the high frequencies were too emphasised and lacked dynamics. This did make listening to reverbs facinating, but the amp sounds much more relaxed and balanced now with a more present and deeper sound stage. I don't know why anyone could say a bad thing about single ended zero feedback tube amps.
There is not much to mod on this amp. Just some capacitor upgrades really, perhaps a larger choke. That's it. I will let you know when I do these upgrades. But at the moment it sounds very nice. There is no going back to transistors now.
One thing to note is the original Chinese tubes sound a bit fuzzy. I have installed NOS Sylvania and Russian tubes which have better focus and sound stage.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.