Hi,

I am a beginner HW designer (student at the moment). I have a little circuit involving LM393 comparator and a thyristor that I have to turn under certain conditions...


The idea is that when that condition is met, the LM393 outputs logical one, which should trigger the thyristor (BT169D). I added a simple transistor stage just to make sure the triggering signal for the thyristor gate has enough voltage and current. However, the thyristor turns on right after the supply voltage is applied to the circuit... Have you got any ideas, how could I fix it? Circuit in the attachments.

I made sure my wiring is correct.
The LM393 does not output a high signal right after the power on. LM393 has a default logical zero output state (I verified it with measurements.)

Thank you in advance

 

Show the rest of the circuit before the LM393.

That chip has an open collector NPN output so the output could be high for an instant too short to measure. (with a meter)

You could try placing a small capacitor from the gate of the SCR to ground.

 

Hi,

I am a beginner HW designer (student at the moment). I have a little circuit involving LM393 comparator and a thyristor that I have to turn under certain conditions...


The idea is that when that condition is met, the LM393 outputs logical one, which should trigger the thyristor (BT169D). I added a simple transistor stage just to make sure the triggering signal for the thyristor gate has enough voltage and current. However, the thyristor turns on right after the supply voltage is applied to the circuit... Have you got any ideas, how could I fix it? Circuit in the attachments.

I made sure my wiring is correct.
The LM393 does not output a high signal right after the power on. LM393 has a default logical zero output state (I verified it with measurements.)

Thank you in advance

try disconnecting the scr gate from the transistor and check if the SCR is still latching on when you power-up. I would put an RC (make a 0.1uF cap in series with say 100ohm resistor) across the SCR to avoid latch up during power up.

 

A thyristor is a power device: if you want to make a latching logic circuit, I would recommend using a SR latch (either made out of two transistors, or two NAND gates) rather than a thyristor.

 

Ok, so more on my circuit...

I am designing a current tester for the 10-100 mA range. If the current is lower than 10 mA, blue led turns on to tell about that, if the current reaches the 100 mA, red led turns on (permanently, until the device is powered of) and disconnects the circuit not to damage it. My testing load is nothing more than a high power potentiometer...

P MOSFET should act like a switch, driven by a thyristor, which is triggered by the LM393 comparator (if the current passes threshold level.) The threshold levels are calibrated according to my needs, the wiring is checked many times, blue led and it`s operation works perfectly.

As you can see, since last time, I removed the transistor stage between the comparator and the thyristor. This step got my red (overcurrent) led to shine. However, my blue led did not light up together with my red led (when there is overcurrent, the red led should be light up by the thyristor, and the opened P MOSFET should result in no current flowing through the main branch, thus lighting up the blue (undercurrent) led.) When there is overcurrent condition and the red led shines, I get 2.8V between P MOSFET source and drain (it should be nearly 5 V for a fully open circuit)

I have carefully read the thyristor datasheet and all the ,,threshold" current and voltage parameters. Given resistor values should have a fair margin to trigger both thyristors. Also I have measured around 0.85V drop across closed thyristor, so the P MOSFET gate-source trigger voltage (minimal -3.5 V) is also satisfied. Do you have any ideas why is the MOSFET not fully latching? Thank you in advance

 

Well, it looks like you have the voltage to the MOSFET gate divided up but it's hard for me to read this schematic.

Just as a side...why do you have a divider net at the plus input to the top Op-Amp both ends from ground?

 

Well, it looks like you have the voltage to the MOSFET gate divided up but it's hard for me to read this schematic.

Just as a side...why do you have a divider net at the plus input to the top Op-Amp both ends from ground?

Well, I am not sure if that makes any difference, but I wanted to prevent any noise from the ground affecting my sense voltage, I wanted to make it fully differential, but like I said, it was an advise from one of the more experienced guys (seems to work anyway...)

 

Ok, back to the MOSFET...

What is that 1k resistor connected from the gate to the cathode of the SCR for?

Is that a pull down, try connecting it directly to ground instead.

 

Ok, back to the MOSFET...

What is that 1k resistor connected from the gate to the cathode of the SCR for?

Is that a pull down, try connecting it directly to ground instead.

That resistor is a datasheet recommendation, I dont quite know what it does actually, but it is for some internal semiconductor stability compensation, but I guess I can try to remove it...

 

So not being terribly experienced in using SCRs I did some research and came up with this.



Just for information.

Changing the resistor configuration probably won't make any difference.

One thing I would consider is the possibility that the SCR cannot pull the MOSFET gate close enough to the 5 volt rail to completely turn off the Transistor.

 

Also consider the effect of placing the load on the Cathode instead of the Anode which is the normal placement.

I spent some time trying to get an answer, but with no luck.

I bring this up because placing a load on the emitter of an NPN transistor provides negative feedback, but quite honestly, I don't know how this changes the behavior of an SCR or if at all.

 

Alright, one more update (circuit in the attachments) I changed the mosfet to IRLR8743 (this one has a lower Vgts parameter) I decreased the resistor values to 100 ohms, resulting in stronger currents everywhere and put back the comparator output transistor stage. This time, when the overcurrent occurs, red led turns on, and the mosfet gate is pulled to 3.6 V (maximum threshold from the datasheet is is just under 2.4V, so 3.6V has to be enough). But my mosfet still does not turn off, what is more, when I lower the current, my red led turns off (while it should stay on permanently...) I have also eliminated the thyristor gate-cathode resistors, so the output transistor has no contact with the cathode. This is driving my crazy... How can a thyristor conduct at first, but then turn off, while the anode is still at vdd and the thyristor conducts 20 mA (minimal holding current is just 1 mA). How can a mosfet be still conducting, while gate voltage is 1.2 V above maximum threshold?

 

Well first off, your new MOSFET is N-Channel...that's not good.

Lowering the load resistor value on the SCR won't do you any good...if anything you would want to increase it.

 

You are right, I can not do it with an N MOSFET, but lets put the MOSFET aside for a bit...

The fundamental problem is that I can not turn on the SCR. I tried it both with a gate-cathode resistor (1k) and without.
I assembled a part of my circuit on the breadboard (only the SCR part.) My wiring is perfect, I am getting good contacts everywhere. When I turn on the transistor, I get solid 4.7 V and around 45 mA of gate trigger signal, which lights up the led. But like I said, when I pull my transistor's base low, the led stops shining, as if the SCR was not even there... When the led is on, I am getting 3.7 V at the SCR's cathode and 15 mA of current through the led. I verified all these numbers with my multimeter. I understand that an SCR is a power component, but the one I have is really designed for a delicate operation (look at the trigger parameters in the datasheet, I am orders of magnitude beyond...) I guess I am clearly missing some SCR operation fundamentals, have you got any ideas? Thank you in advance.

 

Here's an off the wall idea...if the SCR has failed it's possible the transistor is turning on the LED.

The transistor is in follower configuration, so the voltage sounds about right.

 

This configuration should work although I wasn't able to actually test the design.
This eliminates the SCR and uses the LM393 as the latch.

 

This configuration should work although I wasn't able to actually test the design.
This eliminates the SCR and uses the LM393 as the latch.
View attachment 288880

Is D1 also 1N4148? Can I use 2N2222 instead of 2N3904 Thank you in advance for this solution, I will try it ASAP

 

2N2222 is fine. I added R4 and another diode D3, all are 1N4148.

 

After looking the circuit over again I removed D1, R2 and changed R1 to a 2.2K.

 

After looking the circuit over again I removed D1, R2 and changed R1 to a 2.2K.
View attachment 288898

Alright, I made exact same replica of this circuit (I put 2.0 kOhm as R1)

The MOSFET was always open, comparator output state did not made any difference in MOSFET gate voltage (MOSFET did not close, when the comparator raised output to logical 1) That is the main problem...

Led had some strange behavior. Right after the power on, the led would start to shine and keep shining until the power is disconnected (no matter the initial comparator state), but when touched my multimeter probe to the comparator's output, the led went off. The led would turn on again only when the comparator's output reaches logical one (then it latches again and keeps shining until the power is disconnected.) So I guess that's the right direction, the only problem is that the led can not be turned on right after the power on (while the comparator is outputting logical 0, which happens to be around 0.11V)

Since the comparator is driving 3 different branches, maybe it can not output enough current, what do you think if I put a output stage transistor with some 100 ohm resistor across 5V? Is it worth trying?