Remember These?

[Cressy Snr]

There may very well be gains from one of those KK cathode sinks, but I think settling with the TL783 for now, might be the best plan. If we plan some room in our respective chassis' for the K&K cathode sinks, they can be incorporated later on.

[Cressy Snr]

Right!
A couple of mods to the breadboard this afternoon:

First one was a duplicate power tube supply. So now each 13E1 has its own dedicated arrangement.
The supplies are basically half a Mini Beast per tube, but of course the HT is higher. Each one is 2500uF-choke-2500uF. The chokes are the 2 Amp Triad Magnetics coils, as used in the Rozenblit amp; no point re-inventing the wheel. The HT is switched for both supplies, via the 250V, DPDT, 10A switch, mounted on the black acrylic plate at the front.

In the back left of the picture can be seen an EZ81 tube rectifier. I put this in as the best way at the moment to give a slow start to the input/driver stage power supply. There was just enough room on the board to put a small 6V transformer to supply the EZ81 heater. The rectifier is configured as a hybrid bridge. I used an electrical switch back box I had spare from the room renovation. knock out one of the apertures and the resulting hole is the right size for the B9 socket the EZ81 uses. It has knocked 18V off the HT to the input and driver stages but that's of no consequence in the grand scheme of things.

As before, I am only checking the voltages out at the 'mo. Any evaluation of sound will have to wait until later. Interestingly the amp is now dead quiet.

[Cressy Snr]

OK,
As I hinted earlier, the bass was OK..ish, after I'd changed the input/driver stage power supply, but there was a tendency towards unevenness on the bottom end; which was particularly noticeable on slow, walking acoustic basslines on Capitol recordings of Nat King Cole, Frank Sinatra and Dean Martin with various orchestras. This was a tad disappointing, given the excellence of the rest of the frequency range. OK, I know I'm talking hi-fi here rather than music, but sometimes this kind of thinking is needed, when building and testing equipment. It was obvious that the output impedance was a little high and was affecting the FR of the Fanes at one point in the frequency range, putting a peak in the bass that corresponded with the deliberate hump that enabled close-to-wall siting, so it became clear that a tweak to the feedback was needed.

To cut a long story short, the feedback resistor is now 3K3 as opposed to the 4K2 it had been. The result of the increased feedback is excellent and has both evened out, and tightened the bass up very nicely. I decided to use ears rather than scope and sig gen to fine tune the feedback to my own tastes and for the benefit of the Fanes.

With the tweaked feedback and the individually powered output tubes, the degree of ease and authority has increased. This was not unexpected, as exactly the same kinds of improvement happened with the change from my NVA A20 to the upgraded power supply in the NVA clone I built. This usually happens as power supplies get better; you are able to hear more easily to the back of the recording, and at the same time, an increase in the solidity of the soundstage manifests itself. These improvements duly reported for duty with the OTL. We're really cooking with gas now.

That's it I think now, for the tweakery.

[The Stratmangler]

OK, I know I'm talking hi-fi here rather than music .....

Where does one start, and the other take over?
They're one and the same, surely?

Hi-Fi is allegedly a more accurate rendition, but they're both about music ....

[Wolfgang]

To cut a long story short, the feedback resistor is now 3K3 as opposed to the 4K2 it had been. The result of the increased feedback is excellent and has both evened out, and tightened the bass up very nicely. I decided to use ears rather than scope and sig gen to fine tune the feedback to my own tastes and for the benefit of the Fanes.

I wonder if/how this applies to your situation but the amount of NFB and AC or DC coupling made some dramatic difference in combination with certain speakers in case of the 300B SE OTL. The problems here were motorboating and loosing control over the bass at 30-40Hz. Originally it was AC coupled with less capacity for the output caps (3k3uF) and half the NFB and it sounded better like this IMO. But BR connected it DC , doubled the capacity and the NFB (5k1 to 2k2) and could get rid off these problems. My point is: there might be a combination of these three factors that works best for certain speakers and NFB is only one variable. There is a lot of room for tuning with the ears.

Where does one start, and the other take over?
They're one and the same, surely?

Hi-Fi is allegedly a more accurate rendition, but they're both about music ....

To listen to these kind of OTLs in combination with high sensitive single driver speakers is like relearning how good music reproduction at home can sound. It must be experienced in order to feel the difference. But Steve's description is spot on. To be in the studio with the musicians or to sit right in front of a small ensemble (Jazz/Classical music) in a pub, jazz club, or concert hall is how one feels.Instruments sound like "live" as if they were played right in front of you but especially human voices do not only sound "live" but truly reach out to you and touch you almost physically rather than coming out of speakers. With some female singers this effect is almost shocking, as spine tingling as if she starts walking right up to you while she sings.The only other way to experience this is to go to a live concert and have a seat in the first 3 rows. Hearing is believing.

[Ray P]

There may very well be gains from one of those KK cathode sinks, but I think settling with the TL783 for now, might be the best plan. If we plan some room in our respective chassis' for the K&K cathode sinks, they can be incorporated later on.

Yes, I agree Steve. I will build with TL783s first. I've asked Kevin at K&K if he can provide a mechanical drawing of his CCS board so any required mounting holes are pre-drilled/tapped to make subsequent experimentation easier.

[Cressy Snr]

I wonder if/how this applies to your situation but the amount of NFB and AC or DC coupling made some dramatic difference in combination with certain speakers in case of the 300B SE OTL. The problems here were motorboating and loosing control over the bass at 30-40Hz. Originally it was AC coupled with less capacity for the output caps (3k3uF) and half the NFB and it sounded better like this IMO. But BR connected it DC , doubled the capacity and the NFB (5k1 to 2k2) and could get rid off these problems. My point is: there might be a combination of these three factors that works best for certain speakers and NFB is only one variable. There is a lot of room for tuning with the ears.

Yes, ears and measuring equipment are obviously the best way to go. Feedback is one of the most important things with OTLs, especially of the single tube breed I've found; and even more so, when you want to drive 8 Ohm speakers with just one tube. The difference it makes can be make or break; something you really have to get to grips with. If you don't want to do that then don't build an OTL; especially not a single tube one.

I'm not even going to attempt DC coupling, with my amp. I suppose it could be done in theory, by standing the 'out' terminal of the CCS on a negative voltage, then adjusting that voltage until zero appears at the cathode, but I'm not that way inclined, at least not yet.

To listen to these kind of OTLs in combination with high sensitive single driver speakers is like relearning how good music reproduction at home can sound. It must be experienced in order to feel the difference. But Steve's description is spot on. To be in the studio with the musicians or to sit right in front of a small ensemble (Jazz/Classical music) in a pub, jazz club, or concert hall is how one feels.Instruments sound like "live" as if they were played right in front of you but especially human voices do not only sound "live" but truly reach out to you and touch you almost physically rather than coming out of speakers. With some female singers this effect is almost shocking, as spine tingling as if she starts walking right up to you while she sings.The only other way to experience this is to go to a live concert and have a seat in the first 3 rows. Hearing is believing.

I'd go with that, wholeheartedly.

[Cressy Snr]

I'll put up my revised schematic later on, when I've drawn it. Several of the components are now different to the initial ideas I put down.

[Cressy Snr]

Audio cct:

I limited the LF bandwidth ever so slightly at the input stage. Seemed like a good idea as these kinds of amp are wide-open otherwise, and IME, wide-open OTLs are not a good idea.

[Cressy Snr]

Power supplies:



I've drawn the input/driver supply as if it was using a 230-230V isolating transformer. That is what I would have used if I'd had one handy, rather than having to use glow tubes as droppers.

[Ray P]

Is the speaker facing back towards the amplifier in the first schematic part of some enhanced feedback mechanism, or did you just run out of paper

[Ray P]

Thanks for posting though Steve, really helpful.

[Cressy Snr]

Is the speaker facing back towards the amplifier in the first schematic part of some enhanced feedback mechanism, or did you just run out of paper

I ran out of paper and there was no way I was drawing the damn' thing again

[Nick]

I limited the LF bandwidth ever so slightly at the input stage.

In that case maybe you could consider a smaller and possibly better bipass cap where the 1000uf one is?

[Cressy Snr]

I limited the LF bandwidth ever so slightly at the input stage.

In that case maybe you could consider a smaller and possibly better bipass cap where the 1000uf one is?

Thanks Nick, I overlooked that. Far too big (0.3Hz)