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.

What's wrong with this opamp??? Why it breaks easily?

Status
Not open for further replies.

shothand

Newbie level 5
Joined
Mar 17, 2006
Messages
10
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,369
Here are high power version opamp from QinetiQ used in the MEMS vibrometer design project. But it is mentioned that there is a problem with this opamp and it breaks easily. Could some one have a look at it and give me a hint? Thanks so much.

61_1160563816.JPG
 

Re: What's wrong with these opamps??? Why it breaks?

Please re-post the images so they can be seen without downloading the files.
 

Seems that nobody gets an idea yet.
 

the function of the opamp has no problem.maybe the application enviroment has problem.
 

Where did you read/hear that "... it breaks easily"? Do you have a drawing of the circuit that it's used in?
 

did u check the phase margin of this op ? what is P.M. ?
there exist 3 pole in this op
 

Note that you have different gains for inverting and non-inverting path to output !!!... Ve gain is ~(gm*ro)^2 trough output nmos and ~(gm*ro)^3 trough output pmos, and here comes problem with stability.
I think that M20 should be connected to M17.
 

ok, wr is the input circuit, i mean how u make the differential input, bcz i cant reconise any problem from the circuit
regds
 

the function of circuit have no problem. isn't some compensation problem due to pole locations?
and what does 'break easily' mean?
 

notice the vgs of m20 and the current of m20.
Maybe compensation have some questions.
 

I agree with pixel, M20 is not connected right, it must be at M17, otherwise M22, M27 can get open at the same and burn as a consequence of big current.
 

Please define "breaks easily"...

Interesting concept. Tricky to get it to bias over all corners, I'd think.
M20 and M22 share the same gate bias voltage.

Another issue is that there are several paths from the inputs to the output, all superposed. Multi-loop feedback systems are not simple to analyze, but just by looking I can say that there's a kind of nested miller compensation. I don't know if it's intended, or not.
Ignoring M20 M21 M25 M26 you'll have the standard OTA, miller compensation and zero-nulling resistor. M20 M21 M25 and M26 were added as a push-pull intermediate stage, that drives the M27 load. The output stage M27 M22 becomes itself a push-pull, then.

Without going further I'd say that one signal path has orders of magnitude more gain, so I don't see the point of connecting M18 to M21. Just bias M21 with a fixed voltage, and then compensate the three stage amplifier with nested miller capacitors (w/o the resistors).

Do you have a publication reference for that ?

Hope it helps!
 

Thanks for the comment, it helps me to think about it anyway!!!

Here are some of my linear analog design course teacher's friendly comments:

1. It would be better to choose nmos differential pair for the first stage because it will give a larger gain-bandwidth product, which will help on stability. NMOS has a larger mobility so that the gm with the same drain current is larger.
2. Try to choose M=even number for M23, M24, M17, M18 and M19 for layout matching
3. If the output is not connected to input through some feedback, the output stage needs common mode feedback.
4. From the input to the M22 output stage, it is a two-stage opamp, which is easy to be stablized by Miller compensation. But for input to M27, there are 4 stages. Even though the two middle stages has diode-load (rout=1/gm), it hurts the stability.

I don't think there are many people used this push-pull type output stage because of the stability, and not much additional gain you will get. Instead, a simple two-stage opamp would be preferred. Remember the gain comes from gm and Rout. We can try to use cascoded output to increase the gain instead of using push-pull type.


Please do help yourself to raise further findings and issues! I have uploaded the pspice file(winrar). Please note that you have to add models(nmos, pmos, opamp) library as global.
 

Interesting comments, but I don't see very well how to apply them to this circuit...

1. Yes, using a NMOS diff input pair will yield greater gm for the same biasing current.
All other things being equal, the gain-bandwidth will increase. But, all things being
equal, the stability will worsen... in Miller compensation the outpust stage gm has to go up in order to push the non-dominant pole above the unity-gain frequency.
Given more input gm, everything else will need to be adjusted, I think.

2. These devices are rather wide (30um), so M=1 or M=2 does not matter, in my opinion, since each 30um device will be drawn as a multiple finger device, which can then be interdigitated with the others to improve matching.

3. It's a single ended output amplifier w/ high-gain, so , yes, it should always be used in a feedback configuration, as any other high gain amplifier. Since it's single-ended, any feedback will stabilize both differential and common mode levels. I don't see how/why a CMFB would help.

4. I meant essentially the same thing in my previous post. Yes, he/she is right to
say that there are 4 stages. but M25 is diode-connected, so only the 3 high-impedance nodes count for stability. Actually only the dominant and the first non-dominant pole matter really, and all the other non-dominant poles (such as the one in the M25 gate) will just try to make your life a bit more difficult. It may be that M25 and M18 will both appear as doublets in this configuration, in this case you'll have no stability problems from them, but your settling time wil be ruined...

In any case, follow the directions of your teacher, I'd say.

Thanks for the file, but I don't have enough points to dload it. Could you put it on rapidshare ? Is there any pdf describing this somewhere btw ?
 

I have created rapidshare file to be downloaded here:

**broken link removed**


Thanks again for your comments.

Added after 5 hours 58 minutes:

Here is my Prof's answer about the breakdown issue of the opamp. Thanks for
your fork's comments; could you please find the problem with design related
to the breakdown?

Good to see people taking an inquisitive interest in their chosen fields... All your
comments are, of course, valid but none are fatal in real operational applications and, you should find them adequate for conditioning your vibrometer during the exercise.

Your first point (nmos vs. pmos for input diff pair) is valid however, people usually
choose pmos for one reason only - much lower 1/f noise in pmos..

Since you have spent a considerable amount of time trying to identify the flaw, I think it is only reasonable for me to point it out to you...

Set up a X10 amplifier (positive or negative - it does not matter), then drive the input hard with a sine wave, or something, so the output saturates against the rails and, monitor the power supply current when saturated... The problem occurs when the output tries to go below the negative power rail. In practice this can happen quite often, and you could not use these op amps as a comparator !! - the battery would die in no time at all ...

Added after 54 minutes:

Thanks ch1k0 for your comment. Please go ahead and help yourself adding some more if possible.
 

I think M21 should connecto to M17, the output is a push-pull stage,
if M20 gate rising, then via two inverter amply, M27 gate voltage rising, at the same time, M22 gate voltage rising. inverse, M20 gate falling, M22 gate voltage
falling. of cource, the you need reduce the size of M27 to reach the sink or souce the same current. Thankx.
 

I think:
1,the gate of M21 should be connected to M17 to reduce a zero.
2,you should add some control circuit to control quiescent current.
3,you should check phase margin at different output level including quiescent point.
good luck!
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top