5
\$\begingroup\$

This will be a conceptual question. I'm sometimes dealing with data-acquisition of transducers, like strain gauges, accelerometers ect. sort of sensors. Most of these sensors have their own precision amplifiers. So what I mean by the transducer output is the amplified sensor signal. These signals then go to data-acquisition's input amplifier which is simply a differential amplifier ect. But most of the time the transducer outputs are single ended. Sometimes I encounter all sorts of noise, common mode noise ect.

Since differential signalling is more immune to noise, I thought about converting a single ended signalling to a differential signalling as below(I want to implement Figure 2):

enter image description here

So here are my questions.

1-) Some transducers are manufactured and sold as differential signalling transducers. So they are ready to be wired to a differential amplifier. But if one has a transducer and want to use it as differential signalling as in my Figure 2, would that be a wrong treatment? Im asking because If I invert the signal myself to obtain a differential signalling as in Figure 2, then I might introduce noise to the inverted input by interacting it with the inverting opAmp circuit, and that will not be common on both signals. So my first question is: is it a common practice to convert single ended signalling to differential ended signalling(in the aim of noise immunity) where the transducer was actually designed for single ended signalling?

2-) If this method makes sense. Here is the typical inverting opAmp configuration:

enter image description here

I would choose R1 and R2 10k. How does the input impedance of the data-acquisition's differential amplifier have affect on choosing R1 and R2 here? I want the inverting as precise as possible. Is there an opAmp category for that, an example would be great? I dont want use LM741 for instance.

\$\endgroup\$

3 Answers 3

Reset to default
4
\$\begingroup\$

Since differential signalling is more immune to noise

Any signalling is susceptible to noise - it's how your receive amplifier handles those received signals that determines how much immunity can be acquired.

However, you can have a perfect differential amplifier attached to a single ended source (via a properly balanced cable) that has problems. If the output impedance of the hot wire is several tens of ohms compared to the impedance of the 0 volt transmit reference you have what is known as "earth impedance imbalance". Note that I said imbalance.

If noise comes along and "hits" the cable, it will develop a larger signal on the hot output than that developed on the 0 volt reference signal. Here's what I mean for a good scenario: -

enter image description here

The signal source is "perfect" in that it presents the same low impedance for hot wire as 0 volt reference. Clearly, if any noise comes along then it hits both wires in the cable and, because both wires have equal impedance balance to ground, the noise received by the diff amp is equal and can be quite easily cancelled.

If the signal source has an output impedance that isn't zero then there could be a problem that can be overcome by this: -

enter image description here

Now, the impedances are largely the same - the added resistors are chosen to be identical and "swamp" the difference in impedance between hot wire and 0 volt reference. Earth impedance balance will be good and noise will be the same on both received wires (providing your input amplifier has good input earth impedance balance as well).

Adding an inverting stage can make things worse - keep the earth impedance balance at the sending end good and you minimize problems without adding an amplifier. Of course, in extreme circumstances you have to transmit a bigger signal and this can be done (carefully) with a balanced buffer. To keep "balance" (the same for both signals) use an inverting amplifier and a non-inverting amplifier - this largely ensures that the impedance at high frequencies will be equal.

You cannot achieve this using the "original" signal and a buffer amplifier because you have no way of controlling the impedances relative to each other. If it works it's just luck and that's not good engineering.

\$\endgroup\$
10
  • \$\begingroup\$ I think I get the idea. So if I have single ended transducer, I can add some resistor between sensor GND and measurement differential amplifier's GND; same impedances would create same noise and they will be subtracted. But how can I measure the impedance of the hot and so I can add a series resistor to ground connection? \$\endgroup\$
    – floppy380
    Mar 30, 2017 at 12:39
  • 1
    \$\begingroup\$ Usually, you add (say) 100 ohms to each line so that the 100 ohms becomes the dominant impedance. If the transducer output impedance is still dominant (read the data sheet) then it becomes more problematic; you then should buffer the transducer and buffer the 0 volt reference. If buffer amps are the same part then you obtain balance plus you can add 50 or 100 ohms as well. If the DS doesn't tell you then it's probably safest to buffer rather than measure. \$\endgroup\$
    – Andy aka
    Mar 30, 2017 at 12:47
  • \$\begingroup\$ This 100ohm scenario only apply for one traducer hooked up to a channel right? Imagine I have four single ended transducers goes to a four data-acquisition channels by BNC cables. Is it okay to add 100 ohm per each transducer's ground because in that case it will be four 100 ohm resistor in parallel I though? Shoud I than add 400 ohm to each transducer ground. Or simply add 100hm to AIGND in series of the daq system? And how can I add 100 Ohm series in practice to a BNC cable? Should I basically cut it? Im sorry to ask these may sound stupid but I really suffer from practical information. \$\endgroup\$
    – floppy380
    Mar 30, 2017 at 12:56
  • \$\begingroup\$ You MUST treat each transducer as a separate signal and reference connection. You must not common up the references at the receive end. OK that sounds harsh and between 0.5 metres and 100 metres of cable the rules can be bent but if you always stick to the rule above you won't go far wrong. \$\endgroup\$
    – Andy aka
    Mar 30, 2017 at 13:58
  • \$\begingroup\$ But in single ended daq the sensor grounds are all connected together to the analog input ground and to the earth. At the receive end the grounds are connected to the AIGND. Im talking about wiring single ended sensors to a single ended daq board. \$\endgroup\$
    – floppy380
    Mar 31, 2017 at 23:34
3
\$\begingroup\$

The important thing about balanced lines for interference rejection is that the source impedance is matched, not that the voltage be differential.

Due to this fact you can do nearly as well as an active differential stage simply by matching the impedance to ground between the two legs at the transducer.

A resistor connecting the -a leg to ground at the transducer, selected to match the output impedance of the sensor will cause the differential input amplifier to reject significant interference (And does not need any power at the transducer).

The way to think about this stuff is to see it as a Wheatstone bridge where the exciter voltage is the noise and the signal voltage is series injected either into one leg or in anti phase in both legs, from which we can see that providing the resistances are matched you get cancellation even if the wanted signal is only driven onto one leg.

Incidentally, high common mode impedance at the receiver will substantially reduce the impact of small mismatches in source impedance.

If you are working in audio sorts of bandwidths, THAT CORP have an excellent (patented) bootstrapped receiver that is really very good. http://www.thatcorp.com/datashts/THAT_1200-Series_Datasheet.pdf

I would also suggest that reading a paper on this by Bill Whitlock of Jensen transformer may be interesting. https://sound-au.com/articles/balanced-interfaces.pdf

\$\endgroup\$
5
  • \$\begingroup\$ Andy got there first! \$\endgroup\$
    – Dan Mills
    Mar 30, 2017 at 10:15
  • \$\begingroup\$ But your links are awesome, thanks! \$\endgroup\$
    – pipe
    Mar 30, 2017 at 10:53
  • \$\begingroup\$ Nice explanation. Do you have also source/link how I can measure impedance of a line and perform impedance matching? Its only theory everywhere. \$\endgroup\$
    – floppy380
    Mar 30, 2017 at 12:41
  • \$\begingroup\$ Impedance matching is usually a term that applies to transmission lines (Lines that are a significant fraction of a quarter wavelength long at the frequency of interest), I got the impression this was low frequency stuff you were dealing with? To measure the low frequency source impedance, measure the voltage while switching in two different resistive loads then solve the resulting simultaneous eqns. Alternatively replace the resistor with a pot and adjust till the loaded voltage is half the open circuit voltage then disconnect and measure the pot, pot resistance will equal source resistance. \$\endgroup\$
    – Dan Mills
    Mar 30, 2017 at 14:48
  • \$\begingroup\$ i see what you mean. but what scares me to do this is: imagine the source sensor's unknown output+line impedance is 100 ohm, and when I set the pot to 100 ohm at the end of the line I can found the source impedance when the voltage is half. so far good. but then the source will be overloaded by only seeing 200 Ohm? or is it safe? thanks \$\endgroup\$
    – floppy380
    Apr 1, 2017 at 13:09
2
\$\begingroup\$

I do not think you should unbalance it like that. There is no way you will be able to match the impedance of both signals. Whether you really need to is answered by others here.

If you are doing it yourself, you should use a single to differential converter like a LT6350

enter image description here

However, you have to get the power supplies over there and filter them well.

Also check your signal range is within the range of the device.

If this is an industrial application, there may be such conversion devices, pre-amps, available on the market for a suitably overpriced cost.

\$\endgroup\$
2
  • \$\begingroup\$ would this IC overcome the impedance matching issue or still it will remain? \$\endgroup\$
    – floppy380
    Mar 30, 2017 at 12:36
  • \$\begingroup\$ @doncarlos with the appropriate RL is should. \$\endgroup\$
    – Trevor_G
    Mar 30, 2017 at 12:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.