Headroom Voltages Vs Sonics - More headroom isn't always a good thing?

[Electrobumps]

Congrats, you’ve been cute again and disinclined the senior elders from ever helping you again.  Sometimes one knows too little to know what they are hearing, or who they are hearing it from.  All you ask has been covered over and over and over again here.  The research material is deep.  Head into the stacks instead of asking for a thesis at the front desk.  Yeah, it’s ‘bus’.

Fair enough.  I don't know your credentials just as you don't know mine.  We are on completely different spectrums of our approach.  You're obviously extremely technical and I'm from a creative side of engineering. 

I learnt a lot from this post, I wanted to approach it from a sonics point of view. 

Operator VS designer in most cases are extremely different, so when they meet and discuss it can produce great things.

However chastising me for my choice of acceptable spelling does not help anyone.

 

[Michael Tibes]

The regular 1073 does not have a control after the output transformer.

Cheers

Ian

I know, but we could actually imagine using one at that point. And in the console isn't the fader behind the 1073 /1081 in the routing circuit (too lazy to pull out the manual now)?

I'm still interested in what exactly Electrobumps did there that sounded so great 8)

Michael

 

[emrr]

We would blow up snare drums at mix time with 31114’s.  Nice splatter. 

 

[Winston OBoogie]

I know, but we could actually imagine using one at that point. And in the console isn't the fader behind the 1073 /1081 in the routing circuit (too lazy to pull out the manual now)

In the big desks, yes the fader was post the channel amplifier in the routing module.  There was no provision for easily turning down the level from the pre stage into the 1073's output amp as this was a trim control on the rear of the module.

Of course, in a stand alone module application we normally put that trim control as an output level  (fader) on the  front panel.

Cranking the pre gain and trimming the output down for  level is a common trick on snare's and such.  We get more H2, H3 etc from that little, 3 stage pre amp due to the lower % of feedback.  And up to crashing into the 24V rail, it is also monotonic due to the simple resistive loading of the collectors. 
And if the source from the microphone is a hot one, we'll also get some saturation of the 10468 input transformer.

Great for adding thickness, body, tone, whatever...  if suitable for the track.

Maybe not something I would do if recording Daniel Barenboim at the Royal Albert Hall but then, YMMV 

 

[Winston OBoogie]

We would blow up snare drums at mix time with 31114’s.  Nice splatter.

Yep.  Me too 

One think I haven't mentioned is the recovery time from saturation.  It's pretty well established that if there is a clip of a transient with recovery that's of sufficiently short duration (a mS or so) then it is undetectable.
Some (most?) of this older solid-state stuff can be quite slow to recover, plus the signal doesn't clip symmetrically as it might (or should) in a well designed modern power amplifier.

 

[emrr]

Yep.  Me too 

One think I haven't mentioned is the recovery time from saturation.  It's pretty well established that if there is a clip of a transient with recovery that's of sufficiently short duration (a mS or so) then it is undetectable.
Some (most?) of this older solid-state stuff can be quite slow to recover, plus the signal doesn't clip symmetrically as it might (or should) in a well designed modern power amplifier.

Right.  RCA SS stuff will effectively begin to mute when overdriven severely.  You can put a drum into it and it becomes a gated crunch percussive. 

 

[Michael Tibes]

Maybe not something I would do if recording Daniel Barenboim at the Royal Albert Hall but then, YMMV 

Yep, but I guess him and I don't have much in common anyway, apart from going to the same physical therapist.

I love to run my Neves really hot during recording, but adding a pad or better a suitable fader (those old german radio faders should probably do the trick) seems definitely worth trying

Michael

 

[ruffrecords]

However chastising me for my choice of acceptable spelling does not help anyone.

I think chastising is a little strong. Poor spelling of bus is something that happens so often, even here, that it is frequently pointed out. It is meant as an educational thing. Nothing was meant personally. This isn't gearslutz - we don't do personal attacks.

Cheers

Ian

 

[Winston OBoogie]

I love to run my Neves really hot during recording, but adding a pad or better a suitable fader (those old german radio faders should probably do the trick) seems definitely worth trying

Hey Michael,
yes those German faders would do a terrific job of it.  Besides the distortion from the pre stage, you'd then be able to dial in (with the inter amp trim) the amount of output transformer juice in the thing before dialing in the level to your converter with the Maihak or whichever fader.
An easier or less expensive option for an output control would be just a standard 1K log pot across the output.  Should be fine unless you were running the signal all the way from the stage to the back of the Albert Hall. 

 

[Michael Tibes]

Yep, gotta find the boxes with these and those bloody tuchel connectors which seem to cost nearly as much as a U47 nowadys. Great inspiration! I have already been doing that with a V76, but never with a Neve. How silly...

Michael

 

[Winston OBoogie]

Right.  RCA SS stuff will effectively begin to mute when overdriven severely.  You can put a drum into it and it becomes a gated crunch percussive.

I have zero experience with any RCA SS gear but I do know that particular effect from other stuff 

*Edit:  something that just came to mind regarding overdriving solid-state line level stuff:  When I first met Andy Johns (he'd agreed  to endorse Universal Audio plus he and I also had a mutual friendship and relationship with both Jason Bonham and Jimmy), I was asking about some of the Zeppelin stuff he worked on (tons, if not most of it) and he said that the distortion on the main guitar riff in 'Black Dog' was just a DI signal from the Helios desk which fed an 1176 with gain turned up and which then fed a second 1176 on which he just turned down the levels until he got what was needed to feed to tape.  Course, some of that sound is also the compression J-Fet saturating as well as a teeny UTC 'O' series transformer getting the sh*te kicked out of it but still...  No guitar amp involved.

 

[Squeaky]

Thanks for that interesting snippet Winston, I had read somewhere (no idea where - you may have even been the source) that it was a DI signal and that a 1176 was involved but I didn't recall that there were two involved. Isn't the chorus guitar riff on No Quarter another well-known Zeppelin DI track?

Given this is a headroom topic, weren't Helios consoles were famous for their lack thereof? In addition, poor Arny (RIP) copped a lot of flack for the perceived lack of headroom of his reissue.

Certainly not for everything (e.g. not for classical guitar) but I often like to work on the edge of headroom and sometimes find myself gloriously tipping over into the realm of untold saturation and discrete component abuse. But I am a guitarist and I probably think of mic pres a bit too much like valve guitar amps.

 

[ruffrecords]

Given this is a headroom topic, weren't Helios consoles were famous for their lack thereof? In addition, poor Arny (RIP) copped a lot of flack for the perceived lack of headroom of his reissue.

The Helios consoles are very interesting.  They were unbalanced in a lot of places where Neve consoles fro example were balanced e.g. channel and bus inserts. For this reason, the output of the channel amp was usually unbalanced and at a level of around -8dBu in order to provide a decent headroom. Poor Arny's mistake was to try to build these as stand alone modules for people to use at +4dBu but of course this meant they threw away 12dB of headroom - hence the reputation. What he should have done was replace the EQ gain make up amp with the correct Helios design line amp which will drive and transformer and up the internal -8dBu level to +4dBu thus retaining the headroom. I am told Arny was warned of this at the time but did not action it.

The other interesting thing about the Helios channel amp is the bias of the input stage. For maximum headroom you would normally aim to bias the output transistor to close to half the rail voltage but the Helios first stage is biased some way off this point according to the schematic. This means one half of the waveform starts to clip before the other which results in relatively pleasant second harmonic distortion. I do not know if this was accident or design. If you swap the two main bias resistors the output stage becomes biased very close to the mid point. So It may be the preamp was designed to be biased at the mid point but an error in copying the design to the official schematic meant it was built with the bias off centre so to speak which accidentally led to its reputation for a musical sound. All speculation on my part but who knows?

Cheers

ian

 

[Winston OBoogie]

... Isn't the chorus guitar riff on No Quarter another well-known Zeppelin DI track?

...Given this is a headroom topic, weren't Helios consoles were famous for their lack thereof? In addition, poor Arny (RIP) copped a lot of flack for the perceived lack of headroom of his reissue.

Ya know, I don't actually remember talking about 'No Quarter' with Andy (another RIP and one who I personally miss) but def TWO 1176's on Black Dog.  I usually just let the conversation go where Andy felt like going rather than ask specific questions as he was very entertaining,  a great story teller, I'd be almost on the floor laughing at times.  Anyway...

Helios:  The very first desk  made for Olympic  was actually considered as being a bit "off" in terms of headroom by Glyn & Andy Johns, and I've heard Keith Grant  say similar.  I don't know what Olympic's operating level was (Abbey  Road operated lower than the  now accepted  .775v for instance)  but I think  headroom was an issue early on.   
The later desks, after Olympic's Helios 1,  evolved over time and the type 69 ended up with 36V*** running the output amps and then a simple zener reg on board which dropped the voltage down to 24V for the input amp.   

Ian is correct of course in that biasing off centre clips asymmetrically and will emphasise H2.  Whether this was accident or by design in the Helios, I don't know.


*** I've seen 'restored' modules from a very well known Helios desk where, despite the presence of this internal 24V zener reg feeding the input stage, the  whole module was powered by just 24V.    Wrong.

 

[Harpo]

Looking at opamps datasheets, increasing the supply voltage of a sending device in order to increase headroom for whatever reason seems as a receip for failure for the next device in chain when exceeding the abs.max. parts input voltage. "The magnitude of the input voltage must never exceed the magnitude of the supply voltage". Lots of common f.i. +/-15VDC supplied receiving devices (compressors, eqs, ...) don't come with input protection clamping diodes to their supply rails, so some peak voltage of the increased headroom sending device might damage or blow up the following input stage.

 

[JohnRoberts]

Looking at opamps datasheets, increasing the supply voltage of a sending device in order to increase headroom for whatever reason seems as a receip for failure for the next device in chain when exceeding the abs.max. parts input voltage. "The magnitude of the input voltage must never exceed the magnitude of the supply voltage". Lots of common f.i. +/-15VDC supplied receiving devices (compressors, eqs, ...) don't come with input protection clamping diodes to their supply rails, so some peak voltage of the increased headroom sending device might damage or blow up the following input stage.

Most modern ICs incorporate input clamping to prevent chip damage, but are supply limited to roughly 36V due to the typical semiconductor process used. We perceive audio loudness on a log scale so a few volts of power supply increase do not make a significant improvement, while increasing the heat dissipation proportionately and other less good parameters.

JR

 

[emrr]

Helios:  The very first desk  made for Olympic  was actually considered as being a bit "off" in terms of headroom by Glyn & Andy Johns, and I've heard Keith Grant  say similar.  I don't know what Olympic's operating level was (Abbey  Road operated lower than the  now accepted  .775v for instance)  but I think  headroom was an issue early on.   

Same is true with many of the earlier tube preamps as well, when used outside of a console context, and that was the stuff mentally modeled by those designers in terms of accepted headroom.  Standards changed.  Lots of modern complaints about headroom in reproduction REDD47's, etc.  A bit apples to oranges, but worth keeping in mind. 

 

[Winston OBoogie]

Same is true with many of the earlier tube preamps as well, when used outside of a console context, and that was the stuff mentally modeled by those designers in terms of accepted headroom.  Standards changed.  Lots of modern complaints about headroom in reproduction REDD47's, etc.  A bit apples to oranges, but worth keeping in mind.

Yessir, true that.  I've seen, on some of the better custom built valve desks, output amps that are capable of driving higher levels than the channel amps could.  An instance would be the original UA 610 desks which had 1016 amps  for the bus outs. 
Same deal if a desk used Langevin parts where there might be a 5116B on the output . 

Interesting on the REDD47 issues though, didn't know folks had had headroom problems.  Must've built, dunno?  100+ of those buggers myself and never had issues  but  anything's possible in other clones I suppose.

 

[Gertius]

Hi all,
I´m loking at racking an SSL 292 EQ card, but with a +-15V power supply instead of +-18V. Would you see a drawback in general sound/dynamics character, when doing it this way, or is it just a reduced headroom by a few dB (which I could live with).

Thanks & best,
Christian

 

[ruffrecords]

Hi all,
I´m loking at racking an SSL 292 EQ card, but with a +-15V power supply instead of +-18V. Would you see a drawback in general sound/dynamics character, when doing it this way, or is it just a reduced headroom by a few dB (which I could live with).

Thanks & best,
Christian

Just 1.6dB lost headroom is all.

Cheers

Ian