TINA spice simulation gives floating warnings

inkl

Member
2019-10-25 9:47 am
Hey guys!


I'm trying to simulate a simple mic preamp with TINA. Here's the circuit I'm using.


So I created a simulation, here it is.


When I run ERC Analysis it's all green. But when I run DC Analysis -> Table of DC Results, I get 2 warnings:

- Pin 1 of Voltage Pin Vout is floating

- Pin 2 of capacitor C3 is floating.


I tried to change that part multiple times but no luck.

Please help me out here.

Thanks!
 

rayma

Member
2011-04-29 8:37 pm
You'll have to try it, but that's certainly not a loud sound. The mic should be close.
There's something wrong with the gain scale in the graph.
Maybe cutting off the bass (say around 400Hz) will help to reduce noise.
The midband gain of your circuit is around +27dB.
 
Last edited:

PRR

Member
Paid Member
2003-06-12 7:04 pm
Maine USA
www.diyaudio.com
...Here's the circuit I'm using.
- Pin 1 of Voltage Pin Vout is floating
- Pin 2 of capacitor C3 is floating.

That's not the plan in your simulator. And I don't have TINA here to read your TINA file.

But I bet it objects to having one end of a capacitor "floating". SPICE *insists* on figuring the DC voltage on every node before it does any other analysis. But the "floating" end of a cap can float to ANY voltage.

Anyway in real life you would have a LOAD.

Put a reasonable load on that output after the capacitor. Now SPICE's tiny mind will figure the DC voltage here is 0.0000V and be happy.
 

Attachments

  • inkl-42.gif
    inkl-42.gif
    5.4 KB · Views: 135

PRR

Member
Paid Member
2003-06-12 7:04 pm
Maine USA
www.diyaudio.com
> Would this be enough to hear a single water drop? I'm not sure.

Insufficient information. How much sound power in the drop? How far away? In a room or in the open? What are you going INto? High-gain line input, headphones, mike input?

That's why mike amps have gain controls. In this one, make R4 about 1k. This gives gain of 1 to 100. Most jelly-bean opamps will be clean-enough at gain of 100(*). Set up your system, drip your water, and trim VR1 for happy results.

(*) the slower opamps working at gain of 100 may not have the extended treble to cleanly reproduce, say, Ginger Baker's cymbals. However even the antique '741 opamp at gain of 100 would be flat to 10kHz, which really is all you need in most situations, especially when working very-soft where system and ambient noise intrude.
 

inkl

Member
2019-10-25 9:47 am
Let's take SPL 25 dB at the 10 meters distance (32 ').
The idea is to get the input level out of this to connect to a standard line-in.


I tried changing R4 to 1k and then setting VR1 to different levels - it's doesn't change much, still sround -35 dB.


ok, I'll read more about this op amp and see if I can make it better.


Thanks for the hints. I'm still do not understand how this works and I hate this. I'm still reading but I thought I'd make something real.
 

PM2222A

Member
2019-09-29 10:21 pm
I looked at your diagram and noticed that you have the value of R2 set at 10k that the diagram shows should be 100k. I corrected that and set the trans-conductance of the controlled current source you used for the electret capsule model to a more reasonable value of 1mS, up from 100uS and got a gain sweep of around 24 dB. The trans-conductance value should be set to match the capsule you end up using.
 

inkl

Member
2019-10-25 9:47 am
I don't get it... I have almost the same circuit and I get only 13,24uV! Why is that?


[IMGDEAD]https://inus.blob.core.windows.net/temp/OPA134%20preamp%20-%20err.png?sv=2018-03-28&si=ro&sr=b&sig=NwkknQ2LDJ9kmiOCramRjTsqH3G31PMr%2BvvSZ47MQZc%3D[/IMGDEAD]


Could it be the "electret mic source"?


And yes, the gain < 0...?!
 
My version of Tina doesn't seem to have the part marked "Pa" (not sure what that is).

But a couple of things - it there power to the opamp? (it doesn't look connected to me, but ERC would test that).

Also - have you tried taking power to your electret module via the 10k resistor?

I'm not entirely sure how that works in simulation.
 

PRR

Member
Paid Member
2003-06-12 7:04 pm
Maine USA
www.diyaudio.com
I don't get it... I have almost the same circuit and I get only 13,24uV! Why is that?..... And yes, the gain < 0...?!

Is it?

Your *assumed* input, divided by output, seems low. By inspection and by Tony's cross-check.

If you assume/hope the capsule works as specified, you do NOT have to simulate that part. What you really want is the AMPLIFIER response. Since lack of output "could be", lack of input, you usually want at least TWO meters on your workbench. Yes, this gets crowded and costly on a bench, but in simulation meters are free and small. LOOK at the input as well as the output. Maybe the VG VCCS pA assembly doesn't really do anything, or not the right thing. Check Everything. On bench I sometimes have 3 meters reading 5 points, and at least as many in sim.
 

Attachments

  • inkl-43.gif
    inkl-43.gif
    29.7 KB · Views: 75