bradb
Well-known member
Question...
Can we increase the release time to be as fast as the 33609? 50ms?
Can we increase the release time to be as fast as the 33609? 50ms?
2254C build/support thread
- Thread starter Igor
- Start date
Help Support GroupDIY Audio Forum:
Hi Brad, it is not increase, minimal release time in compressor's section is 100ms, according to Neve manual IIRC 
But yes, changing relase time is possible, just check this:
http://i251.photobucket.com/albums/gg291/diy33609/compREC_sw.jpg
and this:
http://i251.photobucket.com/albums/gg291/diy33609/compREC_sw.jpg
Please mind, there's error on Neve's drawing: on compressor recovery switch 6k8 is actually 24k,
it is r'10 on bottom-mid on my schematic:
http://i251.photobucket.com/albums/gg291/diy33609/schem-1.gif
Let me know how do you want to change recovery settings, will post it here.
But yes, changing relase time is possible, just check this:
http://i251.photobucket.com/albums/gg291/diy33609/compREC_sw.jpg
and this:
http://i251.photobucket.com/albums/gg291/diy33609/compREC_sw.jpg
Please mind, there's error on Neve's drawing: on compressor recovery switch 6k8 is actually 24k,
it is r'10 on bottom-mid on my schematic:
http://i251.photobucket.com/albums/gg291/diy33609/schem-1.gif
Let me know how do you want to change recovery settings, will post it here.
bradb
Well-known member
Yes, i ment to say "Can we shorten the fastest release time to something like 50ms...?"
I understand that it can be switched between these values with the position of the release switch, but is there anyway to make the 100ms position something faster like 50ms?
I mention this because i was discussing the 2254 and 33609 with a trusted engineer friend. his experience with the 2254 was that it was a touch slow on drums (i don't want to start a subjective battle over this btw). But I was just curious if we could change this.... hey, since this is DIY... DI-whY-not!
I understand that it can be switched between these values with the position of the release switch, but is there anyway to make the 100ms position something faster like 50ms?
I mention this because i was discussing the 2254 and 33609 with a trusted engineer friend. his experience with the 2254 was that it was a touch slow on drums (i don't want to start a subjective battle over this btw). But I was just curious if we could change this.... hey, since this is DIY... DI-whY-not!
Brad, 2254C sidechain is copy-pasted from 33609. This is one of the things which makes it
more versatile than old good 2254A or E. To fasten first release setting, just decrease R'10 value (say 12-18k)
and slightly increase value of R'11 to ~82k, by same amount, to compensate it on second release setting, or,
leaving R'11 same (75k) will slightly shorten second and third release times (allmost inaudible).
more versatile than old good 2254A or E. To fasten first release setting, just decrease R'10 value (say 12-18k)
and slightly increase value of R'11 to ~82k, by same amount, to compensate it on second release setting, or,
leaving R'11 same (75k) will slightly shorten second and third release times (allmost inaudible).
bradb
Well-known member
I think I understand. Thank you for the explanation. (I may ask a clarifying question later when I actually dive in)
how about attack ratios... can they be quickened?
Youre the best Igor!
thanks
how about attack ratios... can they be quickened?
Youre the best Igor!
thanks
Regarding attack settings. There's R'31 (see sidechain, bottom left of schem), at TR8's emitter.
Originally, it is 560 or 680R (difference between old and new units). Change it to 150R; at first, second and third attack
switch settings, attack will be faster than original, three times at least at first setting.
To faster both attack and release is possible by changing C13' 10uF capacitor as well. Lower is faster both att/release.
One thing, on too fast settings, sidechain can oscillate in some cases...at least, IIRC it did on breadboard
a years ago when I started to play arround 33609. However, IIRC, it was ground or power supply issue and was fixed...
Will play with this when will have a time next week.
Automatic time constants, A1/A2 possible to change as well.
Best way to do this: take unit in studio and play untill it smokes
Well, don't forget to switch off when changing components
Just kidding.
Originally, it is 560 or 680R (difference between old and new units). Change it to 150R; at first, second and third attack
switch settings, attack will be faster than original, three times at least at first setting.
To faster both attack and release is possible by changing C13' 10uF capacitor as well. Lower is faster both att/release.
One thing, on too fast settings, sidechain can oscillate in some cases...at least, IIRC it did on breadboard
a years ago when I started to play arround 33609. However, IIRC, it was ground or power supply issue and was fixed...
Will play with this when will have a time next week.
Automatic time constants, A1/A2 possible to change as well.
Best way to do this: take unit in studio and play untill it smokes
Well, don't forget to switch off when changing components
Just kidding.
substitute
Well-known member
Wondering if the tamura pm1000 output buddy would be appropriate for this circuit. Can't find much in the way of a data sheet.
BOM for power supply (from 33609 thread):
1 LM317T IC2
3 0.1u 50V ceramic cap 5mm C2, C3, C4
2 1N4004 D1, D7
4 1N5400 3A fast rectifier diodes D9, D10, D11, D12
1 50r 1/4w 2% R4
1 196r 1/4w 2% R1
1 3k57 1/4w 2% R5
1 25k 1/4w 2% R2
1 10k mutiturn trimpot
2 100ux50v pitch 3.5mm low esr C5, C7
1 2200-3500u 50v low esr 7.5 or 10 mm pitch C13
1 TL071P single opamp IC1
1 Heatsink for lm317
Regarding the heatsink: preferred type is SK129.
Board can hold 4 types of "big" caps an 3 types of heatsink.
Heatsink (Farnell):
http://uk.farnell.com/jsp/search/browse.jsp;jsessionid=ERENQZVKUFT4VQFIAGJZLTQ?N=0&Ntk=gensearch_001&Ntt=sk129&Ntx=&_requestid=397515
1 LM317T IC2
3 0.1u 50V ceramic cap 5mm C2, C3, C4
2 1N4004 D1, D7
4 1N5400 3A fast rectifier diodes D9, D10, D11, D12
1 50r 1/4w 2% R4
1 196r 1/4w 2% R1
1 3k57 1/4w 2% R5
1 25k 1/4w 2% R2
1 10k mutiturn trimpot
2 100ux50v pitch 3.5mm low esr C5, C7
1 2200-3500u 50v low esr 7.5 or 10 mm pitch C13
1 TL071P single opamp IC1
1 Heatsink for lm317
Regarding the heatsink: preferred type is SK129.
Board can hold 4 types of "big" caps an 3 types of heatsink.
Heatsink (Farnell):
http://uk.farnell.com/jsp/search/browse.jsp;jsessionid=ERENQZVKUFT4VQFIAGJZLTQ?N=0&Ntk=gensearch_001&Ntt=sk129&Ntx=&_requestid=397515
bradb
Well-known member
LM317T Mouser 512-LM317T 1
0.1u 50V ceramic cap 5mm Mouser 80-C322C104M5U5HA 3
1N4004 Mouser 863-1N4004G 2
1N5400 3A fast rectifier diodes Mouser 863-1N5400G 4
50r 1/4w 2% Mouser 271-51-RC 1
196r 1/4w 2% Mouser 271-196-RC 1
3k57 1/4w 2% Mouser 271-3.57K-RC 1
25k 1/4w 2% Mouser 271-24.9K-RC 1
10k mutiturn trimpot Mouser 652-3299W-1-103LF 1
100ux50v pitch 3.5mm low esr Mouser 647-UPS1H101MPD 2
2200-3500u 50v low esr 7.5 or 10 mm pit Mouser 647-UVY1H472MRD 1
TL071P single opamp Mouser 595-TL071CP 1
Heatsink for lm317 SK129 Mouser 567-657-20ABP 1
I changed the "2200-3500u" cap to a 4700 uf / 50V.... i CANNOT find a low ESR version
I just yanked these part numbers from other posts in the 33609 thread. Please double check....
0.1u 50V ceramic cap 5mm Mouser 80-C322C104M5U5HA 3
1N4004 Mouser 863-1N4004G 2
1N5400 3A fast rectifier diodes Mouser 863-1N5400G 4
50r 1/4w 2% Mouser 271-51-RC 1
196r 1/4w 2% Mouser 271-196-RC 1
3k57 1/4w 2% Mouser 271-3.57K-RC 1
25k 1/4w 2% Mouser 271-24.9K-RC 1
10k mutiturn trimpot Mouser 652-3299W-1-103LF 1
100ux50v pitch 3.5mm low esr Mouser 647-UPS1H101MPD 2
2200-3500u 50v low esr 7.5 or 10 mm pit Mouser 647-UVY1H472MRD 1
TL071P single opamp Mouser 595-TL071CP 1
Heatsink for lm317 SK129 Mouser 567-657-20ABP 1
I changed the "2200-3500u" cap to a 4700 uf / 50V.... i CANNOT find a low ESR version
I just yanked these part numbers from other posts in the 33609 thread. Please double check....
substitute
Well-known member
dang yo, so fine!
MikoKensington
Well-known member
Igor, can we get a little more info on the metering? What type should we start with? How was your scale generated? Thanks.
Yap. I used ST7 from Farnell, link:
http://il.farnell.com/anders-electronics/st7/vu-meter-rect-0-200ua/dp/7758200?_requestid=182388
and printed the scale by myself.
Meter connected this way:
200 Ohm 1/4W resistor connected in parallel (across the meter)
Plus of meter goes to 680 Ohm 1/4W resistor
Minus of meter goes to front pannel, to pad marked (-mtr1) (next to bypass switch)
Free leg of 680R resistor goes to front pannel, to pad marked (+mtr1) (next to bypass switch)
All this stuff backlighted from bottom of meter (2xPLCC-2 red leds glued with 5 min sets epoxy).
LED's connected in series.
+VULED pad to + of first LED; - of first LED to + of second LED; - of second LED to -VULED pad.
It is a bit tricky to solder PLCC-2 led's, they are 2x2mm...
First glue them with small amount of epoxy, let glue set 5-10 min.
Glue piece of breadboard 5x10mm on back of meter.
Then pull some pieces from 24awg wire, connect led's in series,
then connect led's to breadboard...
here's terminals easy to connect to wires going from front pannel.
I like this meter. BTW, faster than Sifam several times...shows closer picture to real gain reduction.
Sifams are too slow, they don't show actual gain reduction on fast peaks.
Will post link to printable scale drawing and photos of backlighting stuff in few days.
Scale was generated simply by measuring gain reduction and marking the old scale with pen...as is
I took -20 db as last point of old scale...rest is simple.
http://il.farnell.com/anders-electronics/st7/vu-meter-rect-0-200ua/dp/7758200?_requestid=182388
and printed the scale by myself.
Meter connected this way:
200 Ohm 1/4W resistor connected in parallel (across the meter)
Plus of meter goes to 680 Ohm 1/4W resistor
Minus of meter goes to front pannel, to pad marked (-mtr1) (next to bypass switch)
Free leg of 680R resistor goes to front pannel, to pad marked (+mtr1) (next to bypass switch)
All this stuff backlighted from bottom of meter (2xPLCC-2 red leds glued with 5 min sets epoxy).
LED's connected in series.
+VULED pad to + of first LED; - of first LED to + of second LED; - of second LED to -VULED pad.
It is a bit tricky to solder PLCC-2 led's, they are 2x2mm...
First glue them with small amount of epoxy, let glue set 5-10 min.
Glue piece of breadboard 5x10mm on back of meter.
Then pull some pieces from 24awg wire, connect led's in series,
then connect led's to breadboard...
here's terminals easy to connect to wires going from front pannel.
I like this meter. BTW, faster than Sifam several times...shows closer picture to real gain reduction.
Sifams are too slow, they don't show actual gain reduction on fast peaks.
Will post link to printable scale drawing and photos of backlighting stuff in few days.
Scale was generated simply by measuring gain reduction and marking the old scale with pen...as is
I took -20 db as last point of old scale...rest is simple.
MikoKensington
Well-known member
Well that pretty much sums that up. Thank you, Igor.
juniorhifikit
Well-known member
matta said:
Hi - excuse my ignorance, but I have a couple transformer questions... What would we need a microphone transformer for? What would be a good Carnhill equivalent for the interstage transformer?
Thanks
Signal structure is: 0db input-> -6db after input trafo-> -31db approx. after input pad (diode bridge can't handle line level),
then, current buffers (same level), interstage trafo-> -25dB. Signal comes balanced to bridge compressor,
and output amp is unbalanced. Audio comes out of phase, CV in common phase. Besides buffers are ac-coupled to transformer,
there's still CV swing which we have to distract from audio...transformer take care about this.
Interstage converts balanced signal to unbalanced, and matching impedances as well.
Hope it clear, if someone can explain in normal english wellcome.
then, current buffers (same level), interstage trafo-> -25dB. Signal comes balanced to bridge compressor,
and output amp is unbalanced. Audio comes out of phase, CV in common phase. Besides buffers are ac-coupled to transformer,
there's still CV swing which we have to distract from audio...transformer take care about this.
Interstage converts balanced signal to unbalanced, and matching impedances as well.
Hope it clear, if someone can explain in normal english wellcome.
juniorhifikit
Well-known member
So then the Carnhill VTB9045 is a good candidate for the interstage transformer?
