Author Topic: Can a variable thermal shut off circuit be build?  (Read 4129 times)

0 Members and 1 Guest are viewing this topic.

Offline FriedMuleTopic starter

  • Frequent Contributor
  • **
  • Posts: 807
  • Country: dk
  • Can make even the simplest task look imposible.
Can a variable thermal shut off circuit be build?
« on: July 31, 2019, 07:33:37 pm »
Sometimes when I do test of circuit I would like to be able to shut off the PSU / Load leads if my circuit gets to hot.
Imagine a thermal sensor that are measures a components temperature and if it exceeds the temperature I select, it shall shut of, not the unit but the test leads.
Like if I am testing a DIY circuit and it i.e. gets above 120F after longer time test 3-4 hours into the test.

It is important that the "cut off" part is as "invisible" for the test result as posible, so it do almost not change the power, resistance or anything like that.

Edit: forgot to say that we are talking about max 100V or 75A (max voltage one of my psu can deliver and max ampere an other psu unit can deliver)
« Last Edit: July 31, 2019, 07:37:48 pm by FriedMule »
Even if I appear online is it not necessary so, my computer is on 24/7 even if I am not on.
 

Offline Rick Law

  • Super Contributor
  • ***
  • Posts: 3490
  • Country: us
Re: Can a variable thermal shut off circuit be build?
« Reply #1 on: August 01, 2019, 01:27:19 am »
Sometimes when I do test of circuit I would like to be able to shut off the PSU / Load leads if my circuit gets to hot.
Imagine a thermal sensor that are measures a components temperature and if it exceeds the temperature I select, it shall shut of, not the unit but the test leads.
Like if I am testing a DIY circuit and it i.e. gets above 120F after longer time test 3-4 hours into the test.

It is important that the "cut off" part is as "invisible" for the test result as posible, so it do almost not change the power, resistance or anything like that.

Edit: forgot to say that we are talking about max 100V or 75A (max voltage one of my psu can deliver and max ampere an other psu unit can deliver)

Until someone with more experience comes along and give you a better option, let me share my hobbyist experience...

(Beside, I have to jump in to see what you want to do with 75Amp!!!)

I do what you wrote about all the time but at 12V 2A.  The same can be done by using a better relay rated 100V 75A.  75A is a hell of a lot of current and that is going to be your road block.  Whether you can do it or not would be decided upon if you can find a relay that size.  My full size fridge takes 6.5A (@100V) only, 75A is more than 10 full size refrigerators like mine!?  75Amp (at any voltage) is still a lot for any relay to handle.

You should consider controlling the PSU's power-in instead of power-out.  See the last paragraph.


I use a simple Arduino NANO with LM35z as temperature probe to drive a relay.


- The LM35z is able to measure −55°C to 150°C.  When you are doing just +C temperature, the connection is a lot easier.  You can also offset the LM35z with a couple of diodes so you don't need -Vin as shown in the datasheet (page 16).  The LM35z will give you 10mV/degree C for the Arduino to read.
   Link: http://www.ti.com/lit/ds/symlink/lm35.pdf 

- To "connect" the LM35z as probe, I just tape the LM35z onto what I want to measure.  Typically, I am measuring a 18650 battery, so getting a good thermo-contact is not hard.  You will have to work that out properly.
If what you want to probe is smaller than the LM35z, you are out of luck with this method.

- The Arduino NANO reads the analog port connected to the LM35z and make decisions based on input voltage.  In my case, I also display the temperature on a 20x4 LCD and some times I output it back to the PC for logging.

- I use the regular cheap Arduino relay module like the one discussed in this thread:
   Link:https://www.eevblog.com/forum/beginners/chinese-relay-module-reliability/ 
  The module I use has relays rated for 250V @ 10A, I am only doing 12V@2A only.   Arduino output port alone doesn't have enough power to drive even those small relay modules, the module has its own +5Vin which is necessary.  You will need way way way better relay modules if you are going to do 75A!  You can also use the cheap module to drive another (bigger) module with the 75Amp relay.  I suspect the 75 Amp relay would be your difficulty.
Make sure you read that other thread and understand the reliability and issues with this cheap relay module if you end up using that to drive power-in.

- Alternative is:  Just relay the input side of the PSU -- if the PSU power-in doesn't exceed the "easy to obtain" relay's spec, let the relay control the power-in for the PSU rather than having the relay module handle power-out at 75Amp...

Good luck there...

EDIT: Forgot the links - added now.  (Also added the last section suggesting relaying the PSU input instead of output)
« Last Edit: August 01, 2019, 02:04:41 am by Rick Law »
 

Offline GerryR

  • Frequent Contributor
  • **
  • Posts: 256
  • Country: us
Re: Can a variable thermal shut off circuit be build?
« Reply #2 on: August 01, 2019, 10:25:12 am »
An automotive starter relay can handle the current (12 volt coil) or a motor contactor used in industrial applications (different control coil voltages available).  Thinking about the current, sounds like your getting into the welding arena, so a welder control relay may also be a possibility.  :-//
Still learning; good judgment comes from experience, which comes from bad judgment!!
 

Online Zero999

  • Super Contributor
  • ***
  • Posts: 20363
  • Country: gb
  • 0999
Re: Can a variable thermal shut off circuit be build?
« Reply #3 on: August 01, 2019, 10:32:38 am »
An automotive starter relay can handle the current (12 volt coil) or a motor contactor used in industrial applications (different control coil voltages available).  Thinking about the current, sounds like your getting into the welding arena, so a welder control relay may also be a possibility.  :-//
An automotive relay certainly will not be able to break 100V at 75A.

One thing to note is that the original poster might not be running at 75A and 100V, as they mention two power supplies, but neglect to mention the VA of them both.

Quote from: FriedMule
we are talking about max 100V or 75A (max voltage one of my psu can deliver and max ampere an other psu unit can deliver)
 

Offline FriedMuleTopic starter

  • Frequent Contributor
  • **
  • Posts: 807
  • Country: dk
  • Can make even the simplest task look imposible.
Re: Can a variable thermal shut off circuit be build?
« Reply #4 on: August 01, 2019, 11:05:26 pm »
Yes 75A is an insane amount og current, but nice to have, because, hmm, just nice to have. :-)
The 75A is a 10V unit and I do newer think I'll use that many amps, but why not build so it can be used?

I imagine some sort of turn button that when turned, changes the max temperature on a display and if the real temperature on the probe is going above, then off!
My reason for wanting to "cut" the leads and not the PSU is that I have heard so many times, that: "if you want a reliable and precise result, keep the unit on for at least xx hours"

Even if I appear online is it not necessary so, my computer is on 24/7 even if I am not on.
 

Offline digsys

  • Supporter
  • ****
  • Posts: 2209
  • Country: au
    • DIGSYS
Re: Can a variable thermal shut off circuit be build?
« Reply #5 on: August 02, 2019, 12:11:41 am »
Go solid state - perfect voltage / current for parallel sets of Power FETs, say 10-20A per. Should be able to find Rdson of ~ 1mR. Control it all with 1mA. A bit more work than a simple Kilovac / Gigovac relay (used in EVs), but "cleaner" / more controlled switching.
Hello <tap> <tap> .. is this thing on?
 

Online ledtester

  • Super Contributor
  • ***
  • Posts: 3282
  • Country: us
Re: Can a variable thermal shut off circuit be build?
« Reply #6 on: August 02, 2019, 01:20:54 am »
Go solid state ...

I looked up what a 75A solid state relay costs at Mouser and was shocked.

I suppose paralleling two $10 40 amp SS relays from Amazon isn't a great idea.
 

Offline digsys

  • Supporter
  • ****
  • Posts: 2209
  • Country: au
    • DIGSYS
Re: Can a variable thermal shut off circuit be build?
« Reply #7 on: August 02, 2019, 01:36:07 am »
Quote from: ledtester
I looked up what a 75A solid state relay costs at Mouser and was shocked .... I suppose paralleling two $10 40 amp SS relays from Amazon isn't a great idea
We've used Crydoms (and others), at ~$100 ea, actually not much cheaper than a Kilovac / Gigovac proper DC mechanical relays. BUT, you can make a simple switch yourself, with parallel power FETs, IF cost is important. After all, it is just an On/Off switch.
Hello <tap> <tap> .. is this thing on?
 

Offline FriedMuleTopic starter

  • Frequent Contributor
  • **
  • Posts: 807
  • Country: dk
  • Can make even the simplest task look imposible.
Re: Can a variable thermal shut off circuit be build?
« Reply #8 on: August 02, 2019, 05:23:17 am »
Quote from: ledtester
I looked up what a 75A solid state relay costs at Mouser and was shocked .... I suppose paralleling two $10 40 amp SS relays from Amazon isn't a great idea
We've used Crydoms (and others), at ~$100 ea, actually not much cheaper than a Kilovac / Gigovac proper DC mechanical relays. BUT, you can make a simple switch yourself, with parallel power FETs, IF cost is important. After all, it is just an On/Off switch.
It is always dagerus to say that the price dos not matter, but I'd rather that it WORKS!! then save a few dollars.
Since I am a noob, I'd tryed to find some circuit on the internet, but I havent found any.
A circuit where a turn button changes the number in the display to maybe max 200 C and also can compare the measured temperature with the max choosen value and then shut off.

What about optorelays, do they not have to handle great amount of current and voltage?
Even if I appear online is it not necessary so, my computer is on 24/7 even if I am not on.
 

Offline Rick Law

  • Super Contributor
  • ***
  • Posts: 3490
  • Country: us
Re: Can a variable thermal shut off circuit be build?
« Reply #9 on: August 02, 2019, 06:16:18 am »
The latest reply from the OP said the 75Amp is at only 10V.  Out of curiosity, I did a Amazon search.  At this happy voltage, automotive relay is an option.  The available options are aplenty:

$23.45 USD Pack of 5  (store brand) relays+harnesses rated at 12V 80A/60A High Switching Capability  (I have no idea why they have both 80A and 60A in the description.)
https://www.amazon.com/ONLINE-LED-STORE-Switch-Harness/dp/B01NCA6CAB

For a better known brand
$45.50 USD Each: Bosch Automotive Relay, 12V 75A
https://www.amazon.com/Bosch-332-002-168-0332002168/dp/B0117FAGVO

So, one can indeed do it easily and inexpensively: Use Arduino driving a cheap relay module (5 volt coil) which in turn drive the coil for the 75Amp relay.  Not exactly pretty, but works in theory to drive 75amp.

* * *

Actually, the LM35 makes a rather easy to use probe.  Let me show off my "quick and dirty" temperature probe construct here.  It was really a "quick and dirty" construct, made for "one time use" when I found my newly recovered 18650 (from a dead laptop pack) seem to be charging warmer than I like, so I want it to log the V/I/Temp/Time plus cut off at about 90 degree C.

I made my own "offset circuit" before I ran into the version of LM35 datasheet with an example.  So my offset circuit isn't the same as the one in the datasheet.  Mine has only 1 diode offsetting the LM35's Vgnd at about 1Volt.  With the 5V from the NANO, that leaves me about 4V, just high enough to run the LM35.  I also made my offset board into a "daughter card" for two LM35 probes and plugs directly plug into NANO's 5V and three analog ports - a shared Vgnd of the LM35, and two V-sense for the two probes.

The two LM35's has the connections with extension wires from a dead network cable then shrink-wrap into a probe.

Like most "one time thing", the "one" is a repeating one.  This little "one time" thing is still in use.   The NANO is equipped with 2004 LCD display and a DS3231 real time clock so I can log with date/time.  It is now my regular desk clock with "room" temperature display from the DS3231 module - plus two other temperatures from these two probes that I can just stick anywhere...  Mostly just to see how hot my batteries are on the charger or how hot my desk-lamp is getting...  The relay module is currently reassigned to another experiment.
« Last Edit: August 02, 2019, 06:23:49 am by Rick Law »
 

Offline digsys

  • Supporter
  • ****
  • Posts: 2209
  • Country: au
    • DIGSYS
Re: Can a variable thermal shut off circuit be build?
« Reply #10 on: August 02, 2019, 06:47:40 am »
Quote from: FriedMule
It is always dagerus to say that the price dos not matter, but I'd rather that it WORKS!! then save a few dollars ...
NO ONE suggested that the "cheap(er) price of ~$100 reflected poor quality" !! Crydom is one of the very best solid-state manufacturers out there. I was replying to ledtesters comment = "I looked up what a 75A solid state relay costs at Mouser and was shocked".
I don't know what price ledtester saw or what part he was actually looking at, or what he thought was "expensive". ~$100 is an extremely good price for that type of device !
Quote from: FriedMule
... What about optorelays, do they not have to handle great amount of current and voltage?
That is an "opto-relay". 100V at 75A+
Hello <tap> <tap> .. is this thing on?
 

Online ledtester

  • Super Contributor
  • ***
  • Posts: 3282
  • Country: us
Re: Can a variable thermal shut off circuit be build?
« Reply #11 on: August 02, 2019, 04:10:50 pm »
Yes 75A is an insane amount og current, but nice to have, because, hmm, just nice to have. :-)
The 75A is a 10V unit and I do newer think I'll use that many amps, but why not build so it can be used?

There is something to be said for using a power supply that is sized for your application. You want some headroom, but using a 75A power supply when you only need a couple of amps is foolhardy. Even if the supply has current limiting, 750W is a lot of power, and it's a good idea, especially for beginners, to limit the worst possible outcome to something that's not going to ruin your day. At the very least employ fuses that are appropriate to your application.

Quote
...
I imagine some sort of turn button that when turned, changes the max temperature on a display and if the real temperature on the probe is going above, then off!
...

These things are called "temperature controllers" (and in the consumer world "thermostats") and there are a ton of ready-made bare modules available with a "dial" and display. The difficulty is finding something that will work at the temperatures you want and uses a sensor that is compatible with your application. Thermistors and thermocouples are generally used to measure the temperature of a heatsink / battery / electronic component.

You might also find some pre-made modules by searching for "fan controller". I just recently bought one of these:

https://www.ebay.com/itm/12V-PWM-Fan-Temperature-Speed-Controller-Module-High-Temp-Alarm-5V-Buzzer-PC-CPU/263605466941?hash=item3d601c373d:m:md_Xw7-TdWpozYJ83KekdBw

Haven't had a chance to test it out though.
 

Online CatalinaWOW

  • Super Contributor
  • ***
  • Posts: 5571
  • Country: us
Re: Can a variable thermal shut off circuit be build?
« Reply #12 on: August 02, 2019, 05:28:28 pm »
Your supplies should have a remote turn off option which will do the job nicely.  I know you have said you don't want to turn off the supplies so you can maintain a long on time on your test circuit for "stability", but when you change the load current by that amount all stability is out the window anyway.  Those stability issues arise because of thermal effects primarily and when you dump the load you change the thermal environment completely.

If you decide to stick with your plan as others have said temperature controllers are widely available.

https://www.omega.com/en-us/search/?text=temperature+controller

The ones in the link are pricey, but quality is assured.  I am sure you can find Chineseum ones far cheaper.  Then the only trick is to use the relatively low power relay (only a few hundred watts) in these things to turn off the high power contacter you need to find.
 


Offline FriedMuleTopic starter

  • Frequent Contributor
  • **
  • Posts: 807
  • Country: dk
  • Can make even the simplest task look imposible.
Re: Can a variable thermal shut off circuit be build?
« Reply #14 on: August 02, 2019, 10:30:23 pm »
Thanks a lot, you are all fantastic!
About 75A being to much if I maybe only need a few amp. Yes that is perfectly correct and I do have some 30V 5A and other that I can use for the normal everyday use.
What I am looking for, is the possibility to turn off the PSU no matter what PSU I choose to use, if it is my 30v-5A, 10V-75A, or my 100V-10A.

I'll take a look on the thermostats! :-)
Even if I appear online is it not necessary so, my computer is on 24/7 even if I am not on.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17428
  • Country: us
  • DavidH
Re: Can a variable thermal shut off circuit be build?
« Reply #15 on: August 03, 2019, 03:18:32 am »
We have used the metal can AD590 as the temperature sensor for this sort of application in the past but they are pretty expensive.  The current output is easier to use.

I would build a latching circuit using the temperature sensor, a reference, and an operational amplifier as a precision comparator.  With proper calibration, the trip point can be set with a potentiometer and multimeter.

An external temperature trip might be a useful feature to build into a power supply.
 

Offline FriedMuleTopic starter

  • Frequent Contributor
  • **
  • Posts: 807
  • Country: dk
  • Can make even the simplest task look imposible.
Re: Can a variable thermal shut off circuit be build?
« Reply #16 on: August 03, 2019, 09:24:55 am »
We have used the metal can AD590 as the temperature sensor for this sort of application in the past but they are pretty expensive.  The current output is easier to use.

I would build a latching circuit using the temperature sensor, a reference, and an operational amplifier as a precision comparator.  With proper calibration, the trip point can be set with a potentiometer and multimeter.

An external temperature trip might be a useful feature to build into a power supply.
The dmm idea sounds great, maybe even use one of the dmm displays from Ebay?

Yes and maybe in an electronic load, I am wondering on why that is not standard in the top of the range PSU / Loads.

May I put an enormous strange on your genres help? :-)
Could you maybe try to draw a schematic that I can start my project from?
Even if I appear online is it not necessary so, my computer is on 24/7 even if I am not on.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17428
  • Country: us
  • DavidH
Re: Can a variable thermal shut off circuit be build?
« Reply #17 on: August 03, 2019, 10:42:08 am »
The dmm idea sounds great, maybe even use one of the dmm displays from Ebay?

That could work but I always just connected a multimeter temporarily to set the trip point.

Quote
May I put an enormous strange on your genres help? :-)
Could you maybe try to draw a schematic that I can start my project from?

There is not much to draw.  Filter the signal and reference voltage before the comparator.  Using an operational amplifier makes this easier because of lower input bias current.

When I did it, I latched the output by using the operational amplifier to trigger an SCR which pulled the output to the load disconnect down to ground and kept it that way until reset.
 

Offline FriedMuleTopic starter

  • Frequent Contributor
  • **
  • Posts: 807
  • Country: dk
  • Can make even the simplest task look imposible.
Re: Can a variable thermal shut off circuit be build?
« Reply #18 on: August 03, 2019, 11:11:38 am »
Thanks a lot, I'll try to se if I can draw something that do that:-)
Would you mind looking at it when / if I can draw it?
Even if I appear online is it not necessary so, my computer is on 24/7 even if I am not on.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17428
  • Country: us
  • DavidH
Re: Can a variable thermal shut off circuit be build?
« Reply #19 on: August 03, 2019, 05:17:58 pm »
Thanks a lot, I'll try to se if I can draw something that do that:-)
Would you mind looking at it when / if I can draw it?

Just post it.  Everybody will comment on it.
 

Offline FriedMuleTopic starter

  • Frequent Contributor
  • **
  • Posts: 807
  • Country: dk
  • Can make even the simplest task look imposible.
Re: Can a variable thermal shut off circuit be build?
« Reply #20 on: August 05, 2019, 05:18:58 am »
I have tried to make something that may work, remember I am as noob as they come! :-)
My idea is that the "temp to LCD" shall connectes in some way so the display show max alloveable temperature and by switch, show the current temperature.
Placeholder relay shal go to an NC 100A 100V relay
Even if I appear online is it not necessary so, my computer is on 24/7 even if I am not on.
 

Online ledtester

  • Super Contributor
  • ***
  • Posts: 3282
  • Country: us
Re: Can a variable thermal shut off circuit be build?
« Reply #21 on: August 05, 2019, 06:32:29 am »
Why is pin 3 of the LM358 unconnected?

Temperature doesn't change that fast, so why not just use the uC to activate the relay? In pseudo-code:

Code: [Select]
    while (1) {
         read LM35 using the ADC
         if LM35 temp > threshold temp:
            activate relay using a GPIO pin
            break
    }
 

Offline FriedMuleTopic starter

  • Frequent Contributor
  • **
  • Posts: 807
  • Country: dk
  • Can make even the simplest task look imposible.
Re: Can a variable thermal shut off circuit be build?
« Reply #22 on: August 05, 2019, 06:41:55 am »
Why is pin 3 of the LM358 unconnected?

Temperature doesn't change that fast, so why not just use the uC to activate the relay? In pseudo-code:

Code: [Select]
    while (1) {
         read LM35 using the ADC
         if LM35 temp > threshold temp:
            activate relay using a GPIO pin
            break
    }
Oh sorry, an error! LM358 Pin 3 -> LM35 pin 2

I am not sure about your code question, uC ?
Even if I appear online is it not necessary so, my computer is on 24/7 even if I am not on.
 

Online coppercone2

  • Super Contributor
  • ***
  • Posts: 11340
  • Country: us
  • $
Re: Can a variable thermal shut off circuit be build?
« Reply #23 on: August 05, 2019, 06:52:39 am »
use a contactor triggered by a thermostat (heat sink the parts so you can screw a thermostat into them, it will be more fool proof then trying to make a control system based on thermocouples that is safe, thermostats and contactors are used everywhere
 

Online ledtester

  • Super Contributor
  • ***
  • Posts: 3282
  • Country: us
Re: Can a variable thermal shut off circuit be build?
« Reply #24 on: August 05, 2019, 07:00:28 am »

I am not sure about your code question, uC ?

uC = microcontroller
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf