555 switching a mosfet not producing square wave output

Thread Starter

Guy Eastwood

Joined Mar 13, 2018
3
I've forgotten more about electronics than I remember which is probably why I don't understand why the waveform at the drain of the mosfet isn't square, it actually resembles something charging though why I have no idea.

The output of the 555 is square and a little under 200kHz. If the frequency is dropped by changing the timing cap to, say, 5kHz the waveform at the drain is square as you'd expect.

In the Smartscope attachment the green channel is the gate waveform, the blue is the drain waveform. Can anyone tell me what I've missed?
LabNation_Screenshot0.png 555.jpg
 

wayneh

Joined Sep 9, 2010
16,102
Which MOSFET are you using? You may have too little gate voltage to make a fast switching.
 

RichardO

Joined May 4, 2013
2,273
The FET has a huge amount of capacitance. In all three combinations -- from gate to source, Gate to drain and drain to source. They can add up to thousands of picofarads. These capacitances combine with the 1K resistor to create a long time constant.
 

Thread Starter

Guy Eastwood

Joined Mar 13, 2018
3
Which MOSFET are you using? You may have too little gate voltage to make a fast switching.
It's this chap: https://docs-emea.rs-online.com/webdocs/0d27/0900766b80d2703a.pdf

The FET has a huge amount of capacitance. In all three combinations -- from gate to source, Gate to drain and drain to source. They can add up to thousands of picofarads. These capacitances combine with the 1K resistor to create a long time constant.
Thanks, I did wonder if it were something like that. Will look for strategies to try eliminating the effect if it can be done.
 
Last edited by a moderator:

wayneh

Joined Sep 9, 2010
16,102
Thanks, I did wonder if it were something like that. Will look for strategies to try eliminating the effect if it can be done.
Note that the MOSFET is not a logic level type and is not turned fully on until the gate voltage hits 10V or so. With 5V, its on resistance will be higher than advertised.
 

Thread Starter

Guy Eastwood

Joined Mar 13, 2018
3
Note that the MOSFET is not a logic level type and is not turned fully on until the gate voltage hits 10V or so. With 5V, its on resistance will be higher than advertised.
I'll give it a go at 12v then to see. Perhaps I'm inadvertently operating it in the ohmic region partly.
 

ScottWang

Joined Aug 23, 2012
6,816
1. Use +12V Power.
2. Use +5V Power, use a Vgs 5V logic level MOSFET.
3. Use +5V Power, try the circuit below for ne555, if use cmos type tlc555 then there are no need the D1,D2.

555 switching a mosfet_Guy Eastwood_ScottWang.png
 

ebp

Joined Feb 8, 2018
2,332
There is no problem with the FET being turned on hard enough with the load in use - the waveform clearly shows that. 5 volts is barely adequate for any significant load. You need to use a drive voltage that gets you past the plateau in the gate charge curve. I'd probably aim for about 7 to 8 volts for that particular FET. Any more than that and you are just shoving around useless charge. For fast switching you must use a push-pull driver, and the 555 isn't too bad for that, but you must use very good power supply decoupling for the 555. A resistor in the gate is required to keep the peak current to a reasonable level, otherwise the 555 may overheat. If you were switching a high current load and wanted fast switching it would probably be inadequate and a driver capable of both sourcing and sinking an ampere or so would be required. Yes, a full amp, maybe more. That FET is hugely grossly spectacularly excessive for the task. A 2N7000 would be quite adequate.

The issue is turn-off. The dip in the drain voltage corresponding to the falling edge at the gate is charge being "sucked out" through the drain-gate capacitance. I think Richard is quite correct, it is just that 1k & the FET capacitance is creating a very long time constant. Try halving the load resistance and see what you get. Again, the FET is way too big for this task. It would happily drive a load of half an ohm or less with a 12 V supply (with the caveat of needing higher gate voltage).
 

ebp

Joined Feb 8, 2018
2,332
If you power the 555 with 12 volts, don't be too surprised if you smoke it! You are already being mean to it with the 1k resistor between V+ and the discharge pin, but it should take it. Use at least 10 ohms between the output and the FET gate to keep the current down in the range of 1 A peak. I'm not sure the 555 will survive even then.
 

danadak

Joined Mar 10, 2018
3,569
The mosfet you choose turns on pretty hard at Vgs = 6V, if you want
full ratings at Vgs = 10V.

The boost transistor, pulling Vgs high, solves the turn on problem, but
then you have the turn off problem is a passive RC decay, R5 2.2K in
schematic. You could use a diode from 555 output directly to gate to
aid that as 555 timer can sink a fair amount of current. The diode would
be added as additional component to schematic you show. Ore drop
the value of the 2.2K, although that burns more power by doing so.

I assume you are switching a high current load. To insure MOSFET
switching rapidly to minimize Pdiss you might consider an actual
MOSFET gate driver chip, quite a few variants available. In addition
if switching high current (just the gate C current spikes) grounding very
important in design. To insure other logic, and 555, do not get ground
and supply bounce related issues. Use polymer tantalum for bulk
caps on supply rails. And ceramic disk in parallel.

Regards, Dana.
 
Top