Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

AD822 not rail to rail?

Status
Not open for further replies.

po210

New Member
Hello
I´m using an AD822 to build a simple 2,2 gain differential amplifier with R1=100K, R2=220K. My single supply is 5.20V and in ouput i see linear until output goes to 4.54V.Being a rail to rail op amp I though it would be possible to go until Vcc - 20mv = 5.20V - 20mV or so but no way...I measured 13uA output current.
If I build a simple follower with Vcc=+5V and i put 5volts at input I see output is 4,35V at most.
I need full 0+5V range because this is for an ADC based temperature meter,so temperature range spans the full ADC range.And AD822 is claim to be rail to rail.

Do you know why is this or what could i test? So bad is this chip?
I´m measuring with no load.
 
I'm not familiar with the AD822, but some manufacturers' claims of rail-to-rail output (or input) for their opamps are, shall we say, a tad optimistic.
 
hi,
Looking at the d/s for the AD822 it claims to be rail2rail.

Try reducing the R1 to 10K and the R2 to 22K, use a 100k Load, lets know what you see.
Is the 5.2V supply decoupled OK.??

Are you using the circuits Figs 35/36 of the d/s.??
 
Last edited:
I've now had a look at the d/s and, as Eric says, it claims rail-to-rail. More precisely it claims an output swing within 10mV of the rail when the load current is near zero. The output graph shows that at only a few mA load the output can drop much more. However, as you say you are measuring less than 4.35V at 13uA load current it suggests either there is an additional load of which you are unaware or else the IC is faulty.
 
Last edited:
Thank you very much for your help.
I´ve been measuring with the follower configuration shown on fig 35 in d/sheet and i see it perfectly follows input voltage until 4,2V or so.At that point,op output jumps to 4,5V and stays there.with a load od 220K, output current grows until 19uA,below specifications.
Regarding the amplification figure 36, I see output near zero (tens of millivolts) no matter what input i put, from 0 to 5V.:confused:
So, why this BIG limitation as follower?.I must say, i´ve tested with 3 different chips with same result, all chips bought to same eBay seller.
 
Last edited:
Thank you very much for your help.
I´ve been measuring with the follower configuration shown on fig 35 in d/sheet and i see it perfectly follows input voltage until 4,2V or so.At that point,op output jumps to 4,5V and stays there.with a load od 220K, output current grows until 19uA,below specifications.
Regarding the amplification figure 36, I see output near zero (tens of millivolts) no matter what input i put, from 0 to 5V.:confused:
So, why this BIG limitation as follower?.I must say, i´ve tested with 3 different chips with same result, all chips bought to same eBay seller.

hi,
With Fig36, gain of 2 inverter, if you do not have a negative DC off set voltage on the -Inp , the output will always be 0V for a positive signal input on the -Inp.???
 
Oppss...Figure says "Inverting amplifier"...Sorry
So, after feeding Vin- input with a negative power supply,I see linear amplification again until 4,5 aprox (4,35V)
But exactly same behaviour that follower: when you are in the sorroundings of 4V, small jump until those 4,35V and never reaches 5V
 
Last edited:
Oppss...Figure says "Inverting amplifier"...Sorry
So, after feeding Vin- input with a negative power supply,I see linear amplification again until 4,5 aprox (4,35V)
But exactly same behaviour that follower: when you are in the sorroundings of 4V, small jump until those 4,35V and never reaches 5V

With a Unity gain follower, the Vmax on +Inp is Vs-1, read page #18, ref the Input characteristics...
 
Last edited:
hi,
Try this configuration for unity gain, this should drive to close to +5Vout

EDITED: image 0 to +5Vin
 
Last edited:
Many thanks for help.
After testing your circuit,i see same bad behaviour.Perfect follower until input reach 4,2V or so, then it "clamps" at (now) 4,15V.Why do you think this circuit should be better? (is a differential amplifier with gain 1,as i tested first)
Reading the specs, i thougt they were saying that with a follower closer than +1V to V+ rail there are problems with bandwidth, so signal deforms. But my signal is for a temperature meter,and is very slow, so i thought there would be no problems with that. Now,i think i´m going to try TLV247x because this chip seems quite bad to me,or bad documented at least...
I´m just an hobbyst,and I saw those perfect graphs following the voltage and all was so beatiful...
 
Last edited:
hi.
The reason for the circuit I suggested, is that the +Inp never exceeds 2.5v, so its well within the Vs-1v that is stated in the d/s.

Its puzzling why the output limits at that level.?? I have heard that some eBay semiconductors are not upto the manufacturers specification.

If you are in the UK , you could send me a couple of the AD822 and I will be happy to double check your results.
 
Last edited:
I´m not in UK, but many,many thanks for your offer.
I've checked that if i feed your circuit with other Vs voltages (+9V,+10V) max output is always 1V aprox below Vs.
I think i will try with other chip,although i´m afraid same things could happen,if you can´t trust specifications.
 
I´m not in UK, but many,many thanks for your offer.
I've checked that if i feed your circuit with other Vs voltages (+9V,+10V) max output is always 1V aprox below Vs.
I think i will try with other chip,although i´m afraid same things could happen,if you can´t trust specifications.

hi,
Thanks for the feedback.

I use the MCP6002 dual rail2rail, its Vs max is 5.5V and it gets close to 0v and the +V rail on the output
 

Attachments

  • MCP6002.pdf
    454.9 KB · Views: 208
The proposed resistor values are so low that the output will not be rail-to-rail.
Increase the values of the resistors a lot.
 
hi po210.
My Rin/Rf resistors are far too low as 'agu' has pointed out, try the new ones and let us know.
 
Thank you for your help again.
With 10K resistors for R3/R2 I see same behaviour,but measuring current sourced by op amp I see 190uA.With this level,output could go to only a few tens mV. Trying with R3=R2=200K I reduce current to 9uA,but it happens the same.When output is near 4.35V, it stays there.
 
Last edited:
190uA into 10k is only 1.9V, not 4.35V.
10k resistors is much too low because for 4.35V the current is 435uA.

9uA into 200k is only 1.8V, not 4.35V.

The opamp output is rail-to rail when its output current into the feedback resistor plus load is only 10uA. Then a 4.99V output has a total load of 499k.
 
But 9uA over the R3=200K resistor, plus the 2,5V at negative input makes 4,3V, as measured.
Do you mean I must increment feedback resistor values?
 
But 9uA over the R3=200K resistor, plus the 2,5V at negative input makes 4,3V, as measured.
Do you mean I must increment feedback resistor values?
Sorry, I forgot to look at the schematic where the reference voltage is +2.5V.

For (5V - 2.5V)/10uA the total of the feedback plus load resistances must be 250k or more.
 
After trying with 330K resistors,and >1Mohm as load resistor, still the same.Good follower until 4.3V.
I´ve measured the currents through load and at output: they don't sum up even 20uA.
I think something is hidden in the specification data sheet,that I can't see.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top