decoupling and coupling mosfet drivers / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

decoupling and coupling mosfet drivers

Move Thread

LAN_403

loveHV

Thu Dec 25 2008, 09:55AM

Registered Member #1854 Joined: Wed Dec 10 2008, 06:09PM
Location: West-Flanders Belgium
Posts: 88

i don't understand something about the coupling and decoupling of de mosfet drivers. i find some information on the internet but it's not much. (or im searching on the wrong sites

)

like they first talk about an capacitor from the driver to the GDT (i think thats coupling) and then about a capacitor over the power pins of the mosfet driver, and then over the supplie rails of the driver.

but wich value's should be chosen for the supplie rails, the powerpins, and from driver to GDT.
are there rules for chosing the capacitor values ??

(srry for the bad english i'm belgian

)

LithiumLord

Thu Dec 25 2008, 07:02PM

Registered Member #1739 Joined: Fri Oct 03 2008, 10:05AM
Location: Moscow, Russia
Posts: 261

In fact coupling/decoupling is basicly not what you think it is, if applied to drivers this stands for magnetic (GDT), optical or whatever other way of signal transmission to the switching devices that have to be electrically insulated from the control electronics.
What you are talking about is DC blocking capacitor (can also be referred as a decoupling one however) and power rail bypass. The first one has to be significant enough not to cause strong resonant effects and not cause much of a current limiting, but still small enough for any faster response. The use of it is required to make sure there is no DC component in the voltage (that leads to a current that's infinite for an idealized model as an idealized inductor exhibits purely complex impedance) applied to the GDT primary (can be caused by generally asymmetric signal, and also by the dynamic frequency float). The optimal value depends on the current and expected frequency float speed and can be just taken from any project description whenever you feel the device is close to yours.
The bypass capacitor is used to reduce ringing across the power supply rails of the driver - as the drive currents sometimes reach a few amps peak it's essential to remove any need of driving that much through thin wires with their stray inductance and active resistance. That capacitor has to be rigged straight to the driver supply rails and it's value has no upper limit - just make it big enough for handling such a high current for the time you need without causing much of a voltage drop.

loveHV

Thu Dec 25 2008, 08:40PM

Registered Member #1854 Joined: Wed Dec 10 2008, 06:09PM
Location: West-Flanders Belgium
Posts: 88

oh i think i understand know. ive got trough some site today also and now i understand why my TC was giving poor results
i didn't use any capacitor on my power rails, no capacitor on my powerpins. i only used a capacitor as DC blocker for my GDT.

i know use over the power pins an electrolyc capacitor of 10uF 63V

over the supplie rails of my driver electrolyc 2000uF / 35V and a shottky to reduce the ringing

and over the power rails over the halfbridge i think im gonna use an electrolyc of 220uF / 385V

do you think those values should do or do they need to be adjusted ???

Arcstarter

Fri Dec 26 2008, 12:55AM

Registered Member #1225 Joined: Sat Jan 12 2008, 01:24AM
Location: Beaumont, Texas, USA
Posts: 2253

Well, from what i know decoupling caps are basically there to reduce/eliminate ringing caused by inductance/capacitance on the circuit board. May be wrong.

Decoupling caps should be as close to the power pins as possible, and i have heard of .1 to 1uf. I know there are probably different values, but i just recently learned what they where anyway. I never used them neither, and alot of ucc's and 555 have died, maybe because of that, or maybe something else. The mosfets, however, never died when that happened. It was usually on a breadboard though, with high inductance/capacitance, which is what made me think that it was because of lack of decoupling caps. I now use breadboard only for prototyping, and usually not even that. I now use pre-etched pc boards, that are very simple to use.

loveHV

Fri Dec 26 2008, 09:59AM

Registered Member #1854 Joined: Wed Dec 10 2008, 06:09PM
Location: West-Flanders Belgium
Posts: 88

yeah my ucc's and 555 kept dying also because of the ringing and the tesla coil noise but this time i have prepared myself placed decoupling caps over the ucc's, over the 555 and 2 large ones with a schottky on the supply rails to prevent ringing. i have also aquired an metal box that will help in protecting my circuit against the tesla coil noise.

i also use a breadbord board but if the connections don't make big loops in the air and keeping the wires as short as possible will prevent a lot of inductance (i think)

today im finally gonna get my orderd UCC's and i hope this time my SSTC gives off more sparks

HV Enthusiast

Fri Dec 26 2008, 02:40PM

Registered Member #15 Joined: Thu Feb 02 2006, 01:11PM
Location:
Posts: 3068

A good rule of thumb for bypass capacitors for MOSFET drivers (especially those for higher current) is using 0.1uF ceramics and a 10uF Tantalum for each IC's VCC.

Lower current drivers can probably live with the 0.1uF, or even a 1uF ceramic.

But for high current drivers, this is a neccessity. Used for both energy storage (local) and also decoupling as others have already described.

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob