How to reduce the voltage spikes?

Thread Starter

Nick Long

Joined May 12, 2020
52
Hello again!
I am researching the synchronous switch circuit for vibration control. The figures below show the idea schematic and waveforms for some key nodes. The third figure is the simplified topology for the schematic.

Here, I got some problems with my circuit. When the switches close, the inductor L2 and the capacitor C5 would form an LC loop, and then the voltage on C5 would invert. However, during the switching events, the voltages on many nodes, such as VO, NVO, and the inverting input of Op Amps, would be affected by the inductor spikes, which may destroy these components. Is there a method to solve or alleviate this problem? Thanks a lot!

1628047909558.png
1628048032874.png
1628048335735.png
 

Attachments

Thread Starter

Nick Long

Joined May 12, 2020
52
What is this device supposed to do exactly ?
What is the problem that this device is supposed to solve ?
.
.
.
It is designed for connecting the piezoelectric patch which is attached to the vibrating structure to suppress vibration.
The problem is that during the switching event, the inductor spikes may destroy the Op Amps.
 

LowQCab

Joined Nov 6, 2012
1,096
From what I've seen so far, this doesn't look like a viable solution.
Do You have any Pictures or Drawings that would show what the actual problem is ?
What creates the "Vibration" in the first place, and why is it bad ?
This sounds more like a mechanical design problem.
Also what is the general Frequency-Range of this Vibration ?
.
.
.
 

Thread Starter

Nick Long

Joined May 12, 2020
52
From what I've seen so far, this doesn't look like a viable solution.
Do You have any Pictures or Drawings that would show what the actual problem is ?
What creates the "Vibration" in the first place, and why is it bad ?
This sounds more like a mechanical design problem.
Also what is the general Frequency-Range of this Vibration ?
.
.
.
Thanks for your reply. It is hard to describe this clearly using several sentences. If you are interested in this topic, you can refer to the literature I attached.
 

Attachments

LowQCab

Joined Nov 6, 2012
1,096
I'm not particularly interested in how much money and time have been devoted to
figuring out how to use a very limited box of Tools to solve a problem.
That's why I asked the questions that I did.
I wouldn't use Piezo-Sensors because they are too finicky,
and require precise positioning,
which can only be found by extensive testing of the Structure/Part,
and adhering to rigorous surface-preparation and mounting-requirements.
And, for the lower Frequencies mentioned in the Paper, (50hz),
Piezo-Sensors are almost useless,
You need a Strain-Gauge for that, or an inertia-based "Microphone", (for lack of a better description),
and serious Power on tap to move some serious-weight at a high rate of acceleration.

I try to avoid getting "stuck" on a particular "solution",
there are very few "one-size-fits-all" solutions.

This problem seems to be all about "Resonances",
I say that to differentiate, between random-motions, and actual repetitive "Oscillations",
which BTW, can be engineered-out of the part, or damped with various materials,
or, the natural Frequency of the part can be changed to a range that won't tend to Oscillate.

This device requires Power and possibly Maintenance at some point,
therefore I don't see its value for use in Permanent-Structures,
if that's what it is being used for.

I think what You need to have is a good universal "Magnet-Coil-Weight" device to start off with,
but more info is required on what You are currently using, and why,
before I could comment further.
.
.
.
 

DickCappels

Joined Aug 21, 2008
7,707
What is the purpose of the inductor? I can't quite see exactly how it connects to the circuitry shown.

Perhaps you should disconnect the opamps from the output while you are sending pulses. A diode bridge might be the best way, but relays and MOSFETs are other solutions you might consider.
 

Papabravo

Joined Feb 24, 2006
16,799
With any LC circuit you will get ringing in the absence of damping. So a solution to the general problem of ringing is to compute the appropriate amount of damping, and then apply it. This may or may not solve the problem, especially if the problem happens to be unrelated to LC resonance.
 
Top