I am so deep in these NFB maths right now…
I’ve got the original 5F6A and JTM45 sorted out now, but for the life of me I can’t rationalize why @robrob selected a 15k NFB resistor for the micro. 1+2=4 and it doesn’t seem to provide the proper amount of feedback at the moment, and it’s making me nuts.
As soon as I can grok what’s going on in the Micro NFB and EF80 math, I’ll post the proper resistors for a switched mod.
(Ditto for the switched Micro “lead”/Plexi mod.)
Good question, one I'm even less able to tackle. I looked into this a bit, decided just to try 15K, and found I liked it -- in the '5F6a' 12AY7 setup. Again, I think @joulupukki gets this a lot more than I do -- his comments above are relevant.
What little I know is what Rob says in his 5E3 NFB discussion: "The 56k feedback resistor is correct for a speaker output of 8 ohm and will supply a light level of feedback. You can increase the feedback by reducing the feedback resistor value. A 39k resistor will increase the feedback by 41% and a 20k resistor will give you a Marshall JTM45 2.8 times more feedback. The Fender tweed Princeton amp used 22k feedback resistors and the tweed Harvardused a 56k (both with 8 ohm speakers) so feel free to try anything between 20k to 100k with an 8 ohm speaker tap (10k will give more feedback, civility and clean headroom)." So it sounds like you could tack in a resistor you choose and see how it sounds, then adjust. Or some folks have suggested a pot?
My take on Rob's page is he suggests the 15K because he found it sounded best. I know Rob is very supportive of us micro amp builders. @robrob -- any NFB thoughts?