New 'ultimate' SSL buss comp clone ;-)
- Thread starter SSLtech
- Start date
Help Support GroupDIY Audio Forum:
Purusha
Well-known member
Does anyone of you have a working ultimate clone with additional HP filters?
Do I wire them the same as I would be doing a Gyraf version?
Do I wire them the same as I would be doing a Gyraf version?
radiance
Well-known member
[quote author="Purusha"]Do I wire them the same as I would be doing a Gyraf version?[/quote]
Yes, only you need two side chain boards and a dual pole rotary for this....
Yes, only you need two side chain boards and a dual pole rotary for this....
Purusha
Well-known member
I thought so... thanks.
BTW, have you tried the mode switch yet? I wonder if this picture shows correct wiring?
BTW, have you tried the mode switch yet? I wonder if this picture shows correct wiring?
Purusha
Well-known member
BUMP
khstudio
Well-known member
#1 - I doesn't actually "SOUND" different... it WORKS different, which in turn makes it sound different.
This was discussed earlier in the thread. I don't think I could retype it any better. :wink:
#2 - NO
#3 - I'm DEFINITELY using a 47k... are you removing the wrong one? (I'm NOT sure if this makes a difference or not)
#4 - I'm not sure... you're making me want to recheck mine.
#5 - I'm not using a mode switch in mine... I think I even mentioned that .
Good Luck man... sorry I couldn't be more helpful.
khstudio
Well-known member
Freddy G,
I don't think you'll find one BUT... all you need is the GSSL parts list - TIMES 2
MINUS the parts you don't need twice - like 2 cases, power trannys, switches etc...
It depends on which build your going to do. I chose to load HALF of both boards (the AUDIO SECTION that is) everything else gets doubled EXCEPT for the second front/switch panel. Check out the pics I posted & you'll see.
You should read thru this & any other corresponding thread to get a grip on how it developed & the options you have.
I don't think you'll find one BUT... all you need is the GSSL parts list - TIMES 2
MINUS the parts you don't need twice - like 2 cases, power trannys, switches etc...
It depends on which build your going to do. I chose to load HALF of both boards (the AUDIO SECTION that is) everything else gets doubled EXCEPT for the second front/switch panel. Check out the pics I posted & you'll see.
You should read thru this & any other corresponding thread to get a grip on how it developed & the options you have.
khstudio
Well-known member
[quote author="SSLtech"]Yes, you can load up just the left side and the complete sidechain on each board, then couple the sidechains as described. Remember to use a 2-pole ratio switch, and wire one pole of it to each of the two ratio boards. You need to stuff only one set of attack/decay componente (the tantalum caps and associated high-value resistors on the switch boards).
Keith[/quote]
Keith[/quote]
khstudio
Well-known member
[quote author="Rob Flinn"]Here are the notes for the build. Can the next few builders please give feedback, so that we can amend any errors (hopefully there won't be any)
Ultimate GSSL mods
SEE PAGE 10 OF THREAD FOR AMENDED INSTRUCTIONS ![/quote]
Ultimate GSSL mods
SEE PAGE 10 OF THREAD FOR AMENDED INSTRUCTIONS ![/quote]
After working on this one for some time i brought it to the end, however I still have some minor issues to get sorted out. Mine main concern is heat which builds up on a certain components. Let me briefly inform you of what I've done...
Instead of using two set of boards, i just made another board (slave) in addition to the main (master) board. Master board is wired as usual
and it hosts side chain VCA (2180) and BOTH "audio" VCA'S (202), input and output circuits.
Slave board hosts another side chain, two HPF filters (Steffen), and relay bypass circuit. Six resistors and diode from the control board are included on this slave board, too. I connected CV from this (slave board side chain) to the jumper behind first VCA which is going to the 1k resistor, CV TO VCA'S from the Jacobs schematic.
Filter boards are fed from the "inputs" of 47K (mix resistors) and the outputs of the HPF are fed into 47k junction. There is also a switch that
selects between feeding both sends to a single HPF or splitting the sends
so every side chain gets one send.
All circuits are interconnected following tips from this thread.(page 10)
The unit is compressing and all of the functions appear to be OK, but:
1. 5532 on the filter boards are very hot ~50-60C a bit hotter and i could not keep my finger on it...They don't seem to want to go over this temperature...
2.same goes for the 5534 on the input, slightly cooler...
3.when i switch into oxford mode i have ~6dB of gain reduction more.
if Aarhus= 4db of GR oxford=10db of GR. Is this OK?
Hope some one could help to sort this out . Once we nail this, I'll post
layouts, if someone wants to go this way...
Steffen HPF
GSSL
HI RESO Photo
Instead of using two set of boards, i just made another board (slave) in addition to the main (master) board. Master board is wired as usual
and it hosts side chain VCA (2180) and BOTH "audio" VCA'S (202), input and output circuits.
Slave board hosts another side chain, two HPF filters (Steffen), and relay bypass circuit. Six resistors and diode from the control board are included on this slave board, too. I connected CV from this (slave board side chain) to the jumper behind first VCA which is going to the 1k resistor, CV TO VCA'S from the Jacobs schematic.
Filter boards are fed from the "inputs" of 47K (mix resistors) and the outputs of the HPF are fed into 47k junction. There is also a switch that
selects between feeding both sends to a single HPF or splitting the sends
so every side chain gets one send.
All circuits are interconnected following tips from this thread.(page 10)
The unit is compressing and all of the functions appear to be OK, but:
1. 5532 on the filter boards are very hot ~50-60C a bit hotter and i could not keep my finger on it...They don't seem to want to go over this temperature...
2.same goes for the 5534 on the input, slightly cooler...
3.when i switch into oxford mode i have ~6dB of gain reduction more.
if Aarhus= 4db of GR oxford=10db of GR. Is this OK?
Hope some one could help to sort this out . Once we nail this, I'll post
layouts, if someone wants to go this way...
Steffen HPF
GSSL
HI RESO Photo
mikeyB
Well-known member
[quote author="SSLtech"]Yep.
1xdbx202X or dbx202XT is a set of eight paralleled dbx2151's.
I used four of them for the audio signal path. That makes 32x2151's. Add the two sidechain 2151's, and you're at 34 VCAs for the entire unit.
Overkill? -Probably.
Since the 2151 is a reasonable -but not stunning- VCA, there are benefits. Joe Malone in Australia and PAD in America make 202 footprint units with 4x218x series (a rather better starting point) for similar -if not better-performance.
But I did it because I could. I had the bits, and I used them. I measured a small improvement, but -take note- the BIG STORY is the 'oxford' sidechain. and the REALLY big story is the ability to switch between Oxford and Aarhus modes.
Ask anyone who's built one (theres at least 3 or 4 out there at the moment) and they'll tel you the same thing: Killer. Makes the compressor into a two-personalitled monster, with BOTH personalities highly useful.
Keith[/quote]
Just found the PCB on JLM site but heres what chrisp had to say
"David
The GYRAF clone runs on +/- 15V, which is not enough juice for this board.
In any event, the GYRAF VCA chips are the same 7 pin SILs as those on this board, so you'd not gain very much (forgive the pun) compared to just stuffing the GYRAF PCB itself.
I've built a couple of the GYRAFs, and I know others on this forum who have done so as well - if you having any problems, perhaps we can help?
_________________
Chris P "
Keef - have you changed the voltages that your 202's run on or is there a fundamental difference not allowing the JLM to run on +/- 15V :?:
1xdbx202X or dbx202XT is a set of eight paralleled dbx2151's.
I used four of them for the audio signal path. That makes 32x2151's. Add the two sidechain 2151's, and you're at 34 VCAs for the entire unit.
Overkill? -Probably.
Since the 2151 is a reasonable -but not stunning- VCA, there are benefits. Joe Malone in Australia and PAD in America make 202 footprint units with 4x218x series (a rather better starting point) for similar -if not better-performance.
But I did it because I could. I had the bits, and I used them. I measured a small improvement, but -take note- the BIG STORY is the 'oxford' sidechain. and the REALLY big story is the ability to switch between Oxford and Aarhus modes.
Ask anyone who's built one (theres at least 3 or 4 out there at the moment) and they'll tel you the same thing: Killer. Makes the compressor into a two-personalitled monster, with BOTH personalities highly useful.
Keith[/quote]
Just found the PCB on JLM site but heres what chrisp had to say
"David
The GYRAF clone runs on +/- 15V, which is not enough juice for this board.
In any event, the GYRAF VCA chips are the same 7 pin SILs as those on this board, so you'd not gain very much (forgive the pun) compared to just stuffing the GYRAF PCB itself.
I've built a couple of the GYRAFs, and I know others on this forum who have done so as well - if you having any problems, perhaps we can help?
_________________
Chris P "
Keef - have you changed the voltages that your 202's run on or is there a fundamental difference not allowing the JLM to run on +/- 15V :?:
SSLtech
Well-known member
[quote author="syn"] 3.when i switch into oxford mode i have ~6dB of gain reduction more. if Aarhus= 4db of GR oxford=10db of GR. Is this OK?[/quote]
You have your switch backwards... I'll put money on it. Gain reduction will typically be more severe in Aarhus mode.
With signal fed to only ONE input, they should be exactly the SAME.
Firs test:
Connected across a mixer's stereo output, with a tone slowly sweep-panned from one side to the other, with a typical -3dB or thereabouts pan law, the Aarhus mode will compress more in the middle than at the sides. The Oxford mode will compress more at the sides than in the middle.
Second test:
Same tone fed into both chanels, but REVERSED-POLARITY into one of the channels. Oxford mode will compress the same as one channel only being fed. Aarhus mode will not compress...at all... ever.
Keith
You have your switch backwards... I'll put money on it. Gain reduction will typically be more severe in Aarhus mode.
With signal fed to only ONE input, they should be exactly the SAME.
Firs test:
Connected across a mixer's stereo output, with a tone slowly sweep-panned from one side to the other, with a typical -3dB or thereabouts pan law, the Aarhus mode will compress more in the middle than at the sides. The Oxford mode will compress more at the sides than in the middle.
Second test:
Same tone fed into both chanels, but REVERSED-POLARITY into one of the channels. Oxford mode will compress the same as one channel only being fed. Aarhus mode will not compress...at all... ever.
Keith
Keith
yeah, you were right, the ratio switch was backwards ( the pole that controls Oxford mode). Now that is O.K. Thanks a lot.
"Excessive input current will flow if a differential input voltage in excess of approximately 0.6 V is applied between the inputs, unless
some limiting resistance is used."
The above is from the 5532 data sheet but what does it mean exactly?
Regarding heat which builds up on my 5532, I came to conclusion that
for some reason they draw excessive current, as all voltages appear in good order...
Shall I try to limit current drawn by the 5532 from the PSU?
yeah, you were right, the ratio switch was backwards ( the pole that controls Oxford mode). Now that is O.K. Thanks a lot.
"Excessive input current will flow if a differential input voltage in excess of approximately 0.6 V is applied between the inputs, unless
some limiting resistance is used."
The above is from the 5532 data sheet but what does it mean exactly?
Regarding heat which builds up on my 5532, I came to conclusion that
for some reason they draw excessive current, as all voltages appear in good order...
Shall I try to limit current drawn by the 5532 from the PSU?
recnsci
Well-known member
Vozdra M.
[quote author="syn"]
"Excessive input current will flow if a differential input voltage in excess of approximately 0.6 V is applied between the inputs, unless
some limiting resistance is used."
The above is from the 5532 data sheet but what does it mean exactly?
[/quote]
There are protective diodes across input of 553x. So, if input is overdriven
one or other diode will conduct, thus protecting BE junction of input
transistors from potentially harmful current levels.
Little side note: it just appeared to me that these input protection
diodes limit slew rate somewhat, compared to what input LTP can
provide. Not that that this is of any importance to this thread or
any other thread.
[quote author="syn"]
Regarding heat which builds up on my 5532, I came to conclusion that
for some reason they draw excessive current, as all voltages appear in good order...
Shall I try to limit current drawn by the 5532 from the PSU?[/quote]
Nope, 5532s' (or any other opamps') current draw is determined by
opamp "guts", surrounding circuit and signal levels. Limiting PSUs
current will just starve opamp. In your case there is either something
wrong with opamps or surrounding circuit(s).
cheerz
urosh
[quote author="syn"]
"Excessive input current will flow if a differential input voltage in excess of approximately 0.6 V is applied between the inputs, unless
some limiting resistance is used."
The above is from the 5532 data sheet but what does it mean exactly?
[/quote]
There are protective diodes across input of 553x. So, if input is overdriven
one or other diode will conduct, thus protecting BE junction of input
transistors from potentially harmful current levels.
Little side note: it just appeared to me that these input protection
diodes limit slew rate somewhat, compared to what input LTP can
provide. Not that that this is of any importance to this thread or
any other thread.
[quote author="syn"]
Regarding heat which builds up on my 5532, I came to conclusion that
for some reason they draw excessive current, as all voltages appear in good order...
Shall I try to limit current drawn by the 5532 from the PSU?[/quote]
Nope, 5532s' (or any other opamps') current draw is determined by
opamp "guts", surrounding circuit and signal levels. Limiting PSUs
current will just starve opamp. In your case there is either something
wrong with opamps or surrounding circuit(s).
cheerz
urosh
Dr nEon
Well-known member
Hi all!
I was just about to get started on my second GSSL when I came across this new version of the project . . So I'm holding off from stuffing the board , until I ask these couple of questions..
I have read all 17 pages of this thread a few times, and it seems that the only builds so far are ones which use dbx202s.
I am very keen to explore this too , and want to try out the 'ultimate' version! . I have a stock of THAT2180LA's that I had intended to use in the standard gyraf version.
Has anyone out there built an 'ultimate' just using 2180's ??
I am wondering if there would be as much of an advantage doing the 'differential' mode, keef style , but with the 2180s , i.e only 2 vcas per channel..
Or even keeping it to the more simple method of stuffing left channel components only , on each 'channel' etc..
My main interest is in checking out the sidechain summing issue, which I think I'm gonna love , but I am not sure just how wild I need to go with mult-ing the vcas , since I'm using 2180a's...
Cheers!!
nEon x
I was just about to get started on my second GSSL when I came across this new version of the project . . So I'm holding off from stuffing the board , until I ask these couple of questions..
I have read all 17 pages of this thread a few times, and it seems that the only builds so far are ones which use dbx202s.
I am very keen to explore this too , and want to try out the 'ultimate' version! . I have a stock of THAT2180LA's that I had intended to use in the standard gyraf version.
Has anyone out there built an 'ultimate' just using 2180's ??
I am wondering if there would be as much of an advantage doing the 'differential' mode, keef style , but with the 2180s , i.e only 2 vcas per channel..
Or even keeping it to the more simple method of stuffing left channel components only , on each 'channel' etc..
My main interest is in checking out the sidechain summing issue, which I think I'm gonna love , but I am not sure just how wild I need to go with mult-ing the vcas , since I'm using 2180a's...
Cheers!!
nEon x
radiance
Well-known member
[quote author="Dr nEon"]
I am wondering if there would be as much of an advantage doing the 'differential' mode, keef style , but with the 2180s , i.e only 2 vcas per channel..
[/quote]
Someone correct me if I'm wrong but as I understand it, parallelling 2180's (as in the 202XT's) gives you lower noise where as the differential mode á la Keef gives you lower (or maybe even none) VCA distortion. The idea behind the 2 vca's per channel is that, since one VCA path is "out of phase" (or switched polarity), the distortion is canceled out when both paths are summed together.
I would say, just stuff both left and right channels and put all IC's in sockets. If you don't hear much of a difference you can always take out all IC's from one channel (which are the only expensive parts) and use them for another project.....
I am wondering if there would be as much of an advantage doing the 'differential' mode, keef style , but with the 2180s , i.e only 2 vcas per channel..
[/quote]
Someone correct me if I'm wrong but as I understand it, parallelling 2180's (as in the 202XT's) gives you lower noise where as the differential mode á la Keef gives you lower (or maybe even none) VCA distortion. The idea behind the 2 vca's per channel is that, since one VCA path is "out of phase" (or switched polarity), the distortion is canceled out when both paths are summed together.
I would say, just stuff both left and right channels and put all IC's in sockets. If you don't hear much of a difference you can always take out all IC's from one channel (which are the only expensive parts) and use them for another project.....
