Warning LED when valve stops conducting - a little SS help needed

[Nick]

Transistors never fully turn off. I would consider the resistor on the same side as the LED.

Ok thanks

And consider taking the collector of the PNP to the emitter of the NPN.

[RhythMick]

Transistors never fully turn off. I would consider the resistor on the same side as the LED.

Ok thanks

And consider taking the collector of the PNP to the emitter of the NPN.

will do - couple of test circuits needed I think

[Andrew]

Transistors never fully turn off. I would consider the resistor on the same side as the LED.

Thanks Nick, (and Mick for drawing it up) that's pretty much what I was driving at, just didn't have the time to follow it thru' the a logical conclusion .

[Mike H]

At risk of rudely elbowing into the design think-tanking process, I still like the LM193 comparator idea, so I quickly threw this together.
-
-

[RhythMick]

At risk of rudely elbowing into the design think-tanking process, I still like the LM193 comparator idea, so I quickly threw this together.
-
-

Heater LED Warning.jpg

Thanks Mike - puts my crude efforts to shame

[RhythMick]

At risk of rudely elbowing into the design think-tanking process, I still like the LM193 comparator idea, so I quickly threw this together.
-
-

Heater LED Warning.jpg

If I understand it right, that circuit will light the LED when the valve is conducting and stop it when not. Simply reverse the IN+ and IN- to get the reverse effect ?


EDIT - IGNORE THIS NEXT. I realise you've drawn the sense resistor low side of the valve. I'll get me coat.


Also, if the valve heater fails the resistor voltage (both sides) would rise to the set output. Input common-mode from the datasheet is Vcc -2.0v, so the supply would need to be at least 2.0v higher than the maximum heater voltage to be measured. In my case thats 5v, so the supply should be set to, say 8v. Make sense ?

8.2.2.1 Input Voltage Range
When choosing the input voltage range, the input common mode voltage range (VICR) must be taken in to
account. If temperature operation is below 25°C the VICR can range from 0 V to VCC– 2.0 V. This limits the input
voltage range to as high as VCC– 2.0 V and as low as 0 V. Operation outside of this range can yield incorrect
comparisons.

The following is a list of input voltage situation and their outcomes:
1. When both IN- and IN+ are both within the common-mode range:
a. If IN- is higher than IN+ and the offset voltage, the output is low and the output transistor is sinking
current
b. If IN- is lower than IN+ and the offset voltage, the output is high impedance and the output transistor is
not conducting
2. When IN- is higher than common-mode and IN+ is within common-mode, the output is low and the output
transistor is sinking current
3. When IN+ is higher than common-mode and IN- is within common-mode, the output is high impedance and
the output transistor is not conducting
4. When IN- and IN+ are both higher than common-mode, the output is low and the output transistor is sinking
current

[RhythMick]

I just realised why NONE of these strategies will work for my application.

I'm using virtual cathode resistors. If the valve heater fails the DC board will adjust so the set voltage so appears across the virtual cathode.

Dang.

[RhythMick]

I just realised why NONE of these strategies will work for my application.

I'm using virtual cathode resistors. If the valve heater fails the DC board will adjust so the set voltage so appears across the virtual cathode.

Dang.

Nope, got my own thinking muddled there. The board isn't sensing the voltage across the cathode pins it's sensing the voltage across the sense resistor near the outputs. So when the valve heater fails the DC board will not adjust the output voltage, it will try to put out the same current it was doing. The difference is that the current will run through a much higher resistance, about 400R. I don't know what it would be able to output, probably a maximum of around 7v across 400R so 18mA. Across the 0.12R that's 2mV.

The D44H11 solution won't work I don't think, but the comparator will.

[RhythMick]

Clearly I'm talking broken biscuits and proving how little I know about how these boards work lol. Or maybe I just shouldn't type when I'm tired.

I put a 273R load on the output. The board hit 5v as it is set to, putting out 18mA. The D44H11 pin 1 method won't work though as it stayed at 5v, as expected.

Regardless of my inaccuracies, I think with virtual cathodes in place the comparator method seems the way to go.

[RhythMick]

Building on Mike's idea, I'm thinking of using the dual comparator to measure across the 10R resistors I already have in the cathodes. Would require a 5v supply grounded to the top of the shared cathode.

[Nick]

These sort of isolated supplies can be great for providing a floating supply from a existing DC supply.

https://uk.rs-online.com/web/p/isolated ... s/3965174/

[RhythMick]

These sort of isolated supplies can be great for providing a floating supply from a existing DC supply.

https://uk.rs-online.com/web/p/isolated ... s/3965174/

Thanks Nick

[Mike H]

If I understand it right, that circuit will light the LED when the valve is conducting and stop it when not. Simply reverse the IN+ and IN- to get the reverse effect ?

You might have already sussed, but no, other way round. If + input is higher than - input, then output is off (being a transistor switch), so LED is off. I.e. output is positive (aka, 'high').

[RhythMick]

If I understand it right, that circuit will light the LED when the valve is conducting and stop it when not. Simply reverse the IN+ and IN- to get the reverse effect ?

You might have already sussed, but no, other way round. If + input is higher than - input, then output is off (being a transistor switch), so LED is off. I.e. output is positive (aka, 'high').

Thanks Mike - too many head flips. I'll wrap around that bit later. Easy to reverse if it's wrong

[Mike H]

Building on Mike's idea, I'm thinking of using the dual comparator to measure across the 10R resistors I already have in the cathodes. Would require a 5v supply grounded to the top of the shared cathode.
1550098328019861.jpg

Sorry I forgot these are DHT's. (Heater not on a separate circuit.)

These sort of isolated supplies can be great for providing a floating supply from a existing DC supply.

https://uk.rs-online.com/web/p/isolated ... s/3965174/

Oooo aahhh.

Yes the sense - indicator circuit will need its own lickle PSU. Getting complicated!


How are the filaments heated?