Thank you I will do that

 

I replaced the trim pot with a 10K as instructed.

I tested between The trim pot and pin one on the TL431. The resistance does change from 0 - 10K

I measured In circuit the resistance of the 100K resistor R8 the value measured was 94.3 K

I then applied power and adjusted the trim pot nothing happened in either direction. I then measured the voltage between gnd and pin 1 of the TL431 even adjusting the trim pot it remained at 2.4 volts

I have enclosed the drawing I copied from you and then auto routed the pcb

Not sure what else to check

 

Is the pinout correct?

 

View attachment 81875

Is the pinout correct?

I have checked it against the pin out you sent it looks correct. Did you see the PCB layout

 

Don't you ever breadboard circuits before committing to a PCB? I always do.


I cant even tell if I am looking at the top or bottom of the PCB.

 

not often but I think possibly now is the time to start.

 

Sorry

Don't you ever breadboard circuits before committing to a PCB? I always do.


I cant even tell if I am looking at the top or bottom of the PCB.

Sorry PCB layout is looking at the component side

 

Don't you ever breadboard circuits before committing to a PCB? I always do.


I cant even tell if I am looking at the top or bottom of the PCB.

Mike I did the following tests.

Adjusting the trim

Voltage on the junction of the 100K and the pre-set Max 3.4V Min 2.45V

Voltage on the junction 33K and Trim

Max 2.51 Min 1.87

Voltage on the Pin 1 of The TL431 remained on 2.51 regardless of how I turned the trim.

The incoming Supply voltage was set at 12.25Volts.

I turned the TL431 around and got exactly the same results.

I removed the TL431 from the circuit, the test light did not come on regardless of moving the trim.

 

Mike I did the following tests.

Adjusting the trim

Voltage on the junction of the 100K and the pre-set Max 3.4V Min 2.45V

Voltage on the junction 33K and Trim

Max 2.51 Min 1.87

Voltage on the Pin 1 of The TL431 remained on 2.51 regardless of how I turned the trim.

...

This makes no sense.
You do have all three terminals of the pot hooked up?
Wiper goes to REF input of TL431?
Top goes to the 100K, Bot goes to 26K.
Hysteresis resistor (my R9) goes to wiper/REF and is 2.2megΩ ?

The thing that needs to move when you adjust the pot is the Wiper/REF voltage. The Top and Bot Pot terminal voltage will remains constant unless Vin is changed.

 

This makes no sense. You do have all three terminals of the pot hooked up? Wiper goes to REF input of TL431. Top goes to the 100K, Bot goes to 26K. Hysteresis resistor goes to wiper/REF?

The thing that needs to move when you adjust the pot is the Wiper/REF voltage. The Top and Bot voltage will remains constant.

I have just checked again.

The 100K does go to the top and 33K at the bottom.

With the TL431 out of the circuit I have moved the pot the voltage varies from 2.93 to 3.82 volts.

The top and Bottom voltages Are now constant

I checked and the exact writing is TL431-A

 

...

I turned the TL431 around and got exactly the same results.

I removed the TL431 from the circuit, the test light did not come on regardless of moving the trim.

Does that mean that the TL431 was in backwards? If so, it may be compromised.

With the TL431 removed, the PFET's gate is tied to its own source, so is turned off, so the light would be off.

I remember you saying that the initially, the light came on even at low voltages. That would be consistent with having the TL431 anode and cathode swapped. I'm guessing that you have a damaged TL431

 

Does that mean that the TL431 was in backwards? If so, it may be compromised.

With the TL431 removed, the PFET's gate is tied to its own source, so is turned off, so the light would be off.

I remember you saying that the initially, the light came on even at low voltages. That would be consistent with having the TL431 anode and cathode swapped. I'm guessing that you have a damaged TL431

The Anode is Pin 2 and the cathode is Pin 3, this is the way I had it connected originally. I have now replaced it and Have exactly the same results. The only thing I have not changed is the fet?

For the load I am using a 23 Watt Lamp connected the drain and Gnd, so now I am really out of my depth.

 

I am going to be away from this for a couple of days...

 

I have got to the point now where my knowledge ends. (for now) I have yje circuit working right up to the mosfet. There is always 2 volts on the gate until the voltage drops to 10.6 volts. Then the gate goes to 9 volts. However during this time the Light switched by the mosfet stays on.

I have connected a 23 watt load between the source and negative

 

What is the threshold voltage of the PFET?

 

What is the threshold voltage of the PFET?

this is all very new to me PFETS I mean. I hope I have understood the Datasheet. VGS +- 20V

Vgs(th) min -2 max - 4 volts

 

I have got to the point now where my knowledge ends. (for now) I have yje circuit working right up to the mosfet. There is always 2 volts on the gate until the voltage drops to 10.6 volts. Then the gate goes to 9 volts. However during this time the Light switched by the mosfet stays on.

I have connected a 23 watt load between the source and negative

I had company here all day, today...

You may have a bad PFET or have it wired wrong!



I reran the simulation, this time showing the voltage at the gate of the PFET (same as the cathode of the TL431) as the green trace Vgate. Follow the green trace as the input voltage Vin goes from 13V down to 10V. With the pot centered, the switch point occurs at ~11.3V (look along the plot x-axis).

When Vin>11.3V (left side of the plot pane), Vgate is low (about 2V like you reported), so that the PFET is turned on hard. Note that even at a simulated 12A of load current, Vout=Vin-few mV. Look at the Red and Blue traces to see that.

As Vin gets below ~11.35V, the TL431 begins to turn off, and Vgate begins to rise. Suddenly, because of the positive feedback through R9, the TL431 snaps off the rest of the way, causing the green trace Vgate to snap high, turning off the PFET, causing the blue trace Vout to snap off. You said that Vgate only goes up to ~9V, but the simulation shows the TL431 turning off completely with Vgate pulling all the way to Vin?

On the right side of the plot pane, where Vin<11.3V, with the TL431 turned off, Vgate=Vin and Vout = zero because the PFET turns off, but yours doesn't. Could you have the PFET in upsidedown? Note that the source connects to Vin, and the drain connects to the load Vout.

Take the advice of your frog: Never Ever give up


Ps, if you erroneously swap source and drain on the PFET, it will always conduct through the body diode (cannot shut it off) , which is what your build has been doing.

 

Hi Mike
Thanks for this information, I am just trying to understand something as I think I know where I am going wrong. I have one MOSFET left so don't want to make a mistake.

On the drawing the 2M2 resistor is it connected connected to the source. And the drain is it connected to positive input.

I think this is where my confusion comes in. Like you suggested I have swopped the drain and source. Looking at the above diagram is the out being shown from the source, I am totally confused with these symbols. Sorry

 

...On the drawing the 2M2 resistor is it connected to the source. ...

Should be connected to the drain, see my schematic.



In normal usage of an NFET or PFET, the little arrow, which represents the body-diode inside the FET, should never conduct.

 

Hi Mike

Good news, it is now working perfectly!!! I did have the Source and Drain swopped. I have learnt so much from you and all the help I am really grateful.

I have not had a chance to try to test this but have seen it on an inverter system. When the battery goes low the inverter switches out then the load on the battery is off so the voltage goes back on and the Inverter switches back on.

Will this happen with this circuit.

Than you again

Best wishes

Rodney