• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

LTSpice showing different output DC with same circuit

Willen

Well-Known Member
Thread starter #1
Hi,
I don't know what the LTspice trying to show me. In Transient, if I set 2 second stop time, the osc (voltage booster) showing nothing boosted (just showing around 8V which is its supply voltage). If I set 2 mili-second stop time without changing any parts then I can see the boosted output of more than 20 volts. WHAT I should understand ?

(I made the osc in real world. When I measured output with my DMM, it whows 18V just first half a second then starts to show 8V. Maybe it has high impedance output which cannot drive my DMM?)
 

Attachments

crutschow

Well-Known Member
Most Helpful Member
#2
You've run into a common problem with oscillator simulations.
Circuits with oscillators sometime will not start or start erratically in simulation, since the circuit normally does a DC analysis at the start, and this can put the oscillator in a quasi-stable, non-oscillatory state with no intrinsic noise to start it.
Check the voltage startup at 0V box or the Skip initial operating point box in the transient analysis which eliminates the initial DC analysis and usually starts the oscillator.

1551331384741.png

Don't know why the voltage drops that much with your meter attached. :confused:
 
Last edited:

unclejed613

Well-Known Member
Most Helpful Member
#3
since i don't have LTspice on my work computer, i went and found a schematic for one of these microphones [here].
if you are measuring at the element, the oscillator/rectifier voltage is being sourced through a 1G resistor. at least that's according to that schematic. your DMM has an input resistance of 10M. the condenser element only needs an electrostatic charge on it, so there's barely any charging current from the oscillator/rectifier. it's entirely possible the 10M input resistance of the DMM is loading the circuit down.
 

Willen

Well-Known Member
Thread starter #4
You've run into a common problem with oscillator simulations.
Circuits with oscillators sometime will not start or start erratically in simulation, since the circuit normally does a DC analysis at the start, and this can put the oscillator in a quasi-stable, non-oscillatory state with no intrinsic noise to start it.
Check the voltage startup at 0V box or the Skip initial operating point box in the transient analysis which eliminates the initial DC analysis and usually starts the oscillator.

View attachment 116888

Don't know why the voltage drops that much with your meter attached. :confused:
Hi, Thank you! It worked! But with 2 second stop time, it took me tens of minutes to simulate 100%. Is it normal?

since i don't have LTspice on my work computer, i went and found a schematic for one of these microphones [here].
if you are measuring at the element, the oscillator/rectifier voltage is being sourced through a 1G resistor. at least that's according to that schematic. your DMM has an input resistance of 10M. the condenser element only needs an electrostatic charge on it, so there's barely any charging current from the oscillator/rectifier. it's entirely possible the 10M input resistance of the DMM is loading the circuit down.
I was thinking same, but a member here who donated me the MXL mic said he measured 40V output (after 1Meg resistor or across filter cap) on his DMM. My DIY condenser mic has little lower gain so I am thinking that I made a faulty OSC...
 

crutschow

Well-Known Member
Most Helpful Member
#5
But with 2 second stop time, it took me tens of minutes to simulate 100%. Is it normal?
Yes.
Simulating a high frequency circuit for a long period of time requires a zillion (estimate) calculations which takes a lot of time, even with a modern PC.
My laptop takes about a minute to do 2ms of simulation, so 2 seconds sim time would take about 17 minutes real time.

I measured an oscillator frequency of about 1MHz so for 2s, that's 2 million cycles.
So it calculates about 33 oscillator cycles per real-time second.
And each cycle takes many floating-point calculations.
 

EE World Online Articles

Loading

 
Top