-20db Pad circuits? Various solutions?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Plexibreath

Well-known member
Joined
Jun 3, 2004
Messages
111
Location
Pleasanton, California (SF Bay Area)
I've seen 2 basic kinds:

1) Uses 3 resistors and a DPDT switch.
Where with the pad on, a resistor is bridged across the + and - signals and a resistor is in series on both sides, when off the resistor that bridges across is taken out and the series resistors are shorted across, this seems to be the most common.

2) Uses 5 resistors and a 4PDT switch
Where with the pad on 2 resistors are in series on each side, with a resistor bridging across at the points between the 2 series resistors, the switch just bypasses the entire works. I can only see one advantage here, in the case of more than one level of attenuation this would probably be more expandible. But is there more?

Other than the fact that the first solution looks easier to emplement, is there an advantage to the second solution sound wise or other, that people have gone to that kind of trouble to build it that way. I bring this up as I saw the second solution in a schematic on the Tube Mic Pre thread for the Steve Albini modded Ampex 351 pre.
Also, are there other, possibly better solutions?

Thanks,
Kelley
 
You can get this. It is highly accurate, inexpensive and easy as hell to build....

http://www.bottlehead.com/et/adobespc/foreplay/foreplay_line_stage_kit.htm

Do scroll down for what is called "sweetest whispers"


sweetestinstalled.jpg
 
I would use a fully stepped attenuator and wire in 2 DPDT switches and make the selections. BY using the full attenuator, you could simply reconnect to suit and once you reach your attenuation, you are there. Your Step one will work fine IMHO

PS, a 3rd switch would serve as a phase inversion device..and I am still working on that thing
 
Hi Kelley,

Option one leaves the pad still connected to the input. Therefore it will be a load on whatever the driving device is, microphone or otherwise. If you wish to add a line input pad to a mic input, then this is fine. A mic would like to see 2k or so of input impedance, and if you add this pad to allow for a line input, you would add series resistors that also raise the input impedance to say 10k differential mode...so you get both a pad and higher input impedance for your line sources. If the mic input has an input impedance of say 2k, then adding a constant pad 'load' of 10k or so across this makes little difference. Work out 10k in parallel with 2k for the actual numbers. The Gyraf G9 has this type of pad.

Option two is me guessing more, but the pad described above would not be the best choice for padding a mic input. You may want the pad, but you still want the 2k or so input impedance, or else you're adding noise and also possibly changing the sound as the mic is driving a much higher input impedance. Basically you want to maintain the input impedance, but simply attenuate it so that your pre can handle the levels.

That's all I can give you, I'm sure there's more from the gurus!

Bjorn
 
Guru-esque? hmm...I'm still bashing my head against my G9, so not quite yet, but thank you! My word, you only have 38 posts! Let me guess, RO posts are non-transferrable??

Bjorn
 
[quote author="Mark Burnley"]Bjorn,

That sounded pretty guru-esque to me!

Here's a great pad page:

All About Pads

:thumb:

Mark[/quote]
Mark,

That's a very informative article about Pads! I learned a lot but new questions arise out of it all.

At least I now have titles to add to the circuits I was groping to find discriptions for:

1) The first is called a "U" Pad, which is a Balanced version of an "L" Pad, this circuit is asymetrical.

2) The second is called a "H" Pad, which is a Balanced version of the "T" Pad, this circuit is symetrical.

But then it goes on to describe more:

3) "Bridged-T" Pad and "Balanced Briged-T" Pad. The article is a little bit of a let down here as it only describes the benefits of this circuits as fewer resistors have to change to make it variable, but I'd imagine there would be more to say about it as it's the most complex with 7 resistors.

4) "O" Pad. Here the article introduces this circuit, not with the others, but in the section on determining values. I could use some help here on determining if the writer is just saying it's only use is to make the impedence look better to the microphone.
"...you get to the O configuration only when the pad input impedance is significantly higher than the inpedance bridged across the output (i.e. the following mic preamp). The extra resistor simply shunts down the input impedance to better approximate what the microphone sees with the pad connected to what it sees without the pad. There are still subtle differences, such as any reactive elements present at the preamp input, and these will not be mirrored by the pad."
Uuhh, is that a good thing? Maybe I've had too many of these :guinness: and these :sam: in my wreckless youth so I'm only getting the gist of it all but not the deep understanding.

Where this raises more question than it answers for me is at the beginning of the article he states that there is coloration by the interaction of the microphone with the preamp and that coloration can be minimized "by inserting a pad by paying attention to this detail and designing the pad to mimic the conditions present before it's insertion. The easiest parameter to mimic, and the one that is the biggest contributor is the impedance that the pad presents to the microphone, and the cource impedance that it presents to the microphone preamp."

Now this gets down to the heart of my question. Does mean you can't just slap any 'ol pad circuit in front of an input transformer and be done with it? Do you have to match the pad circuit resistor values to work symbioticly with each specific preamp circuit? Does this explain why almost every pad circuit I've looked at, though the topology looks usualy like a "U" Pad, the resistor values for -20db, though relative to each other, vary widely from preamp to preamp? Is the extra resistor on the mic side of the pad in the "O" configuration really all he's talking about with regards to making the impedence appear to the microphone as if the pad is not on at all? Why have I not seen any "O" Pad configurations used in any of the preamp schematics I've looked at though his presentation of it looks to be the most desirable?

What I am certain of is there's a lot more to a concept I had originaly dismissed as bonehead simple (unless I'm below bonehead), this whole pad business.

Kelley
 
Kelley,

This really is not as complex or mysterious as you are beginning to believe it is.

Ohm's law. That is all.

Attenuators get more complex when one designs it to be variable; the complex resistive networks maintain the correct input/output impedance as the amount of attenuation changes.

The "U" pad is just fine. Design it and use it for a specific, non adjustable amount of attenuation.

Bear with me. :roll:

Let's say the input transformer has a 2K input impedance.

That is the AC inductive load placed across the output of the microphone when it is connected.

The DC resistance of the transformer primary is probably less than 50 ohms; it is the AC inductance of the transformer windings that present a higher impedance (2K) to the audio signal coming from the mic.

Ok, so you want a pad. You need a voltage divider. It needs to be balanced. The mic needs to see ~2K. The input transformer wants a source impedance that is approximately that which the microphone would present if it were connected without a pad.

The U pad handles all of these duties:

Say the bridging resistor is 100 ohms and the two series resistors are 1K. The 100 ohm resistor is in parallel with the existing 2K transformer impedance, which yeilds something slightly under 100 ohms(95 ohms). The pair of 1k resistors in kick the impedance the mic is looking at back up to 2095 ohms... pretty close to 2K.

This resistive divider attenuates by a factor of around 21, which equals about 26db. If that is too much, make the bridging resistor a little bigger and make the series pair of resistors a little smaller until you get what you want.

As far as the U pad affecting the source impedance that the input transformer is looking back at compared to just the microphone, let's take an SM57 which as a source impedance of 300 ohms; you've got that in parallel with about 2095 ohms which gives 262 ohms...pretty close to the original 300.

The pair of phantom power resistors will lower the source impedance a little bit as well.

Will the pad change the sound? Yes, a little bit, but not really enough to warrant not using this arrangement.

Dean
 
Thanks Dean, that really sets it straight. :thumb: I'll be "U"ing it. But one last thing. As appropriate as the "U" Pad is for my purposes, what reason would I need that would drive me to use the "H" Pad or "O" Pad? I have considered more than one step of padding, maybe 2 at the most 3, is this where the other configurations become more favorable? And in those circumstances why would one use one over the other?

Thanks,
Kelley
 
Kelley;
I think you are getting terms mixed up here.
I hope I help not hurt.
You asked about a 20dB pad.
simple attenuation at the input of a mic pre.
Usually you see a switch which adds a 620 ohm in series to each
balanced input wire and a 169 ohm that shorts between the two.
This will give a 20 db pad. But this is not a full H pad
only the front section.

If you need only to pad a mic pre the above will work.
You may need to adjust your values as the input Z of your mic pre.
Does effect the absolete attenuation value.

If you want a variable attenuator to use a a volume control.
Then you need to look for attenuator stuff on the web not "pad"
a pad is usually fixed and a attenuater can be fixed or
variable. An attenuator usually has a constant impedance
on the input and output.

If you want a variable attenuator the scope is beyond the discussion here.
Defind the input and output impedances. Define dB per step.
is it ballanced or unbalanced. Switch you are going to use and number
of poles and sections.

What exacly are you trying to do?
Let us know here.

Some pad stuff below.
http://www.quadesl.com/attenuator.html
http://www.mcsquared.com/hpad.htm
http://www.gyraf.dk/schematics/bridge-attenuator.gif
 
Kelley,
But one last thing. As appropriate as the "U" Pad is for my purposes, what reason would I need that would drive me to use the "H" Pad or "O" Pad? I have considered more than one step of padding, maybe 2 at the most 3, is this where the other configurations become more favorable? And in those circumstances why would one use one over the other?

I would use a rotary or slide switch to select between two or three different bridging resistors in the U configuration, since varying the bridging resistor has the least influence on impedance. This is how my store-bought inline attenuator is set up.

As far as the H pad is concerned, that network adds two more series resistors, so you could have more control over the source impedance that the pad presents to the input transformer. As the link Mark put up said, it can be configured to have the same output(source) impedance as the input impedance, hence the 'symetrical' designation. Since the U pad lowers the original source impedance a bit, you could use the H network to fine tune things a bit.

The O pad just adds a parallel resistor to the front end of the U pad in case the U network one chooses creates a higher input impedance than one wants.

It's all just a basic series/parallel circuit, and should be analyzed and adapted to your specific needs using Ohm's law and a calculator.

Some sonic voodoo may come into play due to the fact that you are somewhat replacing the transformer's inductive load with a resistive one; I can attest to the fact that the sound changes, but it isn't much. Other factors such as the reduced drive level make it harder to discern where the actual tonal difference is occuring.

Dean
 

Latest posts

Back
Top