Quick MOSFET Question / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Quick MOSFET Question

1 2 next

Move Thread

LAN_403

Chris Russell

Fri Jun 06 2008, 09:30AM

... not Russel!
Registered Member #1 Joined: Thu Jan 26 2006, 12:18AM
Location: Tempe, Arizona
Posts: 1052

Hello,

Just a quick question for you folks more familiar with this sort of thing than I am. I'm pondering a simple RF oscillator/amplifier design. I'm using 74HC240, which I know can drive an IRF510 (190pF input capacitance) directly without any trouble, when using several gates in parallel.

I'm wondering if the 74HC240 would be able to drive something a little different... the MTP3055E (600 pF input capacitance), since it's all I have on hand at the moment. Anyone have any additional information I might need to figure out if it's possible? I've tried a few back of the envelope calculations that seem to suggest that it is possible, but I feel that my knowledge in this area is sadly lacking.

Thanks in advance,
Chris

WaveRider

Fri Jun 06 2008, 10:58AM

Registered Member #29 Joined: Fri Feb 03 2006, 09:00AM
Location: Hasselt, Belgium
Posts: 500

Hi Chris,
What frequency are you considering? . Paralleling the buffers (which is what I assume you will do) may work up to a MHz or so. (In the spec sheet, 50pF load per buffer is assumed..)

Give it a go!

Cheers...

Marko

Fri Jun 06 2008, 12:16PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Hello Chris,

74HC240 can only provide 5V which I don't think will be enough for a proper turn-on of a non-logic level mosfet.

If you don't want to use gate drivers, to obtain drive from 5V logic I generally use an inverter from a BS170 MOSFET and 100ohm resistor to +Ugg (12, 15 or etc. V), then amplify it with emitter follower totem pole of transistors. Larger signal transistors like 2N2222 and complementary 2N2907 are excellent for this.

Such drive will work up to 5 or so Mhz for 2-3nF of capacitance max, assuming direct drive.

What frequency are you aiming for?

If it's really high resonant gate drive is a proffered option. It is basically an inductor in series with gate capacitance tuned to resonate with it, and eventually some DC bias.
This is especially easy to do if you have a fixed frequency source.

Ferrite transformers may be used and their leakage inductance used as resonant L.

Avi

Fri Jun 06 2008, 12:33PM

Registered Member #580 Joined: Mon Mar 12 2007, 03:17PM
Location: Melbourne, Australia
Posts: 410

I don't recommend paralleling gates together at high frequencies, i've seen the differences in propagation time make a mess of the signal.

Chris Russell

Fri Jun 06 2008, 09:57PM

... not Russel!
Registered Member #1 Joined: Thu Jan 26 2006, 12:18AM
Location: Tempe, Arizona
Posts: 1052

Thanks for the input.

I should mention that I am running the 74hc240 at 7V, which is above the specified max, but it does work, as long as I have a heat sink of some kind on the package. In this configuration, it has been shown by many people that it can easily drive an irf510 directly, up to 15MHz or so. I'm personally only aiming for no higher than 4MHz, the high end of the 80m ham band.

The chip itself is able to supply a good amount of power, which makes me hopeful. People have used the 74hc240 as a one-chip transmitter, using one gate as an oscillator, one gate as an inverter, and the remaining six in push-pull configuration, and it reportedly gives 200mW or so of output.

I'm tempted to just hook it up and give it a try, but since my budget for this project is capped at $0.00, and I only have the one MTP3055E (I do have many spare 74hc240's, thankfully), I am somewhat concerned that I will end up not fully turning it on, thus causing it to fry in pretty short order, or at least throwing most of my power away as heat.

Marko

Fri Jun 06 2008, 10:24PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

7V is I guess just around the edge, but I would never overvolt an IC like that for anything serious.

You must have a BS170 and pair of just about any small signal complementary transistors over there? BS170 can be driven directly from weak logic.

In case I just had to use HC240, I would go for low input voltage and resonant gate drive to get proper voltage on the MOSFET gate. Lower input voltage and higher Q is more efficient, especially if application is battery powered.

Chris Russell

Sat Jun 07 2008, 12:59AM

... not Russel!
Registered Member #1 Joined: Thu Jan 26 2006, 12:18AM
Location: Tempe, Arizona
Posts: 1052

I have various transistors, but I wasn't able to find any BS170s in my junk box. Is the inversion really necessary? Surely the 74HC240 could drive a 2N2222a directly.

Resonant gate drive is worth a shot, but I'm thinking it's going to be very fiddly trying to wind a ferrite transformer with just the right leakage inductance.

Dago

Sat Jun 07 2008, 09:44AM

Registered Member #538 Joined: Sun Feb 18 2007, 08:33PM
Location: Finland
Posts: 181

Chris Russell wrote ...

I have various transistors, but I wasn't able to find any BS170s in my junk box. Is the inversion really necessary? Surely the 74HC240 could drive a 2N2222a directly.

Resonant gate drive is worth a shot, but I'm thinking it's going to be very fiddly trying to wind a ferrite transformer with just the right leakage inductance.

You can always just add leakage inductance with a series inductor.

Marko

Sat Jun 07 2008, 11:39AM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Hi Chris,

The reason the inversion stage is needed - is that the emitter follower push-pull *can not* amplify voltage - only current - and thus some sort of preamplifier is needed to shift the voltage level. If you connected the push pull stage right on the output of HC240 you would actually get the max output voltage reduced by Ube of the transistors.

Any small mosfet will do, 2N7000, BS 170, BSS123, etc...

There are few other things you can try, but I never found anything better than just a small mosfet;

-bipolar transistor is rather slow in common collector configuration because it is driven deeply into saturation, but with proper biasing and base drive you might just get it to work.

-you can use a higher voltage CMOS IC for level shift. It is slower but should work as preamplifier at 4Mhz I think.

-you could use an operational amplifier if you have one fast enough.

Regarding resonant gate drive:

You *do not* need a ferrite transformer at all, just a small inductor and a resistor will do the job. Resistor should be calculated to give Q of the circuit for required voltage rise.

Higher Q and lower drive voltage is more efficient because Q is inversely proportional to resistance, and by halving the resistance input voltage needs to be halved too to keep the same output, which also reduces dissipated power in the circuit because power is proportional to square of voltage!

So the point of the transformer, apart from insulation, is just to provide low enough output voltage (like lower that HC240 could provide) to drive the tank.

74HC240 goes with supply down to 2V so transformer is probably not going to be needed at all.

Marko

GeordieBoy

Sat Jun 07 2008, 12:43PM

Registered Member #1232 Joined: Wed Jan 16 2008, 10:53PM
Location: Doon tha Toon!
Posts: 881

The resonant gate-drive is, of course, a narrow-band solution. I'm not sure if Chris wants to operate his switching amplifier across several bands?

1 2 next

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