In order to have a quad volume control, I thought of having a single log pot between a DC negative source and 4 gates of 4 matched JFET's as shown above. We're talking audio here, I haven't calculate the level of the input signal yet, but I assumed it's gonna be arround 500mvpk-pk. Will it work? I know I could just buy a 4 gang log pot, but they are hella expensive.
\$\begingroup\$
\$\endgroup\$
1
-
\$\begingroup\$ Consider using a simple dual digital potentiometer such as the DS1882 rather than roll your own. You'll get much better results: datasheets.maximintegrated.com/en/ds/DS1882.pdf \$\endgroup\$– Jack CreaseyJun 26, 2019 at 23:00
Add a comment
|
3 Answers
\$\begingroup\$
\$\endgroup\$
5
First, the \$R_{DS(ON)}\$ vs. \$V_{GS}\$ of a FET depends on a number of quantities which are not normally defined in the datasheet, and which vary from part to part and over temperature. There is not a nice linear, unchanging relationship between \$V_{GS}\$ and \$R_{DS(ON)}\$.
Second, your FETs will never be matched close enough, probably not even if they all came on the same die.
Third, you're not really implementing an audio pot, because the resistance across the assembly won't be constant.
answered Jun 26, 2019 at 20:40
-
\$\begingroup\$ First, the RDS vs. VGS of a FET is not well defined. \$\endgroup\$ Jun 26, 2019 at 20:45
-
\$\begingroup\$ First, the RDS vs. VGS of a FET is not well defined. That's what I'm trying to do, establish a relationship between VGS and RDS, using 4 FET's and a single log pot controlling RDS from the 4 of them using a single negative DC source. I thought of ti because I know quad log pot's will be expensive and trhe 4 pot's might be even less matched than 4 matched FET's. What I wanna do is control 4 RDS from a single spin on a pot. \$\endgroup\$ Jun 26, 2019 at 20:50
-
\$\begingroup\$ I haven't thought about values yet, as I stated, I Haven't calculated the level of the input yet. I just wanna know if it would work. If it would work I'll go thopught a whole lot of calculations and choose the best FET, the resistor between the signal and the drain and the negative soucer. \$\endgroup\$ Jun 26, 2019 at 20:54
-
\$\begingroup\$ Sorry, I misunderstood, you meant that as an internal factor, RDS vs. VGS its not well defined, right? \$\endgroup\$ Jun 26, 2019 at 21:27
-
\$\begingroup\$ I've expanded that statement. There is a mathematical expression for \$R_{DS(ON)}\$ as a function of \$V_{GS}\$, but its also a function of a whole bunch of FET parameters that aren't found in a datasheet, and which you can't depend on being consistent from part to part, or over temperature. \$\endgroup\$ Jun 26, 2019 at 22:21
\$\begingroup\$
\$\endgroup\$
4
No this will distort the audio with quadratic variations for a fixed bias.
You want maybe a four OTA’s for a voltage controlled gain.
This is a JET designed for linear low noise differential amplifiers not log attenuators.
You will get quadratic distortion on the input signal.
< Three bucks for 1 quad log taper pot does not sound as bad as “hella expensive” ;) thumbnail
https://www.digikey.com/product-detail/en/bourns-inc/PTD904-1020K-A103/PTD904-1020K-A103-ND/3534253
-
\$\begingroup\$ 3 american dollars? For a log quad ganger? \$\endgroup\$ Jun 26, 2019 at 22:27
-
\$\begingroup\$ Yup, does that break your bank account? See blue link... \$\endgroup\$ Jun 26, 2019 at 22:28
-
\$\begingroup\$ Haven't come across any such bargain, it's a fair price indeed. Do you know any place where I can order those online? \$\endgroup\$ Jun 26, 2019 at 22:33
-
\$\begingroup\$
\$\endgroup\$
Thank you guys, I just thought of a very simple solution: get 2 dual log pots, drill a hole at the armature on one of them and glue the end of the shaft of the other to the inner plastic part of the of the drilled pot. Just opened a pot up and figured out it ain't a very hard job to do.
