LT1366 circuit not producing the correct output.

Thread Starter

Bod

Joined Sep 18, 2016
317
A very long time ago a thread about and audio circuit. It wasn't a very good thread in terms of my ability to provide good information but that's besides the point. Quite a while later I finished a schematic and got the board produced.
Here's a snippet of the schematic. The whole thing's pretty big and I don't think it's necessary to show it all anyway because I think the issue occurs here:
Capture.JPG
The main circuit isn't mine. It's actually BobTPH's from post #89 of the thread I mentioned, with a few adjustments.
If I simulate my circuit in LTspice and watch the output, it's not the same as shown in BobTPH's post, but it is the correct output and the one I want:
1592482892680.png
If I measure R6 (the output) on my circuit board, this what I get:
IMG_1086.JPG
Its not the best photo since I was holding the phone one handed but either way, that's not the correct output. I'm supplying the board 5V and a 2kHz signal. I would expect it to look similar to as shown above, but it doesn't.
When I was buying the LT1366's I accidentally bought through hole versions, so I though I would also do a breadboard test as well. I set up the schematic shown above on a breadboard and measured the output:
IMG_10870.JPG
This time there's was absolutely nothing on the output.
I've double checked my placements and made sure I don't have a bad connection (by giving everything a little wiggle)
Of course most circuits can work in theory, but not in practise. The original circuit was suggested by someone with far greater knowledge than me which means I don't think that the actual circuit design is the issue.

Can anyone help me fix this? It would be very much appreciated!
Bod
 

AlbertHall

Joined Jun 4, 2014
12,345
Measure the DC voltage on all three signal pins of both amplifiers and report here.

[EDIT] What is connected to the two wires that disappear off the edge of the image below R4?
 

crutschow

Joined Mar 14, 2008
34,285
What is the green screen with the garbled signal?

What is the purpose of the circuit?
The first stage is an inverter with a gain of 0.8 (ignoring the two offscreen wires to R4) and the second is an inverter with a gain of 1.0.
 

MisterBill2

Joined Jan 23, 2018
18,179
When the results are this dismal the first thing to do is to check the voltages at the pins of the ICs. It looks like the DC output voltage level is zero, and with the circuit as shown, it should be half the supply voltage. So the very first check should be the voltages at the power pins , 8 and 4. First, measure across the pins on the device itself, and then relative to the power supply common terminal..
Next, check the input pin voltages , which should all so be close to half the supply voltage. The symptom points first towards no power present.
(Note that the 2 comments appearing before mine actually arrived while I was entering my comment,)
 

Thread Starter

Bod

Joined Sep 18, 2016
317
Measure the DC voltage on all three signal pins of both amplifiers and report here.

[EDIT] What is connected to the two wires that disappear off the edge of the image below R4?
The first chip (left hand one) measures 2.5V across all three signal pins. The right hand chip also measures 2.5V across all three signal pins. I'm going to triple check those because I'm skeptical about those readings.
So for both chips:
Out = 2.5V
A+ = 2.5V
A- = 2.5V
As for the two wires, on my circuit I have a positive and negative peak detector. The output of the first stage of the circuit is flipped so I send that down to the negative peak detector. (Left wire is 0V and right wire is Audio)
What is the green screen with the garbled signal?

What is the purpose of the circuit?
The first stage is an inverter with a gain of 0.8 (ignoring the two offscreen wires to R4) and the second is an inverter with a gain of 1.0.
The green screen is the simulation output from LTspice. I think it looks a bit garbled due to the compression of the image when uploading. It is the signal I'm looking for on my actual circuit.
I'm going to be honest here, my recollection of the circuit isn't perfect because it was a long while ago I made it. What it does is it amplifies the incoming signal (by 0.8 as you pointed out) but also centres the signal at 2.5V so it ranges from 0V - 5V rather than -1 to 1V like a normal audio signal would.
When the results are this dismal the first thing to do is to check the voltages at the pins of the ICs. It looks like the DC output voltage level is zero, and with the circuit as shown, it should be half the supply voltage. So the very first check should be the voltages at the power pins , 8 and 4. First, measure across the pins on the device itself, and then relative to the power supply common terminal..
Next, check the input pin voltages , which should all so be close to half the supply voltage. The symptom points first towards no power present.
(Note that the 2 comments appearing before mine actually arrived while I was entering my comment,)
The input voltages are exactly 5 like I'm supplying it with. The input voltages on the input pins I have mentioned above.

EDIT: I was triple checking voltages and noticed I had placed R5 in the wrong place. Rather than it coming out of the output pin, it was connected to the A- pin. This means the breadboard output is exactly the same ad the one on my circuit board now. The voltages themselves haven't changed and they are all still 2.5V.
 
Last edited:

crutschow

Joined Mar 14, 2008
34,285
If you are not getting the correct output, then there is likely a wiring error.
Check the voltages at all the nodes and see if they all make sense.
 

Thread Starter

Bod

Joined Sep 18, 2016
317
I put this together quickly so excuse the crudeness:
main.png
This image is comparing points on my circuit in LTspice to real life.
The text on the schematic is showing everywhere I tested. Each one of those points is shown in LTspice, but also on my oscilloscope in real life. Hopefully it gives an idea of the odd results I'm getting.
 

Audioguru again

Joined Oct 21, 2019
6,674
Get rid of the breadboard. With the messy wires all over the place and resistors with their wires underneath them, how can you see which connects to what?
 

Thread Starter

Bod

Joined Sep 18, 2016
317
Hello,

Did you think of decoupling?
Put a capacitor of 10 uF accross R12.

Bertus
I didn't do that. I shall add that on soon.
Get rid of the breadboard. With the messy wires all over the place and resistors with their wires underneath them, how can you see which connects to what?
I tried my best to get some good photos. My actual circuit board I don't reallt think is any easier to trace. The traces are thin, it's double sided and the often go under components. For this circuit, luckily there int any traces on the underside. There's many things I need to invest in like leaded solder and even these style of breadboard jumpers. I don't have much money at the moment so I have to make do with what I've got.
Either way, here's a picture of the actual board:
IMG_1101.JPG
There are two vias - The one to the left of the LT1366 goes absolutely nowhere. It just joins the ground plains I believe. The via next to R6 (to the right of the right hand IC) takes the output signal and distributes it off the to the rest of the board.
 

MisterBill2

Joined Jan 23, 2018
18,179
Like I had to tell one tech at one place, "The circuit will not perform as intended unless it is wired according to the provided drawing." He was checking circuit boards that did not yet have all of the parts installed. And none of them worked right.
 

Thread Starter

Bod

Joined Sep 18, 2016
317
Unfortunately that part just seemed to have caught the light badly:
IMG_1102.JPG

Not too long ago I reflowed the other components on the board. I forgot these components so tomorrow I will reflow them to make sure they are seated properly. I'm hoping that will fix it.
 

MisterBill2

Joined Jan 23, 2018
18,179
Unfortunately that part just seemed to have caught the light badly:
View attachment 210094

Not too long ago I reflowed the other components on the board. I forgot these components so tomorrow I will reflow them to make sure they are seated properly. I'm hoping that will fix it.
The trace should have come straight off of the pin for a way and then angled over to the resistor R5. There would be a lower production yield with that layout.
 

Thread Starter

Bod

Joined Sep 18, 2016
317
Measure the resistance between pin 1 and pin 2 of IC2.
It's 8k ohms across pin 1 and pin 2
The trace should have come straight off of the pin for a way and then angled over to the resistor R5. There would be a lower production yield with that layout.
Yeah I understand. There's a lot of wasted space on the board but I don't intent to make anymore of them. They were just for me so I'm not too fussed.
It's mainly because it was auto-routed however I wouldn't have even thought to change it so I'll keep it in mind for my next circuit board.
 
Top