How to find the value of POTs to use?

[alec_t]

Could anyone of you guys explain me how you came up with those values which is what I am wanting to know.

I just chose values which are a compromise between (a) having current through the pots as low as possible to save power and (b) having pot current at least an order of magnitude higher than the input bias current of any following op-amp.

 

[pavjayt]

I ended up using two 100K pots in place of 6.14K (R1,R4) resistors and four 200K pots in place of U1 parallel to 10K pot (U4). This gives the flexibility to set the range and can use for 4 channels. Will it be OK?

thanks

 

[MikeMl]

The problem I see is asymmetric setting of R1 & R4. That's why I asked about a ganged pot in the earlier post. Also why I came up with the dual opamp driver to preserve symmetry.

 

[pavjayt]

Thanks Mike, I understand why you mentioned about ganged POT earlier. But they are very limited in the types they are available. Best thing I can do is to make sure they both are matched as closely as possible whenever the limit is being changed.

As for an off-topic question, is it always better to connect the PCB stand-off holes to the ground plane? For example in my design attached, is it safe to connect the drill holes in the corners to ground plane? Also does the design look good enough or do I need to do more modifications as for the routing to avoid noise/interference. I tried to place all my traces on the top plane as much as possible while routing power lines on the bottom where needed. I know its hard to tell exactly, but visually do you have any suggestions to improve?

thanks

 

[pavjayt]

Paralleling four 50K pots would result in an effective resistance for U1 of 12.5KΩ, so that would necessitate re-figuring everything. Just use four 200K pots in parallel; that makes 50KΩ as before...

Looks like 200K pots are way over priced ($80 a piece) compared to a 50K that you can get for $15. Any other solution that we can use in this case?

thanks

 

[MikeMl]

Go back to driving the ends of the center pot(s) with the push-pull opamp circuit. That way the range pot can be almost anything, and so can the four center pots from which you take the small centering voltage. As long as the total current that flows in the center pots can be supplied by the opamps, you are good to go.

 

[pavjayt]

Go back to driving the ends of the center pot(s) with the push-pull opamp circuit. That way the range pot can be almost anything, and so can the four center pots from which you take the small centering voltage. As long as the total current that flows in the center pots can be supplied by the opamps, you are good to go.

Well, that will be a good design with more flexibility and ease of use on setting the value. Can I use any range pots for the four center pots (in place of U1) or should they be only 10K pots?

 

[MikeMl]

The lower limit is determined by the max current output from the opamp. Likely 10K or higher would fine. Check that the opamps can drive 2.5K (10K/4).

Upper limit determined of the input bias current of the opamp connected to the wiper(s).

 

[pavjayt]

The lower limit is determined by the max current output from the opamp. Likely 10K or higher would fine. Check that the opamps can drive 2.5K (10K/4).

Upper limit determined of the input bias current of the opamp connected to the wiper(s).

Thanks Mike. I am going to use OPA627 op-amps which I guess should be fine with 2.5K total.