Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[steve p]

I need some help on what to look for in the SSL 4000 G EQs
Is there model numbers that go with the colors or a way to tell which EQ it is by looking at it?

Black, Brown, orange??

Can i tell by looking at the knob color?

Thanks
steve

[compasspnt]

Yes, there are differences.  But Keith (ssltech) can answer with great specificity.

Paging Keith, paging Keith...

[bblackwood]

Or Paul Frindle...

[compasspnt]

bblackwood wrote on Thu, 24 November 2005 07:23

Or Paul Frindle...

Of course!  After all, he did actually design them!

[Paul Frindle]

compasspnt wrote on Thu, 24 November 2005 14:21

bblackwood wrote on Thu, 24 November 2005 07:23 Or Paul Frindle... Of course!  After all, he did actually design them!

No, that's not strictly true and I'm sorry if people have somehow assumed this from my past posts? I did not design any version of the SSL console EQ before the G series version. However I did specifically develop the G Series EQ variant as part of (and along with) the rest of the G Series channel electronics.

But sadly after all this time I can't tell you the exact colours, except to say it is the version with the rather more pastel coloured controls and (more obviously) the one with the Bell/Shelf switches re-purposed as X3 freq scale selectors.

It is significantly different from the previous versions and the loss of the Bell/shelf function was a reluctant compromise that had to be made so that the EQ could retrofit the existing E Series console channel module.

The G series upgraded Input stages, EQ, Dynamics and modified Logic cards were originally conceived as retrofit and/or customer optional modules destined for the E Series consoles. It was only later that the G Series console was launched with the new colours and livery, to underscore the initiative - and allow further modification to the centre section electronics etc.

I hope this helps

[ssltech]

huh...? -Is it time to wake up? -I was napping hardcore!  -okay, here's the sequence:

Original SSL EQ had a brown-knob on the LF. It's a pretty straightforward constant-bandwidth design with

[stevieeastend]

Keith,

do you know something about the brand "Westec"? Does this name say anything to you?

cheers
steveeastend

[compasspnt]

Thanks Keith and Paul!

With you two guys around, there are no unanswerable SSL questions!

Fortunately mine has all three versions of EQ.  I did that intentionally for the multiplicity of choice.  I like all three versions for different reasons.  Which one is it that Bob C. likes most again?  And wasn't one somewhere supposed to mimic a Pultec?

And those G preamps definitely were a big improvement!

By the way, somewhere someone was asking about potential upgrades to the stereo buss...any ideas there?

Best regards!

[Paul Frindle]

ssltech wrote on Fri, 25 November 2005 17:23

huh...? -Is it time to wake up? -I was napping hardcore!  -okay, here's the sequence: Paul mentions that it wasn't quite constant-Q, but they certainly dropped the old constant-bandwidth behaviour, and the bandwidth basically used to widen rather a lot depending on how much you boosted or cut, and narrowed as you boosted (or cut) less. Keith

You remember more about it than I do However the gain/Q dependancy was actually the other way round to the way you describe it. In fact the effective Q lessened with lower boosts and cuts and therefore effectively widened the bandwidth for smaller amounts of EQ. The reason for this was to make delicate approaches to EQ more audible and thus encourage the user to employ somewhat less dramatic and more subtle EQ styles.
This was done to address the increasingly popular sonic styles of the time and counter some of the criticisms of the console sounding 'harsh' - due to the narrow bandwidth mid range boosts that had typified earlier production styles. The increase of max boost/cut from 15dB to 18dB was allowable due to reduced noise levels in the G series design.

The LF and HF shelves were also considerably re-worked with extra response poles to create an increased warmth (response overshoot bump) in the LF region and sharper freq definition (MF suck-out) in the HF regions - again to address the emerging styles of the times. BTW this also meant that the shelving sections were slightly unsymmetrical in boost and cut.

Although the G series was certainly a more 'musical' general purpose EQ that responded more naturally than the previous versions, the loss of high Q low gain cuts was a disadvantage when trying to tame resonances (especially percussion) etc.. Therefore some people found it disappointing under some conditions.

Later therefore when I did the Oxford EQ (released from panel constraints) I included 3 different selectable styles of EQ to address these problems. Specifically type 2 has unsymmetrical cut and boost curves to allow low gain very high Q resonance control in cut and overall character correction in boost - IMHO the best of both worlds when doing corrective EQ on percussion and other resonant instruments. The high effective Qs in cut are particularly useful as in a digital system the freq can be very finely and accurately set - and of course the stability of a digital system means it will stay put always

I also recall that I made mods to the mix bus arrangement in the G Series too (sorry to be so vague - it was all a long time ago). The original SSL mix busses were unbalanced and prone to magnetic pick up. We fell over this problem in Germany SWF in Baden baden where the mix busses picked up massive amounts of noise from the massive magnetic fields in the control room. At that time myself and a team flown out from SSL actually modified the whole console in situ to re-use the mix bus ground returns as a balanced references to buck out the magnetic pick up. My memory tells me that this method was installed in the G series console modules.

I also recall intending to change the fundamental mix amp and routing arrangement. The original one switched resistors to ground for the unrouted signals, effectively maintaining high mix amp gains permanently and making the mix buss noise unnecessarily constant regardless of how many channels were routed. This was done originally to maintain mix amp stability. I developed a mix amp that had enhanced stability that would tolerate wide gain ranges, therefore making the unrouted source resistors unnecessary so providing much lower buss noise under normal partial routing situations. But I can't remember if this made it's way into the console before I was taken off the case. Keith may be able to confirm this?

[ssltech]

Paul, It depends how you specify 'bandwidth', I suppose.

I did take a serious look at the 'G' EQ when it came out (we bought a few and installed them, but eventually removed them and I ended up designing my own, the "AAD"...) Using my default definition of the 'bandwidth' as being the distance between the two points at which the response deviates from flat (as opposed to distance between -3dB or any other arbitrary point) then I recall that the G did indeed get wide with taller boosts and deeper cut, not narrower as you might suggest. That was why it didn't do 'deep narrow notch' at all well: the bandwidth got so wide with progressively deepening cut that it started sucking lots of energy from the area around the center frequency, and pulled a lot of the character out along with any (f'rexample) untamable tom-tom ring, an area where the previous versions were more surgically useful, when drum tuning and mic selection/placement (you know, the REAL way to solve the problem!) failed.

So unless I'm remembering very wrongly, I reckon I got it the right way round.

The Oxford EQ is a work of elegance. I also built an analog EQ several years ago (fifteen or more... perhaps the similar vintage to the 'G' design) that -like the Oxford digital EQs that I've played with- offered the use of the (usually redundant) 'Q' -or Bandwidth- control in shelf mode. (In a shelf, by definition the bandwidth should theoretically be infinite, but the slope of the flter -which in bell mode is proportional to the 'Q'- can still be altered, significantly improving the tonal range of options in the way the unit can be set)

Compasspnt, yes there was an EQ-P version. The uptake was miniscule. The curves were said to be "based on the Pultec" -Note that they don't claim they were really anything like the same! -That was a fourth option, but it was heavily based on teh topology of the brown 82E02 and had differing ranges for most of the controls in each band.

Other aftermarket versions that were produced were the AAD (Amazon Audio Developments) EQE-1 and the Maselec (designed and built by Leif Mases as a further development from the AAD after he worked on a couple of albums on mine and decided that he would also like to try and sell a few of his design. We both sold a few, but neither of us got rich doing it!

-A little aside: that was quite some gathering of minds at the console for the first album back in 1989... Leif Mases producing, myself engineering and the house tech who was red-hot on SSL consoles also. -If anything dared to go wrong during the 6 months or so that it took to record and mix, it was a race to see who could call the component number on the board first!!!

I think I might have a few of the unstuffed AAD boards in case anyone ever wants to try a fourth flavour!

The "balanced" buses: -the problem really only started to manifest itself after consoles started to be purchased in ever longer frame-sizes. The unbalanced summing had worked fine for a long time and with small buss lengths, but over 56 channels (unheard of in 1979 when the first E-series went into Ridge Farm in Dorking) was where it really became a problem. the high buss gains and the long antenna-like buses started to become an issue. My standard test to check the bus noise rejection before and after the quasi-balanced buss implementations (which included replacing ALL of the lower buss cards) was to place a Weller soldering iron transformer directly under the bus cards. It was a good way to dial out the noise, though I did find that there was rather little range on the trim pots after the mod was done...

Hell all in all, it was a stellar system: to think that it was designed and built in the 1970s and many of them today still soldier on... Not the cleanest sound -specially the earlier ones- but one of the most easily serviceable -especially so when you consider the incredible power of their automation and that they pioneered recall- consoles. Rock-solid mechanical construction.

The Console at Parr Street Studios www.parrstreetstudios.com still has forty of the AAD EQ cards in the first forty channels, and it's where the last three Coldplay albums have been done, so it can't be all that bad, even though it's a simple design really!

Keith

[Level]

Quote:

If anything dared to go wrong during the 6 months or so that it took to record and mix, it was a race to see who could call the component number on the board first!!!

 

[ssltech]

Oops.

Correction to the url:

http://www.parrstreet.co.uk/

Quote:

Can i tell by looking at the knob color?

Well, the colour is only the cap colour, which can be changed. It's a guide, but the best way is to pull the module and look at the board number.

Keith

[Tim Halligan]

steveeastend wrote on Sat, 26 November 2005 02:53

do you know something about the brand "Westec"? Does this name say anything to you?

IIRC, there were two consoles that appeared on the market at around the same time - or were at least advertised - that appeared outwardly to be knock-offs of SSL desks.

E or G - can't remember.

One was the Westec - the other was the Saje ULN 2.

My (possibly faulty) memory of the Westec blurb was that they were offering automation on the small faders as well as the large.

Cheers,
Tim

[ssltech]

The Westec was in the era of the 'E'. They never sold very many, and the computer and it's software was so integral to the superiority of the SSL in the marketplace that it was silly to try.

SAJE was a French company that insisted on being rather innovative. I don't recall an SSL knockoff, though the SAJE Memory rings a bell as possibly being (from sketchy memory) a TR console at a time when Neve were still years away from having a competing product. If that classifies it as a knockoff then so be it, though it was no 'Clone', niether was the WESTEC, though it looked very reminiscent of an SSL. I remember the routing as having something very odd about it,and perhaps the display for the audio path sequence...

Here's a ULN-2:
http://store1.yimg.com/I/primalgear_1871_5204011

Looks as much like a V-series as an SSL in many ways: no big, illuminated Group/Tape/record buttons etc. \

That's the problem in some ways though: the 4000 series was so staggeringly logically and well laid out that you had to make something VERY similar to have a chance of competing. Colin Sanders had made a console with a very capable dynamics section in EVERY module. Unheard of. Try and sell a console without it from then on however, and you're simply not in the same market segment. SSL dominated so thoroughly in the mid-1980s that you HAD to build something laid out like an SSL or you were kissing that segment of the market goodbye! -Witness the Neve V-series similarity to the SSL 4000 in so many ways.


Keith

[Paul Frindle]

Dear Keith.

I think we may be getting mixed up with our Qs and B/Ws? It is in fact confusing because the effective Q of an EQ is difficult to define from the overall response as the 'flat' through path signal is always present. So we had to define the Q as being that of the resonant section that was being added into the overall response. This Q therefore was the defined by the -3dB points (relative to resonant peak) of the resonant section on it's own - and not the actual overall response of the EQ. The following link from the Oxford EQ detail illustrates my point.

   http://www.sonyoxford.co.uk/pub/plugins-sony/products/eq-ban d.htm

The type 1 curve is most like the original clinical EQs in that the Q of the resonant section (which pokes above the flat line) is pretty constant. This is the most obvious technical interpretation of a parametric EQ that many adopted in the 70's and 80's. This allows quite tight Hi Q (narrow B/W) boosts and dips even at low gains.

The type 3 curve is most like the G series that I designed whilst at SSL (unless I am mistaken and I've truly lost the plot - which is always possible I grant you). In this type the effective Q of the resonant section is varied with cut and boost gain, at low gains it starts at a lower Q (wider B/W) and ends up at higher Qs (relatively narrower peak bandwidth) for higher gains. This behaviour is less 'clinically accurate' and in some ways partially mimicks earlier pre SSL EQ designs where the Q of the resonant sections were not fully decoupled from the gain feedback paths (or indeed the gain itself was actually primarily derived from the Q of the resonant section - i.e. passive LCR circuits etc).
Type 4 is an extreme example of this kind of dependancy which I included in the Oxford because of it's ability to put very low Q and extremely subtle 'sheens' and characters to programme at low EQ gain settings.

That's what I mean

The type 2 with the assymetrical response is my prefered best case curve where resonances need to be tamed but overall response corrections are still required (i.e. especially percussion etc).