Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Switching between several shunt resistors

Status
Not open for further replies.
M

mamech

Full Member level 3
Joined
Nov 9, 2010
Messages
176
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,296
Activity points
3,135
hello

I want to design a circuit that is capable of switching between several shunt resistors to have periodic check of values for each of them.

This idea came to our mind, when we found that the dedicated power ics usually have limited analog channels, and we want to multiplex several shunts.

another problem, is that the minimum current that we want to measure in each shunt is about 1 mA, and this cause in shunt around 10 uv .

We thought in analog multiplexers, but I doubt that they can transmit this AC 10 uV without distorting them. we thought about amplifying the output from shunt using opamp so the analog multiplexer can deal with it, but again, I did not find suitable opamp that can amplify 10 uv with acceptable uncertainty (many opamps measure less than 10uv, but the datasheet mentions that this value can be up to 30uv ), and reasonable price.

are there any suggestions ?
 

Hi,

you surely need differential measurement.
This can be done with analog multiplexers. I don´t expect errors in the range of 10uV...with good layout and well calculated circuit.

For sure if you can amplify the signal before multiplexing it, it will reduce errors.
Use the interactive selection guides for OPAMPs. Most distributors and OPAMP manufacturers have them.

****
For a more detailed answer we need more detailed informations.
Voltages, currents, supplies, frequency, sampling rate, source impedances, circuit, layout, devices....

Klaus
 
  • Like
Reactions: mamech

    M

    mamech

    Points: 2
    Helpful Answer Positive Rating
Do you want a multiplexer to select the signals from multiple shunt resistor circuits, or a multiplexer to change choose one of multiple shunt resistors to be used in a single circuit? Or to put it another way, does the shunt current actually need to pass through the multiplexer?

Passive analog switches should not add significant error to sense voltages, so long as the source impedance is low and you're using differential signaling. Your primary concern is DC offset and CMRR, both of which are basically zero for analog switches (at low frequencies).
 
  • Like
Reactions: mamech

    M

    mamech

    Points: 2
    Helpful Answer Positive Rating
mtwieg said:
Do you want a multiplexer to select the signals from multiple shunt resistor circuits, or a multiplexer to change choose one of multiple shunt resistors to be used in a single circuit? Or to put it another way, does the shunt current actually need to pass through the multiplexer?

Passive analog switches should not add significant error to sense voltages, so long as the source impedance is low and you're using differential signaling. Your primary concern is DC offset and CMRR, both of which are basically zero for analog switches (at low frequencies).
Click to expand...

Actually, my main interest to switch between "differential voltage" that comes from several shunts, so I can select which load will be measured by my power IC. I realized now that DC offset may be a problem because the lowest current can make around 10 uV on each shunt, so I think that adding a DC offset before multiplexer and subtracting it after the multiplexer can do the job.
 

Hi,
I think that adding a DC offset before multiplexer and subtracting it after the multiplexer can do the job.
Click to expand...
This doesn´t reduce offset errors, I rather assume it increases offset errors.

Klaus
 
  • Like
Reactions: mamech

    M

    mamech

    Points: 2
    Helpful Answer Positive Rating
What you probably want, is to measure each shunt
at two timepoints, one where the result is "zero"
(like, switch is off, right before next turnon) and
then another where the voltage is significant and
most stable (dead quiet in a switcher, ain't gonna
happen - but pick the best, perhaps right before
switch turnoff).

There exist integrated shunt-and-sense ICs that
are made for current control. Close-in gain is a
good thing (relative to the idea of long wires, and
muxes, and op amps all lashed up).

A shunt with 10uV on it, do you really care about?
Not really getting what you're after here, sort of
seems like a "health check" but how often does
this really need to be done? And under what
conditions? Maybe just make this a go/no-go test
that only flags if shunt "should" be making serious
output, and doesn't - rather than trying to work
the baby end?
 
  • Like
Reactions: mamech

    M

    mamech

    Points: 2
    Helpful Answer Positive Rating
mamech said:
hello

I want to design a circuit that is capable of switching between several shunt resistors to have periodic check of values for each of them.

This idea came to our mind, when we found that the dedicated power ics usually have limited analog channels, and we want to multiplex several shunts.

another problem, is that the minimum current that we want to measure in each shunt is about 1 mA, and this cause in shunt around 10 uv .

We thought in analog multiplexers, but I doubt that they can transmit this AC 10 uV without distorting them. we thought about amplifying the output from shunt using opamp so the analog multiplexer can deal with it, but again, I did not find suitable opamp that can amplify 10 uv with acceptable uncertainty (many opamps measure less than 10uv, but the datasheet mentions that this value can be up to 30uv ), and reasonable price.

are there any suggestions ?
Click to expand...
The only solution I can think of, is to inject an ac current into the shunt, and then amplify the resulting ac voltage as you are already proposing.

The big advantage of usng ac is that it eliminates dc errors due to thermoelectic effects and amplifier dc drift and offset errors.
Amplifying 10uV of ac should not present that much of a difficulty, particularly if the amplifier is made to reject other noise frequencies.

You can also use some kind of synchronous detection and averaging a great many measurement samples to further reduce random noise effects.

It should be possible to make some very accurate and repeatable measurements at 10uV or less using statistical averaging of many readings.

Injecting an ac current also eliminates any errors due to variable series resistance in the injection process caused by switches or multiplexers.
 
  • Like
Reactions: mamech

    M

    mamech

    Points: 2
    Helpful Answer Positive Rating
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top