Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Overheating detection and shutdown

Status
Not open for further replies.

SnaRf

New Member
Hey all

I am pretty new to electronics, and brand new to the forums.
My highest electronics knowledge is from high school, which is pretty basic..

I am about to embark on building this.
(Circuit attached to this post)
I understand what the whole circuit is supposed to do, but I am having trouble reading it.

I can see that the LM339 comparator has V+ on pin 3 and is grounded on pin 12.
It compares the voltage over pins 4 and 5... I think.

The relay has V+ on pins 5 & 6, and apparently on pin 3 as well (Still looking for a pinout)
(so pin 3 will be used to power the LED's)
When the circuit is powered, pins 7 & 8 will be bridged and pins 3 & 4 and the green LED
will be powered
Thus, R5 would be the resistor to stop the LED's from popping...

So, as the temperature picks up, the resistance in TC1 decreases, allowing more
current to flow through it. The goal of VR1 would be to "bleed" some of the current back
to ground(I think). In doing so, it allows you to customize the amount of current reaching
pin 5 on the LM339. Thus customizing your "high" temperature.

Pin 2 from the LM339 will send a signal to the transistor.
(I am guessing it would be wired up to the collector of the transistor.)
The transistor will then pass the signal on to the relay (pin 1) which will break the
contact between pins 7 & 8 and pins 3 & 4, and make contact between pins 2 & 3
and pins 8 & 9 (pin 9 isn't used though..)
Thus, the green LED will be turned off and the red LED will be turned on.
and the contact will be broken for the "power on signal" wire.

It will then stay in this condition, until the "push to make" switch is pressed to reset the relay,
which will then again break contact between pins 2 & 3 and pins 8 & 9, and make contact
between pins 3 & 4 and pins 7 & 8... right?

What I don't get is the function of R1, R2, R3, R4, Q1, D1 and D2
 

Attachments

  • schematic.gif
    schematic.gif
    12.4 KB · Views: 1,074

Roff

Well-Known Member
See the edited schematic below, with the basic voltage divider equation below. The comparator compares the voltage on pin 5 with the voltage on pin 4. If pin 5 is higher than pin 4, the comparator output goes high, turning on Q1, which activates the relay. Note that the thermistor and the variable resistor form another voltage divider. You adjust the variable resistor until pin 5 is at 2.5 volts when the temperature is at the desired trip point. R1 provides a little hysteresis so the output "snaps", avoiding relay chatter at the trip point.

D1 and D2 protect Q1 from the high voltage spike that would otherwise occur when the current through the relay coil is abruptly interrupted.
 

Attachments

  • schematic_977.gif
    schematic_977.gif
    15.7 KB · Views: 1,005

ChrisP

Member
Point to ponder...

Just for the record, and I am sure that someone will correct me if I am wrong... the circuit operates pretty much as you described, but there is one important consideration that was not made clear in the author's write-up of his design.

The relay in use -- the TQ2E-L2-5V Aromat (TQ2-L2-5V or TQ2H-2L-5V with clinching terminals) -- has polarized coils. This means that they are designed to be connected in circuit in a given direction for standard operation, and will operate in reverse operation when connected with reversed polarity.

According to the Aromat datasheet for the relay, the Pin 1-5 coil is the "SET" coil, with Pin 1 being the + side, and the Pin 6-10 coil is the "RESET" coil, with Pin 10 being the + side. The relay is shipped in its "RESET" condition, which would have the Pin 2-3-4 contact set with Pins 2-3 closed, and with the Pin 7-8-9 contact having Pins 8-9 closed. (See the image below for clarification. The "black" portions indicate "RESET" condition, while the "white" portions indicate "SET" condition.)

**broken link removed**

Normally, energizing the "SET" coil (Pins 1-5) would reverse the above circuits, closing Pins 3-4 while opening Pins 2-3, and closing Pins 7-8 while opening Pins 8-9.

Looking at the schematic that you posted, you can see that the desired effect of energizing that coil will be to close Pins 2-3 and Pins 8-9 -- the exact opposite of what I just described. This is because the relay, as connected in this schematic, is connected in reverse polarity.

My experience is that the latching reliability of this realy type is better when installed with proper polarity, as the magnetic field of the coil is then added to that of the core (the polarizing agent). However, the circuit will work as shown.

I have an extremely similar shutdown circuit that I have built, which uses the relay in its correct polarity, though I used an Omron G6DC5 relay for mine and some other circuit details differ. Post back if you want that circuit as well, and I'll dig it out and post it. Bear in mind that the shutdown effect of the posted device is immediate, with no prior warning.
 

SnaRf

New Member
Hmmm. Interesting.

I have modified the circuit while building it, to use the relay with proper polarity
(Also updated the schematic attached below)

And of course I managed to fry the relay while testing :oops:

So, now a few questions have arisen.
Considering TC1 and VR1 forms a voltage divider, the LM339 will go high when the
voltage to pin 5 reaches 2.5v. Right?

Also, does the thermistor's resistance decrease as it gets hotter?
If so, it *should* be possible to measure the resistance in the thermistor at the shutdown
temperature, and then calculate what the resistance in VR1 should be, then just
measure it as you adjust it instead of adjusting it by trial and error.

Secondly, as I mentioned I managed to toast the relay.
(The following is based on the "corrected" schematic)
While testing, I shorted pin 5 to ground, and the relay switched over.
This then allowed me to close the switch, which reset the relay.
All the LED's lit properly etc.
Then, I accidentally shorted pin 4 to gound and the relay stopped working.
(Not quite sure why...)
:oops:

Hmm, that brings up another question.
Pin 1 should be connected to the + and pin 5 should be grounded for
the relay to become "SET".
So, how exactly will pin 5 be grounded if it connected to the 'emitter' of the transistor
with it's 'collector' connected to a "high" pin 2 on the LM339?

/me feels a bit stupid :)

ChrisP, I'd love to see you schematic as well and yes, this circuit should cut the power with
NO warning as it's the final failsafe.
 

Attachments

  • schematic_639.jpg
    schematic_639.jpg
    57.8 KB · Views: 958

ChrisP

Member
Here's my circuit diagram. If you have ExpressPCB installed, I can send you the .SCH file and library components used as well. Let me know...

There is a lot of similarity between these circuits, though mine includes a warning before shutdown. Most of the components used here are easily substituted by other commonly available items.

**broken link removed**
 

Roff

Well-Known Member
Considering TC1 and VR1 forms a voltage divider, the LM339 will go high when the
voltage to pin 5 reaches 2.5v. Right?
Right.

Also, does the thermistor's resistance decrease as it gets hotter?
If so, it *should* be possible to measure the resistance in the thermistor at the shutdown
temperature, and then calculate what the resistance in VR1 should be, then just
measure it as you adjust it instead of adjusting it by trial and error.
That should work.
Hmm, that brings up another question.
Pin 1 should be connected to the + and pin 5 should be grounded for
the relay to become "SET".
So, how exactly will pin 5 be grounded if it connected to the 'emitter' of the transistor
with it's 'collector' connected to a "high" pin 2 on the LM339?
SnaRf, you seem to be confused about the names of the terminals on transistors. See symbol below.

When the comparator output is low, it (almost) grounds the base of Q1, turning it off. When the comparator output is high, what has actually happened is the internal output transistor in the comparator has turned off, allowing all the current through R4 to flow into the base of Q1, turning it on.
 

Attachments

  • NPN.GIF
    NPN.GIF
    1.5 KB · Views: 906

SnaRf

New Member
Ron H said:
SnaRf, you seem to be confused about the names of the terminals on transistors. See symbol below.
So what you are saying, is that my transistor is soldered the wrong way around?
:shock: lol, you learn something new everyday
No wonder the relay never switched while I was trying to calibrate it :roll:
Thanx.
I'll be getting the new relay today, will let you know if I get it working.

ChrisP said:
If you have ExpressPCB installed, I can send you the .SCH file and library components used as well
I would like that very much, thanx (/me downloads expressPCB)
Also, what did you use to draw that schematic?
Do you have a bigger version of it?
 

ChrisP

Member
The schematic was drawn using ExpressSCH (a component of ExpressPCB). PM me with an e-mail address and I'll send the requisite files...
 

SnaRf

New Member
Well, sourcing the new relay and transistor will be delayed until next year :(

I spent some time last night learning ExpressSCH.
Quite nice, simple but functional :)

ChrisP, I used some of the details from your ExpressSCH file, I hope you don't mind.
Will convert the SCH to a BMP tonight and upload it.
It's Basically just an ExpressSCH version of the circuit (With a bit of modification)
 

ChrisP

Member
No problem!

Note that I intentionally used a relay in my circuit for the alarm LED instead of simply driving the LED via the transistor, as I was building in some basic capability for future circuit mods regarding output devices for the alarm.
 
Status
Not open for further replies.

Latest threads

Top