Help with amplifier. Not amplifying as it should be.

Thread Starter

kris_maher

Joined Apr 24, 2009
90
Ok one thing I'd like to ask before I go off to the lab tomorrow is why for me, at B, the waveform is way less than 1V p-p. It's in mV. Silly question since on my part but perhaps any cause for it...Is it because it's not amplifying for some reason or something??

Also I used my multimeter before placing the resistors, so yes that is 330kOhm. Oh yeh, the packet I take the resistors out of states the value on the label so they should be fine. The 10k is more like 9.8k measured on the multimeter.
 
Last edited:

SgtWookie

Joined Jul 17, 2007
22,230
Ok one thing I'd like to ask before I go off to the lab tomorrow is why for me, at B, the waveform is way less than 1V p-p. It's in mV. Silly question since on my part but perhaps any cause for it...Is it because it's not amplifying for some reason or something??
Did you adjust R11 as I explained how to do several posts back? If R11 is not adjusted properly, the output will be reduced considerably.

Also, it would be helpful to see the output waveforms DC coupled, and indicate which line on the graticule is 0v (ground) potential. AC coupling is convenient, but you lose the ground reference point.

Also I used my multimeter before placing the resistors, so yes that is 330kOhm. Oh yeh, the packet I take the resistors out of states the value on the label so they should be fine. The 10k is more like 9.8k measured on the multimeter.
OK.
 

SgtWookie

Joined Jul 17, 2007
22,230
Kris,
You need to do something with the other channel in the opamp.
Jumper pin 7 to pin 6.
Jumper pin 5 to pin 3.

That will turn the other channel into a unity-gain voltage follower, that will simply track the voltage on the wiper of R11. Otherwise, it may oscillate unpredictably and generate noise.
 

SgtWookie

Joined Jul 17, 2007
22,230
Looks like you may have a problem in the making at rows 17&18 and row 16. Those resistor leads are awfully close to each other.
 

Thread Starter

kris_maher

Joined Apr 24, 2009
90
Ok this is what I did tonight at the uni labs...

First off I've bought a new breadboard dedicated to only the receiver so once that is working I can integrate these two on another board in my own time.

I'm using solid core wires. I also have a 100nF coupling capacitor between V+ and Vee/Ground (between pins 8 and 4).

I'm using a different trimpot that's set at 50%.

And these are the screen shots I took that are attached.

PS: The output was measured as 2.5V in DC via a multimeter. I get a straight line in DC and a sine in AC (still a little choppy but a lot better than before. Choppy as in you can see a mirror of it down below it). Also the resistors are not touching and I got myself a proper trimpot not the other one I had before.

I'll post a shot of the actual system when I get home. Also another thing is it's just a DC line with amplitude at 2.5V. The AC p-p voltage is still not greater than 1V. I've done the whole circuit up till point B, R9 and R10 are also present. Also the 330k resistor is being used, I verified it as 0.328 MegaOhm at the multimeters in the lab.

Can anyone shed light on this one please? thanks
 

Attachments

Last edited:

Thread Starter

kris_maher

Joined Apr 24, 2009
90
Okay on the simulation it works fine, I've checked that out myself with multisim before trying it. With all due respect right now I'd just like to get what I have for the receiver up and running so I can do the programming the micro part myself for the next stage.

Also just to add to my previous post the input "Raw" signal direct from the receiving transducer is being amplified. But not to the extent that I had hoped for with greater than 1V p-p to be put through the comparator.

The picture "Raw" is the signal of the received signal before going through the amp. And "Raw2" is the output of the amplifier. You can see it as being amplified but not to the extent hoped for.

Ideas?
 

Attachments

millwood

Joined Dec 31, 1969
0
there are a lot of things a simulation will NOT answer, some of them we discussed earlier here and in that thread removed as well.
 

steveb

Joined Jul 3, 2008
2,436
Okay on the simulation it works fine, I've checked that out myself with multisim before trying it. With all due respect right now I'd just like to get what I have for the receiver up and running so I can do the programming the micro part myself for the next stage.
I'm just going to throw out a random thought as a stab in the dark.

I quickly scanned through the posts and noticed a lot of checking on the resistors, but didn't notice a mention to check that input capacitor. Are you sure that is a 100 nF capacitor? If this value is not correct you could cut out the signal.

Another thing to consider is the source impedance of the input voltage drive. This impedance will add to that 10K input resistor and reduce the gain. Are you sure that the simulation and the real circuit have the same representative source impedance? I know it's a big stretch to say that your source might have that high a source impedance, but since I don't know what you are driving the circuit with, it can't hurt to ask.
 

Thread Starter

kris_maher

Joined Apr 24, 2009
90
Are you sure that the simulation and the real circuit have the same representative source impedance?
What do you mean by that? You mean the capacitance of the ultrasonic transducer (the AU5550)? That is 1800pF.

The capacitor I'm using as the input and the bypass are the 104M, that is, the 0.1uF capacitor.

In the simulation I put it as 100mV p-p, just as a direct sine wave signal as the input.

Actually steveb your post triggered an idea a moment ago. What I simulated was at 40KHz sine wave 100mV p-p. Yes the sine wave is true in the real thing and it is also at 40KHz in the real thing.

But in the real thing what is not true is that of the 100mV p-p signal input. From the last picture I posted it is more at 10mV p-p which would be a good indicator as to why I am getting such low output voltages. The gain would need to be changed to cater for this. I'll need to look into this...But even so I'm not sure if it would be able to amplify enough.

Also the attachment shows the receiver on the new board.
 

Attachments

Last edited:

steveb

Joined Jul 3, 2008
2,436
What do you mean by that? You mean the capacitance of the ultrasonic transducer (the AU5550)? That is 1800pF.
I believe this is something that should be included in your simulation. Put a 1800 pF capacitor in series with the 100 nF to better represent the interaction of the source and load. If you think about it, the source capacitance of 1.8 nF will combine in series with the input capacitance of 100 nf to generate an effective capacitance of 1.77 nF. So the 100 nF input capacitor is not very relavant here. The 50 times lower input capacitance will drive your cutoff frequency much higher. Calculations show you should still pass the 40 kHz signal with an ideal OPAMP, but your OPAMP has only a 1 MHz gain bandwidth, so you want to be careful here.

Anyway, this may not be your entire issue, but it could be one of a few issues. Even if it turns out to not be an issue, it's a good idea to let the simulation capture as much of the real circuit as possible.
 

Thread Starter

kris_maher

Joined Apr 24, 2009
90
I've decided to overcome the low input voltage and low output voltage signal by making it into a 2-stage amplifier. Using the 2nd amplifier (it's a dual package) as well or using a 2nd LM358.

Just a question if I should still keep the coupling capacitor or not at the start? I've changed the input 10k resistor to 1k to increase the output voltage on the 2nd stage amplifier.

On the simulation it boosts the signal to a reasonable level above 2.5V p-p on ac (from around 160mV output from before where the input signal was 10mV p-p at the start). Also of course I will use the bypass capacitor as well.

EDIT: I didn't notice your above post, I'll check it out now...
 

millwood

Joined Dec 31, 1969
0
But in the real thing what is not true is that of the 100mV p-p signal input. From the last picture I posted it is more at 10mV p-p which would be a good indicator as to why I am getting such low output voltages. The gain would need to be changed to cater for this. I'll need to look into this...But even so I'm not sure if it would be able to amplify enough.

Also the attachment shows the receiver on the new board.
your amp is providing a gain of 20x in real life and 30x on paper. don't you need to ask why?

the first thing I do in debugging is to take measurements against my expectation and try to understand the differences, if any.
 

steveb

Joined Jul 3, 2008
2,436
Just a question if I should still keep the coupling capacitor or not at the start? I've changed the input 10k resistor to 1k to increase the output voltage on the 2nd stage amplifier.
I apologize if I'm misunderstanding. I've tried to catch up on the details, but may be missing something. Are you saying that you are trying to create a gain of 300 with your amplification stage? If so, your OPAMP does not have enough gain bandwidth to do that. You are already pushing the limit with a gain of 30.

GB=1MHz
GB/gain=1MHz/30=33KHz

I'm guessing you probably can not even get a gain of 30 at 40 kHz.
 

Thread Starter

kris_maher

Joined Apr 24, 2009
90
Well I originally intended to have a approximate 5V p-p square wave as the final output I originally came up with in my schematic on the 1st page. Tbh I was trying to see if SgtWookie's receiver would do it, and it did however the output voltage was not what I was after. I do have faster OpAmps with higher bandwidth on me though, namely the CA3130.

However if it's the higher bandwidth models needed then my local electronics store does have the higher-bandwidth model, the LM386.
http://jaycar.com.au/images_uploaded/LM386.PDF

It has a BW of 300KHz. Multisim doesn't have a component for this in its library, and you guys would know a lot better than me, so tell me if this one is alright... I'll test it out tomorrow.
 
Last edited:

steveb

Joined Jul 3, 2008
2,436
It has a BW of 300KHz.
Well, that is the gain bandwidth. To estimate bandwidth, take the gain of your circuit into account. Basically, you multiply gain times the frequency bandwidth and compare to the gain bandwidth spec. For example, if you want a gain of 30, that chip might reach a frequency bandwidth of 10KHz. For 40 kHz, and a gain of 30 you need a gain bandwidth at least 1.2 MHz.

There are many good arguments for using multiple low-gain stages to achieve high gains, and the gain-bandwidth limitation is one. Two stages with a gain of 10 are easier to achieve than one with a gain of 100.
 

Thread Starter

kris_maher

Joined Apr 24, 2009
90
Ok. well I'll try out with the output of that (1st amplifier) being fed into a 2nd amplifier (LM158. Same one) and see how that goes.
 
Top