I will vouch for this one. I've run my DC coupled OTL for 21 years now at +/-160-165 volts and by far the most common tube failure mode has been 6as7g arcover which has usually blown the internal tube and supply rail fuses for that channel. Not a horrible problem, but I've probably lost an aggregate of more than a dozen tubes (defined as having at least one section with an open internal fuse) to this failure mode compared to three or four of everything else. Also, when one or two 6as7g's have gone, I often don't note the failure right away, but the amp starts sounding progressively anemic as the open 6as7g fuses accumulate, usually asymmetrically on the B+ and B-. So, I've belatedly come to realize that it's a little silly to go for the last iota of power if the cost is declining sound and excess tube replacement over time.
Also, I've noticed that the Soviet tubes are more susceptible to this high voltage failure mode as compared to NOS GE 6AS7G's that I have. 6080's have been notably less robust than either of the above in this amp, so I would run them fairly gently at a moderate supply rail voltage if I used them in an OTL application, such as probably not going into positive grid voltage territory.
You should pre ageing NOS 6080 , especially those built in the late 80' ... I purchased Philips ECG labeled WC : they're the worst i 've ever seen without 24 hours pre heating : lots of dust on the plates , grids are "burning" on a bright white light when operated @ 140volts /110mA on my tester 
I'll take that into into thks.
Probably going to purchase a few GE 6080's this weekend (along with some 20 other tubes! ) from a local seller. Tubes are like gold dust here so have to seize the opportunity.
Otherwise I have to have them shipped from Europe which takes a millenium and costs an arm and leg in shipping.
Probably going to purchase a few GE 6080's this weekend (along with some 20 other tubes!
) from a local seller. Tubes are like gold dust here so have to seize the opportunity.Otherwise I have to have them shipped from Europe which takes a millenium and costs an arm and leg in shipping.
Going back to the original thread topic, I've considered using a load of 6CG7's as output tubes. I've got a little Futterman OTL headphone amp on my desk that uses 6N6P output tubes, and there's no reason why someone couldn't scale up the design for more output.
The 6080 is rather ideal for the job though...
The 6080 is rather ideal for the job though...
I use a great big 100uF film cap to decouple any DC to the output. This wouldn't be practical with a big Futterman power amp driving 8 ohm speakers of course.
Practically speaking you can use ANY audio tube in a Futterman style circuit. More tubes in parallel translates into less resistance as tubetvr mentioned. The 6AS7/6080 is a good choice due to it's already low output resistance. Less tube are needed to drive 8 ohms effectively.
For example I was thinking of the 6CG7 where they would be only good for a couple milliwatts of output, but let's say I had 10 tubes in parallel for one channel. That's 20 triodes in action which would bring down the output resistance, and add up to a couple of watts to the output. I didn't say it was practical, just possible.
You'll need at lot of 6N6P's, but they are cheap, and it is possible. I don't think it will be a trivial task though.
To be fair here's the headphone amp I use:
Bijou
Practically speaking you can use ANY audio tube in a Futterman style circuit. More tubes in parallel translates into less resistance as tubetvr mentioned. The 6AS7/6080 is a good choice due to it's already low output resistance. Less tube are needed to drive 8 ohms effectively.
For example I was thinking of the 6CG7 where they would be only good for a couple milliwatts of output, but let's say I had 10 tubes in parallel for one channel. That's 20 triodes in action which would bring down the output resistance, and add up to a couple of watts to the output. I didn't say it was practical, just possible.
You'll need at lot of 6N6P's, but they are cheap, and it is possible. I don't think it will be a trivial task though.
To be fair here's the headphone amp I use:
Bijou
Yes, it is possible and can be made practical if you take care of distributing the current between all tubes, this can be done by matching all tubes and by using cathode resistors for each individual tube to avoid "current hogging" i.e. where one tube takes to high current, thereby failing and thereby making other tubes take to high current and so on...
Even when individual cathode resistors are used the current for parallel connected tubes should be reduced, figures for de-rating where given for regulator tubes in data sheets, (6AS7, 6C33C) typically the current for each tube was only half when 4 or more where used in parallel.
It is my understanding that the reliability problems of most commercial OTL amplifiers are due to a combination of running tubes way above ratings or no or too low de-rating of tubes when parallel connected.
somewhere in the "pictures gallery " file , or here , from Wagtbtoobz, my other french id : [url]http://retro-forum.com/viewtopic.php?t=13883 [/URL] in french ... many pictures and schematic published Cheers
Pierre
Cheers Pierre
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.