I have a circuit going which uses some AD626ANZ amplifiers supplied with a single 5V supply
The output will interface with an ADC which is besides the point right now, the inputs to the amp come from a standard bridge
My problem is the minimum output voltage is way higher than the data sheet tells me (it wouldnt be the first data sheet I have misunderstood!), the minimum output voltage when both inputs are shorted together is around 110mv the data sheet says it should be a minimum of 30mV yet no typical or maximum
The gain of the amp is set at 50 with a 1k resistor, I can double the gain of the amp up to 100 by shorting this resistor out what I don't understand is why the minimum output voltage also doubles when the data sheet states 30mV minimum for all gains yet the reality is that the gain does directly affect the minimum output
I have got some AD627's coming as they have an offset null I was just wondering if anyone can advise me on what determines the minimum voltage output and what I could do to reduce this ideally as close to the 30mV minimum it states in the data sheet
Thanks in advance
Edit
Just done a quick breadboard test
Gain of 10 output minimum = 25mV
Gain of 100 output minimum = 250mV
the minimum swing is directly related to the gain yet the data sheet doesnt say that
for the AD626ANZ . The minimum output voltage is 0.03 volts so it is common for a typical output voltage to be higher than the minimum. You have measured 0.11 volts " when both inputs are shorted together "
The data sheet does not say you should short the two inputs together. Try setting IN+ lower than IN- to get the output to go to its lowest voltage. Then the output should be above the minimum voltage, typically.
I know it doesnt say in the datasheet about shorting out the inputs but in my mind by doing this the difference is zero so the output will be at the lowest it can, in my mind this is!, I am always open to learn so I am interested as to why this is not a good method?
As for making the - input greater than the +input thats something I had already done, I put the + input through a resistor so the tiny drop across it gives -input>+input
I measured the output at about 10mV! which leaves a couple of questions why is it less than the datasheet and the biggest question for me is why is the gain affecting the minimum output when the datasheet clearly shows the same value for gains of 10 and gains of 100
The reality is a gain of ten times more gives a minimum output of ten times more
"I measured the output at about 10mV! which leaves a couple of questions why is it less than the datasheet?"
The data sheet is using a 10k ohm load that pulls up. If you connect the load resitor, it will pull up the output to a voltage that is higher than the voltage when there is no load.
" and the biggest question for me is why is the gain affecting the minimum output when the datasheet clearly shows the same value for gains of 10 and gains of 100"
It shows the same "minimum" value. The value of voltage that you measure will not be exactly the minimum. The output voltage varies with gain. You should not be surprised: different conditions on the circuit give different output voltages. The circuit is not an ideal op amp, it has "Input Offset Voltages" that can be up to 2.9mV. At a gain of 100, the output can respond to that offset with 100 x .0029v = 0.290 volts.
Thanks again for your invaluable input, that has really made the penny drop for me
As you can probably tell using a single supply amp isn't something I have ever done and its proved interesting thus far
It was the datasheet that threw me and I always knew I wasn't reading it right, like a noob I didnt even consider the load characteristics that are there in black and white, more experimenting is required! I always knew there would be an offset to calibrate so theres no show stoppers or anything like that
It will be connected to an 328's ADC which I am not sure of the exact input impedance, it will be high but thinking aloud I have an ATMega328 which has a differential ADC but I know very little about these and how to use them but if I'm using a differential amplifier then should I be looking at using the ADC in differential mode?
"I'm using a differential amplifier then should I be looking at using the ADC in differential mode? "
No. The output of a differential amplifier is a single output. The ADC will be used as a single input ADC to measure the difference indicated by the single output of the differential amplifier.
The ADC measurements can be collected by your sketch software and averaging might be useful so the software can give results that are more useful than a single reading.
My software will be taking an average and using the information to integrate the function as the area under the graph is a parameter of interest,maths is my specialty!
I did some experiments with a 10k load using an AD626, and the minimum output was around the 40-50mV which isn't far from the datasheet claims
So I got hold of some AD627's which are very similar but they have a null terminal
I ground the null terminal to the supply and I read a minimum output voltage of about 5 mV which is just great its a reading of 1 on the ADC so the calibration offset is about as small as it could be!
Is there any downsides to using the null terminal like this?, I certainly cant think of any
REFERENCE TERMINAL
The reference terminal potential defines the zero output voltage
and is especially useful when the load does not share a precise
ground with the rest of the system. It provides a direct means of
injecting a precise offset to the output. The reference terminal is
also useful when amplifying bipolar signals, because it provides
a virtual ground voltage.
The AD627 output voltage is developed with respect to the poten-
tial on the reference terminal; therefore, tying the REF pin to the
appropriate local ground solves many grounding problems. For
optimal CMR, tie the REF pin to a low impedance point.
I won't pretend to understand wtf are local grounding problems?, and I read this as its ideally suited for lifting up the output voltage but I have tested it on a breadboard and this amp works (seems to) better than the AD626
But this area isn't my strong point if I am honest hence me coming here to see what others think, i have an amplifier working perfectly except the minimum output voltage at 130-150mV is higher than I would like it to be, If I can decrease the calibration offset that would be great but its not a problem as such
With an AD627 and the ref pin tied to supply ground I reduce that 150mV to 5mV but I am unsure of the implications of this, as I said I can't see any but that doesn't mean much!
Resinator:
With an AD627 and the ref pin tied to supply ground I reduce that 150mV to 5mV but I am unsure of the implications of this, as I said I can't see any but that doesn't mean much!
You have the output low level to be 0.005 volts using the REF input pin at 0.000 volts. That is good enough for me. Good work. Do you want the output to be closer to 0.000001 volt?
5 mV is certainly good enough for me but then again the 150mV was also easily usable and there would be no problem but as low as possible is better for me and it takes absolutely no doing and costs nothing so why not!
I was just wanting to ask you personally! if I was doing anything wrong or if I was missing anything blindingly obvious!
