"Look Ma, I have a new Xformer. Want a new toaster oven" The FuP6 amp story

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Today gonna try higher Rs on top fet, keeping lower Rs @.1R-.15R. I will also try the cap connection switch as well at these values.

Secondary ideas:
Attach upper cap above jfet, turning it into CCS....I think
Adding 120R resistor from output to V-
Using R085 in lower postition with low to no Rs

I encourage anybody following to try this one thing though. Connect top cap below Rs on upper half and then do same for lower half. After listening, kill power and take upper cap and connect above Rs in upper half. Listen again. Tell me what you hear. TBH, this should work on any fet.

BTW, stock Rs values for my setup are .1, but I am going to play around with this.
 
Official Court Jester
Joined 2003
Paid Member
Today gonna try higher Rs on top fet, keeping lower Rs @.1R-.15R. I will also try the cap connection switch as well at these values.

Secondary ideas:
1.Attach upper cap above jfet, turning it into CCS....I think
2.Adding 120R resistor from output to V-
3.Using R085 in lower postition with low to no Rs
........

1. just disconnect the cap ; I'm lazy to think what trouble you can cause with connecting it to U+

2.can't harm ...... THD spectra , yakyak

3. probably you'll not hear difference
 
No?:D Being the subliinal guy that you are, you either were just trying to make me feel better about the lack of interest by responding with 3rd vote,....or you were giving hint about nother implemetation. I was guessing about a triple cascode or other layouts that would involve 3 transistors. Bjt mirrors have three transistors,..I thnk. Don;t knoe for sure as i am not big into bjt's. Other one was joachim's circuit that Nelson called similar to beast of theousand jfets, also having 3 transistors, if you include the input jfets.

For real do, do i stand to gain anything audible by dropping Rs and cascoding the FE. I know in theory it should be lower noise and also due to lack of degeneration, the jfets will have more gfs/yfs, meaning more drive capablilty and lower Zo. But do you tihnk this will be audible?
 
Last edited:
Official Court Jester
Joined 2003
Paid Member
for moi , it will be better (not just soundwise) , from two reasons :

1.cascoded (as in schm I posted , having a 10R trimpot is nice) Jfets are working in (nearly) constant voltage mode; nearly means - constant sans 1/2 input/output voltage :rofl:
1a. you can make it is per JG schmtc , where they're really working in constant voltage mode ; he's saying that difference is better bass etc. I really never heard that difference , but I didn't stress them enough (load was sort of benign) . Be aware that Pa said , not just once , that if cascode is ref. to gnd (not to source as in JG's) there is some amount of sorta THD cancellation
2.if you are going to use LED array for cascode ref. entire gadget will be shinier . put acrylic cover on amp
3.did I said two? I can give you 10 more , but I'm lazy , still not enough caffeine in my scull
 
Official Court Jester
Joined 2003
Paid Member
just thinking about after lunch coffee makes me think easier ;)

regarding [1] above - cascode(s) bring two major benefits - with Jfets in sweetest voltage and (almost) full Ids range , and serving as garbage umbrella .

whatever happens with rails , much lesser possibility to have it transposed through Jfets to their output

peace of mind , easier mu
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.