padding the output of a preamp? padding values questions..

[Svart]

I have a pair of 1272s that I love dearly but they have a problem, when feeding them with hot mics I literally have to turn the gain almost ALL the way down.. let's say that the dial goes from 1 to 10, i would have to turn them down to like .25 to keep them from clipping the converters following them. I would usually pad the inputs but this creates another problem though, once padded down that far, the signal is usable but the self noise(sounds like pink noise, not hum) starts to get really noticable during quiet passages.

I suppose the only way to make these usable is to pad the outputs.

So my first guess at a 20db pad would be a 600:600 H pad with series values of ~246R and a shunt of 121R

OR

a 600:~2k H pad with series values on the 600R side of 196R, series values of the 2k side being 910R and the shunt being 221R.

Anyone actually try something like this and have any advice or experiences I should watch out for?

Do these values seem ok?

 

[NewYorkDave]

1K A-taper pot at the converter input... Dial in the amount of attenuation you need and rock on.

The output of your preamp is transformer-coupled so it doesn't care if the load is balanced, unbalanced, grounded or floating in space.

 

[Svart]

Thanks NYD we can always count on you!

I'll give it a try but I think I might have to stay balanced due to a 25ft run between the preamps and the converters that unfortunately routes near a large AC area and can't be easily changed right now.

In case of problems, any other suggestions?

I have a client and I won't have much time to cobble together something to work and test it so I'd like as many possibilities as I can in case Murphy's law kicks into full effect if you know what I mean..

:green:

 

[JohnRoberts]

[quote author="Svart"]Thanks NYD we can always count on you!

I'll give it a try but I think I might have to stay balanced due to a 25ft run between the preamps and the converters that unfortunately routes near a large AC area and can't be easily changed right now.

In case of problems, any other suggestions?

I have a client and I won't have much time to cobble together something to work and test it so I'd like as many possibilities as I can in case Murphy's law kicks into full effect if you know what I mean..

:green:[/quote]

You could make the pad balanced by adding two resistors and wiring the pot as a rheostat (variable resiustance).

JR

 

[NewYorkDave]

Yep... You have a certain minimum loss when doing it that way, but it sounds like that wouldn't be a problem in his application.

Two examples:

1
2

 

[skipwave]

I planned to whip up a small box with a pair of pads using Dave's example 1, because I thought I had a pair of nice 1k pots in the drawer. Well, now I only find one pot, so instead I'll use some 12 position rotary switches.

Using the basic formula for a U-pad, I calculated the values for the Rshunt switch in 3db steps. Would anyone be so kind as to check my results?

Also, is there any power dissipated in the Rshunt resistor, when we are merely shorting two opposite polarity signals across it? If so, it seems like it couldn't be the same as the power dissipated in a shunt resistor connected to ground, or is it?

My Rseries resistors are 301 ohms each. Sorry about all the decimal spaces, I should have rounded.

 

[Kit]

I suppose the only way to make these usable is to pad the outputs.

This might sound stupid but.....................maybe you could just move the microphone away from the source. Let some "air" into the mix.

 

[skipwave]

[quote author="Kit"] Let some "air" into the mix.[/quote]

Haven't you read the audiophool sites? Only C37 special lacquer coating adds more "air" to your recordings. :green:

 

[mikep]

you could reduce the gain of the 1272. there is a place in the circuit where you can put an attenuator, between the two gain stages. different sound, the output won't be run as hot, but that is how it was designed. do you have the docs for the 1272?

 

[skipwave]

I should have started a new thread, since I'm building an external box to be used with various preamps. In my case, primarily tube preamps and some fixed gain modules, basically anywhere that there is no convenient place to insert gain or volume controls.

It may also see some use on a 1272, but I understand how an attenuator could be inserted between stages. The schematic makes it easy to see, split into the two stages:

http://members.nuvox.net/~zt.robgrow/circuits/neve/neveba283.html

 

[clintrubber]

FWIW, more attenuation stuff here:

HP Attenuator: 1 to 110 dB, DC-1MHz, 600 Ohm

http://www.hparchive.com/Manuals/HP-350C-350D-Manual.pdf
(8.8 MB)

 

[skipwave]

Anyone care to comment on the power dissipation in the shunt resistor (the potentiometer in Dave's plan)?

It occurred to me that with a variable Rshunt, you could turn it all the way down, shorting out the entire signal. In that case the voltage is nulled, but what of the current? Is the power then dissipated in the series resistors, since one side of those resistors has the full signal applied to it and the other end is effectively at 0v?

 

[clintrubber]

[quote author="skipwave"]I planned to whip up a small box with a pair of pads using Dave's example 1, because I thought I had a pair of nice 1k pots in the drawer. Well, now I only find one pot, so instead I'll use some 12 position rotary switches.[/quote]
Anyone still got the link to those stepped attenuators NYD posted a while back ? I have them myself on paper, but let's add these to this thread as well.
As NYD said, the two circuits he already suggested will do the trick (either with a pot or a single-deck rotary) but if you need or want to go 'further' then a few other drawings from his hand can come in handy as well.

One of them was a 'dual' 600 Ohm attenuator: two rotaries; one for 10dB and the second one for 1 dB steps. (And the circuit had a termination switch for on/off 600 Ohm termination; FWIW, you'll lose the calibration when switching off the 600 termination). For an unbalanced version the switches need to be a two-deck each; three-deck for a balanced version.

That 10dB & 1 dB attenuator is what I've recently been gathering the parts for: cranking a Neve-style preamp, pushing the TXs with healthy signals and then attenuate before the DAW, sounds like your situation from the first post of this thread :thumb:

There was also a one-stage attenuator before (2dB steps ?) IIRIC both NYD & WOB posted circuits. That one is half the work of the first one and most likely enough for what you're looking for (if you didn't go for the single-deck pot-replacement already - as mentioned above).


Regards,

Peter

 

[clintrubber]

[quote author="skipwave"]Anyone care to comment on the power dissipation in the shunt resistor (the potentiometer in Dave's plan)?

It occurred to me that with a variable Rshunt, you could turn it all the way down, shorting out the entire signal. In that case the voltage is nulled, but what of the current? Is the power then dissipated in the series resistors, since one side of those resistors has the full signal applied to it and the other end is effectively at 0v?[/quote]
What is the max signal level coming from the 1272 ? The 1272-core is fed from +24V, add 'TX-influences' to that and from then on it's straight on V=I*R & P = V*V/R = I*I*R...
(using AC-rms values)

But as I understand your question: how to interpret a power-rating for a rheostat, right ? As I can understand it it just holds for the full value and has to be derated when the wiper is not at max value.

Regards,

Peter

 

[skipwave]

http://www.groupdiy.com/index.php?topic=4825&highlight=attenuator

http://www.groupdiy.com/index.php?topic=483&highlight=attenuator

Little did I know PRR posted a variable U-pad plan long ago:
http://www.groupdiy.com/index.php?topic=1860&highlight=attenuator

Lots of good links here:
http://www.groupdiy.com/index.php?topic=2192&highlight=attenuator

 

[clintrubber]

Right, I guess nobody had the impression this thread was the first one about the topic

Nice you compiled those links. The NYD-pics I mentioned are there as well, and then some...

Bye,

Peter