High temp thermometer.

[Syd Rumpo]

On 30/05/2012 16:43, Jamie wrote:

other issues with thermocouple process units is they need to operating
in the same ambient temperature as the device the probe is connected too.
The cold junction reference will miss calculate otherwise.

I don't think so. The 'cold' junction is inside the thermocouple
readout device and its temperature is measured using an absolute device
such as a thermistor, maybe a PRT. The 'cold' junction will often be
made from fairly substantial material to damp any fast temperature
changes and maintain an even temperature (an isothermal block) and to
couple easily to the thermistor.

As long as your thermocouple wire is maintained all the way to the
readout, either with thermocouple wire or appropriate compensating
cable, and providing the connectors are of the correct material, then
all is well.

Cheers

 

[Syd Rumpo]

On 30/05/2012 18:11, Phil Hobbs wrote:

Thermocouple extension wire is normally what's used for longer runs,
because it's cheaper. However, its composition isn't as accurately
controlled as real thermocouple wire, so variations in ambient
temperature along the run can cause errors.

But the main problems with TCs are low sensitivity to temperature and
high sensitivity to EMI.

Cheers

Phil Hobbs

Hi Phil

I've seen plenty of thermocouples connected over several tens of metres
of unscreened cable in *very* noisy industrial environments working well.

Of course long integration times help, but what is often overlooked is
that while their output is small in voltage terms, the source impedance
is very low.

Cheers

 

[Spehro Pefhany]

Thermocouple extension wire is normally what's used for longer runs,
because it's cheaper. However, its composition isn't as accurately
controlled as real thermocouple wire, so variations in ambient
temperature along the run can cause errors.

The temperature differences are typically small in relation to the
"hot" junction vs. ambient, so the errors from extension wire are
minimal. For base metal thermocouples, it's the same alloys that are
used (the limits of error specs are very very old and conservative).
Not so for precious metal thermocouples, which are another beast (very
low output, not necessarily even monotonic output with temperature).

But the main problems with TCs are low sensitivity to temperature and
high sensitivity to EMI.

Cheers

Phil Hobbs

With proper design it's not a problem unless you're going for silly
resolutions. This isn't like your high bandwidth high Z low noise
systems- we typically can LPF the heck out of it. You can't and have
any useful signal left.

 

[Spehro Pefhany]

Thermocouples aren't sensitive to EMI!

RTDs are more sensitive to EMI. You can get substantial self-heating
from a Ge RTD when RFI gets into it.

 

[Jamie]

On 30/05/2012 16:43, Jamie wrote:




I don't think so. The 'cold' junction is inside the thermocouple
readout device and its temperature is measured using an absolute device
such as a thermistor, maybe a PRT. The 'cold' junction will often be
made from fairly substantial material to damp any fast temperature
changes and maintain an even temperature (an isothermal block) and to
couple easily to the thermistor.

As long as your thermocouple wire is maintained all the way to the
readout, either with thermocouple wire or appropriate compensating
cable, and providing the connectors are of the correct material, then
all is well.

Cheers

right, and if the cabinet is 75F+ hotter than outside.. it changes.
I know, we deal with that all the time, if the cabinet cooling isn't
working, in other words, running hot inside verses out side where the
J cables and probes are. there is a differential that creates and error.

There is a need in the summer to calibrate the process controllers
when temps are high in the cabinets, mostly due to the season and
working conditions. All the techs do is simply bring a unit on a roll
cart out and get the offset correction needed for each unit in the
panel. You must remember that much of the couple lead wire is also in
this panel getting heated up.

Yes, we know the cabinets should be cooled, but its not always an option.

Jamie

 

[Jamie]

Thermocouples aren't sensitive to EMI!

No, but the front ends are and they carry RF very nicely.


Jamie

 

[Jamie]

RTDs are more sensitive to EMI. You can get substantial self-heating
from a Ge RTD when RFI gets into it.

RTD's are ok, but don't do well well in high temp apps. At least they
don't suffer from the offset of temp on the two ends.

Jamie

 

[Spehro Pefhany]

Well, it needs good circuit design!

That's why they pay us the big bucks!

 

[Spehro Pefhany]

There is a need in the summer to calibrate the process controllers
when temps are high in the cabinets, mostly due to the season and
working conditions. All the techs do is simply bring a unit on a roll
cart out and get the offset correction needed for each unit in the
panel. You must remember that much of the couple lead wire is also in
this panel getting heated up.

Wouldn't it be more sensible to toss out the poorly performing
controllers?


Best regards,
Spehro Pefhany

 

[Jasen Betts]

BP of ethanol and methanol are 78.37°C/65°C so there's a reasonable
separation. AFAIUI, denaturing uses more than one additive so as to
make the process more difficult to reverse.

close enough to be problematic,
BP of water is 100°C and with by distillation you can't get the water to below
4% in the distillate.

 

[Jamie]

Wouldn't it be more sensible to toss out the poorly performing
controllers?


Best regards,
Spehro Pefhany

it's not the controllers, its the environment.. All controllers have
the same issue of this type.

In one group of controllers they have PTCs on the board for the local
temp to work with the Seebeck calculations for the offset.

THen you have the Seebeck (thermocouple) chips that come with the Cj
with in them. Careful design much be made to make sure this "chip" is
seeing the ambient environment, not some near by hot spot on the board,
in which case,coefficient tables must be tailored per chip, depending
where it sits. Of course, you generally need tables to get very
accurate readings of a standard J/K type any way and this is also due to
the reference not being exact per device.

Good cabinet design dictates that you have balanced air temp moving
through the area of the controllers that matches the out side
environment as close as possible. Most of the time, these types of
controllers are installed away from high heat zones or temperature zones
that are hard to match with the ambient temperature of the cabinet.

Most of the time we don't worry about it because the difference at the
machine may mean 10F offset at the actual probe with temperatures of
750f range. This happens in the summer when cabinets get hot that house
both high power electronics and the process controllers with
insufficient air balancing.

In cases of machines that generate military special products, these
machines must be checked and cataloged on a regular bases and data
attached to the final paper work per job that comes off that machine.
For that, out board process controllers of the same models are used to
plug into the couple, get the real reading, and adjust the offset on the
in board controller to match, the data is them logged.

I guess you have to live it to believe it.

You should try and use your hot air wand some time and heat up a
J/K chip or the PTC, what ever they are using and watch your readings.

Jamie

 

[Spehro Pefhany]

it's not the controllers, its the environment.. All controllers have
the same issue of this type.

No. Some controllers have significantly better compensation than
others. There's more than a 20:1 range, just in industrial class
instruments. Lab stuff can do better again, but it sucks for
industrial use for other reasons.

In one group of controllers they have PTCs on the board for the local
temp to work with the Seebeck calculations for the offset.

THen you have the Seebeck (thermocouple) chips that come with the Cj
with in them. Careful design much be made to make sure this "chip" is
seeing the ambient environment, not some near by hot spot on the board,
in which case,coefficient tables must be tailored per chip, depending
where it sits. Of course, you generally need tables to get very
accurate readings of a standard J/K type any way and this is also due to
the reference not being exact per device.

I've never tried to use that kind of gadget in a controller design,
precisely because of that problem (well, they're also usually too
expensive and single-sourced).

Good cabinet design dictates that you have balanced air temp moving
through the area of the controllers that matches the out side
environment as close as possible. Most of the time, these types of
controllers are installed away from high heat zones or temperature zones
that are hard to match with the ambient temperature of the cabinet.

Most of the time we don't worry about it because the difference at the
machine may mean 10F offset at the actual probe with temperatures of
750f range. This happens in the summer when cabinets get hot that house
both high power electronics and the process controllers with
insufficient air balancing.

That's horrible. It's easy to get 10:1 better with an industrial
plug-in construction, and much better in lab-type instrumentation.

In cases of machines that generate military special products, these
machines must be checked and cataloged on a regular bases and data
attached to the final paper work per job that comes off that machine.
For that, out board process controllers of the same models are used to
plug into the couple, get the real reading, and adjust the offset on the
in board controller to match, the data is them logged.

I guess you have to live it to believe it.

You should try and use your hot air wand some time and heat up a
J/K chip or the PTC, what ever they are using and watch your readings.

Jamie

Okay, it's easy to measure temperature to 0.1°C, the reasons you are
seeing _changing_ errors is that the two cold junctions and the
compensation sensor are not isothermal and/or the sensor is not
tracking changes accurately. That's a sensor and thermal/mechanical
design trade-off.

Of course it's also possible the controllers are fine and that someone
made an error wiring up the panels! I've seen that at least half a
dozen times. I'd be happy to look at the reason for the errors if you
want.



Best regards,
Spehro Pefhany

 

[Jamie]

No. Some controllers have significantly better compensation than
others. There's more than a 20:1 range, just in industrial class
instruments. Lab stuff can do better again, but it sucks for
industrial use for other reasons.




I've never tried to use that kind of gadget in a controller design,
precisely because of that problem (well, they're also usually too
expensive and single-sourced).




That's horrible. It's easy to get 10:1 better with an industrial
plug-in construction, and much better in lab-type instrumentation.




Okay, it's easy to measure temperature to 0.1°C, the reasons you are
seeing _changing_ errors is that the two cold junctions and the
compensation sensor are not isothermal and/or the sensor is not
tracking changes accurately. That's a sensor and thermal/mechanical
design trade-off.

Of course it's also possible the controllers are fine and that someone
made an error wiring up the panels! I've seen that at least half a
dozen times. I'd be happy to look at the reason for the errors if you
want.



Best regards,
Spehro Pefhany

Guess you don't get it. What ever, I had the same argument with QC at
work over this, me and about 10 others joining in. We had to actually
show them the errors of their way.

They too, could not believe what they were seeing, all they could
throw at you was, what they learned in college and to them, that was
enough, but in the real world that does not apply because college does
not teach you about reality. I too learned the same thing at college
years ago, however, experience has served the better way and I do fully
understand why this happens.

There are reasons why abs devices like R type RTD's are used over the
seebeck method when possible. It's obvious to me.

Jamie

 

[Spehro Pefhany]

Guess you don't get it. What ever, I had the same argument with QC at
work over this, me and about 10 others joining in. We had to actually
show them the errors of their way.

They too, could not believe what they were seeing, all they could
throw at you was, what they learned in college and to them, that was
enough, but in the real world that does not apply because college does
not teach you about reality. I too learned the same thing at college
years ago, however, experience has served the better way and I do fully
understand why this happens.

As part of decades of work as an instrument designer, I've worked with
literally hundreds of installations involving temperature control,
including troubleshooting. The deeply theoretical work I did in
University is not particularly useful in this kind of situation- it's
more critical thinking with a bit (okay, quite a lot) of hands-on
Engineering knowledge. I've also designed a number of lines of
temperature controllers, digital, analog and processor-based, so I do
know what I speak of and I know where the instrument specifications
come from (since I wrote them!). The one I'm working on now is capable
of doing (censored) single digit MICRO degrees stability over hours.
It's tested in an environmental chamber over a wide range.

I'm not sure whether you're using a bad instrument/sensor, misusing a
good instrument/sensor, or misinterpreting the specifications, but
something bad is clearly going down at your place and it sounds like
it's costing money.

There are reasons why abs devices like R type RTD's are used over the
seebeck method when possible. It's obvious to me.

Jamie

"R type" refers to a thermocouple (Platinum/Platinum-Rhodium 13%),
useful mostly at very high temperatures. I doubt you have or want
those, they have very tiny output (< 10uV/K vs. more like 55 for type
J).

An RTD is often used at relatively low temperatures when you need
better than a degree C or two stability, and response speed (which
impacts control robustness and response to disturbances) and
mechanical ruggedness are relatively unimportant. The electronic
design for 3 wire RTDs or 4 wire RTDs is quite straightforward.

So, why not throw out the thermocouples and replace them with RTDs?

This sounds like a seriously bad situation you have, honestly.


Best regards,
Spehro Pefhany

 

[John Devereux]

it's not the controllers, its the environment.. All controllers have
the same issue of this type.

In one group of controllers they have PTCs on the board for the local
temp to work with the Seebeck calculations for the offset.

THen you have the Seebeck (thermocouple) chips that come with the Cj
with in them. Careful design much be made to make sure this "chip" is
seeing the ambient environment, not some near by hot spot on the
board, in which case,coefficient tables must be tailored per chip,
depending where it sits. Of course, you generally need tables to get
very accurate readings of a standard J/K type any way and this is also
due to
the reference not being exact per device.

Good cabinet design dictates that you have balanced air temp moving
through the area of the controllers that matches the out side
environment as close as possible. Most of the time, these types of
controllers are installed away from high heat zones or temperature
zones
that are hard to match with the ambient temperature of the cabinet.

Most of the time we don't worry about it because the difference at the
machine may mean 10F offset at the actual probe with temperatures of
750f range. This happens in the summer when cabinets get hot that
house both high power electronics and the process controllers with
insufficient air balancing.

The compensation circuit needs to measure the temperature at a single
specific place: the connector used to terminate the thermocouple (or
extension) wires. So it is not a matter of "seeing the ambient
environment" or keeping the cabinet the same temperature as the
environment. Although this may help of course. There is no reason why a
PCB mounted sensor located at the thermocouple connector cannot be
within a degree or so of the true value.

 

[Spehro Pefhany]

Connecting the extension wires backwards, or using the wrong kind,
would be interesting. He could get some freeze spray and spritz all
the junctions in the cabling, and see what changes.

Backwards will give an error of roughly double the temperature
difference. J-K errors are a bit more subtle- about 20% of the
temperature difference shows up as an error.

It's complicated by the balkanized color codes used worldwide for
T/Cs. If you have an Italian machine bought in the US with Japanese
controls just about anything might be in there.

Helps to have a magnet handy, except for T, S, R, B .

 

[Jamie]

Yeah, he could have the extension wires wrong. The color codes *are*
awful.

You know what, I've work with this shit long enough to know the
difference on wire connections. You guys act like you never leave the
controlled environment. Red = (-), white (+) for J and yellow (+) For K
etc...

It would be nice if you had a work place that could maintain
temperatures inside and out at a even level, but since you obviously
don't work in harsh environments, you wouldn't know.

I'll give you a classic example, we have these unit located in various
places to test polyrad after it has been irradiated. This is to insure
it got irradiated and the process didn't fail. The unit has a fire rode
heater mounted in a 4x4 basic electrical box that has a heat shield
cover the operator can lift, this box also is mounted to a NEMA 4
enclosure that houses the heat controller. after some operating time,
the side of the enclosure gets hot and it heats up the inside and the
process controller gets hotter than the ambient temperature where the J
couple cable is leading, which is a J-Jack on the outside.. a 10+ Bias
must be applied to all these units and the units have to be on for a
while before they are accurate.

We have units that use various venders of controllers to perform this
job from all parts of the world. They all seem to behave the same.

When these units are cold and the firerod reaches temp, they read
correctly, if you have the offset set to 0. when the enclosure gets hot
after an hour of so of just sitting there, which is what they do, all
day, all week etc.. the offset needs to be applied. If we open the door
to the enclosure and let it vent to get balanced, it then does not need
the offset.

The product developing lab has the same issues with controllers that
deionize water. Some of us call them lobster pots, but the controller
sits just below the tank of water and it gets up to 95C, an offset on
those must be applied, too. If you attached a heat controller and let it
sit on the table where It does not get any hotter than the surrounding
environment, then it is with in the spec with 0 offset.

This is all basic stuff to me, I don't understand why it is such a
hard concept for others.

jamie

 

[Ken S. Tucker]

dynamics (at) uniserve.com

business address is
http://www.trak4.com/earco/index.html

Done.[/QUOTE]

Thanks kindly. In other posts, I gather, the subject is your specialty.

A still?

That would be a kind of bench mark (hiccup).
Sold as a "potato purifier" that eliminates Ethyl at a quart
per day might be marketable.

Best regards,
Spehro Pefhany

Cheers
Ken

 

[Ken S. Tucker]

How accurate? What sort of specification did you have in mind?


Some thermocouples do.


But you can't be bothered to say how accurate.


I've interacted with Ken S. Tucker before, and I know that he isn't
going to be ambitious about accuracy. It's useful information in this
context. You may see it as an insult - you do seem to see a lot of
insults.
Bill Sloman, Nijmegen

One problem is that temperature is an ill defined quantity/quality,
much more complex than is usually understood, likely an obsolete
term in General Relativity and Quantum Mechanics, having no general
consensus. The term 'temperature' applies to a bulk of material.
Regards
Ken

 

[Jamie]

That's the American convention. There are different colors for europe
and, I think, Japan.

Obviously something is wrong in your system if you have to recal just
because the cabinet temp changes.

As far as getting out goes, I have laid on asbestos matting on the top
of a live steam valve on a 32,000 horsepower turbine in the middle of
the Gulf of Mexico, in mid-summer, trying to turn a trimpot. I once
enjoyed a 2-day cruise on a container ship down the West coast,
finally locating a loose screw on a terminal block. I don't think
Spehro or I lack experience getting dirty in the field.





Don't get obnoxious without knowing what you're talking about. It can
make you look silly.

You know, maybe JT is correct about you. I really thought you were
better than that. He may have been taking an off sided approach to it
but I can see he made some good points about you.

You do not understand what you think you do about the subject. And
maybe other things you could have been sucking off the group over the years.

My how things come out in the strangest times, you should have kept
quiet.

One day, you'll find yourself in this same situation, and look back
however, by the time you realize that I could of been correct, and I am
in this case, it won't make any difference.

To put it bluntly, you are mistaken this time. You have been a few
other times also. I just did the courteous thing of sitting back and
watching.

Have a glorious time sucking information from others. It seems from
what I've seen here lately, you are good at that. Put out danglers and
have others refine it for you. what a way to get your work done for you
for free. In my eyes, that isn't much different than the free loaders we
have in our society.

If that is what you expect from others? "Feed the pig" so be it.
I feel sorry for some here that have been nice enough to offer you
suggestions that actually helped you. Unfortunately, they too, will be
one day, approach by the likes of you in the same manner. I can only
hope they learn or have already learned their lesson.


There is nothing like actually experiencing life and duties over just
sitting there and theorizing them in your fantasy world.. Does that ring
a bell for you? You really should get out more in the field, on site,
and actually see what goes on out there. It's blurring your vision, but
It does not matter because you've grown accustom to picking fruit off of
what ever tree is in blossom when you need it.


Jamie