Latching comparator help

[MrHeckles]

Hi guys,

I have built the following circuit.

The Vref is 0.01v. The Vin ranges from 0 to maybe 5v at most. What I was intending was a comparator that latched on once the Vin was less than Vref (and in which case the LED would illuminate).

The trouble I have is that when it turns on, the voltage on the + input of the 741 only increases to 0.3v (from 0.01v initial reference). If Vin goes above this, it turns off again.

If I reduce the 380 ohm resistor in value, the LED dims. If I increase the value, the latched + input voltage drops even more.

If I remove the diode on the feedback loop, I cannot set the Vref properly.

One point to note, is that the switch is in there as a reset.

My question is how could this layout be changed to give a greater latched on voltage (so it wont turn off) and also why is the feedback resistor affecting the LED brightness?

Thanks in advance.

 

[audioguru2]

Hi Mr. Heckles,
Try these bananas:

View attachment 38274

 

[indulis]

I would move the LED and current limiting resistor to the collector and add a base resistor to the transistor. Also, while the output is at the minus rail, the output clamp diode is making the poor 741 work really hard!!!

 

[MrHeckles]

I put the output clamp diode on to protect the transistor from having the base sitting at -9v. Does it not need this?

 

[MrHeckles]

Well, I have changed the circuit as per audioguru's diagram.

Now it holds a nice +7.5v or so once turned on.

Some more background. Vin is a feed from a sense resistor from one of the legs of a motor H-bridge, which goes through an absolute value amplifier before getting to the negative input of the 741 comparator.

I find that with this new configuration, the comparator works fine when the motor is stalled, but as soon as it turns freely the comparator turns on, despite the fact that Vin is well above Vref.

I presume that it is due to some voltage spiking somewhere, but how can I stop it?

 

[audioguru2]

Hi Mr. Heckles,
A series resistor then a capacitor to ground would be a spike filter. It would delay an over-current sense.

 

[MrHeckles]

I have put an RC filter in, and it seems to make it work fairly nicely.

Two questions.

1. Is there an upper limit on values for the resistor and/or the capacitor in the RC filter?

2. Will NOT having the clamp diode result in the -ve output of the comparator damaging the base of the transistor?

 

[audioguru2]

1. Is there an upper limit on values for the resistor and/or the capacitor in the RC filter?

A high resistor value will cause an input offset voltage error due to the opamp's input bias current.
A high amount of filtering will cause a delay.

2. Will NOT having the clamp diode result in the -ve output of the comparator damaging the base of the transistor?

The base of the transistor would be damaged by too much reverse voltage if its emitter had a capacitor or something to hold a high voltage.
The 741 opamp is pretty slow. It would take time for its output to slew from -ve up to its operating voltage.
Use the diode.

 

[MrHeckles]

Righto, clamp diode stays in.

I am now having issues with the use of the transmission gate.

The PWM signal for the motor is fed into pin 8 and then from pin 9 to the half bridge driver. Pin 6 is the control pin.

When the comparator is on, pin 6 has a bit over 8v and everything is cool. The PWM pulse goes into the bilateral switch and comes out, and the motor works fine.

However, when the comparator is off, pin 6 is low (-0.7v). The PWM pulse goes into pin 8, and then comes out from pin 9 slightly reduced. The motor will still operate, but at a slightly reduced rate.

I was thinking that the bilateral switch should have stopped the PWM pulse completely when the control pin was low.

Why is it still allowing the signal though?

 

[audioguru2]

You don't have a resistor to ground at pin 6 of the transmission gate, so it never gets a logic low.

 

[MrHeckles]

I have tried that now, but even with a direct connection to ground on pin 6, the bilateral switch is still letting through the signal.

I am most confused.

 

[audioguru2]

Your transmission gate doesn't turn off.
Maybe its pin 6 is blown if you tried the opamp without the clamp diode.
Maybe it is blown by having an input or output voltage that exceeds its supply voltages.
What is the voltage swing of its input signal?
What is the DC voltage of its load?
How much current is it switching through it?

 

[MrHeckles]

Aha, got it.

I was running the transmission gate off the +9v supply. The PWM signal is between 0 and 12v. All fixed, the gate is now running off the 12v supply and shuts it all off nicely.