Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Register Log in
  • Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

zero-volt-drop solid-state switch for thermocouple 4

Status
Not open for further replies.
RyreInc

RyreInc

Electrical
Apr 7, 2011
205
0
16
US
I need to switch between 4 thermocouple inputs every few seconds so we can use one thermometer. First thought is some 555-type ICs and some electromechanical relays. I desire to avoid big mechanical parts, so a solid-state solution is preferred, but a SSR, BJT, or FET would have a voltage drop.

1) Is there a solid state relay or opamp that will precisely follow the input voltage, and allow me to switch between 4 outputs?

2) Will a thermometer designed to take J-type thermocouples work with a low-impedance buffered signal from an opamp?

Thanks!

 
Replies continue below

Recommended for you

Sort by date Sort by votes
That does not compute. Relays are not perfect conductors. If you're measuring voltage, then you're presenting a high impedance load, and the series resistance of a saturated FET should be insignificant.

And if you're really worried about it, why not throw in a gain stage before the switches?

TTFN
faq731-376
7ofakss
 
Upvote 0 Downvote
I think that you'd want a proper thermocouple switching system, perhaps from a company such as Omega.

Voltage drop isn't really the issue (the current can be arbitrarily low). The issue is introducing another half dozen unintended thermocouples into the signal path. A correctly designed purpose built thermocouple switch would be carefully design with appropriate materials to minimize unwanted offsets.
 
Upvote 0 Downvote
So true! But, if you want to roll your own, a simple 4066 4x analogue gate works well. The thermocouple output is unipolar so you don't need to go to DG switches.
Also, it is a one or two dollar component. So it doesn't cost much to learn from that mistake.

Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 
Upvote 0 Downvote
I would convert the signals to 4-20 mA and switch the 4-20. No hesitation. 4-20 mA signals will overcome quite a few Volts of drop and maintain accuracy.

Bill
--------------------
"Why not the best?"
Jimmy Carter
 
Upvote 0 Downvote
But, Bill - that hurts. That is what anyone would do. Making it as cheap as possible is the name of the game here - I think. :) :) :)

Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 
Upvote 0 Downvote
Thanks everyone for alerting me to the potential troubles of the route I'm taking!

The 4066 is the exact type of answer I was looking for (and as a bonus we appear to have some in stock too), thanks Skogsgurra.

Yes, the goal is to make this as cheap and simple as possible, but I will look into these other solutions. (The MiniMux looks neat but far too expensive. The HC4052 apparently needs a binary-coded input to select channels, which is not ideal.)

This circuit will be physically close to the TC reader, and at the same ambient temperature.

If I use the 4066, would an input resistance of 250 ohms be large enough? Or will I need to insert an opamp beforehand?
 
Upvote 0 Downvote
250 ohms! That sounds more like an 4-20 mA receiver. If that is the case, you already have a current output transducer and you can use any switch or relay to switch between the channels.

Are you sure that you only have a thermocouple? Or is it a thermocouple with a signal conditioner? The former is a pair of wires soldered together while the latter is a device with terminals on it.

Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 
Upvote 0 Downvote
250 ohms is the stated ON resistance of the 4066. I am sure we have thermocouples only, no signal conditioner.

I'm looking at the 4016 now, per the 4066's datasheet: "The MC14066B is designed to be pin–for–pin compatible with the MC14016B, but has much lower ON resistance."

Both 4066 and 4016 ON resistances are quite low--am I misinterpreting this parameter?

Either way 250 ohms should be plenty high for a TC measuring 4 ohms, right?
 
Upvote 0 Downvote
Just one last comment regarding reading thermocouples. In my experience, I've used a variety of instruments to measure TC voltages, and have never been very impressed with their performance under (what I'd consider normal) operating conditions, i.e. when the temperature environment of the reader must vary from the 68F/20C norm of the producer's laboratory. Portable/handheld readers have trouble with the tempco of gain setting resistors, and the linearity of their cold junction compensation across any temperature change, even if the temperature is allowed to equalize for up to 8 hours. One of the worst performers I've found is the classic Fluke 2-channel TC reader (go figure, right Skogs?). Whenever the accurate reporting of temperature is required, I've never gone too far wrong by grabbing some ice from the lunchroom freezer, and a styro cup with a little water, to form an ice junction. Put one of your TC's in said cup and read "simultaneously" with the other TC's on the multiplexor, and you'll have at least one point of sanity.
 
Upvote 0 Downvote
Ryre
Sorry, did you mean channel resistance when you wrote input resistance?

Channel resistance is not a problem if your actual input resistance is 1 or 10 Mohms, typical values. The voltage drop would less than one thousandth or one tenthousandth of the TC voltage.

If your input resistance is low, you cannot use the 4066.

BTB
I see what you are referring to. I seldom use TC for low temperatures. NTC or Pt100 there. NTC mostly. They are quite good in my applications (winding temperatures and such).

Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 
Upvote 0 Downvote
The main reason thermocouple thermometers are usually significantly more expensive than RTD thermometers is that the contain an RTD thermometer to provide temperature compensation for the reference junction which is inside the thermometer. A thermocouple thermometer can only measure the difference in temp from the reference junction. So if it doesn't know the reference temp., it can't know the measured temp.
 
Upvote 0 Downvote
Skogsgurra, yes I did mean channel resistance. I was confused since that is the only impedance listed in the datasheet.

Input resistance is measured to be about 2 megaohms, so it shouldn't be an issue.
 
Upvote 0 Downvote
Yes, all set--thank you!

The 4066 seems like it'll work--temperature readings are close enough since we don't need absolute precision, and the IC will be at ambient temperature.
 
Upvote 0 Downvote
One more observation: There can be some crosstalk in 4066 from the control signals to the output, especially during the switching times. According to the data sheets, this may be roughly in the range 10-100mV. This may appear as spikes and/or as a small offset in the output (output voltage slightly different from the input voltage). ( I learned this in the hard way...)
 
Upvote 0 Downvote
Yes, there is a picocoloumb charge transfer on every switching edge. But that will not be a problem if the source or sink resistance is low (source resistance in this case). And it will not be noticed in a system where you do not switch very often and where a display is used to read temperature.

Gunnar Englund
--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.
 
Upvote 0 Downvote
Status
Not open for further replies.

Similar threads

jabba99
  • Locked
  • Question
replicating an old circuit- unexpected voltage drop
2
Replies
23
Views
102
IRstuff
IRstuff
Moonwalker031
  • Locked
  • Question
Zero-Cross detect circuit
Replies
6
Views
28
sunnysky
sunnysky
RyreInc
  • Locked
  • Question
Three single-phase SSRs in place of one three-phase, with zero-cross
Replies
1
Views
11
k75ks
k75ks
Costin Ruja
  • Locked
  • Question
Precision zero crossing for mains metering 2
Replies
3
Views
22
IRstuff
IRstuff
xtal01
  • Locked
  • Question
min NiMH charge voltage and extra parts
Replies
1
Views
36
EnzoAus
EnzoAus
Back
Top