Driving +4 gear into -10 inputs will increase the noise floor.

I'm not real sure about this but in case no one else has an answer I try here.

The noise floor of the +4 device stays the same but the dynamic range of the device is effectively reduced by 14dB, increasing the noise floor by the same amount. (I think this is how it works)

Someone like Kev who understands all the ins and outs of the science side of these issues would be able to tell you why but I do know that gear that runs at +4 sounds better to me.

If this helps at all:

The +4 dBu/dBm standard operating level is an average expressed as 0 VU for 100% modulation. This is based on a 1 VU = 1dB for a continuous sine wave... not complex music. Music can blow this standard out the window.

So whatever your noise floor is (in -dBu... the "low" side) the difference to +4 dBu
is the signal to noise ratio. But the closer you get to the noise floor with your device, the less usable the sound.

But on the upper side you have distortion at the maximum output level. This is 20dB for most pro gear.

If a VU is reading or steady around "0" it is just an average representaion of the ear and could swing out to +15 or 20 resulting in distortion.

Regardless of your cable run, I'd say you are safe as long as you don't stay at +3 VU or more for too long. It's really about finding the furthest point from the noise-floor without distorting.

In radio, they tell us to keep it around 85-90% since there is another processor in the chain keeping everything 95-100%.

Hope this helps a little!

Thanks and....

Considering the raw quantity and level of discussion around such esoteric topics as the nickel ratio in a given transformer, it seems like this is a topic we should understand well since virtually every hunk of gear made must make assumptions about its output level and input calibration, and many of them offer us a trim choice. This would also seem to impact the impedance match if I understand ohm's law (not ohm's opinion, right?).

Thanks for your answers they are helpful but if I lose 14 db of dynamic range with a +4 input (why would the pro standard have less dynamic range???) telll me again why I would do it? (btw, I now default my input sensitivity to +4, but only because its the "pro" standard).

My main question remains a mystery - if I force my preamp to make more gain by matching it to a lower sensitivity input (ie, a +4 input assumes higher input gain, right?) then, since its working harder, there's simply more to go wrong under my non scientific assumption that the closer a device gets to doing nothing the more transparant it will be. :). I do understand that making your gain structure work overtime can be a good thing in the tone dept., and often drive things harder for tone, but I want to stant with bedrock.

This thread risks going in circles and trying to answer too many questions at once will just make it more complicated.

Give me some specifics and one situation and name the gear
sorry to be pedantic but it just has to be that way

Kurt Foster wrote: Driving +4 gear into -10 inputs will increase the noise floor.
I'm not real sure about this but in case no one else has an answer I try here.

sort of kind of :roll:

BUT
it doesn't have to be that way

as I said above
give me a specific situation and I'll do my best to paint a picture

ok then...

Here's a specific case

My Pendulum quartet was outputting at +4 into my rme fireface which I had set at -10. of course 0vu on the output of the quartet would overload the -10 input, so I just turned down the preamp so that the computer vu's would give me 0 db.

Would I have been better off with the attenuation in the trim (rather than earlier in the chain) from an audio quality standpoint?

Re: ok then...

tmcconnell wrote: Would I have been better off with the attenuation in the trim (rather than earlier in the chain) from an audio quality standpoint?

earlier in the chain ??

no time right now to check for facts ... so let Kev make a fool of himself and make some assumptions
:roll:
never assume

but here goes

By lowering the output you probably have introduced some noise that you didn't have to.
and not making use or the Headroom available and dropping the operating level down toward the noise ... -10dB levels ... a difference of around 14dB plus.
should I explain more here ?

A couple of things,
The Pendulum is probably a BIG output unit and is probably at it's best up there.
Try a terminating resistor of 1K or 600ohm and even try 300ohms to put load on the output and get some IRON sound going on.
Then use a simple attenuator to KNOCK back that 20+ back down to where the rme fireface will crash.

as I said no paper work in front of me so I can't be sure of the connections.
6.5mm ?? An XL3 to 6.5mm adapter lead with a resistive divider as a PAD.
Yes there is a possibility of ground loops.

If you are included to spend a little money and time I'd buy a Transformer and set up an INTERFACE with a trim pot so you can dial in the sound you want.

errr
interested ?
Jensen Trafo site will have a drawing of what I mean.

I think I get it now.

Kev thanks - the rme does have a +4 setting so I don't need to go to al the trouble of building an interface. The points I was after are:
1) The pre will perform better closer to its nominal operating level, 2) The output transformer will sound better (iron) into a +4 trimmed load.

I was suffering under the illusion that amplification circuits would generally sound cleaner if they were doing less "work". thanks. t.

I was suffering under the illusion that amplification circuits would generally sound cleaner if they were doing less "work". thanks. t.

For Mosfet Power Amps ... the THD is at it's least just before the amp clips.
So it is not that simple.

I can finish what I was saying if you like.

1) The pre will perform better closer to its nominal operating level

YES

2) The output transformer will sound better (iron) into a +4 trimmed load.

YES
Loaded and perhaps attentuated for the RME

WAIT a minute !!
you have a Pendulum QuartetII ??

see the
http://www.pendulumaudio.com/QuartetIImanual.html
extract
OUTPUT: 0 to 10
The OUTPUT Control is a passive attenuator positioned between the mic preamp tube stage and the other elements of the Quartet II. It can be used in conjunction with the GAIN control to adjust the level that is sent to the EQ, Compressor or Peak Limiter (or a tape machine or hard disk recorder, if the other elements are bypassed). For the cleanest sound, set the OUTPUT control fully clockwise (completely out of the signal path) and use the GAIN control to set the overall signal level. To add more 'harmonic content' by driving the tubes at a higher signal level, set the GAIN control to a higher gain setting and use the OUTPUT attenuator to bring the output level down to a more useable range. Or, use the OUTPUT control to make fine adjustments (within the 3dB range of the GAIN switch settings) to the level sent to a tape machine. Used sparingly, this gentle tube 'overdrive' can create subtle changes in the harmonic balance of the source. Used to excess, it will cause audible distortion. Proceed with caution. With the OUTPUT attenuator set to mid-range, the level of attenuation is 20dB for load impedances greater than 10k.

This solves the level and the impedance to the RME ... doesn't it ?

maybe

Yes. (Actually, the version I has the same control). Its just another place I can attenuate the gain. I think the output transformer of the quartet is on the balanced out, which is after all the stages of the quartet.

None the less, if I attenuate it there, I am still driving the downstream stages of the quartet including the output transformer at well below their nominal ability, so your rules would still apply.

... , if I attenuate it there, I am still driving the downstream stages of the quartet including the output transformer at well below their nominal ability

we can not be completely sure when that output pot sits in the signal chain.
if it is in fact just before the output drive section and not at the beginning of the chain then
NO
the preceding sections will be at nominal level.

This unit is little more complicated than first though as it does seem to have options.

Would be better to see a block diagram of this as it is a bit confusing when just looking a the words.

but looking at the back panel it does make more sense

you have the Quartet?
http://www.pendulumaudio.com/Quartet.html
the manual does suggest some operational configs
http://www.pendulumaudio.com/Quartetmanual.html

extract
Quartet
Max. Output Level: +32dBu into 10k load, Output control at maximum, unbalanced or transformer-balanced
+24dBu into 600 load, Output control at maximum, unbalanced or transformer-balanced

see how the output voltage drops as you present a load .... just like power amps and 8, 4, and 2 ohm loads

Quartet II
extract
DI Preamp
Max. Output Level:+34dBu into 10k load, Output control at maximum, unbalanced or
transformer-balanced
+25dBu into 600 load, Output control at maximum, unbalanced or transformer-balanced

extract
Peak Limiter
Output: +22dBu into 10k load

it is hard being an expert when you haven't seem or touched the unit
:D

Exactly which RME do you have and do you know of a web page that gives me the input specs ?

Back to my attenuator interface gadget
refer to Jensen apps
as089.pdf
as092.pdf
for ideas
more later

yes, the quartet

I'm certain the pre out is the feed to the downstream devices because I have to change the threshold on the compressor if I up the gain onthe output of the pre section.

You've helped a lot. I'm going to start running the daw inputs at the +4 level so I can run all my pre's hotter. I know this might have seemed obvious, but you know, they gave me a choice I didn;t fully understand. And some people would say "just go with your ears" - and I support that fully, but I still like to know what to listen for. :).

Re: yes, the quartet

tmcconnell wrote: I'm certain the pre out is the feed to the downstream devices because I have to change the threshold on the compressor if I up the gain onthe output of the pre section.

:shock:
that seems really stupid ??
please double check
really do need a block diagram
:(

It is important to get the signal level and gain structure correct for all units to work at their nominal levels.

Which RME do you have ... spec sheet location ?