Circuit debugging help please.

[zoulzubazz]

Hello guys the following is the circuit i am trying to debug. It is an impedance estimation circuit. I am using this circuit to pass a constant of 250uA through load R (shown in red, the switches Sw1 and Sw2 are in place to change the load to a different impedance). The XR2206 gives a sinusoidal output which is fed to a grounded Howland current source (VCCS), providing the constant current which i need to push through the load R. a 100 ohm resistor is in place to accurately measure the current through R (which is around 250uA as mentioned earlier). Ch1 and Ch2 are two channels of a oscilloscope. Here is the problem, when both Ch1 and Ch2 are connected, the magnitude of voltage across 100 ohm resistor is to shown close 4mV (which should be 25mV since 250uA is passed through 100ohms). Now when i divide the voltage shown across Ch1 by 250uA the estimate of R is exact, although theoretically current flowing through 100ohms is (4mV/100 and not 250uA). Now when the lead of Ch1 shown in red is disconnected the voltage across 100ohm series resistor is 25mV (confirming a flow of 250uA of current) but the magnitude of voltage across R is elevated resulting in a higher value estimated for R when divided by 250uA. Hope the description is not too convoluted. any help would be much appreciated.
the switches are IC switch package (AD436).
Thanks.

 

[E-design]

In a scope, the ground leads are normally common to one ground, unless you have a battery portable scope with floating separate channels & grounds. If you connect your CH1 probe over the test resistor with the ground clip at the lower end, that point will be shorted to ground when you connect CH2 over the 100Ω resistor. So very little or no current will go through the 100Ω resistor.

 

[zoulzubazz]

Thanks a lot, I did find that the oscilloscope ground was causing trouble, it's a 2 channel DSO btw. I have run into a few further issues. I have completely removed ch1 for the time being and am focusing on current flowing through the 100 ohm resistor hence the voltage across ch2. Now after some scrutiny I am observing a rather strange phenomenon and am unable to figure out what is happening.when I power on the circuit and measuring ac voltage( ac current being passed through R) across the 100ohm resistor using a multimeter and a daq device there is no sign of any credible current flowing through 100 ohm resistor, but after connecting ch2 of the scope across 100 Ohms I see the ac current being passed on the scope and this time the multimeter and Daq also register this current flow. Sometimes even after dis connecting ch2 the multimeter shows the right ac current value but after is re start the circuit the same strange thing happens. Connecting the power supply's earth-ground to the dual supply ground helps and sometimes powering the circuit from a battery resolves this issue ,but not always. I sense there is something strange happening at the circuit ground but am unable to figure out what. Any help would be much appreciated. Thanks a lot.

 

[tckev]

From past experience with analog switches, similar to what you are using, two things come to mind -

1. Ron (the on resistance) of these switches can changes drastically with supply voltage. Doubling in value from the manufacturers published headline value as the supply voltage is reduced is not unusual.

2. Analog switches are touchy about power supply sequencing. To quote from Analog Device on this topic -
"Care must be taken to ensure correct power-supply sequencing.
Incorrect power-supply sequencing can result in the device being subjected to stresses
beyond those listed in the Absolute Maximum Ratings. Always sequence VDD on first
followed by VSS and the logic signals. An external, analog or digital, signal can then be
safely presented to the source or drain of the switch."
****************Other manufacturers have similar statements.***************

I would advise a good look at the switch manufacturers datasheet if you haven't already done so.
And good luck.