Aleph P Attenuation Circuit

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi-

I have only recently started studying this subject, so forgive me if this question is too simplistic. I noticed that on the Aleph P attenuation board, when the relays are switched "off", they connect their respective resistors to ground, as apposed to just leaving them hanging, so to speak. Why is this done? Thank you!
 
The one and only
Joined 2001
Paid Member
The Aleph P, like the BSOZ and the Aleph L (BOZ) circuit, are all
basically current source output circuits, that is to say that
they have a high output impedance and don't care what they're
driving, including a short to ground.

The attenuators for these types of circuits are output shunt-to-
ground, so that the output resistance (to ground) is lowered as
the output level is lowered. The weighting is a modified binary
for the different relay/resistors, and you can parallel any set you
like to get the value you want. We use a microcontroller with a
look-up table to set the values where we want them.

These types of circuits also benefit from variable Source-Source
resistance, and particularly with the BOZ, from some input
attentuators, as some sources are high enough as to overload
the input. You'll find lots of commentary around on the subject
in this forum.

Just post the question, and a couple of glasses of wine later
someone will offer an opinion.

:cool:
 
Boosting up the ADC 0804 to 8 bits ( instead of 6 in the Aleph P ) you can achieve 255 position virtual precision balanced attenuator . The only problem is the oscilation between 2 positions of attenuation , as the value DC on the control pot might not be stable enough for so many attenuation positions.
I used a 4013 connected to add to the base of pot ( which is not directly connected to the circuit ground ) the value of 1/2 step of 256 in DC value referenced to 5V supply , and if oscilating the 4013 corrects the DC value read by the input of ADC , so it does not oscilate anymore between 2 positions ;)

Clever , but this is not genius at all :smash:

Best regards

Anael
 
Yes, i understand. But what i want to know, is there a difference in what Nelson has in his look-up table, and a simple binary row from 00000000 to 11111111 ?

I ask because what a stable ADC circuit with 8 Bit would make i made with a counter and a microcontroller. Now i would like to know whats the advantage of the look-up table and whats the difference for the order of relay switching.
 
Difference

The difference between the binary counter in aleph p 1.0 and the binary lookup table in the aleph p 1.7 is that the counter makes use of all of the values within the 6 bit range whereas the lookup table makes use of a limited number of settings within its 8 bit binary range.
That is the aleph P1.0 has 2^6 positions = 64 possible volume settings;
The aleph p1.7 has 2^8 positions = 255 possible volume settings;
But since it is not a linear range, and some of the settings are very close in terms of absolute attenuation, passlabs have "cherry-picked" :devilr: a number of settings - possible volume settings - and therefore not using all 255 settings. Again since it is not a linear volume control it is the best solution to make a table (a lookup-table) of all the used settings and program it into the microcontroller.

This is my understanding of the aleph p differences. It is now just a question if i'm right or wrong - mr. Pass? :D

Actually i'm not sure how many of the 255 settings are used in the aleph p1.7.

Kongen
 
if i'm not mistaken

if i'm not mistaken they have chosen the settings inorder to get a more linear range of volume settings. I do believe that if you calculate the attenuation for each of the 255 settings in the original circuit some of the figures will be very close... :dodgy:

kongen
 
but what about...

But another thing for me is the absolute attenuation. As I see it the original aleph p attenuator must have different absolute values with different amps - because of the varying input impedances. Can anyone telll me for an example the max attenuation of the original aleph P1.7 circuit when power amp input impedance is 50k compared to one with perhaps only 2k input impedance? :confused:
 
What do you think about my cherry-picking?

I’ve picked 70 out of the 255 possible steps, I find nice looking together.

Do you thing I should look for another setting, or is this one find?

BR

/loovet
 

Attachments

  • volume.gif
    volume.gif
    4.9 KB · Views: 893
Hi,

I would like to use the aleph p 1.7 volume attenuator with my own controller software but I don´t understand to calculate the attenuation with the original restistor values.
I would like to calculate a a look up table.

Can anybody help me with an example?


Daniel

Sorry for my bad english!
 
When you look at the attenuation circuit you can see that it is in fact a voltage divider which
can switch the resistors between being a series resistor and a parallel resistor. You can
then make a spreadsheet using the formula's from here where you make columns from 1 to
254 (but then in binary, so 8 columns) which decide if a resistor is in series or parallel. The
values 0 and 255 are not possible, because then no resistor is in series resp. parallel. Then
choose the values so that you get a curve as displayed in a previous post.

I have also a question: if you want to make it a 9-bit attenuator to increase the maximum
attenuation from 48dB to 54dB, would it be better to add a resistor+relay on the LSB side
(16K) or on the MSB side (31r)?
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.