Hi, mainly for curiosity i would like to learn on an EL84SE amp with local NFB.
Imagine a standard EL84SE, pentode mode, kathode bias. NFB is usually being appplied from the 8 or 16 Ohms tap of the OT to the cathode of the stage in front of the EL84.
And what if the application needs a volume pot ore some kind of tonestack just in front of the EL84?
Which means that there is no NFB of this kind possible in a meaningful way - it would depend on the position of the vol pot.
So - applying NFB directly to, say, the cathode of the EL84? Which would, of course require that the cathode condensor of the EL84 be omitted.
That would lead to
a) current NFB due to the missing cathode condensor
b) NFB including also the transformer.
has someone also done this? How would, for example, the stability of the bias point be affected?
Imagine a standard EL84SE, pentode mode, kathode bias. NFB is usually being appplied from the 8 or 16 Ohms tap of the OT to the cathode of the stage in front of the EL84.
And what if the application needs a volume pot ore some kind of tonestack just in front of the EL84?
Which means that there is no NFB of this kind possible in a meaningful way - it would depend on the position of the vol pot.
So - applying NFB directly to, say, the cathode of the EL84? Which would, of course require that the cathode condensor of the EL84 be omitted.
That would lead to
a) current NFB due to the missing cathode condensor
b) NFB including also the transformer.
has someone also done this? How would, for example, the stability of the bias point be affected?
You would have to flip the usual polarity of the primary winding to get the correct polarity for feedback to the cathode.
The DC bias won't be affected at all, since there is no nfb at DC with transformer coupling.
The DC bias won't be affected at all, since there is no nfb at DC with transformer coupling.
Which means that the speaker will always be loaded by the full anode current, would it not? And the secondary of the transformer as well?
Last edited:
No, most of the tube DC current bypasses the speaker voice coil. But the amount of nfb is not adjustable.
If directly connected to the cathode, the very low DCR of the secondary will short out the cathode bias resistor.You would have to flip the usual polarity of the primary winding to get the correct polarity for feedback to the cathode.
The DC bias won't be affected at all, since there is no nfb at DC with transformer coupling.
Another consideration:
An EL84 using a 5k/8Ohm OT just has a gain around 2 at the secondary.
So the NFB might not be very effective.
An EL84 using a 5k/8Ohm OT just has a gain around 2 at the secondary.
So the NFB might not be very effective.
Last edited:
The schematic in post #3 shows classic CFB, cathode feedback. I use it in several of my DIY amps and it works well. Yes, there will be a little bit of DC across the speaker terminals but the DCR of the transformer winding is just a fraction of that of the speaker so it's usually 40mV or something like that.
The amount of feedback depends on the output stage gain and the turns ratio of the transformer, usually 3-6dB.
The amount of feedback depends on the output stage gain and the turns ratio of the transformer, usually 3-6dB.
I treid it recently as an experiment:
I did the same thing recently on a simple SE EL84 amp using a very cheap Chinese 5k output transformer. It resulted in about 6dB of feedback and these measured results (test signal directly to the grid of the EL84, measuring across an 8R dummy load, with and without the feedback):The Tubelab SSE board (a two stage stereo SE HiFi amp) has provisions to apply feedback from the secondary of the OPT to the cathode of the output tube. This raises the DF and lowers the THD a bit, but also seems to help squeeze a bit more bass out of a cheap OPT. No global feedback is used.
Hi, not exactly an advice, rather rough idea, please take a look at a schematic of one of my best tube amps, it uses FB from secondary to EL34 cathode, OPT is toroidal, good quality: https://www.diyaudio.com/community/threads/my-last-project.78158/post-898445Hi, mainly for curiosity i would like to learn on an EL84SE amp with local NFB.
Imagine a standard EL84SE, pentode mode, kathode bias. NFB is usually being appplied from the 8 or 16 Ohms tap ...
The best way to apply local feedback over an output pentode is to go Schade. But when you say "tone stack", I get the impression this is for a guitar amp. In that case omit the feedback!And what if the application needs a volume pot ore some kind of tonestack just in front of the EL84?
There ist a world outside guitars. basses, keyboards, hifi...
The amp i woult like to try it first is actually a tiny bass amp (a heavily modded Epiphone Valve Junior, ECC83+EL84). That one will profit from NFB including the OT (something better than the original one) because it will also linearize the latter. It actuaslly has NFB added, to the cathode resistor of the 2nd triode. Meanwhile i need the cathode for further sound modification, and i would like to have a vol pot aver the 2nd triode as well. And of course i would like to keep the NFB.
But again, i would first like to improve my knowledege, for further non-guitar amp projects.
BTW: what exactly is "Schade" NFB? Just NFB from the anode to the input grid?
The amp i woult like to try it first is actually a tiny bass amp (a heavily modded Epiphone Valve Junior, ECC83+EL84). That one will profit from NFB including the OT (something better than the original one) because it will also linearize the latter. It actuaslly has NFB added, to the cathode resistor of the 2nd triode. Meanwhile i need the cathode for further sound modification, and i would like to have a vol pot aver the 2nd triode as well. And of course i would like to keep the NFB.
But again, i would first like to improve my knowledege, for further non-guitar amp projects.
BTW: what exactly is "Schade" NFB? Just NFB from the anode to the input grid?
- Home
- Amplifiers
- Tubes / Valves
- EL84SE with local NFB?