3
\$\begingroup\$

I designed this circuit to scale a sine wave with a +/-5 V swing to scale to 0-5 V. Also I'd like to be able to attenuate the input.

That's why the potentiometer is there (connected as a variable resistor). Question: I made a simulation with the potentiometer pin 3 open and a simulation with pin 3 connected to ground. Same result. I've seen both practices in similar situations.

What is good practice to do?

Circuit

\$\endgroup\$
3
  • \$\begingroup\$ What's the value of the pot (and why isn't it on the schematic)? What gain adjustment do you require? \$\endgroup\$
    – Transistor
    May 5 '20 at 19:51
  • \$\begingroup\$ the pot is 10k, must have 'walked away'. In the mean time: change of plans : read below \$\endgroup\$ May 5 '20 at 20:42
  • \$\begingroup\$ It is not a good idea to vary R1 in an inverting amplifier because of two reasons: the function is not linear; the input resistance varies and can become zero. So, in some cases it would be better to connect the potentiometer (as a rheostat) in series with R2. \$\endgroup\$ May 5 '20 at 20:45
5
\$\begingroup\$

schematic

simulate this circuit – Schematic created using CircuitLab

Figure 1. (a) The original circuit. (b) The alternate circuit. Both ignore the non-inverting offset reference for simplicity.

In Figure 1a the gain can be adjusted from \$ - \frac {10k} {20k} \$ to \$ - \frac {10k}{30k} \$, -0.5 to -0.33. It is common to connect the wiper to one end of the pot so that if there's a bad spot on the track that a completely open circuit is avoided.

In Figure 1b the overall circuit gain can be adjusted by R6 from 0 (wiper at bottom) to -0.5 (wiper at top).

Which is better is up to you.

\$\endgroup\$
3
  • \$\begingroup\$ Ok fair is fair :-) You guys are geniuses, i totally forgot about the 'loop' option. Thank you both! (I'll opt for b then) \$\endgroup\$ May 5 '20 at 20:29
  • \$\begingroup\$ Thanks for accepting the answer. We normally recommend that you don't accept until the Earth has rotated once or twice so that the whole of humanity has a chance to answer. You might get some other insights. You can unaccept for now if you wish. \$\endgroup\$
    – Transistor
    May 5 '20 at 20:48
  • \$\begingroup\$ A disadvantage of using a potentiometer as a variable resistor is that the tolerance of the pot is important where when used as a potential divider much less so. If you do use one as a variable resistor it is usually good to put a large resistor in series to restrict the range of operation and swamp any tolerance errors. \$\endgroup\$ May 5 '20 at 20:53
3
\$\begingroup\$

Definitely DO NOT GROUND (at least with the way you have it in your schematic). That changes the circuit by sticking a resistor in parallel with your desired variable resistor. When using a pot as a variable resistor one end pin is just left open.

Why you get the same result whether that end is connected or left open? I don't know. That doesn't make sense.

EDIT: I forgot, but you can turn a pot into a variable resistor by connecting up all three terminals such that the wiper shorts across the resistance to be ignored. So you could connect terminal 2 and 3 together (see (a) in Transistor's answer).

I guess you could connect to GND as shown by @Transistor's (b) example. That has a similar effect to what you described as wanting but sort of changes the "intent" of how you were thinking the pot would be used in your circuit so it didn't occur to me. You should probably pick his answer. It's better.

\$\endgroup\$
4
  • \$\begingroup\$ Thanks, I'm gonna stick to my initial design already. I just had this hunch of grounding the pot, but found it strange the sim wasn't displaying what I expected. So, leaving the pin open :-) \$\endgroup\$ May 5 '20 at 19:56
  • \$\begingroup\$ DKN, you may have misunderstood the question. See my answer. Maybe I did? \$\endgroup\$
    – Transistor
    May 5 '20 at 20:13
  • \$\begingroup\$ @Transistor I don't think you misunderstood either. Just alternate approaches to use the pot that also result in variable gain? \$\endgroup\$
    – DKNguyen
    May 5 '20 at 20:17
  • \$\begingroup\$ "You should probably pick his answer. It's better." Just to let you know, you are too kind! Thanks for helping me out! \$\endgroup\$ May 5 '20 at 20:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.