|
Post by Mr Whippy on May 8, 2018 18:55:29 GMT
Aye. It can. I've been wondering how long Fane has been going. I should have a look. Our 50's Ferguson radiogram had an 8" 3 ohm one in. My brother's large 60's portable that I often had the back off, also had one in.
|
|
|
Post by MartinT on May 8, 2018 19:23:02 GMT
They go back to the 1950s.
|
|
|
Post by Mr Whippy on May 8, 2018 19:27:32 GMT
Oh. I'm having doubts now. The one in the radiogram might have been an Elac.
Yes. I think it was.
|
|
|
Post by steveeb on May 30, 2018 21:31:01 GMT
Steve, how did you find the comparison between your initial cabinet-on-the-floor and then with a bit of air beneath on the legs?
I wonder whether you had useful bass reinforcement from the floor versus driver nearer head height?
|
|
steve
Rank: Trio
Posts: 206
|
Post by steve on Apr 4, 2019 16:23:07 GMT
Steve, how did you find the comparison between your initial cabinet-on-the-floor and then with a bit of air beneath on the legs? I wonder whether you had useful bass reinforcement from the floor versus driver nearer head height? Nearly a year late but the bass with the legs on the cabs is fine.
|
|
steve
Rank: Trio
Posts: 206
|
Post by steve on Apr 4, 2019 16:27:04 GMT
A single valve OTL has now been added to the system. The NVA clone was fine but it was a bit too powerful in the end, when feeding a pair of speakers with 98dB/W efficiency. The OTL is 1WPC. It has a lovely synergy with the speakers.
|
|
|
Post by MartinT on Apr 4, 2019 19:37:57 GMT
You're brave going OTL. I remember even the Croft OTLs being a bit suspect for reliability.
Is it an established circuit or your own design?
Gorgeous finish, by the way.
|
|
steve
Rank: Trio
Posts: 206
|
Post by steve on Apr 4, 2019 21:22:51 GMT
It's my own design.
The design is basically a power cathode follower output stage. The output valve bias is held solid by a LM317HVT regulator chip configured as a constant current sink. The output stage is locked at 475mA by the sink, which is placed in the 13E1 cathode. Input and driver stage is a Russian 6H30pi twin triode, chosen for its superb linearity and ability to operate at (for valves) a relatively low plate voltage.
I claim no originality for the design as the power cathode follower output stage was proposed by John Broskie, who designed his around a 6C33C Russian triode. He drove his proposed circuit with an Aikido configured 6SN7 cascade as input and driver. I used a 13E1 as the output stage as I had a pair doing nothing.
The output stage is capacitor coupled, to keep DC out of the speakers.
|
|
|
Post by MartinT on Apr 5, 2019 8:16:57 GMT
Elegant. Well done for getting it working.
|
|
steve
Rank: Trio
Posts: 206
|
Post by steve on Apr 5, 2019 9:49:48 GMT
Part of the problem with OTLs is the totem-pole output stage. A conventional "Futterman" output transformerless amplifier has a sort of white cathode follower output, which means basically one tube configured as a cathode follower (output taken from the cathode where the cathode "follows" what is happening on the tube's grid) sitting on top of a conventionally configured tube outputting at its anode. The two signals are therefore out of phase so that one pushes and the other pulls, so we get a push pull output at the centre connection between the two tubes but without an output transformer being required. Balance is critical at that centre point because the thing is directly coupled to the speakers. Now transistor amps do this as a matter of course, but not with 340V between the rails (neg 170V and pos 170V) as is often the case with PL509 TV line output pentodes, or "sweep tubes" as the Americans call them.
If one half of the output stage fails then you end up with either +170V or -170V DC running through the speaker voice coils. Cap coupling is a problem with this system, because high value electrolytic caps would be needed, because of the filter it would create with the low impedance of the speaker at AC. If the 'wrong' side failed then the cap could explode, so bi-polar configured electrolytics would be essential. This arguably would cause more signal degradation than using an output transformer, defeating the whole object of the exercise.
So, without a cap on the output, some form of protection is needed. These range from simple HT fuses as used perfectly safely by Bruce Rozenblit to sophisticated monitoring systems, which watch for DC imbalance and shut down the HT in milliseconds if a fault is detected. Push-pull output OTLs are difficult to implement, but not impossible. Then there are the "circlotron" OTLs as manufactured by Atma Sphere, which are a whole other concept.
Single ended OTLs are easier, and the constant current sink in the output valve cathode goes a long way to making them friendly and easy to live with. The Transcendent Sound "Mini Beast" being a good example of a reliable single ended OTL. Though it uses chokes in the tail of its cathode follower output rather than solid state parts, the idea is the same. The Mini Beast uses three parallel connected EL509 tubes per channel as its power cathode follower and puts out 4 Watts. My single 13E1 and those using a single 6C33C are the ultimate minimalist expression of the single ended OTL concept and speakers below 97dB efficient need not apply. A single 13E1 will drive an 8 Ohm load but a single 6C33C needs a 15 Ohm load to make any useful noise.
|
|
|
Post by MartinT on Apr 5, 2019 11:53:48 GMT
Thanks for the explanation. I saw/heard a Futtermans once and couldn't remember the name.
When I ordered a Croft Series 4 Glenn rang me and tried to switch my order to an OTL. He had some issues with Audio T and didn't want them to supply me. I'm glad I didn't switch, the OTL would have been a poor match for the Cambridge R4 speakers I was running at the time.
The 4 (later upgraded to a 4S) gave me good service with my Croft SIP preamp.
|
|
steve
Rank: Trio
Posts: 206
|
Post by steve on Apr 5, 2019 12:34:54 GMT
Thanks for the explanation. I saw/heard a Futtermans once and couldn't remember the name..... The doomsday scenario I talked about with Futterman type OTLs blowing up speakers, is unlikely as output valves are parallelled up, sometimes as many as 18 per pos and neg bank and one tube failure is not going to be a catastrophic problem, nevertheless there would still be a DC offset at the speaker, which is undesirable whichever way you look at it considering how much a good speaker costs, so either fuses on each pair of tubes or DC detectors are still an essential precaution. Tube fails, turn off immediately is the drill. These days, push pull Futterman output stages have had all their problems ironed out by respected engineers such as Bruce Rozenblit of Transcendent Sound. His kits are good value for money and the ones I've heard sound beautiful. Nevertheless an OTL still needs a bit more investment in periodic maintenance and checking than a conventional valve amp or a solid state job.
|
|
|
Post by ChrisB on Apr 5, 2019 18:39:11 GMT
That's very nice Steve, thanks for showing it to us.
|
|
|
Post by MartinT on Apr 6, 2019 5:29:36 GMT
|
|
|
Post by steveeb on May 6, 2019 8:52:41 GMT
Hey Steve! Look what the lovely delivery man delivered! Managed to get them in the house while the wife was out, left one opened on the table. Reaction was "they're F-ing big speakers!" - understandable considering what she's used to looking at:
|
|
|
Post by MartinT on May 6, 2019 9:04:46 GMT
LOL! What's the plan for them, Steve?
|
|
|
Post by steveeb on May 6, 2019 9:39:47 GMT
Initially they'll go in a box to Steve's plan, get a feel for the size and sound and something to experiment with. But Steve's format would work for me though I have been imagining a low box on the floor, like a stage monitor. That would give off axis listening but maybe an unacceptable bass hump and unacceptable footprint. Maybe add an up firing tweeter sans Blackwoods (for the presentation, not HF extension). Another alternative train of thought is a stubby column with the driver mounted on an angled top panel.
|
|
steve
Rank: Trio
Posts: 206
|
Post by steve on May 17, 2019 21:57:16 GMT
Initially they'll go in a box to Steve's plan, get a feel for the size and sound and something to experiment with. But Steve's format would work for me though I have been imagining a low box on the floor, like a stage monitor. That would give off axis listening but maybe an unacceptable bass hump and unacceptable footprint. Maybe add an up firing tweeter sans Blackwoods (for the presentation, not HF extension). Another alternative train of thought is a stubby column with the driver mounted on an angled top panel. Look forward to seeing these getting built
|
|
|
Post by dubulup on May 26, 2019 15:23:27 GMT
Amazing story...valve guy goes SS to re-baseline and progresses back to his roots with simplicity this time; single tube OTL!!
Great journey!
Who here has heard an inverted OTL?
|
|
|
Post by MartinT on May 26, 2019 15:48:03 GMT
Can't say I have. What's the topology?
|
|