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

Official Court Jester
Joined 2003
Paid Member
Got T-Bed-ed Berserk M2 , aka Babelfish M25 back , after week long test with Dynaudio Contour 1.8 Mk.II

several boyz , besides buddy with Dyns , had opportunity to hear it and all were more than satisfied , even thrilled

it seems that I'm going to need several pcbs more , to put them in cases :clown:

will try to do battery of same THD tests already done , but this time with enabled Borbely White Cathode JFet follower , just to see how much of improvement we can gain with already gutsy performance

also will try to wrap my head around some additional/other software , combined with my headache giving EMU1212 , to rec. polarity of harmonics in THD spectra

last but not least ...... doing yearly big cleaning and tidying up of my workshop ........ got to state of too giant steps , to hop in chair :clown:

got some new arrivals , besides heavy iron and copper for Westminster xovers (not for me , don't like quasi horns) , got some new pcbs ....... some to populate ( at last !) and send , some really new - SEF/Babelfish M2 pcbs , also intended for SUFI-SIT (THF51-S, hope to get them soon )

life is fun
 

Attachments

  • WP_20180728_16_16_22_Pro.jpg
    WP_20180728_16_16_22_Pro.jpg
    70.9 KB · Views: 927
  • WP_20180728_16_35_52_Pro-(1.jpg
    WP_20180728_16_35_52_Pro-(1.jpg
    104 KB · Views: 937
Member
Joined 2003
Paid Member
Russell , where is possible to see porn of your spks ?

OK, here's the porn...can make more of anything if of interest in particular.
 

Attachments

  • IMG_4702[1].JPG
    IMG_4702[1].JPG
    63.9 KB · Views: 265
  • IMG_4699[1].JPG
    IMG_4699[1].JPG
    72 KB · Views: 249
  • IMG_4695[1].JPG
    IMG_4695[1].JPG
    77.3 KB · Views: 265
  • IMG_4694[1].JPG
    IMG_4694[1].JPG
    102.4 KB · Views: 268
  • IMG_4693[1].JPG
    IMG_4693[1].JPG
    83.2 KB · Views: 285
  • IMG_4692[1].JPG
    IMG_4692[1].JPG
    92 KB · Views: 312
  • IMG_4691[1].JPG
    IMG_4691[1].JPG
    108.7 KB · Views: 310
  • IMG_4688[1].JPG
    IMG_4688[1].JPG
    57.1 KB · Views: 883
  • IMG_4687[1].JPG
    IMG_4687[1].JPG
    68.3 KB · Views: 926
  • IMG_4685[1].JPG
    IMG_4685[1].JPG
    70.2 KB · Views: 947
Official Court Jester
Joined 2003
Paid Member
as I just wrote to Russellc in e-mail :

yesterday finally settled for increased number of ordered bugs (ACS Halls ) , also sent order for best possible solution I can get (securely) for P channel mosfet - Infineon IRFP9140


as written on DiyA , especially by Ilimzn , 9140 is better paired with 240 , than 9240 is ....... so much better symmetry


in this case , I'm just shooting for equal Berserk in both directions ....... up and down :clown:

of course that I bugged Pa with few questions about Infineon , he said that their datasheet is carbon copy of IRF , so , go figure

anyway , to have it here , sort of black on white (that's literally translation from Serbian :rofl: ) , I found Papa's post which I'm recalling in last few days:

You can check your Mosfets easily enough, the IRF parts have a
shelf in the transconductance vs frequency curves. If you need some
hints you can look at the Mosfet testing article at

PassDiy

Of course it makes a difference, but it is minimized in follower applications.

so , verdict
 
Member
Joined 2003
Paid Member
VOT has x 0ver and compensation put together by member on Audiokarma. Tried many variations of x overs, including original 500k Altec. Haven't found one superior to this. Didn't try Hiraga version, thought it was charged large format CD and 1505 horn? Maybe I should.

Russellc
 
Member
Joined 2003
Paid Member
Link to DiyA stuff:

yesterday finally settled for increased number of ordered bugs (ACS Halls ) , also sent order for best possible solution I can get (securely) for P channel mosfet - Infineon IRFP9140


as written on DiyA , especially by Ilimzn , 9140 is better paired with 240 , than 9240 is ....... so much better symmetry

Russellc
 
Official Court Jester
Joined 2003
Paid Member
comparison : regular (N-N) buffer vs. Borbely's WC buffer

as explained in post #1 (and few other posts ) , I made provision for implementing two types of buffers in amp front end ;

transition from one to other is simple - pulling two 0R resistors from pin sockets , and putting 27R in one place and anything between 470K to 1M in other place, and finally adding one 220nF capacitor

in simulation and in real world , these two buffers are mostly the same in performance when load is benign , while Borbely's White Cathode Follower transposed in JFet domain is having serious advantages with lower loads ..... say 1K and less

anyway , proof is in da pudding , so I finally made some measurements , comparing these two.

find enclosed two part article-Bible for Greedy Boyz , from E. Borbely

always worth reading
 

Attachments

  • JFETS THE NEW FRONTIER, PART 2.pdf
    121.7 KB · Views: 144
  • JFETS THE NEW FRONTIER, PART 1.pdf
    541.3 KB · Views: 187
Last edited:
Official Court Jester
Joined 2003
Paid Member
so , measurements ;

just 1 and 10W , in 4 and in 8R load , regular buffer and Borbely buffer

read captions on pictures to know what's what

conclusion ....... OS , transposed back to buffer through Edcor ..... isn't such/enough heavy load to invoke any particular benefit from Borbely's buffer , EXCEPT slight increase in 2nd harmonic , thus making sound little sweeter

didn't made measurements with any software capable to show harmonic's phase ; for that I need first to repair damn EMU1212 , having Lout channel damaged from some reason ...... proggie is demanding both measuring route through one channel out-in and ref loop-back through second channel

bummer , but who cares ...... fun is all mine ........ we don't need stinkin' 'puters to make amps :rofl:

again - ignore everything left from 1K

edit : maybe the reason why there is no bigger difference is because existing regular buffer is already overkill ..... having 6JFets paralleled up and 6 JFets paralleled down , with 21mA Iq
 

Attachments

  • B M25 10W 8 R regular buffer.png
    B M25 10W 8 R regular buffer.png
    66.9 KB · Views: 179
  • B M25 10W 8 R Borbely buffer.png
    B M25 10W 8 R Borbely buffer.png
    67.1 KB · Views: 170
  • B M25 10W 4 R regular buffer.png
    B M25 10W 4 R regular buffer.png
    63.7 KB · Views: 173
  • B M25 10W 4 R Borbely buffer.png
    B M25 10W 4 R Borbely buffer.png
    65.7 KB · Views: 1,297
  • B M25 1W 8 R regular buffer.png
    B M25 1W 8 R regular buffer.png
    67.4 KB · Views: 1,310
  • B M25 1W 8 R Borbely buffer.png
    B M25 1W 8 R Borbely buffer.png
    68.1 KB · Views: 1,331
  • B M25 1W 4 R regular buffer.png
    B M25 1W 4 R regular buffer.png
    65.8 KB · Views: 1,343
  • B M25 1W 4 R Borbely buffer.png
    B M25 1W 4 R Borbely buffer.png
    66.3 KB · Views: 1,377
Last edited:
Member
Joined 2003
Paid Member
so , measurements ;

just 1 and 10W , in 4 and in 8R load , regular buffer and Borbely buffer

read captions on pictures to know what's what

conclusion ....... OS , transposed back to buffer through Edcor ..... isn't such/enough heavy load to invoke any particular benefit from Borbely's buffer , EXCEPT slight increase in 2nd harmonic , thus making sound little sweeter

didn't made measurements with any software capable to show harmonic's phase ; for that I need first to repair damn EMU1212 , having Lout channel damaged from some reason ...... proggie is demanding both measuring route through one channel out-in and ref loop-back through second channel

bummer , but who cares ...... fun is all mine ........ we don't need stinkin' 'puters to make amps :rofl:

again - ignore everything left from 1K

edit : maybe the reason why there is no bigger difference is because existing regular buffer is already overkill ..... having 6JFets paralleled up and 6 JFets paralleled down , with 21mA Iq

Alrighty, here I go again displaying level of ignorance on world wide web...I notice in these graphs, the big spike at 1K...then a little to right a small thin spike, followed by another small thin spike to the right of that.

Then , I read where you said the Borbely doo dads increased H2, and I noticed that the first thin line to the right of big spike was taller than standard M2 version.

Now, from that I wondered if first thin spike to the right was 2nd harmonic, and next one to right is 3rd?

I hope so.

Befuddled,

Russellc
 
Official Court Jester
Joined 2003
Paid Member
Final pcb and schematic

OK , suddenly found my self in cloud , while sitting on the porch and preparing for hiking ; too much thunders these days , so better stay and brew some more coffee

to repeat - pcbs are common for Babelfish M25 and SissySIT amps

so , I've done them (files) :
 

Attachments

  • Babelfish-M25-schm-for-DiyA.jpg
    Babelfish-M25-schm-for-DiyA.jpg
    197.4 KB · Views: 787
  • pcb sshot.jpg
    pcb sshot.jpg
    285.1 KB · Views: 817
Last edited: