I use it in the DC Coupled OTL, but what I do differently than solid state amplifiers, through most of the audio band, is to start with a nearly constant open loop gain into load out to 20 khz, and then apply a fairly moderate amount of feedback (less than 30 db) across the whole audio range. I also have designed my Iron Lawbreaker speakers to have nearly a constant input impedance from 30 hz up to nearly 500 hz (a bit tricky with a BR design) which I believe with the OTL, results in amplifier distortion spectra that is less variable with frequency through the bass and midrange - this appears to significantly improve the sonic transparency of the amplifier. Having the OTL's damping factor be over 100 through the audio range assists here.
Most SS power amp feedback systems start cutting the open loop gain back at 6db/octave starting at some low audible frequency to avoid instability. This tends to improve LF distortion and damping, although I have noticed the the Crown K2 I have is actually more neutral in the bass than the solid state AB amplifiers using feedback I have. this issue is probably mitigated for non-switching amplifiers by operating the output stage farther into Class A. However, for the HF, both parameters are generally significantly inferior which I believe shows up in the IM distortion when reproducing more complex music.
Have fun criticizing this!
Last edited:
I had actually bought about 35 6LF6's for what I planned to build as two 'ultimate OTL' monoblocks capable of 150-200 watts each (I had bought enough 6LF6's to replace all the original tubes). However, I thought I would prototype my circuit ideas with a 'starter' circuit using 6AS7G's which were available in the late '80's for a buck or two (I also bought a 'lifetime supply' of about 100 6AS7G's in that price range later), in case I had output stage reliability issues during development. It turned out that I liked the 'starter' result so much that I eventually sold all the 6LF6's to Richardson Electronics for a fair profit.
Last edited:
Hi -
Here's a link, with it all f'ared up in its own f'arplace.
http://www.diyaudio.com/forums/attachments/tubes-valves/21746-photo-gallery-imag0006r4.jpg
It's actually about 60W/channel. All the aluminum-top caps are 70uf polypropylene types. Only the output stages uses 'lytics as filters in their supplies. The input and driver stages are hidden behind the big computer 'lytics in this photo. It's all balanced triode amplification , except for the PP triode output stage - this helps keep DC offset reasonable (<25mV) with the moderate amount of feedback I use. I built it in an old instrumentation chassis that I stripped down, except for a front extension I later added for the power toroid, etc. Due to the relatively large amount of filter capacitance, I have to use a delay power up sequence to avoid tripping circuit breakers. I actually ran this amplifier for about 10 years into Chartwell LS3/5 A's without problems, except once when a tape recorder preamp supply started motorboating at about 1hz which of course was passed through this amplifier and toasted a woofer VC before I could reach the volume control.
I later built two 125W monoblocks based on the same basic circuit using 12 6AS7G's for each and a 1400VA toroidal transformer in each which I wound the secondaries on, just like in the OTL pictured with a 700VA transformer, but both eventually ran into power supply issues probably because I was using some slightly underrated MOSFETs for the turn on circuit instead of relays. They worked beautifully - I could power both of these amps on at the same time not dipping the AC until one day an oversized AC transient came along - still plan to fix these two soon.....
Here's a link, with it all f'ared up in its own f'arplace.
http://www.diyaudio.com/forums/attachments/tubes-valves/21746-photo-gallery-imag0006r4.jpg
It's actually about 60W/channel. All the aluminum-top caps are 70uf polypropylene types. Only the output stages uses 'lytics as filters in their supplies. The input and driver stages are hidden behind the big computer 'lytics in this photo. It's all balanced triode amplification , except for the PP triode output stage - this helps keep DC offset reasonable (<25mV) with the moderate amount of feedback I use. I built it in an old instrumentation chassis that I stripped down, except for a front extension I later added for the power toroid, etc. Due to the relatively large amount of filter capacitance, I have to use a delay power up sequence to avoid tripping circuit breakers. I actually ran this amplifier for about 10 years into Chartwell LS3/5 A's without problems, except once when a tape recorder preamp supply started motorboating at about 1hz which of course was passed through this amplifier and toasted a woofer VC before I could reach the volume control.
I later built two 125W monoblocks based on the same basic circuit using 12 6AS7G's for each and a 1400VA toroidal transformer in each which I wound the secondaries on, just like in the OTL pictured with a 700VA transformer, but both eventually ran into power supply issues probably because I was using some slightly underrated MOSFETs for the turn on circuit instead of relays. They worked beautifully - I could power both of these amps on at the same time not dipping the AC until one day an oversized AC transient came along - still plan to fix these two soon.....
Last edited:
wow that's pretty bad ***. any pics of its next to a rule ror something that better reveals its size?
DAC? No DAC for the Linn, Denon MC and CJ preamp! Then you just need another Eicho magic eye FM tuner. OOPS, their DAC would be blown out too.
Better all stock up on those Russian KT-88s as they may get embargo'd as weapons material.
This is what all those survivalists stocking food and medical supplies don't understand. They should be spending their money on vacuum tubes so they can listen to the post-fallout emergency instructions.
I was born the same year the transistor was invented..........
One thing that I could never understands is, many tube enthusiasts say that they like the sound that tubes impart.
Why would any audio purist go for an altered sound ?
Having said that, there are many bad SS amps... that goes without saying and, there are bad tube amps as well..
There are also very good amps of both types of course..
Let's just make sure were are comparing apples and apples.
One thing that I could never understands is, many tube enthusiasts say that they like the sound that tubes impart.
Why would any audio purist go for an altered sound ?
Having said that, there are many bad SS amps... that goes without saying and, there are bad tube amps as well..
There are also very good amps of both types of course..
Let's just make sure were are comparing apples and apples.
Problem is, as revealed by the few tests ever made back in the days, many tubes will fail due to EMP too. And if tubes don't fail most passive components will fail anyway.
Better just store dozens of battery-powered transistor radios inside a faraday-shielded box if you REALLY care about this issue.
It's a good thing that tubes impart less sound than transistors. There's just something about having three nonlinear high-K junction capacitances per device that doesn't do sound justice.
One way to put it is that listening to tubes is analogous to looking throught a clear window with a perhaps a bit of optical distortion at one end, while listening to transistors is like looking through a window that has been dirtblasted and has random local distortions.
One way to put it is that listening to tubes is analogous to looking throught a clear window with a perhaps a bit of optical distortion at one end, while listening to transistors is like looking through a window that has been dirtblasted and has random local distortions.
Last edited:
The first DIY radios were solid state, and probably safe from EMP. I guess I'll be dusting off my crystal radio. Now where is my cat's whisker.
Last edited:
Err, dont think so .....
Please post if possible a bit of data to back up such claims, like your OTL at Zmin of 4 ohms , thd of bandwidth and output vs power, less look thru that looking glass or is it prism .. ...
Last edited:
If you'd switch the words 'tubes' and 'transistors' around I would agree. ;-)
Well, small signal the OTL is flat 1hz out to past 100khz @ -3db into 8 ohms. It'll make it out to 35-40 khz at full power into 8 Ohms. I already mentioned that it responds to DC (where I have it rolled off to unity gain). There's no specific load resistance where it will 'stop working'. For instance, with only 1 output tube a side instead of 8, it will still drive 8 ohms sounding decent until it clips - just at a much lower power. Might be a good way to save on output tubes, though if all you need is a watt or so.
Damping factor is 100 or higher across the audio bandwidth and it's 60 watts/channel into 8 ohms. I've tweaked damping factor to over 200 up to 20 khz in the past, but it has probably drifted down from there. IAC, few solid state amps can match that high frequency damping factor so their highs tend to lose impact and detail by comparison. Full power distortion is about 0.2% across the audio bandwidth. I could have made it much lower but didn't want to play SS games with feedback, as I have mentioned.
It's a fully balanced triode topology, except for the totem pole output stage so supply ripple is a nonfactor (also due to the size of the supply). And, as everybody knows, triodes have fewer high order distortion products compared to solid state devices or even pentodes, so there's less IM grunge to worry about.
IAC, it's much cleaner sounding than any SS amp I have. Just imagine - all the solid state dirtblasting from crappy junction capacitances is all missing.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.