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.

Opamp not working properly

Status
Not open for further replies.

ants

Full Member level 4
Joined
Feb 20, 2010
Messages
199
Helped
1
Reputation
2
Reaction score
0
Trophy points
1,296
Location
UK
Activity points
3,006
I had a board made up recently using the OPA548 opamp. I can't get the opamp to work properly, I've stripped everything else off the board and the amp is causing the problem. I've checked the soldering, I've checked that the pins are getting power from the rails. I just can't work out what is going on.

As i've now stripped everything the input is a signal from a signal generator. I find that no matter what voltage I input the opamp pin only takes a tiny amount of that voltage. I have also found that the input pin seems somehow to be connected to the V- pin (it is a dual supply). The chip has a current limit which I set to give up to 5A, the feedack resistors set the gain to 2. I didn't put the caps exactly on the pins a mistake that shouldn't cause this much of a problem.

Essentially I input 5V sine and get a sine wave of about 300mV superimposed on -11V. What is going on? Any suggestions?

Thanks,

Ant.
 

I've attached the schematic for my circuit, the 555 timer and 2 voltage regulators have been taken off and all of their components, I've shorted some pads to ensure I have the right voltages getting on to the pins.

I've been left thinking something is wrong with the opamp, but I have a feeling it is one of those times when I'm thinking that because I can't find the real problem.

Before I took all of the other components off the opamp seemed to devour the 555 signal so nothing much changed after I took them off.

Regards,

Ant.
 

Are you sure you are doing the right thing with the current limit pin? It would be a resistor to V- but you have a pot between 0V and V- and you tap a voltage off that. It may still work, but it seems a strange hybrid between forcing a voltage in to the Ilim pin with a voltage or using a resistor to V-. I would disconnect the 0V end of the pot.

What are your supply voltages? I thought they were +/-5V bit you mentioned -11V in your first post.

The output voltage range of the opamp is quite limited: V+-4.1 to V- + 3.7 at 3A. Are you expecting too much of it?

Keith.
 

According to the datasheet, setting the current limit by a voltage is correct, too.

Did you disconnect the load?

Besides a possible damaged OP, confused pin order would be a plausible explanation.
 

FvM said:
According to the datasheet, setting the current limit by a voltage is correct, too.

True, but this is a hybrid of both - it will make it a lot more difficult to know what you have set it to than using either a voltage or a resistance.

Keith.
 

I took components off piece meal and wondered about the current limit pin, so I took it off and then put a resistor across and then ended up connecting it cirectly to V-, but with no luck.

I am using 2 PP3 batteries connected to give a dual supply, just for testing, so my supply was -11V and +7V. They tend to run down quick but I keep an eye on them. I've tried a range of signal input voltages. I also tried connecting it to a mains powered supply, that was +25 and -25V, I was getting -30V and +20V on the supply pins. No luck.

The load on the opamp was the transformer primary which is 4ohms. I didn't connect anything to the secondary just tested it with a scope.

I've now taken the chip off the board I had made up in Eagle and have spent the last hour putting it on a stripboard. I haven't quite finished so haven't tested it yet. I wish I hadn't made the stripboard quite so messy I've confused myself already, however I will double check it before testing. This should give me a better idea of what is wrong.

Thanks,

Ant.

Added after 1 minutes:

I should add IC3 is the transformer.
 

The output transformer is basically a DC short and can cause problems. Start with disconnected load. Check if the DC output voltage is sufficient low to avoid an excessive DC current.
 

I've got the opamp working on the stripboard. So that means my layout has caused a problem. I'm looking at the schematic that I had made up and it looks fine, i've moved all of the components to check they are actually connected to the opamp. I will have to remake the circuit on a spare board now. I've attached pics of the board from Eagle. The only thing I've definitely done different is put the caps much closer to the pins. Would that make much difference?

I've had the transformer working with a different amplifier, it is a miniature audio amplifier 4 ohms to 1.2Kohms. I'm not exactly sure how or why an excessive DC current is a problem but it doesn't have a problem with 4ohms and 1.2A in alternating current.

Thanks,

Ant.
 

I reckon it has something to do with my cap layout. It just doesn't look good having a line like I have done. It wasn't a good idea having the 555 timer working off a dual supply I think too.

The caps are top left downwards
C3,
+BYP
+BYP2
-BYP2
-BYP
CV
C2
C1
REG-2

For example the ILimit pot is also connected to C2 the 555 ground power bypass cap. It just cant be good although I don't know why.
 

The layout isn't great but it is difficult to say if it caused your problem. In general when you place components for layout you should place then to minimise track length, not to make them look tidy.

Keith
 

Yes I think that is sensible advice.

I did make notes on this board and they generally said it was an awful design and awful layout. As it was only my second board I guess this kind of result was a bit inevitable. It does somewhat make me wonder why I learnt Eagle, it took around 2 weeks, and then I had to wait a week for the board to come back. I threw the chip on a stripboard in about an hour and it worked!

Regards.
 

I understand what you are saying, but in the long run I think you will be learning essential skills. Laying out a board like that would only take a few hours when you are used to it, including creating new footprints.

Stripboard can be useful when you need to experiment, provided the circuit can work on stripboard - no 1GHz synthesizers! However, if you are reasonably confident the circuit will work, with maybe just a few component value tweeks, then go straight to PCB if that will be required in the end anyway.

I rarely use stripboard, but am currently doing just that. It is a circuit where the requirements are ill defined, the datasheets vague and the Spice models non-existent.

Keith
 

Yeah, I think Eagle will be useful in the end. Perhaps because it will allow me to shrink my boards in size. I can just about make my own devices/footprints now.

What is your current project? I doubt I could help but I'd be interested to know what it is. I have my long term project which is a piezo driver and tend to test other bits and pieces as they crop up.
 

As my work is for paying clients it is commercially sensitive, so I cannot disclose what it is. Sorry.

Keith
 

I understand. It doesn't surprise me that you make your living from electronics. I'd say electronics is a project for me, its more than a hobby but not my living.
 

I didn't yet check the layout details. But I doubt, that the discussed problems, particularly related to trace length/component distance would showup as reported. They may cause e.g. parasitic oscillations but hardly a comletely different DC bias. It still looks to me like wrong or missing connections. I would trace the problem by checking the voltage level at each circuit pin against expected values.

Some circuit details look also dubious to me, e.g. the volume potentiometer that is shorting the 555 output. I thought it was a schematic drawing fault, but it's also in the PCB.

P.S.: As a general suggestion, I would always try to use an (almost) continuous ground plane with medium/high speed circuits. You get even RF pre-drilled "vero" boards with a ground plane at the top.
 

I've done some testing with the circuit on a stripboard. At a guess I'd say the line of caps was the problem and I probably got confused and put one of them on the wrong way. I certainly put one of the tantalums on the wrong way because it popped quite loudly. I will check the pot and 555 output.

Added after 16 minutes:

I was hoping to use the VOLUME pot as a volume control to change the opamp input signal. It doesn't seem to be working. Should I disconnect the pots earth lead?

The setup is 555 output -> pot with earth -> opamp input.
 

You need to connect one end of the pot to the 555, the other end to ground and the wiped to the opamp.

Keith
 

Ok cheers I've now done that. I've got almost the identical circuit working on a stripboard. The main difference is that instead of having the 555 timer on dual supply its single supply with its own separate power supply. I'm going to start making up a new circuit in Eagle based upon the stripboard. I'm just going to put it down as a learning experience and not try to unpick exactly what was wrong.

Thanks for all the comments in this thread.

Ant.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top