Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

MOSFET circuit question

Status
Not open for further replies.
when noise occurs Q1 in is 1 Q1c is low the Q2D out is 1 then Q3-d out is 0 and LEd is On and. same ime Q3 out =0 drives Q4d =1 so if relay is between RELAY and 0V(GND) it is ON at the same time, but if between RELAY and +V it is inverted , assuming correct coil polarity in each case. so the schematic is correct.

If put in parallel with another sensor driver, the LED is ON externally driven.
...same ime Q3 out =0 drives Q4d =1...... Q4 gate is connected to Q2 source which is HIGH during detection... meaning Q4 is not conducting (p-channel high gate = not conducting), so Q4 drain is low.

Is this wrong? Where?

And because Q4 is p-channel I think the relay must be connected between Q4 drain and ground.
 
Q2&3 are NOT enhancement mode so Vgs is 0 for ON , Q4 is enhancement mode so Vgs is -ve for ON,


yet both are inverters. very rookie design...

Yes Relay MUSt be between Out and GND
 
Q2&3 are NOT enhancement mode so Vgs is 0 for ON , Q4 is enhancement mode so Vgs is -ve for ON,


yet both are inverters. very rookie design...

Yes Relay MUSt be between Out and GND
Q2&3 are NOT enhancement mode so Vgs is 0 for ON... both Q2 and Q3 are n-channel: when Q1 is conducting Q2 gate is low and hence Q2 is not conducting; hence Q4 gate is high hence Q4 (p-channel) not conducting. Hence relay is not powered. Right?
 
no
 
Update to post #8: also eliminate R3 and D4. What is the value of V+?

ak
I would like V+ to be between 5Vdc and 15Vdc, but once at that voltage there would be no variation (for example if I connect it to 12V then V+ remains there). If that is not sustainable then between 12Vdc and 15Vdc.
 
Q2&3 are NOT enhancement mode so Vgs is 0 for ON , Q4 is enhancement mode so Vgs is -ve for ON,
yet both are inverters. very rookie design...
Yes Relay MUSt be between Out and GND
All MOSFETs in the design are enhancement mode.
 
12 V to 15 V is more than enough to turn on Q4 fully. 5 V might not be enough, depending on the part. Its datasheet should have a plot of Rdson versus Vgs.

ak
Right, I will check these parameters: the reason I am posting this here is to get a better understanding of its working because I will be making a pcb using non-smd parts for the MOSFET(s). Q4 BSS84 can not supply enough current for the relays I intend to use.
Therefor too I might very well use 15Vdc for V+
 
Frankly I still do not understand the explanation of Tony Stewart.
I don't believe it was valid
I hope to get an answer on my question in post 23 and 27.
I wouldn't worry about it.

Right, I will check these parameters: the reason I am posting this here is to get a better understanding of its working because I will be making a pcb using non-smd parts for the MOSFET(s). Q4 BSS84 can not supply enough current for the relays I intend to use.
Therefor too I might very well use 15Vdc for V+
If you're making your own circuit, then you can put in any parts that you want, and you're not limited by that schematic you posted. So what are the actual constraints of your design? 15V for Vin? What is the current required for the relay?
 
If you're making your own circuit, then you can put in any parts that you want, and you're not limited by that schematic you posted. So what are the actual constraints of your design? 15V for Vin? What is the current required for the relay?
You are very right!
My initial approach is flawed, next time when I start a design I will not unthinkingly adopt someone elses schematic.
The actual design constraints are
1. V+ (Vcc) to be 15Vdc (with if possible variance between 12Vdc and very maximum 18Vdc, with no varaiation in supply voltage once a pcb is connected)
2. relays of 12Vdc to 15Vdc (I would us a V+ adapted to the relay to be used at the time of connection), EDIT: less then 1A continuous load
3. input signals are 20kHz pwm for detection to be active
 
Q4 BSS84 can not supply enough current for the relays I intend to use. Therefor too I might very well use 15Vdc for V+
Increasing the operating voltage will not increase the current capability of the output transistor. First, select the relay. Determine the peak voltage and current required by the relay coil, double both numbers, and select a transistor with at least those parameters.

There are many more n-channel than p-channel parts available, and the n's have better performance per cost. Because the input is transformer isolated, you can turn the entire circuit "upside down", with a 2N2907 or 3906 or 4403 input transistor and an n-channel relay driver.

What is your expected relay coil current?

ak
 
Here is my hack:
46a.png
Input comes from 1A 10% PWM on one wire going through donut on CT. Sensitivity is adjusted by R1, the CT burden. I simplified the current limiting to the base of Q1. I played a bit with the time constant R3,R4 C1. M1 is almost any modern PFET. The relay is the standard automotive sugar-cube like made by Bosch. D6 is a LED. D4 is any Si rectifier used as a snubber.

Note the time constants, especially due to the inductance of the relay coil (see the yellow trace).
 
Increasing the operating voltage will not increase the current capability of the output transistor.
You are right but I worded it incorrectly: I intend to use another MOSFET than the BSS84 because it does not allow high enough current. And I would choose 12 to 15Vdc because that is the type of relay I would like to choose. Those two issues are separate.

First, select the relay. Determine the peak voltage and current required by the relay coil, double both numbers, and select a transistor with at least those parameters.
There are many more n-channel than p-channel parts available, and the n's have better performance per cost. Because the input is transformer isolated, you can turn the entire circuit "upside down", with a 2N2907 or 3906 or 4403 input transistor and an n-channel relay driver.
Interesting approach. I will adopt your idea and design a little circuit with that intent.

What is your expected relay coil current?
ak
100mA I guess, but I might drive other loads as well, yet to be decided.

Thank you for your great ideas,
Erik
 
Here is my hack:
View attachment 106143
Input comes from 1A 10% PWM on one wire going through donut on CT. Sensitivity is adjusted by R1, the CT burden. I simplified the current limiting to the base of Q1. I played a bit with the time constant R3,R4 C1. M1 is almost any modern PFET. The relay is the standard automotive sugar-cube like made by Bosch. D6 is a LED. D4 is any Si rectifier used as a snubber.

Note the time constants, especially due to the inductance of the relay coil (see the yellow trace).
This is a very interesting post, I will take my time to study it,

thks,
Erik
 
12 V to 15 V is more than enough to turn on Q4 fully. 5 V might not be enough, depending on the part. Its datasheet should have a plot of Rdson versus Vgs.

ak
The graphs on a datasheet are for a "typical" one that might be much better than the one that is used. The printed 10V gate voltage should be used so that any Mosfet that meets its worst case ratings will work properly.
 
D2 and D3 form a 1.2 V clipper, sorta kinda like a zener. If there is a possibility of a large input transient, the diodes clip the max voltage to limit the peak current into the Q1 base. A 2N2222 easily can handle a 10 mA base current, requiring a 10.6 V input. If you think the transformer secondary will exceed that, put the diodes back in.

ak
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top