Dual gang pot

[Jaguarjoe]

I discovered a new thing, it's called "thinking"

I'd like to build a Wein Bridge oscillator but don't want to sacrifice my oldest male to get a 10 turn dual gang pot.

I thought of using 2 single 10t pots geared, or cog belted together.

Then I thought of a pair of 1000 step digital pots but I wonder how well they would track.

This has severely impacted my last functioning brain cell so I'm asking you guys for input and/or other ideas.

Thanks,

Joe

 

[MikeMl]

Ganging two digital pots to the same digital signals should work ok to make a Wein Bridge. The oscillator will jump to discrete frequencies rather than tune smoothly, however. You might be able to find a "stereo digital pot" that has two digital pots already ganged.

If I was building a Wein bridge, I would try to find a ganged 365pf variable capacitor such as would be in an AM radio...

 

[Roff]

Wien bridge oscillators don't use pots, they use tracking rheostats. Good tracking results in less work for the AGC loop. Here is a dual digital rheostat that looks pretty darn good. Keep in mind that the frequency is inversely proportional to resistance, so, with a linear pot, the R-to-F curve will be hyperbolic. You want to use a resistance range of about 10:1 or 11:1, and switch ranges by switching capacitors.

The problem with 365pF variable caps is, that is the high end. To get a decade of tuning, you would have to go down to 36pF, which will generally be seriously affected by parasitic capacitances on the board.
The digital pot that I referenced has 80pF parasitic capacitance to GND on the wiper. One of the rheostats could have that pin grounded, making the capacitance a nonissue, but the other one has to float. 80pF is not insignificant in this application.

 

[Jaguarjoe]

It's moot but I chose a potentiometer because I can tie one end of the pot to the wiper. This will make a rheostat with the additional benefit of not bouncing to infinity if/when the wiper becomes noisy and bounces off the track.

I say moot because that ADN part is ideal for my needs.

Thanks Roff!

Joe

 

[Roff]

It's moot but I chose a potentiometer because I can tie one end of the pot to the wiper. This will make a rheostat with the additional benefit of not bouncing to infinity if/when the wiper becomes noisy and bounces off the track.

I say moot because that ADN part is ideal for my needs.

Thanks Roff!

Joe

First of all, digital pots do not have wiper bounce. Have you decided to go with mechanical dual pots?

I could say a lot more about the advantages and disadvantages of digital pots vs rheostats, and digital tuning vs mehanical (ganged pots) tuning, but if you really pursue this, you will probably learn most of what I could say the hard way, but it is probably the best way.

Also, you might find that digital tuning is too coarse (especially on the low resistance, high frequency end of the range).

 

[Dean Huster]

Consider the Heathkit IG-18 audio oscillator. It used decade switches for frequency setting. You can do the same with 2P10T rotary switches, one for each decade you wish to control. Lots of cheap 1% metal film resistors and six switches will give you six decades of control down to 1 Hz for up to 1 MHz (almost) maximum frequency. And cheap dual pot can provide a fine-tune control.

 

[audioguru]

My home-made Wien Bridge oscillator (it is 49 years old!) uses a dual 365pF variable capacitor for tuning. It has built-in trimmer capacitors so I can adjust for stray capacitance.

At first its level was set by an incandescent light bulb. The level bounced a lot when the frequency was changed so I smoothed it with some LEDs as limiters.
I tried a Jfet for amplitude stability but its fairly low distortion was still too high.

 

[Jaguarjoe]

Amazingly, I rifled through my box of used parts and found a Bourns 100k/10t pot which should work well.

 

[Dean Huster]

I made a cap-tuned one as well, also using a dual 365pF cap. Nice thing about the cap is that with equipment that use one (e.g., the Hewlett-Packard 200CD audio oscillator), you can bend plates to make the frequency readout dial deadly accurate. No amount of resistive tweeks can properly change a pot's tracking within its stops. I did that back in 1972 with a 200CD that was off a bit. My results would have impressed anyone in Palo Alto.

But I wonder if the double-helix resistive winding of a -- well, let's use a tradename to aptly describe it -- a Helipot would cause the oscillator to have upper frequency limitations.