bblackwood
Active member
- Joined
- Jun 3, 2004
- Messages
- 33
The real question is - if you're listening at 90dB SPL or so, can you hear it when you stop the tape?
If so, it's too noisy, imo.
If so, it's too noisy, imo.
signal to noise in the GSSL comps.
- Thread starter j.hall
- Start date
Help Support GroupDIY Audio Forum:
[quote author="Kev"]
I meant something a little more main stream and a little more analog from one end to the other
I can't remember the designers name but I think I did exchange an email or had a conversation on a forum ... these have a few tricks inside that put them just left of centre in the world of comps
[/quote]
GML 8900
I meant something a little more main stream and a little more analog from one end to the other
I can't remember the designers name but I think I did exchange an email or had a conversation on a forum ... these have a few tricks inside that put them just left of centre in the world of comps
[/quote]
GML 8900
bblackwood
Active member
- Joined
- Jun 3, 2004
- Messages
- 33
[quote author="deveng"]
Just for your information this is the specs on the SSL version of the SSL stereo bus compressor:
Audio Performance -
Input XLR Electronically Balanced, Impedance 20 Kohm
Output Output XLR Electronically Balanced, Impedance 40 ohm
Frequency Response 20Hz to 20 kHz ±0.2dB
Dynamic Range >110dB
Noise <-85dBu (-90dBu typical)
RMS 20Hz to 22kHz at output, input terminated with 50 ohm resistor [/quote]
The SSL is close to 25dB quieter than the clone?
Yikes!
Just for your information this is the specs on the SSL version of the SSL stereo bus compressor:
Audio Performance -
Input XLR Electronically Balanced, Impedance 20 Kohm
Output Output XLR Electronically Balanced, Impedance 40 ohm
Frequency Response 20Hz to 20 kHz ±0.2dB
Dynamic Range >110dB
Noise <-85dBu (-90dBu typical)
RMS 20Hz to 22kHz at output, input terminated with 50 ohm resistor [/quote]
The SSL is close to 25dB quieter than the clone?
Yikes!
Kev
Well-known member
[quote author="j.hall"]GML 8900[/quote]
:green: :green: :green:
:roll:
such a show off is our John
:green: :green: :green:
:roll:
such a show off is our John
[quote author="Kev"][quote author="j.hall"]GML 8900[/quote]
:green: :green: :green:
:roll:
such a show off is our John[/quote]
it's actually jason.....and i'm not showing off, you asked for real world comps that achieve 90 to 100 db of s/n ratio....
even the original SSL that has been copied here can do -90...................
empirical labs distressor
crane song trakker
pendulum
chandler
avalon
innerTube audio
manley
purple
smart
millenia
fred forssell
the list goes on......
if all of those various designers and manufacturers can pull it off.....then i think arguing the fact is moot.
:green: :green: :green:
:roll:
such a show off is our John[/quote]
it's actually jason.....and i'm not showing off, you asked for real world comps that achieve 90 to 100 db of s/n ratio....
even the original SSL that has been copied here can do -90...................
empirical labs distressor
crane song trakker
pendulum
chandler
avalon
innerTube audio
manley
purple
smart
millenia
fred forssell
the list goes on......
if all of those various designers and manufacturers can pull it off.....then i think arguing the fact is moot.
NewYorkDave
Well-known member
Let's bring a little logic into the discussion.
You measure your SSL clone at 60dB S/N.
Others who've built the same report that they're getting S/N of 80-85dB.
The original SSL unit achieves 85-90dB.
Obviously, something is wrong either with your clone or your measurements. So it stands to reason that there's a problem that can be fixed. But if you insist on a 100dB S/N, why waste your time with cloning a design that's only capable of 90dB S/N on a good day?
By the way, have you measured S/N on your other pieces of gear using the Terrasonde and the same methodology? Do they all meet spec, or are you just taking the manufacturers' word for it?
You measure your SSL clone at 60dB S/N.
Others who've built the same report that they're getting S/N of 80-85dB.
The original SSL unit achieves 85-90dB.
Obviously, something is wrong either with your clone or your measurements. So it stands to reason that there's a problem that can be fixed. But if you insist on a 100dB S/N, why waste your time with cloning a design that's only capable of 90dB S/N on a good day?
By the way, have you measured S/N on your other pieces of gear using the Terrasonde and the same methodology? Do they all meet spec, or are you just taking the manufacturers' word for it?
[quote author="NewYorkDave"]Let's bring a little logic into the discussion.
You measure your SSL clone at 60dB S/N.
[/quote]
-65 at unity gain, -50 at full gain.
that was my original question......how can i fix this?
fair enough....
i don't recall insisting that this unit must be at -100, i think i just mentioned that that would be ideal. -90 would be a VAST improvement over -65
YES i have measured them. some are better then others....but some of my comps are better then others as well. if that makes any sense.
overall they all meet the manufacturers spec. on average.
You measure your SSL clone at 60dB S/N.
[/quote]
-65 at unity gain, -50 at full gain.
that was my original question......how can i fix this?
fair enough....
i don't recall insisting that this unit must be at -100, i think i just mentioned that that would be ideal. -90 would be a VAST improvement over -65
YES i have measured them. some are better then others....but some of my comps are better then others as well. if that makes any sense.
overall they all meet the manufacturers spec. on average.
I think i measured around 85-90 on my GSSL with that rightmark shiz, but I've since heard that the app aint so good and soundcard factors come in etc (tested on a Pulsar2 Pro)...anyway, all i know is It's dead quiet and pumps!
I take it u used metal resistors and sorted a proper grounding scheme? I had to isolate the pcb from the metal chassis as the copper runs right around all 4 mounting holes (i used the washer thingies used on vregs)...
I take it u used metal resistors and sorted a proper grounding scheme? I had to isolate the pcb from the metal chassis as the copper runs right around all 4 mounting holes (i used the washer thingies used on vregs)...
NewYorkDave
Well-known member
Do you measure the same noise level on both channels?
Have you tried swapping out the opamps?
Did you use metal-film resistors?
What's the spectral distribution of the noise?
Have you tried swapping out the opamps?
Did you use metal-film resistors?
What's the spectral distribution of the noise?
ChrisA
Well-known member
I built a few ssl clones with high grade caps and resistors... they are all very quiet...I would suggest maybe looking into what parts you used to make your clone....
Kev
Well-known member
[quote author="j.hall"]it's actually jason.....[/quote]
sorry jason
too many names to remember, from too many forums
there is a John Hall ... I used to work with one
... I'm getting old and I make mistakes
I said main stream so people could set up a similar test and compare results that's all.
[quote author="NYdave"]What's the spectral distribution of the noise?[/quote]
It could be components or it could be the method of mounting and location of power transformer. A picture of the insides could help us to see how things have been done.
sorry jason
too many names to remember, from too many forums
there is a John Hall ... I used to work with one
... I'm getting old and I make mistakes
I said main stream so people could set up a similar test and compare results that's all.
[quote author="NYdave"]What's the spectral distribution of the noise?[/quote]
It could be components or it could be the method of mounting and location of power transformer. A picture of the insides could help us to see how things have been done.
Mundox
Well-known member
The first one I made had a self noise of about -60dbfs! That's horrible, but then I had printed/etched the pcb manually, and used a smaller box.
The last one I made has a self noise of about -90dbfs! All Pro.
That's progress! :grin: [/i]
edit: tested on "Mac the Scope" on G4 powerbook.
The last one I made has a self noise of about -90dbfs! All Pro.
That's progress! :grin: [/i]
edit: tested on "Mac the Scope" on G4 powerbook.
Just finished my first SSL clone buildt with Gustav pcb's and dbx1252 vca's.
I used the 27K resistor for unity gain and modified the vca control voltage for correkt ratios.
I've measured S/N with a newly calibrated Neutrik A2D and got 99dB unweighted and 114dB Aweighted.
And thats at unitygain
So no problems here.
I'am not sure the terrasonde can measure these low values, or output the 22dBu needed to drive the ssl to 1% thd, so thats maybe why you get your results.
Thanks to all off you here for making these projects available
I used the 27K resistor for unity gain and modified the vca control voltage for correkt ratios.
I've measured S/N with a newly calibrated Neutrik A2D and got 99dB unweighted and 114dB Aweighted.
And thats at unitygain
So no problems here.
I'am not sure the terrasonde can measure these low values, or output the 22dBu needed to drive the ssl to 1% thd, so thats maybe why you get your results.
Thanks to all off you here for making these projects available
Thanks, Repairman :thumb:
..and welcome to "The Lab"..!
Jakob E.
..and welcome to "The Lab"..!
Jakob E.
Svart
Well-known member
i dunno but I built two pairs of GSSL, the first two using resistors within 5% of the values, but identical to each other) because these were free metal films.. the second pair was done with exact value metal 1% resistors. the pairs work different from each other. the exact value comps when set to 2:1 barely compress at all, the within 5% pair when set to 2:1 compress like the other ones set to 4:1..
a very small difference in value seems to make a much larger difference in operation.
a very small difference in value seems to make a much larger difference in operation.
radiance
Well-known member
I did not meassure anything, just using my ears.
My SSL clone is more quiet than my Focusrite Red 3 compressor.
(I don't know wheiter this is a good thing or not)
Can someone explain to me how to meassure the signal to noise ratio?
I have no test equipment, only Logic audio with a motu 828 soundcard.
Do I just hook the thing up to an input of the 828 and watch the input meter?
My SSL clone is more quiet than my Focusrite Red 3 compressor.
(I don't know wheiter this is a good thing or not)
Can someone explain to me how to meassure the signal to noise ratio?
I have no test equipment, only Logic audio with a motu 828 soundcard.
Do I just hook the thing up to an input of the 828 and watch the input meter?
[quote author="NewYorkDave"]Do you measure the same noise level on both channels?
[/quote]
sorry guys, i've had the flu and gettingon the net hasn't been that important the past few days.....feeling better so here i am.
anyway....yes, same noise on both sides
no, like i said, i've been sick....so i'm ready to try whatever now.
1% metal films through out.
i did sweeps from 20 to 20k....is that what you mean?
BRepairman wrote:[/b]
i'm pretty sure the unit performs the test properly. but assuming it doesn't, the noise floor of the box with nothing going through it at unity gain is -65dB.....it doesn't take much test equipment to get that right.
hey Kev, no big deal on the name.......
i'll take a pic later when i have time
i isolated the PCB with rubber washers, the toroid is on the far side of the case while the PCB is on the other.
[/quote]
sorry guys, i've had the flu and gettingon the net hasn't been that important the past few days.....feeling better so here i am.
anyway....yes, same noise on both sides
no, like i said, i've been sick....so i'm ready to try whatever now.
1% metal films through out.
i did sweeps from 20 to 20k....is that what you mean?
BRepairman wrote:[/b]
i'm pretty sure the unit performs the test properly. but assuming it doesn't, the noise floor of the box with nothing going through it at unity gain is -65dB.....it doesn't take much test equipment to get that right.
hey Kev, no big deal on the name.......
i'll take a pic later when i have time
i isolated the PCB with rubber washers, the toroid is on the far side of the case while the PCB is on the other.
Kev
Well-known member
[quote author="j.hall"]
i did sweeps from 20 to 20k....is that what you mean?[/quote]
I think Dave may have been looking for a specific noise like a mains induced hum or buzz.
By looking at the spectrum ... or even zoom in on the output with lots of gain you may idenity the noise as something specific.
Some like to use a high gain power amp with a probe to zoom in on noises. Many specific noises are recognisable and therefore can lead to a solution.
i did sweeps from 20 to 20k....is that what you mean?[/quote]
I think Dave may have been looking for a specific noise like a mains induced hum or buzz.
By looking at the spectrum ... or even zoom in on the output with lots of gain you may idenity the noise as something specific.
Some like to use a high gain power amp with a probe to zoom in on noises. Many specific noises are recognisable and therefore can lead to a solution.
