Millivolt voltage sensor circuits

Thread Starter

SmokerinCB

Joined May 13, 2021
10
Good adfternoon. I'm working on my smoker, my COVID project, and I've got it looking like a time machine. Digital thermometers everywhere and a PID controller. But I need help with one last project. However, my electronics background is limited. I made crystal radios when I was a kid and I took tube theory in college. So I know enough to hurt myself and burn up a few ICs. Here's my project. My smoker has a pilot light which is kept on by a thermocouple. The pilot light is buried in the bowels of the smoker. The only way that I know that it's on is by watching the temperature increase, I can't see the pilot light. I want to make a circuit that will light an LED when the pilot light is on. The gas valve solenoid holding voltage is about 6 millivolts and it releases at about 3 millivolts. Max voltage output of the thermocouple is about 32 millivolts. The gas valve is a White-Rodgers 36C03. I'd like the LED to go on and off at about 8 millivolts. My smoker has a regulated 12 VDC power supply (LM7812). It powers the digital thermometers. I've been looking at voltage sensor circuits. I'm thinking that I could compare the output of a 12 volt powered 50k ohm/20 ohm voltage divider with the output of the thermocouple. From what I see in the literature, an LM741 won't work. I'd need to amplify the millivolts to volts. Could I use an LM358 for that? Or, maybe, could I use an LM358 for the whole thing? Or anything else? If any of you have a moment of boredom that you'd like to spend on helping me with this project it would be much appreciated.
Thanks.
SmokerinCB
 

Sensacell

Joined Jun 19, 2012
2,821
You can do it simply with a good modern rail-to-rail opamp.

My favorite is the MCP6024

The key points are:

Common-mode input range includes ground.
Output goes all the way to ground
Really low input offset voltage.

Forget the jellybean opamps like the LM741 / LM358, these just won't work for this application.
 

Thread Starter

SmokerinCB

Joined May 13, 2021
10
Good morning ..... sorry abouit the late response but I've been out of town ..... got home last night ..... I looked at the MCP6024 ..... it contains 4 op amps .... are you suggesting that I connect them in series ? ..... with my voltage divider circuit output going to Vina- and the thermocouple output going to Vinb+ ? .... this is where I'm going to need some help .....
 

schmitt trigger

Joined Jul 12, 2010
438
Many semiconductor companies have application notes which shows circuits that not only have low offset and high gain, but include cold-junction compensation.

True, this may be an overkill for your project. But the key here is that you study the circuit to understand the requirements of thermocouple measurements and how to satisfy those requirements
 

Audioguru again

Joined Oct 21, 2019
3,175
You need only one opamp as an amplifier but maybe a 2nd opamp can be the comparator.
An MCP6024 has 4 opamps. An MCP6042 has 2 opamps. An MCP6021 has 1 opamp. Look at its datasheet found in Google.
 

Irving

Joined Jan 30, 2016
1,500
The gas valve solenoid holding voltage is about 6 millivolts and it releases at about 3 millivolts. Max voltage output of the thermocouple is about 32 millivolts. The gas valve is a White-Rodgers 36C03. I'd like the LED to go on and off at about 8 millivolts.
Not sure where you got that info from. The 36C03U uses a H-06 thermocouple that generates 30mV approx at 1400degF and 22mV at 800degF. You cannot connect anything else to the thermocouple without potentially preventing it working correctly; How were you measuring the voltages you quoted?
 

Thread Starter

SmokerinCB

Joined May 13, 2021
10
Good morning all ..... schmitt trigger .... the purpose of the thermocouple is to hold open the solenoid valve that provides gas to the pilot light ... nothing more than that .... pretty basic ..... Audioguru again ..... I noticed that and I've seen circuits where two, and sometimes three, op amps are connected in series with the last one switching an LED .... Irving .... I got my data from a reading the thermocouple output with a digital multimeter .... it's not the original thermocouple .... it's one that I picked up at my local hardware store ..... it meets the requirements of my gas valve and it works well ..... but one thing that I have not yet done is to determine the maximum load I can put on that thermocouple before the solenoid valve releases ..... I need to do that and I'll do that today ..... question, how much load will an op amp put on that thermocouple? ..... I don't know that answer ....
 

Irving

Joined Jan 30, 2016
1,500
question, how much load will an op amp put on that thermocouple? ..... I don't know that answer ....
Very little, but its going to need some signal conditioning as its going to be quite noisy. But maybe that's the the wrong question to ask. How are you going to connect to the thermocouple when it's mated to the gas valve as its a sealed device?

Also, not wanting to be a nanny here - you would be completely invalidating any insurances or warranty you have as you are interfering with a safety device. In the UK at least this would be illegal and I'm pretty sure that would be true in the US too...
 

Thread Starter

SmokerinCB

Joined May 13, 2021
10
Good afternoon all ..... sghioto and Irving..... thanks for the info on the op amp load .... that makes this project a little easier .... as for the noise issue .... would a capacitor across the thermocouple output fix that ? ..... I'm going to need some help with that part of this project .... well, I'm going to need help with the whole thing ..... as for connecting to the thermocouple, that's going to be easy ..... I found this thing called a thermocouple adaptor ..... p/n TPI A115 .... got it on the internet .... it's a little more than an inch long ... the male end screws into the gas valve where the thermocouple would be screwed in .... and the thermocouple screws into the female end of the adaptor .... it's real purpose in life is to be used by those in the heating industry to determine thermocouple output in an installed system without having to take things apart .... it has test points for a technician to connect a meter to which I plan on using to connect my circuit to ..... the test point is the plus side of the thermocouple output ..... the negative side is the copper tube covering the positive lead ..... when the thermocouple is screwed into the gas valve, the gas valve becomes part of the negative side .... I'll just attach the circuit to a screw in the gas valve ..... as for the safety issue, I won't be adding any power to the thermocouple but I will be taking a very small amount out .... if I take too much power out of the thermocouple then the gas valve solenoid will loose it's holding current and will shut off the gas to the pilot light and, also, to the burner .... and the fire goes out ..... also, when I use my smoker, it's outside on the driveway ..... so I'm OK with this set up ..... but I appreciate your concern ..... thank you ....
 

Irving

Joined Jan 30, 2016
1,500
I found this thing called a thermocouple adaptor ..... p/n TPI A115 .
Well you learn something new every day! Never come across one of those before, but that's a perfect solution.

Is your 12v supply for the thermometers floating or common to the smoker 'ground'?

So here's my suggested solution...
I'm using a instrumentation amplifier, much better than op-amps for low-mV requirements and completely stable, compared to a 2 or 3 stage op-amp solution. This is configured (by R5) for a gain of 250, giving an output of 0 - 8.2V across the thermocouple range and about 2v at your 8mV operating point. RV1 allows you to set a trip point from 0 - 23mV.

If the ground (-ve) side of the thermocouple and the ground side of your 12v supply are common and well tied together (i.e. a definite 0v between them from a DC and AC/RF perspective) then you can possibly omit R4, C4 and C5 and connect pin 2 of U1 directly to ground.

While you can build this on perf-board a proper double-sided PCB would be preferable.

1621771802644.png
 
Last edited:

Thread Starter

SmokerinCB

Joined May 13, 2021
10
Irving ..... good morning ..... I'm glad that you are happy with my little find .... the thermocouple adaptor .... and, thank you very much for the circuit ..... I appreciate your bent toward simplicity ..... as for my 12 volt supply, at present it's floating ..... I've had no need to ground it ..... but there's no reason that I can't connect the negative side to to my smoker ground ..... I would attach it directly to the gas control valve housing ..... in close proximity to the thermocouple adaptor .... that should ensure a good AC/DC/RF ground .... as for construction, I will use perf-board .... I don't have the materials or skills to make a double-sided PCB ..... oh, should I be concerned about shielding ? ..... I can see that I'm going to have to order some parts ...... I'll do that today .... thanks again .... I'll be out of town for about a week but I should have some kind of progress report late next week ......
 

Thread Starter

SmokerinCB

Joined May 13, 2021
10
Irving ..... good evening ..... parts should be in maybe Saturday .... had a hard time accumulating three 470 nanofarad caps within 1% .... but I did .... so we're good to go .... also it seems that AD623AN may be obsolete ... atleast the's what my first choice for a parts supplier told me ... and no "plug and play" alternative ..... so I went to vendor B and got one .... actually I ordered two .... building my inventory .... anyway .... thinking that I'd really like to have two LEDs telling me the status of the pilot light .... red is pilot light out and green is good .... so I found a LCC 120 in my parts box .... thinking about putting that in your circuit where the LED is .... will have to change the resistor value .... and using that part of your circuit to switch the SSR .... the NC pins would go to the red LED and the NO pins would go to the green LED .... thought and comments, please .....
 

Irving

Joined Jan 30, 2016
1,500
Strange, AD623ANZ is in stock at all my suppliers, 1000's+ of them. Certainly not obsolete, though getting on in years, but still my 'go to' for general use. We buy them by the 100 each year for student experimentation into muscle sensing eg build your own heart monitor (ECG) or control a TV remote by flexing a bicep (EMG).

The LCC120 is one option for a Red/Green output, though overkill (unless you wanted to drive something else as well). I'd use a red/green bidirectional LED (or just wire red and green LEDs in parallel, one reversed, i.e anode of red to cathode of green and vice versa). Connect the cathode green/anode red to the output of the 311 and the other ends to the junction of a 620ohm and a 910ohm resistor wired from +12v to ground. When output of 311 is low (input > set point) the green will light and when high (input < set point) red will light.
 
Last edited:

Thread Starter

SmokerinCB

Joined May 13, 2021
10
Irving .... good morning .... sorry about my delay in responding .... life, as it is wont to do, often side-tracks the best laid plans of mice and men .... I've got a three week sailing trip coming up .... as part of the crew ..... and preparation is taking up more of my time than I expected that it would ..... as to my AD623 problem ..... I neglected to enter the complete part number ..... I didn't enter the suffix ..... my bad .... I appreciate your simplified solution to my red / green LED output issue .... I like that ...... and I've got the parts in hand ....... I will incorporate that change ..... and, yes, the parts that I had to order are in ..... is the AN623 that ESD sensitive ? ..... it came triple wrapped ..... I'm working in my garage, with the door open, and the humidity here in Florida is in the 60's plus ..... I hope to be able to, atleast, breadboard the circuit before I depart ..... will report back as progress is made ....
 

Irving

Joined Jan 30, 2016
1,500
Irving .... good morning .... sorry about my delay in responding .... life, as it is wont to do, often side-tracks the best laid plans of mice and men .... I've got a three week sailing trip coming up .... as part of the crew ..... and preparation is taking up more of my time than I expected that it would ..... as to my AD623 problem ..... I neglected to enter the complete part number ..... I didn't enter the suffix ..... my bad .... I appreciate your simplified solution to my red / green LED output issue .... I like that ...... and I've got the parts in hand ....... I will incorporate that change ..... and, yes, the parts that I had to order are in ..... is the AN623 that ESD sensitive ? ..... it came triple wrapped ..... I'm working in my garage, with the door open, and the humidity here in Florida is in the 60's plus ..... I hope to be able to, atleast, breadboard the circuit before I depart ..... will report back as progress is made ....
You're welcome, happy to help.

The AD623 is a bipolar device so not overly ESD sensitive but doesn't hurt to take reasonable precautions. As I said previously we use them for student projects and haven't noticed any significant failures from ESD.

It was 70% humidity overnight here in London but has dropped to about 60% and is 25degC (75degF) now.

Hope all goes to plan, enjoy the sailing!
 

Thread Starter

SmokerinCB

Joined May 13, 2021
10
Irving .... good morning ..... well, it is still morning here ..... got a window of opportunity to breadboard your circuit .... it works well ..... thank you .... I've got RV1 adjusted to trip at 4 mV .... I like it ...... however, I couldn't get your suggested red / green LED circuit to work .... I tried every possible configuration of LED and resistor ..... so I'm back to your original one LED circuit ...... then I made some voltage and polarity measurements ..... with that one LED, and R2, in the circuit ...... with 12 volts applied to the circuit and the thermocouple cold I put the DVM meter plus lead on pin 7 of the LM311 and the negative lead on ground and I got + 10.4 volts ..... between +12 and pin 7 I got -.64 volts with with meter plus lead on +12 and the negative lead on pin 7 ..... reversing the meter leads I got +.35 volts ..... then I heated the thermocouple up to get the circuit to switch and took more readings ..... with the meter plus lead on +12 and the negative lead on pin 7 I got +11.7 volts .... between pin 7 and ground I got +.29 volts with the plus lead on pin 7 and the negative lead on ground .... reversing the leads I got -.49 volts ..... I've spent a number of hours on the internet searching for an acceptable LED circuit but no luck ..... I'm feeling like I might have to go back to my LCC 120 idea ..... any and all suggestions will be appreciated ....
 

Irving

Joined Jan 30, 2016
1,500
My bad, I forgot the LM311 only pulls down, it doesn't pull up. This arrangement should work (I think I've got the colours the right way round!).

Edited,
1623672172418.pngEdited to increase R8 to 100k.
 
Last edited:

Thread Starter

SmokerinCB

Joined May 13, 2021
10
Irving ..... good afternoon ..... thanks for the circuit ..... I don't have any 2N3904s but I do have a few BC337s .... a little overkill but they're in my parts box ..... and I noted your increase of R8 to 100k ..... will make that change .... don't know when I'll be able to get back to this but maybe tonight .... I'll keep you informed .... stay tuned .....
 
Top