This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
I'm currently designing an amplifier circuit for a photodiode. The photodiode uses a trans impedance amplifier initially but, to keep the bandwidth up, this has fairly low gain. In order to boost the signal later on I've decided to use an instrumentation amplifier.
In choosing one of these I've been looking for low noise, high stability and low thermal drift. I came across the INA326 previously and thought to apply it here: it has some very clever current mirroring going on which removes the need for internal trimmed resistors. However, in comparing it with e.g. the INA103, I found that it underperforms on most all figures of merit.
If that's true then fine, I can use the INA103 instead, but I think I must be misunderstanding the INA326's strengths. Surely the added internal complexity of the current mirroring circuitry must have been done for a good reason, but so far as I can see all it achieved was to limit the bandwidth by 4 orders of magnitude, which makes me think I must be missing something.
So my question is, other than the past-the-rail output, what advantages does the INA326 have over more conventional instrumentation amplifiers (particularly in unity / low gain systems)?
Thanks in advance!
Charles
Charles,
INA326 and INA103 could hardly be more different from one another:
For photodiode application, you typicaly need the lowest possible input bias current, IB, (INA103 has maximum IB of 12uA vs INA326 max IB of +/-2nA) but I take it you use low IB tran-impedance amplifier in front of the INA so it might be fine.
INA326 is a true single supply instrumentation amplifier with the input voltage range beyond its rails, the output voltage within 10mV of either rail and can operate from 2.7V to 5.5V supplies. INA103, on the other hand, input common-mode voltage is 4V from either rail, the output voltage range is 3.5V from either rail and its total power supply range is from 18V to 50V.
In Gain of 1, INA326 and INA103 bandwidth is 10kHz vs 6MHz. For more detailed info on the required photodiode bandwidth, please read the three-part blog by our colleague, John Caldwell:
In addition, you must be aware that INA103 uses classic three-op amp topology that in higher gains has limited input common- mode voltage vs output swing because of the saturation of the first stage outputs.
For more detailed info on required phtodiode bandwidth, please read the three-part blog: Technical articles
If you need further assistance, please provide us with the preliminary schematic of your application.
Morning, here's the more detailed plan I mentioned.
(EDIT: my first reply seems to have disappeared. It said more or less "Thanks very much for your advice, and in particular for pointing out some of the INA103's limitations that I'd missed. I'll get back to you tomorrow morning with a more detailed spec.")
To first give you some background, this system's purpose is to amplify the output of a trans impedance amplifier and to subtract a DC reference level from it. The system should have a BW of >500 kHz, ideally 800kHz for changes in the PD signal. The DC reference has no bandwidth requirements. It is vital that the system exhibits high stability and linearity. The design of the trans impedance amplifier is fixed (the opamp in this is a OP27) and cannot be changed unless there's a very good reason. The rest of the system is what I'm designing now. The whole lot has to operate in an environment with fairly high levels of EMI, which is why the high CMRR of instrumentation amplifiers is attractive. Ground loops are endemic in our laboratory so this circuit must not add more and must resist their effects.
Here's a rough schematic of what I'm thinking.
The choice of instrumentation amplifiers A,B and C was the original subject of this thread, but I'm open to a redesign of the system if it need it.
My thoughts so far:
So far I've considered the INA326 for amplifier B and INA103s or INA129s for A and C (and B). I hadn't noticed the high bias current of the INA103 which may rule it out for amplifiers A and B. Again, I'm open to suggestions for other chips.
Thanks for all the help!
Charles,
Below please find couple of approaches you may take to gain up the voltage coming from the trans-impedance amplifier.
The first one shown in Fig 1, uses a difference amplifier to gain up the signal by G=20 while the second circuit shown in Fig 2 does the same employing instrumentation amplifier like INA128.
Please be aware that to accommodate 20V output at VF1 using OP27, you would have to use -5V and +25V supplies (-5V required by the input voltage range while +25V require by the output voltage range of OP27). In addition, to accommodate 20V output, INA128 would have to be powered from supply greater than 21V and that is the reason I lowered the input voltage range from 0V to 0.5V in order to use +/-15V supplies.
If you plan to use different power supplies, you must make sure that you operate instrumentation amplifier in a linear region; and it is NOT enough to look at the input and output voltage range as defined in the datasheet because of potential non-linearity of the outputs of internal op amps. For that reason, you have to download the Vcm vs Vout calculator show below using following link: