hi luca,
Downloaded your AD623.asc file, it is all mixed up.

I will try to correct it, if possible.

E

 

hi luca,
Delete all the files for the AD623 on your computer.

Download and unzip this Desktop.zip to your Desktop.
You should see the unzipped Desktop Folder, open it up and double-click on the AD623_BaseW1.asc file name.

This will run LTspice and open the asc file.

Let me know when you have it running.

E
BTW: There is a gif image, in the folder, showing the sim result

 

hi luca,
Delete all the files for the AD623 on your computer.

Download and unzip this Desktop.zip to your Desktop.
You should see the unzipped Desktop Folder, open it up and double-click on the AD623_BaseW1.asc file name.

This will run LTspice and open the asc file.

Let me know when you have it running.

E
BTW: There is a gif image, in the folder, showing the sim result

Thanks! The circuit works perfectly. Had I made a mistake in drawing the circuit or importing the spice model of the component? :/

Can you tell me why when I simulate I display a graph that starts at 5V and not a 0V ? I set PWL (0 0 .1 10m). It is not the load cells because connecting directly between in+ and in- a voltage generator that simulates the behavior of the cells the problem persists. Probably the problem is that I'm still not very familiar with the simulation on LTSpice.

 

hi luca,
Well done.
Your image does not show the full circuit.?
Please post your asc file and I can check.
E

Update:
Don't leave on the circuit, odd bits of unused circuit.

 

hi luca,
You do not measure a Load Cell output at +in.
You have to measure from +in to -in.

Do the following actions.
Place the Mouse pointer/cursor on the +in, press down the left Mouse key and hold it Down while moving the cursor pointer to the -in, then release the Mouse button.

This will then display the plot for +in to -in Voltage

Do you follow OK.?
E

 

hi luca,
You do not measure a Load Cell output at +in.
You have to measure from +in to -in.

Do the following actions.
Place the Mouse pointer/cursor on the +in, press down the left Mouse key and hold it Down while moving the cursor pointer to the -in, then release the Mouse button.

This will then display the plot for +in to -in Voltage

Do you follow OK.?
E

Perfect! I really thank you very much for the support and information you provided.

 

hi luca,
Pleased to hear you have been successful.
Is the project now completed.?
E

 

hi luca,
Pleased to hear you have been successful.
Is the project now completed.?
E

The schematic is complete and it will be realized in a few weeks ... so there's a chance I'll have additional questions in the future. Will I open a new discussion or continue with this one?
I ask beacause maybe this could be a problem if you have to close a discussion when you finish it, as is usually done in forums of this type.

Thank you.

 

hi luca,
If it is the same amplifier Topic, this thread will be OK.

E

 

Hi everyone! I hope this section of the forum is the correct one for the question I am about to ask.

I should make a circuit that amplifies an analog signal [0 ... 500uV] into [0 ... 5V]. We have lm339n in the lab, but I read that it is a comparator and cannot be used as an amplifier, so I opted for an op07.
The questions I ask are twofold:

1) what op amp should I use for this purpose?
If you know of integrated circuits or modules that perform this task let me know

2) is the circuit I made (which I simulated and it works) ok?
I think using a 100kohm resistor and a 10 ohm resistor can be a problem as they have very different values.

Thank

Hi there,

Not sure if anyone mentioned this yet but, you can use an LM339 as an op amp if you provide a pullup resistor on the output. A comparator internally is much like an op amp it works almost the same except there is usually no output drive circuit so you usually need a pullup resistor.
It may not be exactly the same, but it can be used as an op amp depending on the application, but then again when you choose an actual op amp it has to fit the application also as i will explain next.

The main problem though is not the comparator it is the input offset voltage. Because you need to go from 500uv to 5.000000v that will mean a gain of 1000 which although a comparator should be able to do that and an op amp can the input offset also gets amplified and this leads to an output offset error which can be significant if the right op amp is not used.

For example, for an LM358 op amp which has a plus or minus 2mv input offset voltage spec (might be 3mv too) with a gain of 1000 that means even with a zero volt input the output can be as high as 2 volts or as low as -2 volts. That would be unacceptable in most applications.
What this kind of application requires is a low input offset voltage op amp one that has maybe 10uv input offset, and that would mean a chopper stabilized op amp. That is an op amp that continuously monitors it's own output offset voltage and makes adjustments in order to keep the error low, and many can keep the equivalent input offset voltage down to 100uv or even 10uv.
If you look around the web you will find one easily. You do have to keep an eye on the bandwidth too though as many of these op amps are rather slow compared to op amps made for high frequency.

I dont know what kind of specifications your circuit has to meet but i would think you need a low input offset op amp of some kind because the general purpose type will not work very well here, and the LM339 although can work as an op amp will have a rather large input offset as well.
Note that the spec on this can change over temperature also, so you may measure little error at 20 degrees C but as the temperature rises so could the input offset, depending again on the op amp.

Also be aware that spice models do not always show the effects of an input offset or grossly underestimate the real life input offset so it looks better on paper.

 

Hi there,

Not sure if anyone mentioned this yet but, you can use an LM339 as an op amp if you provide a pullup resistor on the output. A comparator internally is much like an op amp it works almost the same except there is usually no output drive circuit so you usually need a pullup resistor.
It may not be exactly the same, but it can be used as an op amp depending on the application, but then again when you choose an actual op amp it has to fit the application also as i will explain next.

The main problem though is not the comparator it is the input offset voltage. Because you need to go from 500uv to 5.000000v that will mean a gain of 1000 which although a comparator should be able to do that and an op amp can the input offset also gets amplified and this leads to an output offset error which can be significant if the right op amp is not used.

For example, for an LM358 op amp which has a plus or minus 2mv input offset voltage spec (might be 3mv too) with a gain of 1000 that means even with a zero volt input the output can be as high as 2 volts or as low as -2 volts. That would be unacceptable in most applications.
What this kind of application requires is a low input offset voltage op amp one that has maybe 10uv input offset, and that would mean a chopper stabilized op amp. That is an op amp that continuously monitors it's own output offset voltage and makes adjustments in order to keep the error low, and many can keep the equivalent input offset voltage down to 100uv or even 10uv.
If you look around the web you will find one easily. You do have to keep an eye on the bandwidth too though as many of these op amps are rather slow compared to op amps made for high frequency.

I dont know what kind of specifications your circuit has to meet but i would think you need a low input offset op amp of some kind because the general purpose type will not work very well here, and the LM339 although can work as an op amp will have a rather large input offset as well.
Note that the spec on this can change over temperature also, so you may measure little error at 20 degrees C but as the temperature rises so could the input offset, depending again on the op amp.

Also be aware that spice models do not always show the effects of an input offset or grossly underestimate the real life input offset so it looks better on paper.

Hi! I just read your reply now and thank you very much for the advice. So due to the open collector of the comparator I need to use a Pull-Up Resistor. Can you give me some advice on how to size/calculate it?

I will probably make the application with a precision differential amplifier, but I would still like to know the process in case of LM339, even though I won't be using it.

Thanks

 

Not sure if anyone mentioned this yet but, you can use an LM339 as an op amp if you provide a pullup resistor on the output.

No-one has mentioned that for a good reason.
Since a comparator does not have internal compensation to help make it stable in feedback circuits like an op amp, using an LM339 for that purpose is quite problematic.

 

An LM339 quad or LM393 dual comparator also has a weak minimum output low current of only 6mA when its output saturation is an awful +1.5V.

 

No-one has mentioned that for a good reason.
Since a comparator does not have internal compensation to help make it stable in feedback circuits like an op amp, using an LM339 for that purpose is quite problematic.

Hello,

Well that's funny because Texas Instruments did mention it in at least one of its app notes.
But he said that was all he had, and where i come from when we are stuck using certain parts we find a way to make it work.
I faced this challenge many times because the head guy at a company i worked for did not want to increase the stock on hand supply so we often had to work with what was already in stock for a new design.
I would of course rather use an LM358 but if all you have is an LM339 you may be able to use it. There will be limitations of course.
There is also an app note somewhere that uses an NPN transistor on the output to increase output current.

TI App Note:

 

Hi! I just read your reply now and thank you very much for the advice. So due to the open collector of the comparator I need to use a Pull-Up Resistor. Can you give me some advice on how to size/calculate it?

I will probably make the application with a precision differential amplifier, but I would still like to know the process in case of LM339, even though I won't be using it.

Thanks

Hello,

Well the pullup can not be too high or too low, but you can start by looking up the sink current of the output and maybe design for half of that.
So if the sink current was 10ma and the supply voltage 10 volts, so maybe 2k to 5k. One design uses 15k however that is in the reply just before this one.

 

Well that's funny because Texas Instruments did mention it in at least one of its app notes.

Below is my LTspice simulation of the (funny) circuit;

That large peak in the response seems problematic to me.
(Of course the real circuit may respond better, but I have my doubts that it will be significantly different).
And the peak gets worse as the gain is reduced.

Also you are limited with the load you can drive with a 15kΩ resistor driving the output.

And the LM339 is not specified for noise, which likely is important here for the large gain the TS needs.

 

My LTspice does not have a model of an LM339 and yours probably also does not.
The impedance of the 0.5uF capacitor is only 10.7 ohms at 30kHz so I doubt there will be any output signal. An opamp has a very low output impedance so it will oscillate. At the 200kHz of your other simulation the capacitor impedance is only 1.6 ohms.

The circuit is on my hard drive and says, "very low frequency amplifier".

 

Below is my LTspice simulation of the (funny) circuit;

That large peak in the response seems problematic to me.
(Of course the real circuit may respond better, but I have my doubts that it will be significantly different).
And the peak gets worse as the gain is reduced.

Also you are limited with the load you can drive with a 15kΩ resistor driving the output.

And the LM339 is not specified for noise, which likely is important here for the large gain the TS needs.

View attachment 266570

Hello,

Yes, as mentioned, there will be limitations. You can look for app notes that have better circuits for this.
We can not expect to get 10MHz bandwidth and 0.001 percent THD from this device (ha ha). it comes down to what we 'have' to do not what we 'want' to do when working with a limited parts stock. It requires more ingenuity not defeatism.

 

My LTspice does not have a model of an LM339 and yours probably also does not.
The impedance of the 0.5uF capacitor is only 10.7 ohms at 30kHz so I doubt there will be any output signal. An opamp has a very low output impedance so it will oscillate. At the 200kHz of your other simulation the capacitor impedance is only 1.6 ohms.

The circuit is on my hard drive and says, "very low frequency amplifier".

Hi,

Yes there will be limitations, and if they are too tight then it cant be used as is. Look for some other app notes there are better circuits this was just one example. There may be mods that can be done too to make it better.
The point is though that if we are handed an LM358 (your favorite op amp ha ha) we have to use an LM358 we cant use anything else.
Of course we all hope we can when we need to.
I used an LM358 for one power supply design but had to go to a faster device to get faster response, but i had that option at that time. If i didnt have that option i would have been stuck with the LM358 and would have to find a way to speed it up.

 

Hello again,

Here is a simulation of a modified circuit.
Note this would have to be breadboarded to test completely.
Also note the slight gain change to make it 100, and a small input offset to center the AC sine output at 1/2 of Vcc.
The input is 10mv peak at 1kHz, sine.
In the AC plot the horizontal red line is the -3db point from max which is at 0Hz.