@Papabravo
So in this post I was told that the IRF 4905 is not is it not a very good logic level P channel mosfet.
https://forum.allaboutcircuits.com/threads/555-voltage-booster.187006/page-5#post-1748651
so how about the FQP27P06, is it any better?

 

so how about the FQP27P06, is it any better?

Not really.

It's easy to determine if it is a logic-level device.
Just look at the Vgs(th) voltage, and the Vgs used to determine the ON-resistance.
The max Vgs(th) is usually 2V or less, and the Vgs to test the ON-resistance is 5V or less.

Below is an example of one:

 

@Papabravo
So in this post I was told that the IRF 4905 is not is it not a very good logic level P channel mosfet.
https://forum.allaboutcircuits.com/threads/555-voltage-booster.187006/page-5#post-1748651
so how about the FQP27P06, is it any better?

My original assessment was based on the datasheet which showed Vgs(th) between -2V to -4V and rds(on) measured at Vgs = -10V, which would not be possible in a 5V system. Go back and check the edit to that post where a simulation with a model of unverified connection with reality shows that it might be OK.

https://forum.allaboutcircuits.com/posts/1748892/

 

It's rare that one would actually need a logic-level P-channel MOSFET, as it would imply that a large load connected between 5V and 0V was being switched.
A P-channel MOSFET is most often used to switch a load on a higher voltage supply, in which case it would not be a problem to use a standard MOSFET.
For instance, FQP27P06 can switch 20A - when would one need to switch a 20A load on the 5V supply?

 

It's rare that one would actually need a logic-level P-channel MOSFET, as it would imply that a large load connected between 5V and 0V was being switched.

It really depends upon the load current being switched, when the worst-case Vgs is no more that 1V above the maximum Vgs(th), (which is typically 4V for a standard MOSFET).

 

My biggest concern is the on ohmage with a gate voltage of five volts.

 

My biggest concern is the on ohmage with a gate voltage of five volts.

For the PMOS device the gate voltage is negative with respect to the source. When the gate voltage is at ground, it is at -5V with respect to the source.

 

I think I understand, if the push button is pressed the LED will light.

​

 

I think I understand, if the push button is pressed the LED will light.

View attachment 271952​

Yes, that is if there is enough current from the +5V source to actually light the LED.

 

Most 5 Volt power sources are in excess of 1 amp. I calculate 2.9ma.Not too much current I should think.

 

Most 5 Volt power sources are in excess of 1 amp. I calculate 2.9ma.Not too much current I should think.

The power supply is NOT the issue. The forward voltage drop of the LED and the 1K resistor are the issue, because 2.9 ma will hardly allow a standard LED to glow brightly. You'd need a high efficiency LED to do that.

 

I have ZVP2016 logic level MOSFET which works pretty good.

Gate-Source Threshold Voltage VGS(th): -1.5 to -3.5 V, | ID=-1mA, VDS= VGS
Static Drain-Source On-State Resistance (1) RDS(on) 5 Ω | VGS=-10V, ID=-500m

 

I have ZVP2016 logic level MOSFET which works pretty good.

Gate-Source Threshold Voltage VGS(th): -1.5 to -3.5 V, | ID=-1mA, VDS= VGS
Static Drain-Source On-State Resistance (1) RDS(on) 5 Ω | VGS=-10V, ID=-500m

That is in the same ballpark as the BSS84 which did not perform particularly well in the simulation. It had an imputed rds(on) of 6.25Ω and dropped almost volt driving a 200 ma load.

 

That is in the same ballpark as the BSS84 which did not perform particularly well in the simulation. It had an imputed rds(on) of 6.25Ω and dropped almost volt driving a 200 ma load.

Link to this simulation? I've been using them for basic stuff so the high resistance has been negligible. The N channel is a more acceptable 2Ω

 

Yes, that is if there is enough current from the +5V source to actually light the LED.

weren't you questioning whether the five volts had enough current to drive an LED?

The power supply is NOT the issue. The forward voltage drop of the LED and the 1K resistor are the issue, because 2.9 ma will hardly allow a standard LED to glow brightly. You'd need a high efficiency LED to do that.

current to the LED is not important it is only an indicator nothing more. My experience with LED suggests that one milliamp would light it. If that's all I wanted to do which is frequently enough.

 

weren't you questioning whether the five volts had enough current to drive an LED?

NO! A powerful supply cannot force more current into a 1K resistor than Ohms law will allow.

current to the LED is not important it is only an indicator nothing more. My experience with LED suggests that one milliamp would light it. If that's all I wanted to do which is frequently enough.

That approach doesn't work too well for products that need to be produced in volume; for breadboards you can do whatever you want.

 

The threshold voltage of a Mosfet is when it is almost turned OFF. Don't you want to feed the gate-source with a voltage that will turn all of them (the Vgs is a wide range of voltages up to 10V) ON?

 

So if I wanted to test one out I would use this scheme and then measure the voltage on the drain to calculate the IDS resistance?

​

 

Your 1k resistor is not in the Mosfet circuit.
5V/100 ohms is a current of only 50mA. If the current is higher then a Mosfet that needs 10V of gate to source voltage to conduct well will work poorly.

 

So if I wanted to test one out I would use this scheme and then measure the voltage on the drain to calculate the IDS resistance?

View attachment 272044​

Yes. NOTE: The drain current is equal to the current through the 100Ω resistor, which is Vd/100. So (5-Vd)/(Vd/100) will give the resistance of the channel.