0
\$\begingroup\$

Below is an LTSpice simulation of two attempts at a Mosfet bilateral switch or 'transmission gate'. The plot shows the switch (onoff) going from 0 to 5v (in blue) and a sine wave audio signal being moved up or down by the two circuits.

What I was hoping to see is the sine wave, only between the blue lines, and nothing either side.

Can anyone tell me what is wrong with these two circuits?

I know that I can use a CD4066 etc., or a relay, but now I've become obsessed with getting a working bilateral switch circuit.

enter image description here

\$\endgroup\$
  • \$\begingroup\$ What are the VDD/GND voltages to the 2 inverters? Also, bring the audio voltage down to VDD/2 +- 0.5 volt, thus well inside the rails. \$\endgroup\$ – analogsystemsrf Aug 28 '18 at 15:59
  • \$\begingroup\$ The idea is to switch these from a 74 series chip, so 'onoff' is 0 or 5V. the audio signal is plus or minus 0.7v. Is this the problem, do I need to DC shift the audio signal so that it is above ground? The plot shows the transmission gate allowing the original zero centred AC Signal through correctly, it's when the switch is off that it goes all funny. \$\endgroup\$ – Richard Aug 28 '18 at 16:43
  • 1
    \$\begingroup\$ You show two circuits and the top one won't work because the substrate connections need to connect to power rails and not their respective sources. \$\endgroup\$ – Andy aka Aug 28 '18 at 17:00
  • \$\begingroup\$ Are you saying that I need a 4 pin mosfet? I've seen diagrams like that. They don't appear to exist in small signal varieties. Am I doing something unusual here? Further, I've no idea how to simulate that in LTSpice, or where to get someone elses attempt at it. \$\endgroup\$ – Richard Aug 28 '18 at 17:02
  • 1
    \$\begingroup\$ Vt or Vgs(th) =1.5V on both with matched RdsOn 50~100 Ohms \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Aug 28 '18 at 20:20
0
\$\begingroup\$

After the helpful comments above I managed to crack it. So here for the good of humanity is an LTSpice simulation of a MOSFET transmission gate, switching on and off a sine wave sweep from 20Hz to 20KHz over two seconds. If anyone can find a complimentary P-Channel MOSFET to the BSS83 that has a substrate pin, I would be eternally grateful. Also if anyone could provide better parameters for the pmos model than the ones shown in the simulation below, please do :)

Note there are some small transients on switching which may cause popping, but this is surprisingly clean switching, better than is likely from a relay. There is some signal attenuation, but again, surprisingly little. I think this is actually viable as a cheap, fairly low distortion dual polarity audio switch. I am super pleased that I decided to stick with this, and post here for help.

enter image description here

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ Adding a 68nF capacitor to ground parallel to R3 removes the transients, making this clean switching. \$\endgroup\$ – Richard Aug 29 '18 at 0:08
  • \$\begingroup\$ Make that more like 2.2nF \$\endgroup\$ – Richard Aug 29 '18 at 0:15
  • \$\begingroup\$ MIC94030 MIC94031 MIC94050 MIC94051 are smd p-channel fets with substrate pin, may be complimentary to BSS83 \$\endgroup\$ – Richard Aug 29 '18 at 9:19

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.