help with u-pad

[salomonander]

hey
i need help with designing a mic pad for some preamps that dont have volume control.
all my mics have 200Ohm output impedance. the preamps are 2k. so thats all fine. from my research i gathered that i should use a u-pad.
i also have lots of 4pole 5position rotary switches to burn. i need help with finding the values and getting the wiring right though.

i couldn't find a u-pad calculator anywhere. so i assume i have to use an l-pad calculator? but how will the values translate to the u-pad.?

will i have to switch all resistors in order to get the impedance and attenuation right? or is there a smart way to use less resistors/poles? im asking this since on a commercially bought rack i have (using the same amps: v72) there are some resistors on the xlr inputs while the rotary switch only has one pole... so only one resistor gets changed.... im wondering if this is a compromise or some clever engineering i dont understand

thanks alot

 

[pucho812]

read this...

http://www.uneeda-audio.com/pads/

 

[salomonander]

read this...

http://www.uneeda-audio.com/pads/

im trying to get my head around it but its complicated for a math looser like me. whish there was a l-pad calculator that would let me input my source and load impedance plus my desired attenuation and simply do it for me...

i checked my professionally made v72 rack and the apply a u-pad with fixed serial resistors. only the shunt resistor is switched via a single pole rotary. am i right in that this is a very bad solution messing with impedance in basically every setting?

 

[pucho812]

here read this too

http://setxac.com/wp-content/uploads/2013/12/audio-pad-attenuators.pdf

 

[bruce0]

hey
i need help with designing a mic pad for some preamps that dont have volume control.
all my mics have 200Ohm output impedance. the preamps are 2k. so thats all fine. from my research i gathered that i should use a u-pad.
i also have lots of 4pole 5position rotary switches to burn. i need help with finding the values and getting the wiring right though.

i couldn't find a u-pad calculator anywhere. so i assume i have to use an l-pad calculator? but how will the values translate to the u-pad.?

will i have to switch all resistors in order to get the impedance and attenuation right? or is there a smart way to use less resistors/poles? im asking this since on a commercially bought rack i have (using the same amps: v72) there are some resistors on the xlr inputs while the rotary switch only has one pole... so only one resistor gets changed.... im wondering if this is a compromise or some clever engineering i dont understand

thanks alot

An L pad is just an unbalanced U pad.  The "in line" resistor would be split half on each leg of the U and the shunt resistor would be left the same.  This would give you the same attenuation and the same impedance but balanced.

There is a calculator here for H and T pads.  (WARNING - this calculator gets incorrect dB of attenuation figures for balanced pads)  It should be useful in understanding the above, and you can compare values and resistors, and set your output impedance to the shunt resistor value and poof it is a U pad.  Just don't trust the dB attenuation on balanced pads, attenuation value is wrong but the proportions and such are approximately right and it could help you understand how the pad varies impedances (it is really quite simple once you get the hang of it, they are just resistors set up as voltage dividers)

If you look at a line to mic attenuator pad (approx 40dB with say 10k or 20K input impedance), you will notice that the shunt resistor is very small values, and varying this value a little varies the attenuation a lot.  In a U pad, the impedance that the mic input sees is just that shunt, so you can imagine the "engineering" trick is just that you can vary that shunt by a little, it doesn't change the impedance seen by the amp much, but changes the attenuation a lot.

Play with the calculator this might make more sense.  In the calculator the shunts are R2 "legs" are all the other resistors.

 

[PRR]

Bruce was tping while I was drawing....

> mics have 200 Ohm output impedance.

And expect to drive >2K load.

> the preamps are 2k.

And expect to see a 200 Ohm source.

There's really only one way this can happen.

First ignore the balance and get an L-pad. Input has to be 2K, output has to be 200r. So try a 2K resistor tapped at 200 ohms. Say 1,800r and 200r. Input is obviously pretty near 2K. Output is clearly under 200.

A 200r tap on a 2,000r resistor is 1/10 or 20dB.

You can NOT get less attenuation and still meet these impedance goals.

While I could imagine a 40dB pad for Line In, I don't see a point in cutting 20dB-25dB-30dB-35dB pads for mike in. You have another gain-control somewhere in the preamp, use it.

We want 200r but really have 200||2K (preamp). That would be 10% low, so we make the 200 10% high, 220r.

So you have L-pad as 1800+220. To split it for a balanced U-pad you do 1800/2= 900 each side, still with 220 at the end.

40dB is a 100:1 ratio. We still need the ~200 facing the preamp. The series arm of the L is 99 times higher, or 20K. Split two 10K for U-pad. The input is 20K which is (oddly??) a very nice value for Line In.

 

[salomonander]

Hey guys

thanks a lot for your help!
i think its getting clearer now. problem is that i would need rather small attenuation steps... certainly less than 20db. so if everything less than 20db is a compromise i might better forget about this. my preamps do not have any gain control. they are fixed 34db. plus the 200Ohm to 2k seems like a perfect fit and i dont want to mess with that.
maybe its a better idea to put a pot after the amps? i do need something to prevent my converters from clipping. i do have lots of 10k stereo alps lying around. not sure how they will work with the low impedance of the preamps output (35Ohm).... any ideas? my converters have 10k bridging inputs.

 

[bruce0]

I don't think you can keep the pad output impedance as low as 200 and get pad values of less than 16dB or so without dropping the input impedance of the pad below 2k which would be a compromise because in general for "bridging" signal transfer you want the output impedance to be 1/10 or less of the input impedance it is driving.

This might be different if you had a transformer couple microphone output and a transformer coupled mic input.  In that case I wonder if you could use an impedance matching approach (the way it was back with tubes and iron signal transfer).  So say an SM57 into a transformer input on a tube pre might be a different situation. ( I am a bit out of my depth on impedance matching so apologies if this is wrong or a half truth).  Anyway if you were impedance matching it is easier to get small pad values though you might need an H-pad. (There is an article out there about SM57's and how they like something more like a 600-700 ohm load, just google SM57 load impedance)

But in general transformer-less mic head amps want to see a preamp input of at least 10x their output impedance (the mic spec's almost always show output impedance and often show expected load or input impedance clearly.)  Loading them more than that I think but too much load on the microphone output amp, and raising the impedance into your Mic Pre i think add's noise.

 

[andre tchmil]

im asking this since on a commercially bought rack i have (using the same amps: v72) there are some resistors on the xlr inputs while the rotary switch only has one pole... so only one resistor gets changed.... im wondering if this is a compromise or some clever engineering i dont understand

thanks a lot

Are you sure these fixed resistors are there for the attenuator ?

 

[salomonander]

im asking this since on a commercially bought rack i have (using the same amps: v72) there are some resistors on the xlr inputs while the rotary switch only has one pole... so only one resistor gets changed.... im wondering if this is a compromise or some clever engineering i dont understand

thanks a lot

Are you sure these fixed resistors are there for the attenuator ?

im by no means an expert. but there certainly is attenuation happening. since the parallel resistor the only element that is switchable it has to be attenuation. btw...

 

[andre tchmil]

a picture maybe ?

If that company , is the company I'm thinking of , then the only attenuation is on the output ( 34dB)

Maybe you can make a drawing also ?

 

[salomonander]

i'll check it out when im in the studio tomorrow. maybe im wrong and the attenuation really sits on the output. i guess it would be fine then?

 

[salomonander]

ok... i couldn't resist. the studio is only a block away.
yes, im certain that the pad sits on the input. the output goes straight to the xlr outputs. the input goes through this half assed pad that is not even bypassed in the "no attenuation" setting. dont know if we are talking about the same company.

 

[salomonander]

I don't think you can keep the pad output impedance as low as 200 and get pad values of less than 16dB or so without dropping the input impedance of the pad below 2k which would be a compromise because in general for "bridging" signal transfer you want the output impedance to be 1/10 or less of the input impedance it is driving.

This might be different if you had a transformer couple microphone output and a transformer coupled mic input.  In that case I wonder if you could use an impedance matching approach (the way it was back with tubes and iron signal transfer).  So say an SM57 into a transformer input on a tube pre might be a different situation. ( I am a bit out of my depth on impedance matching so apologies if this is wrong or a half truth).  Anyway if you were impedance matching it is easier to get small pad values though you might need an H-pad. (There is an article out there about SM57's and how they like something more like a 600-700 ohm load, just google SM57 load impedance)

But in general transformer-less mic head amps want to see a preamp input of at least 10x their output impedance (the mic spec's almost always show output impedance and often show expected load or input impedance clearly.)  Loading them more than that I think but too much load on the microphone output amp, and raising the impedance into your Mic Pre i think add's noise.

hey. all my mics are transformer coupled and have output of 200Ohm. all my preamps have transformers on the input with 2k. cheers

 

[bruce0]

Maybe  I'm wrong but can't the original poster use a power transfer approach? In which case he doesn't need the impedance to be 10 times the output impedance. Instead he can build pads that match impedances which are much easier to do in small values.

That is assuming he continues to use transformer balanced outputs on the Mics and transformer balanced inputs on the pre's