I am building a filament extruder, and have built my own heating element out of nichrome wire. the heater has an aluminium case, and is wraped around an aluminium pipe to increase heat capacity. I have a K-type thermocouple that screws into the aluminium, but its giving me weird values.
-when the power supply to the heater is off, the thermocouple reads correctly, verified with my phone.
-when I turn the heater power supply on, the thermocouple reads 0.0 degrees C.
-if I gound the aluminium pipe, (and hence the thermocouple), the thermocouple reads 4 degrees lower than the real temperature.
-there is no continuity between the nichrome wire and the case (and hence thermocouple)
-the voltage across the thermocouple leads is not affected by making theses changes (grounding the case etc)
-when the power supply is on, I get 0.2 V across the case and ground, but there is no continuity between the two

I am suspicious that there is some inductor or magnetic field at play that is messing with the results, but I cant work out how that would be happening. the nichrome wire is wrapped around a mica core, which is then wrapped around the aluminium case. (picture and drawing shown below)

thanks for any help!

 

First question would be is the thermocouple a grounded or ungrounded type. I am assuming a sheath type thermocouple. So no power applied TC removed what resistance do you see between thermocouple + and - leads and the tc sheath.

Types of TC junctions:


Next question becomes what are you using to actually measure the TC output? Would you be going TC to a temperature controller or a type of other signal conditioning? The reason I ask is when I have seen these symptoms it is normally attributed to a temp controller or a temp transducer being for one type TC and using another type TC as to grounded or ungrounded junction.

That is my first thought and since I see nothing about how things are wired like a schematic and no narrative as to controller or signal conditioning based on your description these are my first thoughts. I have seen the symptoms before.

Ron

 

In addition to the questions in #2: What is the power supply for the heating element? Is it AC or DC? How is it controlled?

 

@Reloadron @KeithWalker
the resistance of the TC is 10.3 Ohms, and i dont know how to check if its grounded, isolated or exposed.
I am using a MAX6675 to convert the thermocouple voltages into a readable signal that i can send to my raspberry pi pico. I also dont know how to check to see what type of thermocouple this is for. in the future I will connect a pwm controller to the heater so i can control it with my raspberry pi based on the signals coming in. the supply for the heater is 65V and it is 45 Ohms.

 

If the thermocouple is ungrounded, then it may help to connect the TC sheath to whatever is a "ground" connection on the TC temperature display device. All it takes is a few millivolts of noise to completely disrupt TC temperature readings. To see if it is connected to the case or not, use the OHMS function. with the thermocouple disconnected from everything, one probe on the TC wires and the other probe on the stainless steel sheath It should show open circuit.. If not, then it is probably connected to the case (Grounded type TC)

 

@MisterBill2 when I test continuity between the thermocouple leads and the case, I get a resistance of about 1.6 mega Ohms. does this mean it is ungrounded?

 

No, that means inadequately isolated from ground. Given that condition the solution becomes more challenging, at best.
One choice would be to switch off the heat to read the temperature for a short time. Is the temperature signal used for temperature control by switching the heater on and off?? I suggest that it might be the heater power source noise causing the problem. How about a simple mains powered transformer to power the heater? That could be a 120/240 transformer with the 240 side connected to the 120 volt mains. No switcher supply noise generated.

 

if I gound the aluminium pipe, (and hence the thermocouple), the thermocouple reads 4 degrees lower than the real temperature.

This is with the power on, correct?

when the power supply to the heater is off, the thermocouple reads correctly, verified with my phone.

How do you verify temperature with your phone?
Are you using a thermal camera accessory?
Are you verifying temperature at steady state?

The MAX6675 that you're using is a pretty basic solution that doesn't employ any signal biasing (or if it does, I didn't see it in the datasheet). An error of 4 degrees in the presence of EMI might be the best you can reasonably expect from it. More accurate ways of reading thermocouples are described here:
https://www.ti.com/lit/pdf/sbaa355
https://www.ti.com/lit/pdf/sbaa274
If you open those documents take note of the external biasing circuits which help the measurement to eliminate EMI-induced offset error.

 

Using a clean AC power source instead of a switcher power supply will certainly reduce the ambient noise level. AND it is probably a good idea to include some sort of temperature control that can switch the heater on and off.
Also, using an inverter to power a resistance type heater is terribly inefficient.

 

OK now I see what you are doing. I have one of those MAX6675 modules. While I have never used it with the thermocouple it came with I have used the module along with an Arduino Uno uC and it actually worked well and stable using a Type K ungrounded thermocouple. Never used it with a Rasbery Pi though. I can tell you the included TC you have is a Grounded Junction type. This allows a path for noise into your measurement plane.

Based on your description and symptoms my best guess at this point is you are picking up noise. So the question now is how are you powering your heaters?

Ron

 

OK now I see what you are doing. I have one of those MAX6675 modules. While I have never used it with the thermocouple it came with I have used the module along with an Arduino Uno uC and it actually worked well and stable using a Type K ungrounded thermocouple. Never used it with a Rasbery Pi though. I can tell you the included TC you have is a Grounded Junction type. This allows a path for noise into your measurement plane.

Based on your description and symptoms my best guess at this point is you are picking up noise. So the question now is how are you powering your heaters?

Ron

Looking at post #4,the TS is using a "step up converter to provide 65 volts from a 12 volt DC source. I have suggested that it was a noise source, which nobody else believes, evidently. I did suggest using a mains powered transformer to provide a similar voltage but nice clean AC. I mentioned that a 240 to 120 transformer would also provide close to the same voltage connected to step 120 volts down to 60 volts. Way too simple, I guess. and who ever heard of a step-up converter producing any noise??

 

Looking at post #4,the TS is using a "step up converter to provide 65 volts from a 12 volt DC source. I have suggested that it was a noise source, which nobody else believes, evidently. I did suggest using a mains powered transformer to provide a similar voltage but nice clean AC. I mentioned that a 240 to 120 transformer would also provide close to the same voltage connected to step 120 volts down to 60 volts. Way too simple, I guess. and who ever heard of a step-up converter producing any noise??

As to noise I agree and have no idea why the thread starter is going about things this way. Normally when I incorporate heaters in a design I like off the shelf solutions. I can see where this is a custom application but I would consider either a transformer which depending on current could be large or another option, I would not use DC on my elements or a boost converter to get my power.

Ron

 

cleaner power might even be the ultimate fix.

 

I have suggested that it was a noise source, which nobody else believes, evidently.

Are there replies to this thread which I can't see? The only people in the discussion since you showed up (that I can see) other than the TS and yourself is me and Ron. Which of us are you insinuating is an idiot?

 

OK now I see what you are doing. I have one of those MAX6675 modules. While I have never used it with the thermocouple it came with I have used the module along with an Arduino Uno uC and it actually worked well and stable using a Type K ungrounded thermocouple. Never used it with a Rasbery Pi though. I can tell you the included TC you have is a Grounded Junction type. This allows a path for noise into your measurement plane.

Based on your description and symptoms my best guess at this point is you are picking up noise. So the question now is how are you powering your heaters?

Ron

I am powering my heaters with 65V DC, coming from a step up converter, specifically this one https://www.ebay.com.au/itm/304601203523?hash=item46eba58343:g:Mt0AAOSw~Whi5139
what do you mean by my measurement plane?

Rob

 

Using a clean AC power source instead of a switcher power supply will certainly reduce the ambient noise level. AND it is probably a good idea to include some sort of temperature control that can switch the heater on and off.
Also, using an inverter to power a resistance type heater is terribly inefficient.

I am not really experienced with AC power, and all my other components work with DC. In the future I think I will learn more about AC, but for the time being, i wont be using it. I do agree that it could be noise, however the TC reads a stable temperature without the supply turned on, and when its turned on, it is stable, only 4 degrees lower. I do plan to implement a temperature control system, but that is not my problem right now. What do you mean by inverter? and how is it inefficient?

I am willing to learn, if anyone is willing to explain some concepts as to what might be going on, just dont put too much complex show off words that I wont understand.

Rob

 

As to noise I agree and have no idea why the thread starter is going about things this way. Normally when I incorporate heaters in a design I like off the shelf solutions. I can see where this is a custom application but I would consider either a transformer which depending on current could be large or another option, I would not use DC on my elements or a boost converter to get my power.

Ron

why wouldnt you use DC power or a step up converter? I have seen plenty of nichrome wite heaters powered with DC? what is the advantage of AC? Off the shelf components arent really available where I live, but im also trying to learn about things rather than just having them work. in the case of the heater, I couldnt find any options in my country that were under $100 USD, and my one cost $5 USD.

 

The boost converter you are using states 93 % efficiency when boosting 30 to 40V (ratio of 1.33)
You are boosting 12 to 65v (ratio 5.4) Higher conversion ratios tend to make for less efficiency, I would imagine 75% or less.

To deliver 93.8 W to your heater, you need to draw approx 125 W from your 12 V power source.
You need to heatsink 30 W of power loss from your very toasty boost converter.

 

The boost converter you are using states 93 % efficiency when boosting 30 to 40V (ratio of 1.33)
You are boosting 12 to 65v (ratio 5.4) Higher conversion ratios tend to make for less efficiency, I would imagine 75% or less.

To deliver 93.8 W to your heater, you need to draw approx 125 W from your 12 V power source.
You need to heatsink 30 W of power loss from your very toasty boost converter.

that explains my 10A fuse blowing. do you have any other solutions?

 

that explains my 10A fuse blowing. do you have any other solutions?

Different heaters or different power source.

Generally, heaters are massive power hogs in any system, you want them matched to the power source to allow for simple and efficient driving.
In your system, the heaters are very poorly matched to the power source.