ENERGY AMPLIFICATION

[Dave45]

If you turn the transistor on and off quickly, it will do a spiffy job of destroying the transistor.

Lol Mark you crack me up.
I really like the spiffy part LoL

Its set up just like any other buck converter except for the recycle windings, tell me why it will kill the transistor and a solution.

I do enjoy your input you know a helluva lot more about electronics than me.
What do you do for a living if you dont mind I ask, I dont blame you if you dont want to answer on a public forum.

dave 

[MarkE]

Lol Mark you crack me up.
I really like the spiffy part LoL

Its set up just like any other buck converter except for the recycle windings, tell me why it will kill the transistor and a solution.

I do enjoy your input you know a helluva lot more about electronics than me.
What do you do for a living if you dont mind I ask, I dont blame you if you dont want to answer on a public forum.

dave

Dave I have explained this to you before:  V = L*di/dt.  When you switch off a transistor that carries current coming through a coil, the voltage at the junction of the coil and the transistor swings as needed to maintain the current.  When you insert another coil in series with the diode the current can only build up in that second coil as the integral of the V*T product across that second coil.  So do a little math under the simplified assumption of a linear current change in the second inductor:

Coil 1 >> coil 2 100uH.  Current 1A.  Turn-off time of the transistor:  100ns.   Parasitic capacitance  = 0.

In 100ns you need a delta I in the second coil of 1A.  V = L*di/T V= 100uH*1A/100nS V= 1000V.

This is only an example.  To determine the real voltage swing you need to plug in the real inductance and current and transistor response.

You can make the circuit survive by:

a. Using a high enough voltage transistor to survive the inductive kick-back, or

b. You can clamp,  or

c. You can snub, or

d. You can slow down the transition, or

e. You can use a combination or methods, or

f. You can fry your transistors.

People build circuits that self-destruct all the time when they switch too fast for the amount of inductance in the wiring loop that consists of: the connection from the transistor to the diode and the diode back to whatever closes the loop with the other end of the switched inductor.  This may be for example in a relay or solenoid driver the other end of the coil itself, or it may be the opposite end  of a power supply from that coil, in which case the inductance through the supply is an issue usually solved with capacitors. 

In your case you not only have the wiring, but you are inserting a wound inductor.  If the wound second inductor is a tightly coupled winding to the first inductor with the appropriate orientation, then the mutual inductances will cancel, but you will still have to deal with the leakage and wiring inductances.

What the appropriate fix is for your circuit depends entirely on what you are trying to do.  I suggest to you that controlling voltage by wasting energy probably won't help with your ultimate goal.

[forest]

MarkE


Can you post a good source of information step by step how to build snubber circuit to protect mosfets/transistors from inductive kickback ? I knew that in relay coils it is usually used diode across coil terminals. Funny thing with that diode (is it called freewheeling diode?) - when I added it to my circuit current consumption from battery rised from 200mA to 4A !!! Just that is interesting....but the snubber would help really I think. Another strange issue is my IR1405 mosfet survive 80V spikes, hmm.....

[lost_bro]

MarkE


Can you post a good source of information step by step how to build snubber circuit to protect mosfets/transistors from inductive kickback ? I knew that in relay coils it is usually used diode across coil terminals. Funny thing with that diode (is it called freewheeling diode?) - when I added it to my circuit current consumption from battery rised from 200mA to 4A !!! Just that is interesting....but the snubber would help really I think. Another strange issue is my IR1405 mosfet survive 80V spikes, hmm.....

Hope this helps out a bit....

Take care, peace
lost_bro

[a.king21]

forest and markE
You can also  use an NE2 bulb between collector and emitter if your transistor can stand 100 volts.
You can use 2 ne2 bulbs in series if your transistor can take 200 volts etc, etc, etc.


Done it loads of times.