Ring modulator LTSpice simulation

Before breadboarding the circuit again I thought it would be a good idea to simulate it to see what effects to expect when replacing pots and caps.

The circuit as found on the Yusynth pages (http://yusynth.net/Modular/EN/RINGMOD/index.html)

The LM1496-model was found here: 

https://forum.allaboutcircuits.com/threads/mc1496-monolithic-balanced-modulator-spice-model.47212/

With the pots centered, two 10Vpp input waves result in a 8Vpp output.

Signal and carrier

Green is output

With the 220Ohm pot fully turned to one of the sides, we get this:

 

With the 500Ohm pot fully turned to one of the sides, we get this:

It looks like the wave with the longest wavelength has gotten a DC offset

Now, if we replace the 220Ohm pot with a 500 one we get the exact same waveform when centered, byt at one of the sides we get this:

If we reduce the wiper to half way between center and one of the sides we get back to what we saw with 220Ohm:

It is thus likely that the 220Ohm potentiometer may be replaced with a 500Ohm one.

Cap biasing

The 1uF input cap on the left looks like it has one side DC biased between -3V and -4V:

The 1uF input cap on the right looks like it has one side DC biased around +7.5V

As for the 100uF cap, it is biased at around 6.5V. I cannot see any changes to the output whether I use a 200uF, 100uF, 10uF or no cap here, not with a combination of 1MHz and 10kHz, nor with the combination 1kHz and 10Hz. I think I will try using a 100uF ceramic even with the capacitance dropped to much lower by the DC bias issues such caps have.

Now, these are just simulations of course, so things may be different in practice, but it looks promising - I may be able to standardise on 500Ohm pots and use ceramics instead of electrolytics for the other caps. 

UPDATE: I have even used 500Ohm pots on my prototype so I will be fine :-D

Ceramic capacitors and DC biasing

I am considering replacing many of the high value capacitors in my circuits, which are usually electrolytic caps, with ceramics.

There is however an issue -  ceramic caps have a strange property: if they are DC biased, e.g. normally have a certain voltage across them, the effective capacitance changes:

https://community.infineon.com/t5/Knowledge-Base-Articles/DC-Bias-characteristic-of-Multilayer-Ceramic-Capacitor-MLCC/ta-p/250035

The capacitance change is relative to the capacitor size, so 1206 caps do not change as much as 0603 for example. The voltage rating however, has little effect.

As an example, if one side of the cap is connected to ground, and the other has a signal that swings around 4V, the effective capacitance will be just 30% of the rated value if we're using an 0603 cap. Using a 1206 instead gives around 70%.

C0G caps are not as affected by this as X7R, X5R, but in return they are bigger and not readily available in as high capacitances.

https://www.kyocera-avx.com/docs/techinfo/CeramicCapacitors/mlcc-dc-bias-characteristics.pdf