About Jim Williams Oscillator

DVBmaster

Member
2012-03-22 7:20 pm
Hi all!
I have some doubts about a project by Jim Williams. I am speking about the oscillator implemented in DC1858A evaluation board by LT.

The link is: https://www.analog.com/media/en/dsp-documentation/evaluation-kit-manuals/dc1858af.pdf

I am planning to build one with the additional feature of variable frequency control (10 Hz to 100kHz).
I know this could be very challenging. The main idea to achieve this is to transform the resonant R C as follows:
- R1/R8 becomes 950 ohm resistor + 10k dual-gang potentiometer
- C7/C9 is split into four caps 1.5u, 0.15n, 1.5n, 0.15n defining as many as frequency ranges

Apart from the fact that I will not probably be able to maintain the same THD performance across all frequency ranges, will I be able to keep the trimmer (R4) setting FIXED to ensure oscillation at all frequencies between 10 Hz and 100 kHz?

By looking at the schematics, R10/C10 and R11/C11 need to be adjusted to shift the cutoff frequency of the output filters according to the selected frequency range...
In the DC1858 manual there are some hints about how the output filter is adjusted according to the oscillation frequency.

The same manual shows how to change the amplitude of the generated sine wave. What I do not fully understand is at page 3, Table 1. With the configuration fout=2 kHz, Vout=18 Vpp (last row in Table 1), the designers also change C11, increasing the cutoff frequency of the output filter.... Do you have an idea about why? C10 and R10 are not changed.... Is it for stability reasons?
 

PRR

Member
Paid Member
2003-06-12 7:04 pm
Maine USA
www.diyaudio.com
The AGC gain can trim about 1%. Dual-pot tracking is unlikely to be 5%. I suspect, as proposed, it won't stay in oscillation. If the AGC range is made bigger, photo-resistor distortion starts to show.

This one is Jim's obsessive attempt at low-low THD at single frequency for testing ADCs. There are other Wien-bridges optimized for variable frequency.
 

DVBmaster

Member
2012-03-22 7:20 pm
Is the Vactec VTL5C10 optocoupler available?

It is hard to find due to its discontinuation. The NSL-32SR3 exists as possible replacement and maybe some circuit tuning.


I do not understand the concerns about using this oscillator in wideband fashion. Jim also provided a variable frequency version with ouptocoupler and as non linear resistive element. He used a 2k ohm dual gang pot and 3 cap pairs defining 3 frequency ranges...
 

BSST

Member
2018-07-07 2:21 am
I've not investigated the Vactec optocoupler but the device cited in the Williams/Hoover ap-note is available at Digikey: NSL-32SR3 Advanced Photonix | Isolators | DigiKey

This type of photocell optocoupler has relatively slow control response; I don't know if they are fast enough to serve as the gain control element for an oscillator at 100kHz. I've seen only FET or analog multipliers in 100kHz oscillators. They may be usable--- this is only a cautionary comment.
 

BSST

Member
2018-07-07 2:21 am
It is hard to find due to its discontinuation. The NSL-32SR3 exists as possible replacement and maybe some circuit tuning.


I do not understand the concerns about using this oscillator in wideband fashion. Jim also provided a variable frequency version with ouptocoupler and as non linear resistive element. He used a 2k ohm dual gang pot and 3 cap pairs defining 3 frequency ranges...

I just now discovered your earlier post.

Can you point me toward Williams' mutl-range design? Many thanks!
 

DVBmaster

Member
2012-03-22 7:20 pm
I just now discovered your earlier post.

Can you point me toward Williams' mutl-range design? Many thanks!


Sure! Here's the link:
https://www.analog.com/media/en/technical-documentation/application-notes/an43f.pdf
Starting from page 29. In the first example (Figure 40) Jim still uses the lamps. The first version of a tunable oscillator is at page 32, Figure 47. Finally, in Figure 48, common mode suppressin is added.

Some of the options for the non-linear resistors are, as far as I know:
- Lamps
- FET, needing a trimmer to properly reduce 2nd harmonic distortion
- Optocuplers with resistive output
- A transconductance amplifier