Heating operational amplifier OP284FSZ-REEL7

[Glebiys]

Hello all,

Task: to make a square wave with a frequency of 1 kHz in the voltage range (-12V - + 12V), using a PWM signal with voltage (0V-5V).

Circuit:


Originally LM258WYDT (ST) was used as an operational amplifier, it did not heat up, but since it was not from the Rail-to-Rail series, there was a 1.5V drop at the output, which did not allow us to give out a voltage close to 12V. So I decided to switch to Rail-to-Rail, for these purposes I found OP284FSZ-REEL7 (AnalogDevices). After replacing with this amplifier, the value of the output voltage became close to 12V (+ -11.85V), but the operational amplifier began to heat up (after 5 minutes its temperature was about 45-60 degrees).
I did not find a short circuit, but noticed that at the output of the voltage divider (R1, R2), instead of 2.5V, the voltage became 4.5V. On the LM258WYDT, this voltage was 2.5V. Perhaps there are elements on this input of the operational amplifier that should lift it?

Unfortunately, at the moment I do not have an oscilloscope and I can not see the output signal.

My suspicions:

• The amplifier is not designed for such a load or, conversely, it is not enough for it, and it begins to oscillate.
• Is this an accurate amplifier, perhaps additional elements are needed for its operation?
• The amplifier is damaged.

Parameters of currents from datasheets:

LM258WYDT

• Input current: 5mA (DC)
• Source output current (Isource): 20-60mA
• Output sink current (Isink): 10-20mA

OP284FSZ-REEL7

• Supply current: 2.25mA
• Output current: 10mA

Question: What could be the reason?

[TerminalJack505]

You might try attaching your circuit schematic again.  I think you need to have 5 posts or more before you can attach files to your posts.

[Glebiys]

TerminalJack505, thank you for the answer!

[DuPe]

it is no good idea to use an linear op amp as comparator and even worse to use it without negative feedback.
Try a real comparator (http://www.ti.com/lit/ds/symlink/lm397.pdf) with recommened feedback circuit given on page 1.
cheers
Peter

[Glebiys]

@DuPe, thank you for the answer!

At this stage, I have already created a printed circuit board, and it is impossible to remake it. I first encountered operational amplifiers. In the next version of the board, I will think about an alternative.

There is information:

The specification of the amplifier indicates the value of the Input voltage difference: 0.6 V, as I understand it, this is the maximum difference between the inverted and non-inverted input.

In my case, 2.4V (with a divider of 100k-100k) should come at the inverted input, and 5.1V should come at the non-inverted input. The difference between these voltages is 2.7V, which is clearly greater than 0.6V.

I see 4.5V, instead of 2.4V, which leads to the idea that he is trying to raise the voltage to 0.6V and because of this starts to heat up.

Current must be limited. The non-inverted input is connected directly to the microcontroller, which can output up to 20mA, and the inverted input to it comes with a divider, here I do not know if it can produce a current of more than 5mA, which destroy the opamp.

[TerminalJack505]

Does the IC have more than one op amp?  If so, you can't leave any unused inputs disconnected.  You should connect all unused op amps as shown in the attachment.

[wraper]

it is no good idea to use an linear op amp as comparator and even worse to use it without negative feedback.

There is nothing wrong with it as such. Suggesting op-amp should be necessarily operated with negative feedback is stupid. Though some op-amps may recover from saturation not quite gracefully. Many of "comparators" are simply a bit simplified op-amps with open collector/drain output. And you can also use them as op-amps if you like.
EDIT: and LM397 you suggested is exactly that. Simply op-amp with open collector output. Open collector output is useful when you say want to convert input voltage to something not related to input voltage or power rails. Say larger voltage into low voltage logic. Opposite to what OP wants to achieve.

[wraper]

The specification of the amplifier indicates the value of the Input voltage difference: 0.6 V, as I understand it, this is the maximum difference between the inverted and non-inverted input.

This. Many op-amps and "comparators" have inputs which cannot accept large differential signal.

[DuPe]

 maybe I was not clear enough: Your circuit cannot work. You are torturing to death any op amp with this.
op amp is continuosly in open loop amplification and input overdrive.

[wraper]

maybe I was not clear enough: Your circuit cannot work. You are torturing to death any op amp with this.
op amp is continuosly in open loop amplification and input overdrive.

Again, there is nothing wrong as such with using op amp with open loop or positive feedback. Though some may not perform well. But it's certainly not torturing.

[DuPe]

maybe you are right wraper. But why then does it heat up that much?

[wraper]

maybe you are right wraper. But why then does it heat up that much?

As minimum differential input voltage is exceeded. Once that's fixed, we can dig further. If op amp recovers from saturation poorly, it can increase heat dissipation. Transition can be happening with oscillations as well, adding some positive feedback can help with that.

[Vovk_Z]

maybe I was not clear enough: Your circuit cannot work. You are torturing to death any op amp with this.
op amp is continuosly in open loop amplification and input overdrive.

Not any opamp but most of, you are right.

[Jwillis]

 wraper is correct.
If you look at the LM258W the Differential input voltage ca be as high as the supply voltage . The OP284 is only 0.6 V .
But your not totally in trouble  because it also states "For input voltages greater than 0.6 V, the input current should be limited to less than 5 mA to prevent degradation or destruction of the input devices."  Page 6   note 1  of maximum ratings .  You can put series resistors on the inputs to limit the current .  1K should do the trick.

[Glebiys]

@Jwillis, thank you for the answer!

I already created a board, so, unfortunately, adding a resistor at the moment will not work. This can be done in the next version of the board.

Then what are the options:
1) Change the value of the resistor on the voltage divider so that the voltage difference is in the range of 0.6 V.
2) Buy another operational amplifier that corresponds to a large difference in inputs.
The LM258W came up, but it's not Rail-To-Rail.

[Jwillis]

@Jwillis, thank you for the answer!

I already created a board, so, unfortunately, adding a resistor at the moment will not work. This can be done in the next version of the board.

Then what are the options:
1) Change the value of the resistor on the voltage divider so that the voltage difference is in the range of 0.6 V.
2) Buy another operational amplifier that corresponds to a large difference in inputs.
The LM258W came up, but it's not Rail-To-Rail.

I've only worked with a very few op amps and mostly for high voltages. I'm sure some one else would be able to say what would be a more suitable Op amp .
Just changing the voltage divider at the inverting input won't help  because it has current limit with the 100K resistor. It's more on the non inverting input . How much current is the PWM putting out.

[wraper]

@Jwillis, thank you for the answer!

I already created a board, so, unfortunately, adding a resistor at the moment will not work. This can be done in the next version of the board.

Resistor will do nothing since input current is already limited by high resistance of divider connected to negative input. You should use different op-amp or comparator if you manage to find one with push-pull rail-to-rail output in the same package and high enough supply voltage.

[Kleinstein]

The current limiting at the negative input should be enough. However the OPA284 seem to no like the circuit. There is nothing absolutely wrong with the circuit, though the OPA284 is not a good choice with the limited input range.

I have found a similar problem with the TLE2071: it does not like to have it's output driven hard negative and than consumes a lot more current (some 20 mA) and gets warm  .  Looks like the OPA284 may have a similar problem.

So the best is to use a different OP or comparator. My suggestion would be OPA2170. It is not fully rail to rail at the input, but at the output.

[TerminalJack505]

The OP284 has two op amps so the other op amp may be causing the problem.  How is it hooked up?

[wraper]

My suggestion would be OPA2170. It is not fully rail to rail at the input, but at the output.

Nope, don't use OPA2170. It has back to back diodes between inputs.

[Glebiys]

The second operational amplifier just hung in the air, I understand that this is not good.

Today I purchased another LMC6482AIMX/NOPB op-amp. Soldering it and starting it, it began to heat up very much and the output was only 1V, instead of 12V. It is Rail-To-Rail (input/output), the allowable voltage difference is Vsupply. But its difference was in CMOS technology instead of Bipolar.

[Kleinstein]

The LMC6482 has an absolute maximum supply of 16 V. So a +-12 V supply is just too much and can blow the chip.

Unused OPs should be connected in a way they don't cause trouble. For the LM258/358 leaving the inputs open is OK, but this is an exception. Other OPs are usually better connected as a follower and the + input to a neutral level.

[Vovk_Z]

TS circuit needs a comparator instead of opamp, I think, so I don't see a reason to change different opamps, if he can put something like LM311, LM2903 instead of.
Because comparators "like" to work with large differential voltage (but not opamps), and without negative feedback.

[Kleinstein]

Using an LM258 is no problem - the input stage is actually not that different from am LM393 / LM2903.
There are not that many comparators with rail to rail push pull outputs that work with a +-12V supply. If high speed is not needed an OP may be OK as an alternative.
If just used as a kind of digital level shifter, one could consider a CMOS switch chip like DG419 - though probably a different pin-out.