Why doesn't the thermocouple work correctly?

Go To Last Post
22 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hi guys
I'm asking my question here and hope to find my problem. about a year ago I started to make my soldering iron station. It's based on Hakko 951 series.

 

 

 

I designed a GUI and a PCB for it. about after a month, I gave up the project because I was busy until now. Everything was fine save one part. thermocouple! I don't know why but it doesn't work correctly. the problem is that it doesn't give correct output.

 

How does this thermocouple work?
It's simple. A solder tip contained a heater in series with a thermocouple. when the tip gets hot, the thermocouple should give us a voltage between 0 to 9 mV due to the heat you apply to it or the built-in heater make(when it's turned off) and I amplify this voltage to measure the temp. Look at this picture for more details:

 

 

How do I configured the tip (thermocouple + heater)?
Here is the schematic I used:

 

 

I need to add that the circuit was used 1 - 2 years ago by this guy and it had worked successfully.

 

http://dangerousprototypes.com/forum/viewtopic.php?f=56&t=5264&start=180#p58264

 

What's the problem?
At first I have to say it gives me a voltage between 0 -3 mV (although the output changes due to the temp change linearly) instead of 0-9mV(alone problem is that this range is smaller than I expected). And second it doesn't work correctly when I add it to the above circuit. E.G. It gives me about 300mV at the output at room temperature. If I remove the tip from the circuit and measure the output of the tip, it gives me an output due to the temp.

 

How do I measure the output?
For measuring the output when it is unconnected to the circuit: I connect one lead(wire) of the tip to the red probe of my multimeter and another to the black one. if you apply heat to the tip, you would see that DMM show increasing the voltage between 0 to 3mV.(while I expect to see an increase between 0 to 9mV ).

 For measuring the output when it is connected to the circuit:
Like above, each prob to one lead but as I said the output isn't correct.

 

In your opinion why doesn't the tip give me the correct output?

Please feel free to ask any question

This topic has a solution.

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

Last Edited: Wed. Sep 14, 2016 - 05:29 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Cold Junction compensation seems to be missing

 

Also, as with most T/C's, the working voltage span is from single digit millivolts to high double digit millivolt values - maybe three digits, even.

 

Even with a decent instrumentation amplifier, it would be better to use several gain stages.

 

Though... Analog Devices makes a couple good thermocouple amplifiers.  External cold junction compensation is still required, though!

You can avoid reality, for a while.  But you can't avoid the consequences of reality! - C.W. Livingston

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Analog Devices makes a couple good thermocouple amplifiers.  External cold junction compensation is still required, though!

 What about the AD8494/AD8495/AD8496/AD8497 type? They have internal cold junction compensation, unfortunately the temp range seems too small for a soldering iron.
 

I guess they make others??

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

 

 A solder tip contained a heater in series with a thermocouple

 

 

I've never taken one apart, or studied their schematics, but that isn't how I would design it.

 

JC 

Last Edited: Sun. Aug 21, 2016 - 09:35 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

My concerns about this setup: we know where the thermocouple "hot" junction is, but where is the "cold" junction actually located?  That makes all the difference (pun intended) in getting an accurate temp reading.  What are the thermocouple metals and what various intermediate metals are in the total measurement circuit?

Tom Pappano
Tulsa, Oklahoma

Last Edited: Sun. Aug 21, 2016 - 11:29 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

T12 have thermocouple in series with the heater with a voltage of around 9mV at 450 degrees Celsius.
Perhaps reference to the " heat to the tip" applied in terms of temperature would put a light on the result.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

We use the max31855 type K thermocouple amp to measure exhaust temperature... adafruit sells one on a little breakout board. The range is up to 1800 deg F. Solder tip is maintained around 650 deg F, so a temp controller on the heater based on the tc would work good. Good irons are expensive, so this might be a good idea.

 

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I think just me have this problem and don't know why.

guys, how can I make a "cold junction" for this?(not software, I'm talking a hardware solution)

tpappano wrote:

  What are the thermocouple metals and what various intermediate metals are in the total measurement circuit?

I don't know what kind of metal the legs(pins) of the amplifier IC has but it's connected on the PCB(copper).indecision

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Is it a max6675 or a max31855? You can dl the ds from maxim ic.

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

First of all, you say that you observe a 0-3mv signal swing as you apply heat to the tip and that you expected 0-9mv.  Why do you expect 0-9mv?  The voltage generated by a thermocouple depend upon what the metals are.

 

We have to do a lot of guessing about what you have going on, but for example let's say the cord to the iron is copper wire.  Inside the handle you have a heating element, probably made of a nickel-chromium alloy.  From your picture one end of the element might be welded to the, "tip" which might be made of copper.  This welded junction would be the "hot" junction.  The other end of the heating element at some point must be connected to one of the copper wires in the cord and the other copper wire probably connects to the copper "tip".  The "cold" junction is where the copper cord wire connects to the heating element.  The "temperature" voltage you measure from the cord depends upon the temperature difference between the cold and hot junctions, which could essentially be nichrome/copper thermocouples.  The actual voltage depends upon the "unknown" composition of these alloys.

 

Your circuit may be working just fine already, you just don't know for sure what voltage you should see because of the unknown thermocouple alloys.

 

The power contacts on the cartridge probably don't matter because the rule of intermediate metals cancels their thermocouple effects.

 

Tom Pappano
Tulsa, Oklahoma

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I would guess that he has a "K" type thermocouple. There are lots of charts: https://www.google.com/#q=k+type...

 

Measuring the very small signal reliably is a bit of a trick. Best bet is to get a "thermocouple amplifier" like Bob suggests. Also be careful you're not shorting the signal out.
 

274,207,281-1 The largest known Mersenne Prime

Measure twice, cry, go back to the hardware store

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

@tpappano

But everything I'm doing is just like the things that the others in that topic have done and there should not be any problem. the problem is that I don't know how to find the problem! what a philosophic sentence I said!wink

I measured the output of the handle by my DMM and there is no problem.

Torby wrote:

I would guess that he has a "K" type thermocouple. There are lots of charts: https://www.google.com/#q=k+type...

 

Measuring the very small signal reliably is a bit of a trick. Best bet is to get a "thermocouple amplifier" like Bob suggests. Also be careful you're not shorting the signal out.
 

He has or it has?angel

Tom

Everything Bob suggested is available for me:

 

http://shop.aftabrayaneh.com/Thermocouple_Amplifier_Digital_MAX6675.html?search=%D8%AA%D8%B1%D9%85%D9%88%DA%A9%D9%88%D9%BE%D9%84

 

http://shop.aftabrayaneh.com/Thermocouple_Amplifier_MAX31855.html?search=%D8%AA%D8%B1%D9%85%D9%88%DA%A9%D9%88%D9%BE%D9%84

 

http://shop.aftabrayaneh.com/CJMCU_K-type_Thermocouple_ADS1120_16-bit_ADC.html?search=%D8%AA%D8%B1%D9%85%D9%88%DA%A9%D9%88%D9%BE%D9%84

 

But I think/feel this won't solve the problem.

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If I am getting the point, both DMM and circuit red 0..3mV instead of 0..9mV.

Would assume   thermocouple is a culprit then.  Would run a test with a separate thermocouple from the DMM kit. Or any else known good one.

Is that T12 genuine (=expensive)?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Get a known good meter with a tc input and a good tc. Measure the diff between what the tc on the meter says and what the display says. Lets says the display is off by 100 deg. Write a note and tape it near the display "Set temp 100 deg higher than you want".

 

 

Imagecraft compiler user

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I just did four experiments/tests.

1.
 I applied heat by a lighted match to the tip and it showed at output 1.7mV and
 it went down after several second.

 

http://img4.uplood.fr/free/joad_20160825_174651.jpg

2.
 Again I inserted the tip into the handle and again applied the heat by match and
 I got the same result like the first experiment/test.

 

http://img4.uplood.fr/free/z6ll_20160825_175032.jpg

3.
 again I soldered the handle to my controller board and didn't turn it on and applied
the heat but didn't get any result. just 0V! just like this:

 

http://img4.uplood.fr/free/lnva_20160825_185403.jpg

4.
 I plugged in the circuit and measured the two lead of the handle and found out there
was 334mV. ok, I applied the heat and measured the voltage just like above and the result was 335mV and it went up to 336 after around 20 second and after a minute it went up to 337. no change again!

it's interesting to me. instead of going down, it goes up!

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

Last Edited: Thu. Aug 25, 2016 - 06:30 PM
This reply has been marked as the solution. 
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

Wash thoroughly the PCB from any residues of flux and dirt, check empty PCB for integrity.
Double check passives, especially C19. Swap semiconductors and opamp to the new ones.
Stop ruining tip with open flame.

Last Edited: Thu. Aug 25, 2016 - 03:34 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I did two tests/experiments again:

1. I desoldered the handle(its terminals) from the board and plugged in the board and measured the voltage of the pads that I solder the terminals of the handle on it and it was 320mV.

2. Again I soldered the terminals but this time in reverse direction(the green wire to GND and black to the +Heater unlike earlier that I soldered the green terminal to the +Heater and the black terminal to the GND) and everything was ok but the output unlike earlier is negative. e.g. instead of showing +1.7mV it showed -1.7mV.

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Rohalamin wrote:

1. I desoldered the handle(its terminals) from the board and plugged in the board and measured the voltage of the pads that I solder the terminals of the handle on it and it was 320mV.

This proves D3 is biased forward and voltage looks legit.

Rohalamin wrote:

 2. Again I soldered the terminals but this time in reverse direction(the green wire to GND and black to the +Heater unlike earlier that I soldered the green terminal to the +Heater and the black terminal to the GND) and everything was ok but the output unlike earlier is negative. e.g. instead of showing +1.7mV it showed -1.7mV.

 So TC itself is OK. Now it's time to find out why TC does not effectively shorts "Heater +" terminal to "GND" when not reversed.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Thanks dear Kas

I replaced the Op-Amp to a new one and now it's working.yes now the problem is that why it doesn't give me 0-9mV at output. currently it gives me 0-2mV at output. everything is working just like a watch doyes except this part.no

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

Last Edited: Wed. Sep 14, 2016 - 05:37 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Rohalamin wrote:

 everything is working just like a watch do

 

... your soldering iron tells the time? WOW! laugh

 

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

valusoft wrote:
...

... your soldering iron tells the time? WOW! laugh

 

Ross

I mean my soldering Iron works very well as a watch work precisely.wink

"One's value is inherent; money is not inherent"

 

Chuck, you are in my heart!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yes I know that Amin... I was "pulling  your leg", meaning having a joke at your expense. I am pleased that your soldering iron is working well.

 

Cheers,

 

Ross

 

Ross McKenzie ValuSoft Melbourne Australia