Beginner question : 555 timer pin 4

[RhythMick]

Just for completeness I built the circuit using separate lm293 chips. Works just the same.

During that test I noticed it's behaving as an AND. Only if both thermistors raise their temperature (lower resistance) does it go high and let the 555 flash

 

[(*steve*)]

You are right, and it is operating correctly.

However, you also have your logic reversed, so the correct operation is not what you require :-(

The output will be low if either comparator is pulled low. So for it to go high, BOTH thermistors have to register "hot".

This is called inverted logic, and with it, OR becomes AND.

If you can draw your circuit for me (just how you are using the comparator for one channel is sufficient) then I will be able to suggest something.

 

[RhythMick]

Thanks Steve, however I will draw the circuit tomorrow. Sitting in bed playing with breadboard, holding thermistors between my fingers to make lights flash has drawn enough strange looks from SWMBO for tonight.

Yes I think I have my logic twisted. Starting again...

I want 555/pin4 to go HIGH when EITHER of the thermistors registers a temperature above it's threshold. Truth table (0 = LOW)...

O1 O2 P4
0 0 0
0 1 1
1 0 1
1 1 1

A bit of googling and it seems the answer is as simple as putting each output (from it's pullup resistor) through a diode, join them at a 10k to ground and take pin 4 from the top of that resistor.

I'll try that in the morning - if I go rummaging for diodes now I'll be in trouble.

Thanks for prompting me to revisit the logic.

 

[RhythMick]

Just to finish this one off, combining the inputs (each with its own pull-up resistor) through a pair of diodes to a pull-down resistor works perfectly. The result is the OR gate I wanted.

 

[(*steve*)]

That is pretty much what I was going to suggest. Well done

 

[RhythMick]

Many thanks Steve. More than half of the fun with this is working out the answer. Sometimes I don't know the right terminology to ask the right questions, but getting there. Thanks for your help.

 

[RhythMick]

This is the circuit I've ended up with. I haven't breadboarded this exact one yet, will do that later and confirm. Comments welcome.

 

[Audioguru]

An LM317 must have an input capacitor and an output capacitor, you have none. An LM317 should have a resistor between the output and its ADJ pin of 120 ohms but yours are much higher and variable.

The output of an LM293 comparator can drive to ground but is disconnected when it goes "high". It needs a pullup resistor from its output to the positive supply voltage to make its output go high.

 

[(*steve*)]

You don't need the voltage regulators. You can just compare the voltage you let from a potential divider using a thermistor with that obtained from a trimpot.

In both cases you are comparing the ratio between a voltage and the supply rail. The comparison remains valid even if the supply rail changes.

 

[RhythMick]

Thanks for your comments.

The pull-ups for the comparator outputs are 10ks to the 9v rail. I've shown then together on the right by the WL/WR connection points. They join there through the diodes to the 10k pull down which forms the OR gate. It's drawn that way for convenience is all.

The LM317s have 3500uF immediately before them. The datasheet says an input is only needed if the filler cap is remote. The top 2 317s have 47uF after them ... placed remotely next to the fans - do they also need a cap next to the 317?

The lower 317 output cap I am missing thanks.

 

[RhythMick]

You don't need the voltage regulators. You can just compare the voltage you let from a potential divider using a thermistor with that obtained from a trimpot.

In both cases you are comparing the ratio between a voltage and the supply rail. The comparison remains valid even if the supply rail changes.

Thanks Steve. The first 2 LM317s are being used to vary the voltage to the fans based on the thermistor resistance.

Interesting that I may not need the 3rd, I'll experiment. Thanks.

 

[RhythMick]

An LM317 must have an input capacitor and an output capacitor, you have none. An LM317 should have a resistor between the output and its ADJ pin of 120 ohms but yours are much higher and variable.

The output of an LM293 comparator can drive to ground but is disconnected when it goes "high". It needs a pullup resistor from its output to the positive supply voltage to make its output go high.

Oh and the regs are LM317L for which the datasheet says minimum current 2.5mA. the fixed resistor is chosen to provide 2.6mA. the variable resistors (thermistors) are valid there because the fans draw more than the minimum current.

 

[(*steve*)]

Thanks Steve. The first 2 LM317s are being used to vary the voltage to the fans based on the thermistor resistance.

Aaaagh. If you're doing that, it may pay to place a 120Ω resistor in series with a 100uF capacitor in parallel with the thermistor. That will ensure that the fan starts at full speed. It may help if you find the fan doesn't start at all on a very cold day.

 

[RhythMick]

Aaaagh. If you're doing that, it may pay to place a 120Ω resistor in series with a 100uF capacitor in parallel with the thermistor. That will ensure that the fan starts at full speed. It may help if you find the fan doesn't start at all on a very cold day.

That's a good thought Steve, though I didn't get your description of what's needed? I think it's in parallel with the thermistor is a 120R in series with a 100uF? So if it's cold the thermistor resistance will be high. The 120R will prevent the resistance rising so high that it takes the output voltage too low.

At 25C the thermistor will be 2k2, but the 120R in parallel will dominate, in fact it will dominate the thermistor at all temperatures. Maybe the idea is right but with a different value. I'll have to play I think. If the temperature is that low I think I'm happy with the fans not starting. The copper sheet and heatsinks should be plenty to cope anyway - I'm just looking for the fans to kick in if the temp rises.

 

[(*steve*)]

The capacitor in series with the resistor means that the 120Ω resistor only dominates until the capacitor charges.

 

[Audioguru]

Your schematic is a fuzzy Jpeg file type with its tiny parts so far apart that details like the circles with dots inside (diodes?) are not seen. A PNG file type is recommended since it makes extremely clear schematics.

When you use LM317L ICs then the schematic should show it since the "L" has a different datasheet from a normal LM317.

Why do you need diodes to make an OR gate? The comparators have open collectors so that they make an OR gate.

 

[RhythMick]

Apologies for the poor format I'll choose PNG next time.

My understanding (always happy to learn) is...

Each output needs a pull-up resistor. When the the temperature is below threshold the output will be low, but the pull-up ensures that when temperature goes up the output goes high. If either output is high then pin 4 will go high. Without the diodes that's just a voltage divider. Breadboarded this and the circuit works.

 

[(*steve*)]

Why do you need diodes to make an OR gate? The comparators have open collectors so that they make an OR gate.

Perhaps you should read the thread from the beginning.

 

[RhythMick]

The capacitor in series with the resistor means that the 120Ω resistor only dominates until the capacitor charges.

I see thanks Steve. Once charged up the capacitor blocks DC. Nice. I'm not yet used to thinking about it in that way. Very useful insight.

 

[RhythMick]

And talk about being a newbie... I just discovered stripboard ! I've been drilling 1mm holes in epoxy board and failing to line up the holes properly. Sigh... Feel like a right numpty...