# Being Frugal: Harvesting Energy from Transients?

## Do you support Dr. Evil in his quest for world domination?

• ### Yes, but only because I don't want a nuke delivered to my doorstep.

• Total voters
1

#### dr.evil

Joined Aug 18, 2010
80
People!

I have to launch a nuke into space, since I don't know when and what I want to annihilate, yet..... I want the control circuits to be operational for as long as possible, and that can only be done by saving on electricity.

I have the following test circuit

And this is the scope readout

Now I have only two questions:

1. How to harvest the energy in the transient? Any suggestion would be highly appreciated.

2. How can the transients showing up on the base of the switching transistor be eliminated? I know it can be "suppressed", but I'm not interested in wasting energy, I'd rather the transients didn't reach the base.

Evilly yours
Dr. Evil.

#### wayneh

Joined Sep 9, 2010
17,153
1. How to harvest the energy in the transient?
Turn off V1 and leave the energy it wastes un-consumed.

#### DickCappels

Joined Aug 21, 2008
7,555
I studied transients once...

...and concluded that there is little hope for recovery.

#### dr.evil

Joined Aug 18, 2010
80
Turn off V1 and leave the energy it wastes un-consumed.
Yes, and how do you suggest that I go about that? Yet another transistor?

#### dr.evil

Joined Aug 18, 2010
80
I studied transients once...

...and concluded that there is little hope for recovery.
Dick, good pun, but seriously - are you trying to say that it can't be done?

#### Alec_t

Joined Sep 17, 2013
12,154
What is the impedance of the V1 source?
If V1 is supposedly 'square', why is the blue trace not showing that?

#### dr.evil

Joined Aug 18, 2010
80
What is the impedance of the V1 source?
If V1 is supposedly 'square', why is the blue trace not showing that?
Alec, Yes, I was asking my self the same question.

The V1 is the output of a 555 timer (pin 3), I don't know the impedance of it.

I guess the reason for the 'squarish' looking signal is a combination of transients, and the rise/fall times of the astable timer. I'm running it at 168KhZ with a calculated duty cycle of 1/10.000, all told is putting a strain on the timer (I guess).

#### dr.evil

Joined Aug 18, 2010
80
OK, so I made a few alterations to the circuit, and I tried to make some efficiency calculations, maybe some resident mathematical aficionado can confirm if my logic is on the right path. I seem to get ~2200% efficiency when I only look the difference between power (R1) through the coil (528uW) and (R2) output (118mW) ), which can't be correct, or maybe there is something wrong with the construction.

I added a diode, cap and a resistor - so the circuit looks like this now (And in the meantime I had to switch from circuitlab.com to schematics.com, as it was constantly demanding money to continue)

NOTE: Wrong Q1 symbol, it's still a PNP ;-)

And this is the scope dump

And if anyone is interested, the (sort of) calculation is attached.

#### Attachments

• 59.9 KB Views: 0
Last edited:

#### wayneh

Joined Sep 9, 2010
17,153
You've drawn the PNP transistor as an NPN.
Your circuit reminds me of an inductor tester I built, it pulses the inductor and makes it ring (LC with C1). Counting the rings tells you about the inductor status.
What is the purpose of this circuit?

#### dr.evil

Joined Aug 18, 2010
80
You've drawn the PNP transistor as an NPN.
Your circuit reminds me of an inductor tester I built, it pulses the inductor and makes it ring (LC with C1). Counting the rings tells you about the inductor status.
What is the purpose of this circuit?
PNP, I know - But thanks, I wrote a note below (now moved up before schematic)

Purpose: It's just a test. If a coil regurgitates the power you put in, and you harvest that, you should get close to 100% efficiency (or 200% depending on your calculation), in any case you put "one" in, and get the work done for half the price, if you follow me?

#### dr.evil

Joined Aug 18, 2010
80
Your circuit reminds me of an inductor tester I built, it pulses the inductor and makes it ring (LC with C1). Counting the rings tells you about the inductor status.
Wayneh, what do you mean by "status"? If you want to measure the inductance, I found this simple and useful tip:
http://www.dos4ever.com/inductor/inductor.html

#### wayneh

Joined Sep 9, 2010
17,153
Wayneh, what do you mean by "status"?
An unbroken coil will produce a large ring count. See this for more.

#### dr.evil

Joined Aug 18, 2010
80
An unbroken coil will produce a large ring count. See this for more.
I see now, more rings mean higher Q. Few or no rings indicates a problem in that component. It's cool test tool.

#### wayneh

Joined Sep 9, 2010
17,153
I see now, more rings mean higher Q. Few or no rings indicates a problem in that component. It's cool test tool.
I built it to try to diagnose a CRT TV. I suspected the TV needed a new LOPT but didn't want to buy one without testing the old one. I ultimately gave up on the TV. I was mad and wanted to show it who's boss, and worked on it far too long. When you can succeed in a repair and end up with something you have to pay to haul away, I lose interest.