i've built a simple preamp to replace a passive stepped attenuator,
sound has much improved.
now i want more,so build a mark2
the pre amp is merely an opamp input buffer then balance pot then volume pot then line driver.
the inputs to + and - on each op amp are impedance matched for sorce resistance and capacitance, and have unity gain.
now, i'm thinking of investing in four buf634 to use as unity buffers, and not using opamps at all.
the specs for buf634 configured as unity buffer, wide bw seem to be vastly better than even opamps i've used such as opa627 and ad843.
has anyone tried this, ie buf634 on its own not opa627+buf etc
I'c just like to know if its worthwhile as each buf634 ius abour £8 from rs and I'll probably do some improved pcb and power supply work at the same time, so it could prove expensive (but interesting!)
just an idea
sound has much improved.
now i want more,so build a mark2
the pre amp is merely an opamp input buffer then balance pot then volume pot then line driver.
the inputs to + and - on each op amp are impedance matched for sorce resistance and capacitance, and have unity gain.
now, i'm thinking of investing in four buf634 to use as unity buffers, and not using opamps at all.
the specs for buf634 configured as unity buffer, wide bw seem to be vastly better than even opamps i've used such as opa627 and ad843.
has anyone tried this, ie buf634 on its own not opa627+buf etc
I'c just like to know if its worthwhile as each buf634 ius abour £8 from rs and I'll probably do some improved pcb and power supply work at the same time, so it could prove expensive (but interesting!)
just an idea
thanks, its good to know that buf634 is a useable chip,
i'll need 4 as i work in stereo so (one for input buffer and one for output line drive( and maybe headphones two (bonus))*2(for stereo)
=4
at present have lm317/lm337 split supplt with plenty of bypass caps dotted around. have been reading some of walt jung's papers and he has a schematic for super low noise, but it seems a little ott.
i was thinking more along lines of localised regulation. ie +-20v central psu regulated then next to chip using to92 lm's and silvermica caps just to make things super smooth.
have found that opamp class a bias is worthwhile and certainly feedback impedance matching.
i've noticed that buf634 in wide bw mode is already class a
the circuit should become a lot less component intensive
the daunting prospect for the future is applying this concept to my active xo if the new preamp is a sucess.
thanks again
i'll need 4 as i work in stereo so (one for input buffer and one for output line drive( and maybe headphones two (bonus))*2(for stereo)
=4
at present have lm317/lm337 split supplt with plenty of bypass caps dotted around. have been reading some of walt jung's papers and he has a schematic for super low noise, but it seems a little ott.
i was thinking more along lines of localised regulation. ie +-20v central psu regulated then next to chip using to92 lm's and silvermica caps just to make things super smooth.
have found that opamp class a bias is worthwhile and certainly feedback impedance matching.
i've noticed that buf634 in wide bw mode is already class a
the circuit should become a lot less component intensive
the daunting prospect for the future is applying this concept to my active xo if the new preamp is a sucess.
thanks again
active xo
i've done a lot of work on my active xo
its mostly home built
20 year ago i bought some crimson electric 2 way active xo boards after getting fed up with inductors and capacitors which seemed very vague.
i was working for wilmslow audio at the time so all pieces could be played with freely.
the only thing we didn't have was proper accoustic measurement equipment. anyway speakers sounded ok with stock 24db lr at 2.1khz and i left them at that only swapping some tl081 for ne5534.
now, with modern computer, soundcard, mike and software its possible to get all sorts of information.
thus the crossover has evolved. firstly good input buffer. opa unity gain.
then high pass 24db lr set at -6db 1800hz with no further eq for my present tweeters, north d28-06s
also low pass 12db butterworth set at -3db 1600 hz, this with bassmid accoustic rolloff gives close to lr 1800hz lp.
also the phase seems to have come out just right to take into account voice coil offset. (had to use allpass filters before when it was standard 24db lr high and low pass at same frequency.
then bass channel goes to a linkwitz transform circuit to give total Qtc of 0.495 at 16 hz.
then opamp based baffle step eq set to +5 db at approx 330 hz, bit hard to get this right . used trimpots instead of fixed resistors so i could measure, fiddle , listen to an infinite variety of steps.
then a simple pot for level adjustment. the bass driver is very efficient so it needs bringing down by about 3db. I'm not happy with this bit and will eventually adjusted bassa mp gain to suit.
i cant play these speakers at full volume as spl is near 112db with 100w amps and my room is small.
of course all opamps have good psu but general layout could be improved. now working on last 10% using salen key impedance matching as ad743 datasheet etc.
best to start with general flow diagram with required stages. its supprising though how much the sound can be improved by paying attention to fine circuit details.
mostly built on tripad board, no noise or hum problems, and no oscillation even with opa627,you must us 100 ohm ouput resistors on opamps even for short cable runs.
i'll post circuit diagram latter, just got in from work and having a beer.
i've done a lot of work on my active xo
its mostly home built
20 year ago i bought some crimson electric 2 way active xo boards after getting fed up with inductors and capacitors which seemed very vague.
i was working for wilmslow audio at the time so all pieces could be played with freely.
the only thing we didn't have was proper accoustic measurement equipment. anyway speakers sounded ok with stock 24db lr at 2.1khz and i left them at that only swapping some tl081 for ne5534.
now, with modern computer, soundcard, mike and software its possible to get all sorts of information.
thus the crossover has evolved. firstly good input buffer. opa unity gain.
then high pass 24db lr set at -6db 1800hz with no further eq for my present tweeters, north d28-06s
also low pass 12db butterworth set at -3db 1600 hz, this with bassmid accoustic rolloff gives close to lr 1800hz lp.
also the phase seems to have come out just right to take into account voice coil offset. (had to use allpass filters before when it was standard 24db lr high and low pass at same frequency.
then bass channel goes to a linkwitz transform circuit to give total Qtc of 0.495 at 16 hz.
then opamp based baffle step eq set to +5 db at approx 330 hz, bit hard to get this right . used trimpots instead of fixed resistors so i could measure, fiddle , listen to an infinite variety of steps.
then a simple pot for level adjustment. the bass driver is very efficient so it needs bringing down by about 3db. I'm not happy with this bit and will eventually adjusted bassa mp gain to suit.
i cant play these speakers at full volume as spl is near 112db with 100w amps and my room is small.
of course all opamps have good psu but general layout could be improved. now working on last 10% using salen key impedance matching as ad743 datasheet etc.
best to start with general flow diagram with required stages. its supprising though how much the sound can be improved by paying attention to fine circuit details.
mostly built on tripad board, no noise or hum problems, and no oscillation even with opa627,you must us 100 ohm ouput resistors on opamps even for short cable runs.
i'll post circuit diagram latter, just got in from work and having a beer.
( or buy his PCB ) , or use directly this circuit, but with DC servo. I mean, that not exist better circuit at this time - Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.