ZM Idiot
just realized that upper mosfet can be easily Schaded ; just addition of one resistor on pcb, and finding nice logic level mosfet in place of IRF510
classic case of not seeing the forest for the trees
damn 3rd amp on same pcb
another one to try and do all measurements ............
anyway, after first morning coffee, up to SSE's, to hear what he's saying about B. XA252 SIT ..... am I deluded or what
just realized that upper mosfet can be easily Schaded ; just addition of one resistor on pcb, and finding nice logic level mosfet in place of IRF510
classic case of not seeing the forest for the trees
damn 3rd amp on same pcb
another one to try and do all measurements ............
anyway, after first morning coffee, up to SSE's, to hear what he's saying about B. XA252 SIT ..... am I deluded or what
Last edited:
it passed (as if Pass made it
) with flying colors... and then someBabelfish XA252 gives you all details .........so you can form your own Opinion
Babelfish XA252 SIT gives you ready Opinion, with all details, and it doesn't care what you think
go figure
Thank you for sharing the fruits of your labor with us all!
This is exciting news - I'm especially curious how the OS of the M25 R.2 will compare when wrapped in a feedback loop with this FE.
I might have missed it in the F4 Beast thread, but apart from power handling, do the big FETs exhibit a different performance/character than a 240/9140 pair? The transconductance numbers seem a lot higher - e.g., 38 S at Id = 20 A vs 6.9 S at Id = 12 A, but I realize that's not a direct comparison due to the different test current.
This is exciting news - I'm especially curious how the OS of the M25 R.2 will compare when wrapped in a feedback loop with this FE.
I might have missed it in the F4 Beast thread, but apart from power handling, do the big FETs exhibit a different performance/character than a 240/9140 pair? The transconductance numbers seem a lot higher - e.g., 38 S at Id = 20 A vs 6.9 S at Id = 12 A, but I realize that's not a direct comparison due to the different test current.
generally, THD is notch lower when going Biguns
that from start, and up where little ones are going out of steam - difference getting bigger
though, always worth observing bigger picture - what speakers, how big SPL and, in the end, what Watts
we all know that with little ones (suddenly TO247 little
) we already have exquisite amps
catch is - with Biguns - driving them appropriately
that from start, and up where little ones are going out of steam - difference getting bigger
though, always worth observing bigger picture - what speakers, how big SPL and, in the end, what Watts
we all know that with little ones (suddenly TO247 little
) we already have exquisite ampscatch is - with Biguns - driving them appropriately
we all know that with little ones (suddenly TO247 little
Indeed. I'll likely try out a version with the 240/9240s in my drawer before splurging for the $$$ IXYS.
...... and I figured few tiny details, how to make Schade iteration
now, just need few logic level TO220 mosfets , to put N pucks back ....... an will inform ya
few days
Silly ZM
because I don't have adequate logic level N Mosfets in workshop, so no go before ..... Tuesdaythen I realized that I have DN2540N5
whoohoo!
Hammer Time !!
(brain is indeed strange creature ........ how it works, if it works, when it works)
logic level mosfets - having treshold Ugs almost half, comparing to IRF510
remember - it goes on top/aditive to Schaded Mosfet Ugs, so if you have 3V5 or so of IRF510, that's already more than negative rail Mosfet is having, so goodbye Ugs symmetry
not so important here , biasing with CM as is with Optobias of SissySIT and others, but why chewing another volt or two from max possible swing ......
Schade is having nothing with CCS biased follower, when you have both series resistor and DG resistor
in Ole world , for Schade you did calc on Rout of preceding stage ..... then deciphering what value is feedback resistor
here - best to have low Rout of preceding stage, then set everything by series and feedback resistor ..... if DG resistor can be titled that way
anyway, effect is same
remember - it goes on top/aditive to Schaded Mosfet Ugs, so if you have 3V5 or so of IRF510, that's already more than negative rail Mosfet is having, so goodbye Ugs symmetry
not so important here , biasing with CM as is with Optobias of SissySIT and others, but why chewing another volt or two from max possible swing ......
Schade is having nothing with CCS biased follower, when you have both series resistor and DG resistor
in Ole world , for Schade you did calc on Rout of preceding stage ..... then deciphering what value is feedback resistor
here - best to have low Rout of preceding stage, then set everything by series and feedback resistor ..... if DG resistor can be titled that way
anyway, effect is same
Babelfish XA252 SET
as my pal Indra, member of same Pantheon and evidently same Caste to which I belongs, asked ....... here is schm of Babelfish XA252 SET
fancyschmancy name for Schade iteration of full Mos OS, where upper output Mosfet is Schaded, so practically being emulation of SIT
gate resistor R50 increased to 750R, introduced new resistor R58 (18K)
SIT related parts (two rectangles of short dash lines) mounted - as for SIT iteration, but instead of M1 - IRF510 (we are using for SIT), placed DN2540
R46 changed to 33R
we need around 3V voltage level at left (inner) side of R50, so result as is - DN2540 gate voltage is 3-4V-ish (fine tune of R46 possible to achieve full symmetry of M1 and M3 gate voltages, trivial task) ; these being referenced to output node
all parts titles ref. to attached schm
once when I have final pcb, containing all 3 amps , schmtcs with proper nomenclature will be posted
I could avoid T8-Q9 30mA CCS, and route lower end of R46 to output node (M2 source), but as I already had CCS in place, why not using it
yes, as Indra said - Papa's fave gigglywiggly graphs are saying that 2nd is neg. in phase ; that not posted, just simulated, but no reason for doubt
so, schm:
as my pal Indra, member of same Pantheon and evidently same Caste to which I belongs, asked ....... here is schm of Babelfish XA252 SET
fancyschmancy name for Schade iteration of full Mos OS, where upper output Mosfet is Schaded, so practically being emulation of SIT
gate resistor R50 increased to 750R, introduced new resistor R58 (18K)
SIT related parts (two rectangles of short dash lines) mounted - as for SIT iteration, but instead of M1 - IRF510 (we are using for SIT), placed DN2540
R46 changed to 33R
we need around 3V voltage level at left (inner) side of R50, so result as is - DN2540 gate voltage is 3-4V-ish (fine tune of R46 possible to achieve full symmetry of M1 and M3 gate voltages, trivial task) ; these being referenced to output node
all parts titles ref. to attached schm
once when I have final pcb, containing all 3 amps , schmtcs with proper nomenclature will be posted
I could avoid T8-Q9 30mA CCS, and route lower end of R46 to output node (M2 source), but as I already had CCS in place, why not using it
yes, as Indra said - Papa's fave gigglywiggly graphs are saying that 2nd is neg. in phase ; that not posted, just simulated, but no reason for doubt
so, schm:
Attachments
Last edited:
Babelfish XA252 SET - graphs , 8R load, BFR=22R
as titled
rails as previously: +/-30Vdc, Iq 1A7-ish
as titled
rails as previously: +/-30Vdc, Iq 1A7-ish
