Babelfish M25, AKA M2 on steroids, AKA M2-XA25 bstrd AKA M2 gone Berserk

[Zen Mod]

now , on to measurements

first 25W/8R

at that level (40Vpp) there is I think is some compression of sine crests , easily read in THD figure

remember crappy caps

Papa sez 25W/8R @ 1% THD for regular M2

 

[Zen Mod]

then Papa sez 40W/4R @ 1% THD

let's see that

 

[Zen Mod]

let's try 50W/4R , can we get 1% THD ?

 

[Zen Mod]

I simmed some 21W as magic number on 2R , while still being in A class ;

let's see that :

 

[Zen Mod]

that's it for tonight

will try on Monday to observe Klunk!! region via scope , across inline resistor or simply across rail wire ;

I'm not overly optimistic , caps being on sluggish side , but I'll try

in any cse , simply observing THD behavior , I think it is safe saying that simulations are confirmed , bloody thing being in A class in this brackets:

25W/8R
43W/4R
21W/2R

few watts up , or down .......


YouTube

 

[Zen Mod]

just to add Posters

then maybe get some sleep .......

🙂

(seen exquisite ones in Mr. MR's listening room)

 

[vdi_nenna]

 

[Russellc]

that day I was able to complement just 10W tests .... couldn't get more at 8R , due to EMU output swing (non)capability

so :

10W/8R

10W/4R

10W/2R

What sort of measuring tools and setup are you using for these measurements? Might be helpful for "Adjusting P3 thread".....

Thanks,

Russellc

 

[Zen Mod]

I was realy gnawing about that through entire thread ... EMU1212 and interface box,shown up

 

[Zen Mod]

is amp that scary looking , so no comments ?

 

[indra1]

Have you tried listening with reverse speaker polarity connection to confirm H2 polarity?

 

[Zen Mod]

nope

frankly , wasn't in situation to have that much concentration ..... or easy-ness , generally

too frenzy period ...... luckily , calming .......

anyway , that's trivial , part of later fun

 

[indra1]

I would expect significant difference due to single ended jfet buffer.

 

[Zen Mod]

this SE JFet buffer is having same THD spectra as complementary one

I've put plenty of thinking in that

disclaimer - I did use 2SK170 model instead of 2SK2145 (half) , but , what the heck

 

[aspringv]

I'm super curious about the hall effect sensors. I'll need to get home and spend some time with some data sheets so I can get around that part of the circuit.

'bravo!' is in order though! Not everyday you get to create a new bias mechanism!

 

[Zen Mod]

indra is Da Man

me , mere muscles , as usual

 

[indra1]

... did use 2SK170 model instead of 2SK2145 (half) , ...

Most similar model to 2SK2145 on standard jfet would be 2SK117. Compared to the K170/J74 complementary pair, the K117 SE buffer with ideal parts simulate to 33% lower H2 and 66% lower H3. I also did not see a CBL for WCF topology on the sim file. On a build however, phase of H2 will be dependent on parts selection.

As you said anyway, it's a trivial matter of reversing speaker polarity to confirm and choose desired connection. Have fun. 🙂

 

[Zen Mod]

........ WC iteration is havin opposite K2

 

[indra1]

The WCF with 2SK117 sim to 35 dB lower distortion. Perhaps it is also rewarding to listen to both polarities without CBL. Just to make sure, sometimes the one with less distortion sounds too flat or dry. 🙂

 

[Zen Mod]

yup

though , I came with sims exactly to point where I was fed up ..... and from where I needed Real McCay

and what that McCay is ..... 🙂