Buongiorno,
Vorrei sapere se qualcuno ha mai riscontrato perdita di circa 2dB a 20 khz quando il potenziometro del volume ( 100K log) in ingresso ad un ampli non è al massimo.
Questo difetto l'ho riscontro in qualunque progetto.
Grazie in anticipo per l'aiuto.
Vorrei sapere se qualcuno ha mai riscontrato perdita di circa 2dB a 20 khz quando il potenziometro del volume ( 100K log) in ingresso ad un ampli non è al massimo.
Questo difetto l'ho riscontro in qualunque progetto.
Grazie in anticipo per l'aiuto.
This is an English speaking forum. Always include a translation if you wish to post in another language.
Thank you.
Google translate:
Questo è un forum di lingua inglese. Includi sempre una traduzione se desideri pubblicare in un'altra lingua.
Grazie.
Band problems with amplifier input potentiometer.
Good morning,
I would like to know if anyone has ever experienced a loss of about 2dB at 20 khz when the volume pot (100K log) input to an amp is not at maximum.
I have found this defect in any project.
Thanks in advance for the help.
*Band problems with amplifier input potentiometer.*
Sorry, here the translation.
Good morning,
I would like to know if anyone has ever experienced a loss of about 2dB at 20 khz when the volume pot (100K log) input to an amp is not at maximum.
I have found this defect in all my project.
Thanks in advance for the help.
Sorry, here the translation.
Good morning,
I would like to know if anyone has ever experienced a loss of about 2dB at 20 khz when the volume pot (100K log) input to an amp is not at maximum.
I have found this defect in all my project.
Thanks in advance for the help.
Thank you and welcome to the forums.
Did you happen to do a Google translation or is it that good this time?
Did you happen to do a Google translation or is it that good this time?
Hi
In a line preamp it always happen at -6dB of attenuation ( the worst condition) if the circuit has an high value of capacitance on input.
Mainly depend on tube type and cable capacitance
If you put the 10/22 kohm pot probably it is much better
Which type are the stuff involved?
Walter
In a line preamp it always happen at -6dB of attenuation ( the worst condition) if the circuit has an high value of capacitance on input.
Mainly depend on tube type and cable capacitance
If you put the 10/22 kohm pot probably it is much better
Which type are the stuff involved?
Walter
Last edited:
Yes it is typically a problem with tubes with high miller capacitance, using a 50K pot or stepped attenuator reduces the worst case source impedance the grid sees to 12.5K or so - in most cases this will allow you to stay close to flat to 20kHz even at the -6dB where this problem is worst.
What input tube are you using?
I use mostly 50K and 100K stepped attenuators in tube gear, and 10K/20K/50K in solid state gear..
What input tube are you using?
I use mostly 50K and 100K stepped attenuators in tube gear, and 10K/20K/50K in solid state gear..
The basic math is that a pot at 50% resistance "shows" 25% output impedance because top and bottom track halves are in parallel as far as impedance goes.
With that, you can determine RC treble loss, R being the above suggested and C the amplifier input capacitance.
With that, you can determine RC treble loss, R being the above suggested and C the amplifier input capacitance.
Thanks for all your suggestions, in my case the input tube is 6SL7 with triodes parallel and, probabily using a 6SN7 or some other tube with lower gain.
I will try decreasing attenuator value or using stepper attenuator because change of tube isn't possible in my case for lower gain
Thx.
I will try decreasing attenuator value or using stepper attenuator because change of tube isn't possible in my case for lower gain
Thx.
Parallel 6SL7 triodes:
grid to plate capacitance, Cgp, is 1.5pF
u = 70 (gain can not exceed 70, that is the worse case to determine Miller Effect Capacitance).
grid to cathode capacitance is 1.5pF
Miller Effect Capacitance is (U +1) x Cgp = 71 x 1.5pF = 106.5pF
add the Cgc 1.5 pF to 106.5pF = 108pF
parallel capacitance two triodes is 2X = 216pF
At 20kHz, 216pF has capacitive reactance of 36,841 Ohms.
The wiper of a 100k potentiometer at the -6dB attenuation setting is 25k best case (only if the driving signal source is 0 Ohms output impedance).
For a 100k pot, the -3dB point is 29,473Hz, and the -1dB point is 14.737Hz (best case, it could be worse).
A 2 stage amplifier frequency response also has to include the frequency response of the 6SL7 plate circuit, the next stage grid circuit, next stage plate circuit including the output transformer.
If each one of those factors is -1dB at 14.7kHz, the total amplifier will be -4dB at 14.7kHz.
A 3 stage amplifier will have more points that have high frequency roll off.
A 50k potentiometer and a fairly low source impedance to the potentiometer will improve the situation.
grid to plate capacitance, Cgp, is 1.5pF
u = 70 (gain can not exceed 70, that is the worse case to determine Miller Effect Capacitance).
grid to cathode capacitance is 1.5pF
Miller Effect Capacitance is (U +1) x Cgp = 71 x 1.5pF = 106.5pF
add the Cgc 1.5 pF to 106.5pF = 108pF
parallel capacitance two triodes is 2X = 216pF
At 20kHz, 216pF has capacitive reactance of 36,841 Ohms.
The wiper of a 100k potentiometer at the -6dB attenuation setting is 25k best case (only if the driving signal source is 0 Ohms output impedance).
For a 100k pot, the -3dB point is 29,473Hz, and the -1dB point is 14.737Hz (best case, it could be worse).
A 2 stage amplifier frequency response also has to include the frequency response of the 6SL7 plate circuit, the next stage grid circuit, next stage plate circuit including the output transformer.
If each one of those factors is -1dB at 14.7kHz, the total amplifier will be -4dB at 14.7kHz.
A 3 stage amplifier will have more points that have high frequency roll off.
A 50k potentiometer and a fairly low source impedance to the potentiometer will improve the situation.
Thanks 6A3sESTATE, for this explanation very usefull to calculate the Miller effect. Now I try with lower potentiometer, optimizing in the same time following stages (if possibile).
Other possibility I think, is to use tube with lower u (6SN7) and/or eliminate the triode parallel.
Ok, now I try...
Other possibility I think, is to use tube with lower u (6SN7) and/or eliminate the triode parallel.
Ok, now I try...