Calculating Resistor Values for a Log Pot

[Gold]

I did some measurements to find resistor values for the log threshold pot on the DAOC. I picked five overlapping ranges and measured. I used a tone at a fixed level and adjusted the pot to see gain reduction in 1dB steps. I measured the pot, disconnected from the board, after every step. I used a +21dBu,+16dBu, +4dBu,-8dBu and -12dBu.

The good news is that where the ranges overlapped the measurements agreed fairly closely. At the top of the throw they agreed less so but were not way off.

My question is: Is there any way to see if my measurements agree with the theory? Do I need more information?

 

[JohnRoberts]

Not sure I understand exactly what you're doing but if you have a calculator with log function or excel spread sheet you can convert dB to a fraction or ratio.

JR

 

[Gold]

Using the method described above I came up with for +16dBu. Total measured resistance of the pot was 470k :
-1dB=8k23
-2dB=9k5
-3dB=11k08
-4dB=12k75
-5dB=15k36
-6dB=18k16
-7dB=21k
-8dB=24k3
-9db=29k2
-10db=33k8
-11dB=40k
-12dB=48k5

I would like to know if there is a way see if these measurements agree with the "theory". I don't know if I have given enough information or if what I'm asking is possible.

 

[JohnRoberts]

No I still have inadequate information. I still have no idea what a DAOC is?  Or what circuit topology the 470k pot is used in.  I can't even guess from the series of numbers you listed.



You can calculate dB ratio using a scientific calculator or excel spread sheet.

to convert decibels to a voltage ratio, the ratio equals 10 raised to the power (dB/20)

this is 10^x  button on a scientific calculator.. in excel spreadsheet type " =POWER(10, n) "  with n=dB/20

Good luck...

JR

 

[Gold]

Thanks. It appears I don't know enough to ask the right questions. It's not the first time... I'll try doing a little more homework. Or I could give up, make the attenuator and move on.

 

[PRR]

> no idea what a DAOC is?

Seems to be "Dark Age of Camelot".

 

[Gold]

It's the Dual Analag Opto Compressor http://www.groupdiy.com/index.php?topic=26585.  The pot I am referring to is RV102.

I think this thread should die a peaceful death until I have a better idea of what question I am actually asking.

 

[[silent:arts]]

if you just want the log pot made by a switch use the Excel sheet I posted for the PM:
http://www.silentarts.de/DIY/PM660/LogPotToSwitch.xls

 

[Gold]

if you just want the log pot made by a switch use the Excel sheet I posted for the PM:
http://www.silentarts.de/DIY/PM660/LogPotToSwitch.xls

Well, not exactly. I will make the switch using my measurements as this is the behavior I want from the switch. Every switch step reduces output by 1db for a constant input level. I want to know the math behind the measurements I took. The more I think about it the harder it looks.

How are the steps laid out in the excel spreadsheet? What criteria did you use to determine them?

 

[PRR]

> It's the Dual Analag Opto Compressor

Darn. Catapults are more fun.

No, there's no useful "theory" for RV102. The photo-resistor is only slightly linear, and there's no convenient approximation. Then you must work out the R102/VTL5C4-R attenuation ratio: trivial but not linear. We may assume that V102 is linear for practical purpose. But VTL5C4-LED is highly non-linear: off up to 1.5V and saturated by 1.7V. And it eats raw audio: action will be different on Sine than on speech-music. But there is NFB around the VTL5C4, so once it starts doing something it gets somewhat predictable.

Then if you replace the VTL5C4, several factors change more than a dB. In think it may drift from winter to summer even in a "controlled" room.

The question is not mathematically unsolvable, does not invite Godel's issues. However you can get "answers" in an hour with the meter, or spend days collecting the many sets of data needed to begin to solve the equations...

 

[Afarawayland]

http://sound.westhost.com/project01.htm

Perhaps the question has already been answered, but I found this the other day.

"Take a 100k linear pot (VOL), and connect a 15k resistor (R) as shown above, to achieve the curve shown. It should be a straight line, but is actually still far more logarithmic than a standard log pot. For stereo, use a dual-gang pot and treat both sections the same way. Use of a 1% resistor for R is recommended. Different values can be used for the pot, but keep the ratio between 6:1 to 10:1 between the value of VOL and R respectively. While 6.7:1 is close to a real log curve, it may still allow excessive sensitivity at low levels. Higher ratios than 10:1 can be used, but will cause excessive loading of the driving stage, or necessitate the use of a pot whose resistance is too high."

 

[Gold]

However you can get "answers" in an hour with the meter, or spend days collecting the many sets of data needed to begin to solve the equations...

That's good enough for me. I think I'll do another set of measurements for kicks. I still don't know what the compression ratio is. Instead of measuring gain reduction with a fixed input I'll get a few dB of gain reduction going and increase the input in a few different ranges to see what I get.

Then with all this information I will make the switch. I will invariably pick the wrong way and have to do it over.

 

[JohnRoberts]

http://sound.westhost.com/project01.htm

Perhaps the question has already been answered, but I found this the other day.

"Take a 100k linear pot (VOL), and connect a 15k resistor (R) as shown above, to achieve the curve shown. It should be a straight line, but is actually still far more logarithmic than a standard log pot. For stereo, use a dual-gang pot and treat both sections the same way. Use of a 1% resistor for R is recommended. Different values can be used for the pot, but keep the ratio between 6:1 to 10:1 between the value of VOL and R respectively. While 6.7:1 is close to a real log curve, it may still allow excessive sensitivity at low levels. Higher ratios than 10:1 can be used, but will cause excessive loading of the driving stage, or necessitate the use of a pot whose resistance is too high."

If those tricks were reliable, there would never be any reason to make audio taper pots.

Using a 1% resistor in combination with a typical screened resistive element that is nominally 20% tolerance seems optimistic to me.

Screened pots can deliver reasonable ratiometric tracking but there is a reason they come in so many different flavors.

JR

 

[PRR]

> I found this the other day.

Yes, it pulls the mid-rotation down from -6dB to say -15dB, which may be useful if you like to see knobs near center and ample reserve gain.

It violates one property of an Ideal Potentiometer: the source loading changes radically. (Normal pots also have some load-change effect, but not 100K:13K.)

It gives good resolution for -10dB to -20dB, but gets mighty cramped when high attenuation is needed. A classic 10% audio pot is a bit kinked -20dB to -25dB, but very much smoother around -30dB (the setting needed when a strong AM radio signal is played near background volume).

Note that this makes a potentiometer. And yes, it can fake a Reverse Audio potentiometer. It will not make a Reverse Audio Rheostat, a part commonly needed in mike preamps.

West's Fig 3 is interesting, and could even be wrapped around a stage like DOAC's V101A, but does not drop-in on this circuit.

It is no real use for Gold. His threshold action is not a simple function of attenuation. And his source, a tube plate, does not take kindly to wild load variation. In particular, he may want one threshold in the start of a track and another for the big finish. Ideally he can change threshold in a quiet passage, when the limiter is not limiting. But with loaded pots, any real change in threshold pot setting also changes main-path gain (though the NFB around V101A makes it moderately solid).

The cost of good audio pots is not bothering Gold. I think he could afford them when needed. But if I  understand his first post: he does NOT want continuous action, he wants repeatable action. He may test one week, do the actual job another week, with different jobs between. The exact "dB" is not so important as being able to get the SAME setting again. Hence a switch and fixed resistors. All this pot-pad stuff is useful in ad-hoc manual trim-to-taste listening, sloppy for precision transfers.

 

[Gold]

The exact "dB" is not so important as being able to get the SAME setting again. Hence a switch and fixed resistors.

Correct. Luckily for me the sloppyness of phono pickup response and degradation of acetate references should swamp any tempco differences in the vactrol. Sometimes it's best not to know where the bodies are buried.