Proper methods to isolate test/measure circuits

Thread Starter

spmaddox

Joined Nov 29, 2018
49
Hi everyone,

I'm working on a design in which I have multiple circuits for which I'm monitoring voltage as a means to detect state change of each circuit, specifically the state (open or closed) of the switch(es) on the circuit. At present the circuits share a common 12V DC source and are all grounded to the same grounding plane. Additionally the ICs I am using to enable/perform state change detection and voltage measurement (digipot, comparator, ADC) share the same 12V DC source and grounding plane though the source is stepped down to 5V and 3.3V via voltage dividers to supply the correct voltage for each IC.

I'm relatively new to hardware design though after the time i've spent learning I'm not surprised by the fact that I am seeing state changes in one or more circuits having an impact on the voltages i'm reading on the other circuits. I'm now looking at ways I can mitigate this such that the state changes on a circuit does not have an impact on the other circuits.

Using the diagram i've attached as a reference if the switch on Test Circuit 1 opens or closes I need to insure that the voltage on Test Circuit 2 and Test Circuit 3 does not change as a result.

Is the proper method for doing this to place an optoisolator between the 12V source and each circuit? Early on I had the supply and ground for the circuits on a separate supply/ground from the comparator, digipot and ADC and was not able to correctly measure voltages until I put them all on a common supply/ground. I'm curious if putting the optoisolator in place creates a situation where I need to put the comparator, digipot and ADC on the same, now isolated, supply/ground as each circuit or if there is something that I am missing.

Are there other options for keeping the circuits isolated but on a common supply/ground?

A couple of points about the diagram;

1. The portion of each circuit between the junction of each circuit to the 12V supply and the TVS are external to my device; my device provides the 12V on the "output" terminal and the "input" terminal begins just ahead of the TVS.
2. The external portion of each of these circuits may or may not include one or both of the external resistors and they may be anywhere from 0K to 5.6K ohm resistors; i've colored them red to indicate they may or may not be present and i've labeled them as Ext R# to denote they are external to my device. My device is designed to accommodate these variances.

Any advice or guidance is greatly appreciated.

thx - sean


Example Circuit Schematic.jpg
 

KeithWalker

Joined Jul 10, 2017
3,063
Hi everyone,

I'm working on a design in which I have multiple circuits for which I'm monitoring voltage as a means to detect state change of each circuit, specifically the state (open or closed) of the switch(es) on the circuit. At present the circuits share a common 12V DC source and are all grounded to the same grounding plane. Additionally the ICs I am using to enable/perform state change detection and voltage measurement (digipot, comparator, ADC) share the same 12V DC source and grounding plane though the source is stepped down to 5V and 3.3V via voltage dividers to supply the correct voltage for each IC.

I'm relatively new to hardware design though after the time i've spent learning I'm not surprised by the fact that I am seeing state changes in one or more circuits having an impact on the voltages i'm reading on the other circuits. I'm now looking at ways I can mitigate this such that the state changes on a circuit does not have an impact on the other circuits.

Using the diagram i've attached as a reference if the switch on Test Circuit 1 opens or closes I need to insure that the voltage on Test Circuit 2 and Test Circuit 3 does not change as a result.

Is the proper method for doing this to place an optoisolator between the 12V source and each circuit? Early on I had the supply and ground for the circuits on a separate supply/ground from the comparator, digipot and ADC and was not able to correctly measure voltages until I put them all on a common supply/ground. I'm curious if putting the optoisolator in place creates a situation where I need to put the comparator, digipot and ADC on the same, now isolated, supply/ground as each circuit or if there is something that I am missing.

Are there other options for keeping the circuits isolated but on a common supply/ground?

A couple of points about the diagram;

1. The portion of each circuit between the junction of each circuit to the 12V supply and the TVS are external to my device; my device provides the 12V on the "output" terminal and the "input" terminal begins just ahead of the TVS.
2. The external portion of each of these circuits may or may not include one or both of the external resistors and they may be anywhere from 0K to 5.6K ohm resistors; i've colored them red to indicate they may or may not be present and i've labeled them as Ext R# to denote they are external to my device. My device is designed to accommodate these variances.

Any advice or guidance is greatly appreciated.

thx - sean


View attachment 188234
I see no reason why the circuits would interfere with each other unless you have a poorly regulated 12 volt supply. Does the voltage on this supply change when switches are opened and closed?
 

ronsimpson

Joined Oct 7, 2019
2,986
Comment about your "comparator".
Most can measure voltages from slightly below ground to the supply voltage-2V. (very dependent on what part you have)
The digi-pot can send 5V to the comparator. With a 3.3V supply the part probably can't see any voltage above 2.5V. See the data sheet.
 

Thread Starter

spmaddox

Joined Nov 29, 2018
49
Hi Keith - the 12V comes from my bench power supply and I do not have a voltage regulator in place between it and these circuits.

At the moment I have the bench power supply set to 3.3V as I was testing some other things and at that level with all of the circuits closed the voltage i read at the junctions between the circuits and the voltage source is 3.29V. As I open each circuit the voltage increases slightly and I end up back at 3.3V when all 3 of the circuits are open. Circuit 1 doesn't actually have Ext R1 or Ext R2, Circuit 2 has Ext R4 @ 2K ohms and does not have Ext R3 while Circuit 3 has both Ext R5 & Ext R6 with both being 5.6K ohms. I should have done the drawing showing the actual configuration of these instead of the template/general circuit design. I do see V change by a different amount with the state change of each circuit which tells me its the resistors on the circuits causing the V change.

Same as what I witnessed when the bench power supply is at 12V, the V change each time was just proportionally more.
 

Thread Starter

spmaddox

Joined Nov 29, 2018
49
Hi Ron - yep the 3.3V supply to the comparator in the diagram is an oversight on my end as I've been working with both a 3.3V and 12V/5V model as the V being supplied to the test/measure circuits. In my 12V/5V model the VCC for the comparator is 5V. In my 3.3V model R1 is present as 100 ohm and R2 is not present, i don't need to voltage divide the voltage for the comparator in that setup. The comparator for both the 3.3V and the 12V/5V model is the ADCMP393...I just didn't update the voltage in the diagram I shared.
 

KeithWalker

Joined Jul 10, 2017
3,063
Hi Keith - the 12V comes from my bench power supply and I do not have a voltage regulator in place between it and these circuits.

At the moment I have the bench power supply set to 3.3V as I was testing some other things and at that level with all of the circuits closed the voltage i read at the junctions between the circuits and the voltage source is 3.29V. As I open each circuit the voltage increases slightly and I end up back at 3.3V when all 3 of the circuits are open. Circuit 1 doesn't actually have Ext R1 or Ext R2, Circuit 2 has Ext R4 @ 2K ohms and does not have Ext R3 while Circuit 3 has both Ext R5 & Ext R6 with both being 5.6K ohms. I should have done the drawing showing the actual configuration of these instead of the template/general circuit design. I do see V change by a different amount with the state change of each circuit which tells me its the resistors on the circuits causing the V change.

Same as what I witnessed when the bench power supply is at 12V, the V change each time was just proportionally more.
Then that is what is causing your inter-action problems. You need to make sure that the supply is stable for each circuit.
Regards, Keith
 

Thread Starter

spmaddox

Joined Nov 29, 2018
49
Hi Keith - so a 12V to 12V voltage regulator? Should I put one between the bench power supply and the feed to all of the circuits or one on each circuit? Once I move this to a PCB I would think that the transformer I put in place to take external power and output 12V would eliminate the need for the regulator correct? Assuming transformer puts out a stable well regulated 12V.
 

KeithWalker

Joined Jul 10, 2017
3,063
Hi Keith - so a 12V to 12V voltage regulator? Should I put one between the bench power supply and the feed to all of the circuits or one on each circuit? Once I move this to a PCB I would think that the transformer I put in place to take external power and output 12V would eliminate the need for the regulator correct? Assuming transformer puts out a stable well regulated 12V.
If you are going to power the three circuits from the same power supply you must make sure that it is very well regulated. That is; when any or all of the switches close, the supply voltage does not change.
Ideally, each circuit should have it's own voltage regulator to ensure that there is no interaction:

Supply.jpg
 
Top