Question about stepped balanced input attenuator

[rotation]

Hi!

I need stepped attenuator for the input of my tube preamp, because simple att.ala Jensen is not enough. I got some switches as a samples from Palazzo (they are good quality, just feel is a bit hard), they are 3 poles, 8, 12 and 24 positiond, MBB with stop pins..
I know this was discussed many many times, but i still don't get it. I'm even more confused than i was.
I found this page, see balanced variant:
http://www.tkk.fi/Misc/Electronics/circuits/line_to_mic.html
Is this correct? What is my input and output Z for attenuator? 150-200 ohm and 2k? If i know this than it's easy to calculate resistor values i guess.
And what is the thing about "constant impedance"?
I have schematic for attenuator Mr Hinson posted here, but it only have 6 steps, i would like to have 8 or 12.
I searched many many threads, some time ago even asked here, but i still don't get it. This is really frustrating me.
Can anyone help me with schematic or some links where i could calculate values myself? I also need to know what is the name of CORRECT balanced attenuators that we use at the input of mic preamps.

Miha

 

[abbey road d enfer]

[quote author="rotation"]Hi!

I need stepped attenuator for the input of my tube preamp, because simple att.ala Jensen is not enough. I got some switches as a samples from Palazzo (they are good quality, just feel is a bit hard), they are 3 poles, 8, 12 and 24 positiond, MBB with stop pins..
I know this was discussed many many times, but i still don't get it. I'm even more confused than i was.
I found this page, see balanced variant:
http://www.tkk.fi/Misc/Electronics/circuits/line_to_mic.html
Is this correct? [/quote]
Yes, this is correct

What is my input and output Z for attenuator? 150-200 ohm and 2k?

That is an acceptable starting point

If i know this than it's easy to calculate resistor values i guess.

You are right, it is not very difficult in terms of mathematics, but you will find that there is an infinite number of solutions to this simple equation.

And what is the thing about "constant impedance"?

That's where things start to become a little more difficult. Basically, you want the input to present the same impedance whatever the attenuation. This is what makes the decision between the infinite number of possible results. And then you have to tweak the values in order to use standardized resistors.
But the main problem is that the actual input impeadnce of your tube pre is not constant over the whole freaquency spectrum; it may well be 2k at mid frequencies, but it will very likely be almost ten times lower at low frequencies. So, if you calculate your attenuator for 2k load, the resulting frequency response will be truncated at low frequencies. That's the reason why vintage tube pres used constant impedance attenuators, designed with 200 (or 150) ohms source impeadance and the same load impeadnce. That would make sure that the input transformer saw a low enough impedance, which may not be the case if you assume 2k.
A quick way around is to make sure that the shunt resistor is never higher than the lowest impedance of the transformer. This is easy for attenuations of 20 dB or more. The real tricky zone is between 0 and 20 dB, in particular 6dB is a real pain, because, if you want to make the shunt resistor of the same value as the xfmr, you end up with a very low resulting input impedance, which you don't want because it loads the microphone too much.
The Input impedance is equal to R1 + R2*Rin/R2+Rin.

I have schematic for attenuator Mr Hinson posted here, but it only have 6 steps, i would like to have 8 or 12.
I searched many many threads, some time ago even asked here, but i still don't get it. This is really frustrating me.
Can anyone help me with schematic or some links where i could calculate values myself?

The link above gives you the right formula for calculation of the attenuators, but, in order to translate it to balanced circuits, you have to split R1 in two halves, one in the hot (in-phase) input, the other in the cold (out-of-phase) input. You have to make a list of the attenuation values you want, make the calculations and check the input impedance.

I also need to know what is the name of CORRECT balanced attenuators that we use at the input of mic preamps.

Miha

The correct name is just that: balanced attenuator. But you have to figure out that most of the documentation on those is related to constant-impedance matching attenuators, i.e. 50 ohm input to 50 ohm output (for RF circuits) or 600 ohm input to 600 ohm output (for telephone land lines). Audio is such a tiny branch of electronics that we have to work it out by ourselves.

 

[guavatone]

I was working on this and my Attention drifted and I can't remember where I left off. It may help. I had some trouble working out formulas for Real world values to find actual atten. but you should be able to get ballpark figures.

http://www.guavatone.com/free/Matten.xls

If anyone wants to contribute to corrections, please do.

 

[rotation]

Thanks, now i understand this a bit better
To calculate an example i helped myself with this program:
http://radiomagonline.com/notebook/passive_attenuators/
Rin-150, Rout-2000, attenuation is 6dB
What i get is:
R1:-483
R2:2609
R3:733
Why i get - for R1? Is this ok?
It is impossible to get values like this, is there a way to calculate them to something more standard?

Another option would be to calculate more position for this attenuator:
http://twin-x.com/groupdiy/displayimage.php?album=random&cat=10184&pos=-1683
It uses all standard values and is intended for use in similar type of preamp i make. Can anyone help me do this? I would like to have 12 instead of 6 steps.

Miha[/img]

 

[guavatone]

I am not sure. But I would use an balanced L which makes a U and the parallel resistors go on mic side and the preamp is connected to the shunt.

did u see this yet?
http://www.uneeda-audio.com/pads/

 

[rotation]

Yes, i've seen it before. It says i just find nearest standard value for r, that's it.
I now know how to calculate values for l and u pads. It's very easy to understand, the problem are different names and some wrong explanations in posts.
Can anyone please explain how Mr.Hinson calculated his pad and why is Rshunt smaller than Rin? I think i read somewhere he built this attenuator to change the sound of preamp..

Miha

 

[guavatone]

You may be confusing Hinson with Oliver's Tele V series pad.