Re: [funwithtransistors] gate capacitance idea


Dec 5 3:10 AM

 


----------------------------

#1048 Dec 5 3:10 AM

I'm wondering if it would make sense to use a pair of BJTs to sink the

residual capacitance in the mosfet that will drive the power circuit in

my injector tester. The thinking goes like this. As I reduce the

resistor value across from the gate to source, the wave form obviously

becomes more square. At some point the current passed through this

resistor will become higher than I want and also start to effect the

performance of the mosfet. The idea here is to achieve a .0002 second

square pulse. The 555 has a positive and a negative output phase. I

am using the positive phase to drive the N channel mosfet. If I can

use the negative phase to drive a pnp transistor which in turn drives an

npn which sinks the the gate capacitance directly to ground, that should

"square it up" nicely. Just a thought........... I have a box chock

full of bjts ......mostly npn.



There is an Russian proverb...... "To a man who's only tool is a

hammer, every problem looks like a nail"



I love Russian proverbs...... exquisite irony is embodied in

nearly all of them. My favorite is this:



Under Capitalism, man exploits his fellow man..... Under Socialism the

reverse is true ........





Howard



Pardon the off topic humor...............



----------------------------

#1049 Dec 5 5:27 AM

Hi Howard,  Just how “square” do you really need the output pulse to be?  In other words what should be the rise time and the fall time of that 200 usec pulse? (Rise time is typically measured from 10% to 90% and fall time from 90% to 10% of the stepped pulse.)  My guess is that you would want them to be less than 10% of the pulse width (measured at FWHM).  Considering that you are driving an inductive load (I.E., the injector), I suspect that the you may want to use a current driver.  You do not want to over-think the subtleties of the driver before knowing how the inductive load is going to alter the driver’s behavior.   Make the rise time too fast and you will have overshoot, ringing, etc., and this will lead you down the road to unpleasant circuit consequences.  Jerry F.  From: funwithtransistors@yahoogroups.com [mailto:funwithtransistors@yahoogroups.com] On Behalf Of StoneToolSent: Tuesday, December 04, 2012 7:11 PMTo: funwithtransistors@yahoogroups.comSubject: [funwithtransistors] gate capacitance idea



----------------------------

#1052 Dec 5 5:10 PM

--- On Wed, 12/5/12, Gerald Feldman gfeldman2904@...> wrote:

> From: Gerald Feldman gfeldman2904@...>

> Subject: RE: [funwithtransistors] gate capacitance idea

> To: funwithtransistors@yahoogroups.com

> Date: Wednesday, December 5, 2012, 12:27 AM

>

> Hi Howard, .Just how .square. do you really need the output pulse

> to be?. In other words what should be the rise time and the fall

> time of that 200 usec pulse? (Rise time is typically measured from

> 10% to 90% and fall time from 90% to 10% of the stepped pulse.).

> My guess is that you would want them to be less than 10% of the

> pulse width (measured at FWHM).. Considering that you are driving

> an inductive load (I.E., the injector), I suspect that the you may

> want to use a current driver.. You do not want to over-think the

> subtleties of the driver before knowing how the inductive load is

> going to alter the driver.s behavior.. .Make the rise time too fast

> and you will have overshoot, ringing, etc., and this will lead you

> down the road to unpleasant circuit consequences. .Jerry F. .



I've had some success, when driving NPN BJTs from logic gates, of

bypassing the current limiting resistor, from the output of the

logic gate to the base of the BJT, with a capacitor. The capacitor

should be similar in value to the base to emitter capacitance of

the transistor (few pF up to a few dozen pF for large power

transistors).



Of course, before you go too far down that path, the proper thing to do

would be to simulate the circuit, using something like SPICE (or

LTSpice).



In my case, I was driving a high power LED with a power transistor,

at somewhere near a 10 MHz speed, so I didn't want any pulse stretching

or roll-off effects.



Dave



----------------------------

#1056 Dec 6 7:11 AM

This is all going well over my head............being a

newbie............ It's fascinating stuff. Today while driving back

from Bozeman Mt (about 90 miles from home), I found myself playing with

circuits in my head. Designing entirely unrelated things. It's a

world where I can build things for pennies that once were entirely out

of reach. For example I am looking at setting up the steering on my

4wd/4 wheel steer forklift (28K pound machine) which I previously

converted to use micro switches and hydraulics, and is now giving

problems, to operate on proximity sensors and mosfets instead of micro

switches directly tied to solenoid valves. I've ordered some triacs to

do some AC switching for another project. One thing is very

clear......I'm gonna make a LOT of mistakes before it's over ;-) Many

of the circuits I look at make no sense to me.......... but I'm

gradually finding the logic. There are only so many ways of doing things.



years ago I was working with an engineer and he was catigating me

constantly because I didn't know hydraulic symbols and notation.......

though I understand hydraulics very well. I drew up plans including

electrical and hydraulic systems, and were it not for my notes, it would

have been difficult to sort out what was hydraulic and what was

electrical. At that level many of the components function in roughly

he same manner, and I tended to look at them in the same way. We had

some interesting discussions and became friends. We spoke the same

language............. when he died it was a great loss to me!



Howard > --- On Wed, 12/5/12, Gerald Feldmangfeldman2904@...> wrote:

>

>> From: Gerald Feldmangfeldman2904@...>

>> Subject: RE: [funwithtransistors] gate capacitance idea

>> To: funwithtransistors@yahoogroups.com

>> Date: Wednesday, December 5, 2012, 12:27 AM

>>

>> Hi Howard, Just how .square. do you really need the output pulse

>> to be? In other words what should be the rise time and the fall

>> time of that 200 usec pulse? (Rise time is typically measured from

>> 10% to 90% and fall time from 90% to 10% of the stepped pulse.)

>> My guess is that you would want them to be less than 10% of the

>> pulse width (measured at FWHM). Considering that you are driving

>> an inductive load (I.E., the injector), I suspect that the you may

>> want to use a current driver. You do not want to over-think the

>> subtleties of the driver before knowing how the inductive load is

>> going to alter the driver.s behavior. Make the rise time too fast

>> and you will have overshoot, ringing, etc., and this will lead you

>> down the road to unpleasant circuit consequences. Jerry F.

> I've had some success, when driving NPN BJTs from logic gates, of

> bypassing the current limiting resistor, from the output of the

> logic gate to the base of the BJT, with a capacitor. The capacitor

> should be similar in value to the base to emitter capacitance of

> the transistor (few pF up to a few dozen pF for large power

> transistors).

>

> Of course, before you go too far down that path, the proper thing to do

> would be to simulate the circuit, using something like SPICE (or

> LTSpice).

>

> In my case, I was driving a high power LED with a power transistor,

> at somewhere near a 10 MHz speed, so I didn't want any pulse stretching

> or roll-off effects.

>

> Dave

>

>

---------------

>

> Always keep electrons and holes flowing in opposite directions.Yahoo! Groups Links

>

>

>

>







This Site's Privacy Policy
Google's privacy policies

S
e
n
i
o
r
T
u
b
e
.
o
r
g