Hey all, an update is that I have not shared the files publicly yet, simply because I have not fully tested the boards. Gotta populate a PSU and do some checks first, then I'll make any necessary revisions before sharing the files. I know my limits as far as how busy I get and don't wish to hold everyone up waiting for me, so once the files are ready I will be posting them up so anyone interested in ordering them in quantity to pass along can handle that as they feel comfortable. I have too many irons in the fire and I'm nowhere near organised enough to set all the ordering and distribution up.
If anybody has any suggestions on an easy way to set up ordering and such for boards that tracks stock and shipping info I'd love a chat off thread, as I do wish to set up an order form and commerce stuff eventually to distribute boards.
If anybody has any suggestions on an easy way to set up ordering and such for boards that tracks stock and shipping info I'd love a chat off thread, as I do wish to set up an order form and commerce stuff eventually to distribute boards.
I was doing some simulations and loadline calculations, and came up with a simple common cathode version 6V6 line stage that looks quite good.
I found that THD is very low if you can employ a 400V B+ and a 15k ohm plate resistor (preferably wirewound 12W).
I adjusted the value of Rk to get Ia = 20mA.
Has anyone tried crazy operating points like this?
I found that THD is very low if you can employ a 400V B+ and a 15k ohm plate resistor (preferably wirewound 12W).
I adjusted the value of Rk to get Ia = 20mA.
Has anyone tried crazy operating points like this?
Well, the idea is that if the ra of the 6V6-triode is about 2k ohms, raising the anode load to more than 5x the ra would make a more linear circuit (also increase gain, but that's a different story).
A 5k ohm load is 2.5x the 6V6-triode ra, which would allow for more power out, but not as linear.
The question is, of course, how does it sound? Well, that I don't know, which was why I was asking.
A 5k ohm load is 2.5x the 6V6-triode ra, which would allow for more power out, but not as linear.
The question is, of course, how does it sound? Well, that I don't know, which was why I was asking.
You know, it's a funny thing... Many times I have been in a situation where I listen to a new design compared with a similar previous design, like recently of one phono preamp vs. an older one. I always seem to prefer the one with higher THD, at least at first. I've seen that happen with other people too. A bit of added H2 sounds nice. The only drawback is if you listen to big orchestral works (Wagner, Mahler) or dynamic recordings of solo piano works (Ravel Gaspard de la Nuit or Lizst B-Minor Sonata). Then you hear the mud caused by intermodulating frequencies. But if your tastes tend towards classic small-group jazz combos and chamber music, that extra H2 can be really nice...
True. There is a catch though, less THD comes along with higher plate load & more gain in the common cathode version which also adds more noise. Best is to arrange the system gain structure with low voltage gain in the power amp. Adequate to reach max power after three o'clock on the knob.
So perhaps... Apply a little NFB?
I only need a line stage with about 3x gain for my current setup. I had some PCBs made for a noval twin-triode (12AT7, 6DJ8, etc.) as a stereo pair of 'anode followers' (common cathode with shunt NFB) each with a MOSFET source follower. The NFB goes from the output of the source follower to the grid of the triode. I need to wire that up with a decent PSU and see if the NFB destroys the sound quality (as so many claim it will).
Why not take a 6V6-triode with ca. 9x gain and reduce that to 4x gain by employing plate-grid NFB? Will the magic disappear? What exactly is the magic?
There. I made another project for myself.
I only need a line stage with about 3x gain for my current setup. I had some PCBs made for a noval twin-triode (12AT7, 6DJ8, etc.) as a stereo pair of 'anode followers' (common cathode with shunt NFB) each with a MOSFET source follower. The NFB goes from the output of the source follower to the grid of the triode. I need to wire that up with a decent PSU and see if the NFB destroys the sound quality (as so many claim it will).
Why not take a 6V6-triode with ca. 9x gain and reduce that to 4x gain by employing plate-grid NFB? Will the magic disappear? What exactly is the magic?
There. I made another project for myself.
I like it 
Even if it's wasteful.
I would follow it though. This is where it could get interesting. Replace the 15k resistor with a 6SN7, and use the other half as a follower... How about that? Maybe lower the B+ to 350V as to not beat the crap out of the 6SN7 or parallel the sections and use two... If you're going to use a 6V6 as a preamp tube, why stop there?
Even if it's wasteful.I would follow it though. This is where it could get interesting. Replace the 15k resistor with a 6SN7, and use the other half as a follower... How about that? Maybe lower the B+ to 350V as to not beat the crap out of the 6SN7 or parallel the sections and use two... If you're going to use a 6V6 as a preamp tube, why stop there?
I think the reason to use a 6V6 as a preamp tube is for its particular, and presumably desirable, 'sound'.If you're going to use a 6V6 as a preamp tube, why stop there?
That begs a few questions.
- What is it about a 6V6-triode that makes that desirable 'sound'?
- Does putting a follower after the 6V6 make it lose that desirable 'sound'?
- Does changing its operating points make it lose that desirable 'sound'?
- Would plate-grid (parallel applied) negative feedback make it lose that desirable 'sound'?
I'll just need to build the dang thing and see what I find. Hopefully pretty soon. I'm working on finishing up a pair of speakers atm... (AmpsLab Osprey-BR)
A little Gnfb is enough to cause it to lost some sparkles and details.
You can hear the difference when different output tube is used.
https://www.diyaudio.com/community/attachments/ta-guitar-png.612552/
You can hear the difference when different output tube is used.
https://www.diyaudio.com/community/attachments/ta-guitar-png.612552/
That example in the video uses a power amp that includes the output transformer inside its global NFB loop.
Is there a difference if the NFB loop has only one reactive element (the output capacitor) so there are no crazy phase shifts and multiple poles and zeros such as you'd find in a tube power amp?
Is there a difference if the NFB loop has only one reactive element (the output capacitor) so there are no crazy phase shifts and multiple poles and zeros such as you'd find in a tube power amp?
Very little Gnfb, is about 1 db, the volume drops very little, and then pentode mode sounds more alive then triode mode, as pentode has no internal FB.
The added sparkle or details is very subjective, but you get to hear it when no nfb is applied.
I have been working on driver for OTL amp for last ten years, with one particular topology and found only a particular driver tube sounds right to my ears. It maybe that 6v6 sounds right to most people ears. Gnfb usually control damping, and if it affects details or sparkle then that something else..?
The added sparkle or details is very subjective, but you get to hear it when no nfb is applied.
I have been working on driver for OTL amp for last ten years, with one particular topology and found only a particular driver tube sounds right to my ears. It maybe that 6v6 sounds right to most people ears. Gnfb usually control damping, and if it affects details or sparkle then that something else..?
Last edited:
For example there are 2 tubes in the driver, they work as a pairs, usually choosen to compliment each other for best listening. In case this is solo 6v6 then it must sound right by itself as no other tube is going to affect or add to its own sound signature, even under FB conditions. in either case only the desirable sound signature is driving the output or amp.