[Diy_efi] MC3334

[Mike]

Having driven several IRF840's in parallel on inverters, I can
tell you that even one 840 or 740 has a fair gate capacitance,
and even when we had a small number we really had to insert a
high transient current capacit driver...

The instantaneous current is given by i=Cdv/dt

this can be simplified in terms of this drive by:-

C = Gate capacitance
dv = Gate voltage drive (ie. Measure range from v max to 0 on scope)
dt = period of time over which it actually changes.

So you can see that a short transition results in a large current,
your driver cct may not be capable of supplying this transition,
hence the fet switches slowly hence it heats up !

the other thing is Drain Source Diode - acts like a zener ! ! ! !

and this can draw/snub a fair back emf *also* causing heat, if
the 320v is measured on a slow timebase cro, as the real peak
may be higher - but then again the device hasn't failed has it,
so its more then likely the main heat contributor is slow turnon
and *of course* slow turn off as well !

In my experience wtih a nice fast crisp driver the mosfet should
run about 30 - 40 deg in that scenario, (more or less)

Suggest use a higher voltage mosfet and/or put in a snubber
cct and definitely a meatier gate driver, Motorola make several
or use a pnp/npn pair - like bc547 and compliment with a gate
zener just to be safe,

rgds

Mike



At 01:57 PM 12/9/2002 -0300, you wrote:
>
>
>I am currently working on a very simple ignition module based on the
>MC3334 High Energy Ignition Coil Driver,from Motorola.This is a
>rather old chip,in fact,Motorola has taken the datasheet away from its 
>intranet,but I know for sure that the chip is still in production.They sent 
>me the datasheet.
>I am driving a 4 mH primary ignition coil with a IRF740 mosfet on an 
>aluminium heatsink big enough to fit on one side of the distributor,and
>50% dwell time from an inductive 50 Hz signal (which represents 750 rpm
>in a 4 cylinder engine).
>The mosfet is getting too hot after 3 minutes of switching.
>Current peak : 5,3 A
>Reverse voltage peak : 320 V
>Dwell time : 51 %
>Captor inductive : 5o Hz sinusoidal signal.
>The IRF740 is set in conduction through a 200 ohms 5 watts resistor
>from a 12 Volt bat.,and the MC3334 pulls the gate down to shut it off
>at the end of the dwell time(roughly 10 ms at 50 Hz),to induce the magnetic 
>field on the secondary winding of the coil,connected to the plug which will 
>produce the spark,and keeps it shut off untill the next
>captor signall is received.
>But a IRF740 drain scope read shows that what should be a fairly flat line 
>(or a mild upward slope to 0,2 Volts)at ground level for the first couple of 
>milliseconds of the dwell time,shows an upward slope from ground to 1,5 
>volts,which I believe springs from the fact that the IRF740 is not in full 
>conduction,therefore,getting hot.
>Has anyone built a project on this chip?
>I will appreciate any help I gan get.
>Thanks in advance.
>
>
>
>_________________________________________________________________
>Chat with friends online, try MSN Messenger: http://messenger.msn.com
>
>
>_______________________________________________
>Diy_efi mailing list
>Diy_efi at diy-efi.org
>http://www.diy-efi.org/mailman/listinfo/diy_efi
>
>

_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi