Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

[SOLVED] OP Amp biasing question

Status
Not open for further replies.

obrien135

Full Member level 5
Joined
Nov 10, 2009
Messages
240
Helped
5
Reputation
10
Reaction score
5
Trophy points
1,298
Location
Connecticut
Activity points
3,259
Hello again,

For troubleshooting purposes, I was wondering about the following pertaining to op amp biasing:

If the + input of an inverting amplifier was not connected to a voltage divider on a single supply version of the circuit, and just left hanging, would you get anything on the output at all when an input is applied? Would it just be pure DC?

A picture of the circuit with the flaw is attached.

Thank you very much for all the help you've been giving me so far on this forum. It is very helpful.

George

 

As the transistors are not biased they are not at all in saturation and shouldn't get any output.
 
It depends on the type of opamp but it will most likely drive hard towards either the positive or negative power rail.

Keith
 
Thank you for answering my post. But , I don't understand how that could be. If the transisters are not in saturation, then I would think they are in the active region, and so there WOULD be an output, although not the output you want. Wound't there be some sort of messy signal other than pure DC on the output? Can you explain a little bit further please?

---------- Post added at 23:35 ---------- Previous post was at 23:34 ----------

OK, now I see what you guys are talking about. Thanks again.
 

As I said, it depends on the opamp. A bipolar opamp would need a bias current. The floating input cannot get any bias current, the inverting input can. So there is a mis-balance which will drive the output towards the rails. For a CMOS opamp the floating input could be anywhere, also causing a mis-balance on the input which will be amplified and again drive the output towards the rails. It could float around and pick up noise and produce a noisy, rail-rail output.

Keith.
 
Hi,

But i doubt as the CMOS non-inverting input (floating input) should have a noise voltage which provides Vgs -Vth such that the input transistor should be turned ON. Otherwise no current flow will be there in the circuit. Similarly the Vin on the inverting terminal could be a small signal voltage which isn't enough to bias the other input too. So there could be a situation where no current flow occurs in the circuit. So how the input signal be amplified?
 

The floating input can be at any input voltage compared to the non-inverting one. Most likely it will pick up mains hum and you will see 50Hz on the output. It will not be amplified - my mistake - because the circuit is a follower but it could reach the voltage rails. The bipolar case is different - the lack of bias current will force the output to towards one rail or the other.

Keith.

[edit] Actually, it may amplify with a CMOS opamp - it all depends on where Vin is connected to (if anywhere).
 

I went to the location where the circuit in question is located, this morning. I determined that the voltage divider is connected properly to the + input. And the output deflects up or down when I connect the input to + supply or ground. I don't presently have a signal generator to hook up to it. But in it's quiescent state the output rests at close to the + supply. Is this normal or is it because I didn't use an offset null circuit? I am trying to determine if the oscillator I built to drive it is working or if it's the op amp circuit that's not right.

Thank you all again.

George
 

I would suspect the offset present at the input. For the transistors in quiescent state and offset cancelled circuit the output DC voltage should be at midpoint the power supply, which will enable us to provide maximum output swing.
 

If I used an op amp that didn't have provision for an offset null circuit, would I still have the same problem? Or would the design of such an IC be such that it wouldn't be a problem? I was thinking about trying the LM7171. It has the same pinout except for the offset null pins, which it doesn't have.
 

I am very sorry, but I do not understand the whole discussion. Are you really discussing the behaviour of an opamp with one input not connected - that means floating ? What do you expect from a FET if the gate is not connected to anything?

....... I determined that the voltage divider is connected properly to the + input. And the output deflects up or down when I connect the input to + supply or ground.
.


What do you expect in such a case? For both alternatives (pos. input connected to +supply or ground) you do not get an appropriate bias point for single supply operation.
Or, didn´t I understand the real problem? May be.
LvW
 

Thank you for your post Sir,

To attempt to answer your question, I will state that I didn't connect the input to the plus supply and then ground for biasing purposes. I just touched to to those points one at a time momentarily to see if the op amp was working or if it was blown. I wanted to see if the output moved at all. I don't have a signal generator handy right now. I have one somewhere that outputs audio frequencies, but I wasn't able to find it this morning. Last night I was suspecting that the + input was not connected to the voltage divider for biasing, but this morning I determined that it was, so I shifted the discussion to the question of why is my output near the positive rail, and whether I need to impliment an offset null circuit or use an op amp that doesn't require an offset null circuit, if there is such a thing. I hope this clears things up a little bit.

Thank you,

George

---------- Post added at 16:34 ---------- Previous post was at 15:55 ----------

Should I post these later questions in a new post?
 

While I am sure most opamps are better than the 741, I suggest you pick something a bit slower than the LM7171 otherwise I fear you will end up with an oscillator. You haven't said what gain you need or what accuracy so it is difficult to advise if you need offset nulling.

Keith
 

Even now, I don`t understand why you are dicussing "offset". For my opinion, this parameter comes into the play (if any!) only after a correct bias point within the linear opamp range has been realized.
Therefore, in order to see if the basic task is solved the whole circuit including power supply values and dimensioning of the voltage divider should be presented. Only then one could discuss offset voltages that only slightly shift the operating point.
Or didn´t I get the problem?
LvW
 

As another source of uncertainty, you didn't specify resistor values or supply voltage. Also the OP type hasn't been said, not even FET or bipolar. But you are asking about neccessity of offset nulling. Ask meaningful questions and you'll get reasonable answers.

For the time being, I assume, that no offset nulling is required, unless the circuit has a very high gain.
 

Hi Keith,

It is an LM318 with unity gain (Rf = 10Kohms and Ri = 10Kohms) and the voltage divider is two 10K's. The supply is 13.8VDC. The operating frequency is 10MHz. I haven't tried the LM7171 yet, but I thought I might , as an alternative because it doesn't require an offset null circuit and I wouldn't have to buy or install the parts.

Does this shed any light on it? If I stick with the LM318, should I use an offset null circuit? I hope this is a reasonable question.

George
 

If I stick with the LM318, should I use an offset null circuit?
You don't need it. LM7171 is most likely a better choice, if you plan to intend large signal operation at 10 MHz. LM318 allow has only a few hundred mV output due to slew rate limitation. But it requires a well considered circuit layout.
 
Just to add, I think you must have some circuit connections problems if your results aren't making sense.

Keith
 

Hi Keith,

It is an LM318 with unity gain (Rf = 10Kohms and Ri = 10Kohms) and the voltage divider is two 10K's. The supply is 13.8VDC. The operating frequency is 10MHz.......
George

George, your circuit only can work if you feed the input signal into the first resistor through a capacitor.
Are you aware of this? Otherwise, the dc voltage at the non-inv. input is not amplified by unity which is necessary for single supply.
(Now you understand perhaps why we have requested some more information and why I was surprised about the offset discussion)
Regards
LvW
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top