Re: [time-nuts] HP5061B Versus HP5071 Cesium Line Frequencies
Hi
Any real crystal you buy will have a tolerance on the angle. In the case of a crystal cut for turn
the temperature will be a bit different and you will match your oven to it. If you attempt a zero
angle cut, you will never really hit it and there is no way to compensate for the problem.
Bob
On Jun 2, 2017, at 3:19 PM, Donald E. Pauly trojancowboy@gmail.com wrote:
A cut at that angle has no turn over temperature. The zero temperature coefficient point is 25°. Its temperature coefficient everywhere else is positive.
On Friday, June 2, 2017, Bob kb8tq <kb8tq@n1k.org mailto:kb8tq@n1k.org> wrote:
Hi
If you are going to use an oven, it’s better to run it at the turn temperature of
the crystal. That would put you above 50C for an AT and a bit higher still for an SC.
Bob
On Jun 2, 2017, at 2:09 PM, Donald E. Pauly <trojancowboy@gmail.com javascript:;> wrote:
https://www.febo.com/pipermail/time-nuts/2017-May/105566.html https://www.febo.com/pipermail/time-nuts/2017-May/105566.html
If we build this circuit it would be a bench model not designed to be
inside a hot chassis. It would be able to lock ± 5° C of 25° C. My
idea of an oven is to keep the crystal and oscillator at 25° C ±0.001
°C with 60 second warm up/cool down time.
πθ°μΩω±√·Γλ
WB0KVV
---------- Forwarded message ----------
From: Bob kb8tq <kb8tq@n1k.org javascript:;>
Date: Fri, Jun 2, 2017 at 5:57 AM
Subject: Re: [time-nuts] HP5061B Versus HP5071 Cesium Line Frequencies
To: Discussion of precise time and frequency measurement <time-nuts@febo.com javascript:;>
Hi
I would suggest you check a few real crystals over the 20 to 40C range ….
With all the “stuff” in a 5061, it will change (rise) at least 10C
after turn on.
Bob
time-nuts mailing list -- time-nuts@febo.com javascript:;
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
That is not true. I say that thermal coolers have made ovens obsolete. A
zero temperature coefficient at room temperature is easier to hit than a
zero temperature at the upper turnover point when such a thing exists. See
curve 0 in Figure 6 at https://coloradocrystal.com/applications/ .
πθ°μΩω±√·Γ
WB0KVV
On Friday, June 2, 2017, Bob kb8tq <kb8tq@n1k.org
javascript:_e(%7B%7D,'cvml','kb8tq@n1k.org');> wrote:
Hi
Any real crystal you buy will have a tolerance on the angle. In the case
of a crystal cut for turn
the temperature will be a bit different and you will match your oven to
it. If you attempt a zero
angle cut, you will never really hit it and there is no way to compensate
for the problem.
Bob
On Jun 2, 2017, at 3:19 PM, Donald E. Pauly trojancowboy@gmail.com
wrote:
A cut at that angle has no turn over temperature. The zero temperature
coefficient point is 25°. Its temperature coefficient everywhere else is
positive.
On Friday, June 2, 2017, Bob kb8tq kb8tq@n1k.org wrote:
Hi
If you are going to use an oven, it’s better to run it at the turn
temperature of
the crystal. That would put you above 50C for an AT and a bit higher
still for an SC.
Bob
You have a fundamental misunderstanding of the AT curve family. See
my QBASIC plot at
http://gonascent.com/papers/hp/hp5061/photos/newxtl.jpg . The
commonly described AT cut is shown as the largest sine wave in the
blue rectangle. The left side of the rectangle is -55°C, the center
is 25° C and the right side is 105° C. The bottom of the rectangle is
-16 ppm and the top is +16 ppm.
Main Cut
Temp Freq
-55° C -16 ppm
-15° C +16 ppm
+25° C ±0 ppm
+65° C -16 ppm
105° C +16 ppm
You can get a lower turnover point of 24° C and an upper turnover
point of 26° C. Their amplitude would be °±0.250 ppb. As the turnover
points approach each other, their amplitude approaches zero. The line
joining all the turnover points is y= -8·x^3. The zero temperature
for 25° is y=4·x^3. Practical tolerance these days is on the order of
0.1 minutes of arc. This is within the width of the traces in the
graph.
You are way off on your 0° to 50° C crystal.
["Umm …. errr … it’s quite easy to get a +/- 2 ppm 0-50C AT cut
including the tolerance on the cut angle."]
Temp Freq
0° C -0.488 ppb (lower limit)
12.5° C +0.488 ppb (lower turning point)
25° C ±0
37.5° C -0.488 ppb (upper turning point)
50° C +0.488 ppb (upper limit)
As I claimed, a Thermal Electric Cooler has never been used to build a
crystal oscillator. In the 50s, TEC efficiencies were on the order of
1% and were useless. The Soviets made coolers more practical in the
70s with better materials. I saw one used at Telemation that was able
to measure dew point by condensing water vapor on a mirror. It looks
like efficiencies have now improved to 33% or so.
It was only in the early 70s that Analog Devices invented the AD590
solid state temperature sensor. It made thermister bridges obsolete.
Switching amplifiers are required to drive thermal coolers if you want
to preserve efficiency.
πθ°μΩω±√·Γλ
WB0KVV
---------- Forwarded message ----------
From: Bob kb8tq kb8tq@n1k.org
Date: Fri, Jun 2, 2017 at 12:22 PM
Subject: Re: HP5061B Versus HP5071 Cesium Line Frequencies
To: "Donald E. Pauly" trojancowboy@gmail.com
Cc: "rward0@aol.com" rward0@aol.com, time-nuts time-nuts@febo.com
Hi
Any real crystal you buy will have a tolerance on the angle. In the
case of a crystal cut for turn the temperature will be a bit different
and you will match your oven to it. If you attempt a zero angle cut,
you will never really hit it and there is no way to compensate for the
problem.
Bob
On Jun 2, 2017, at 3:19 PM, Donald E. Pauly trojancowboy@gmail.com wrote:
A cut at that angle has no turn over temperature. The zero temperature
coefficient point is 25°. Its temperature coefficient everywhere else
is positive.
On Friday, June 2, 2017, Bob kb8tq kb8tq@n1k.org wrote:
Hi
If you are going to use an oven, it’s better to run it at the turn temperature of
the crystal. That would put you above 50C for an AT and a bit higher still for an SC.
Bob
On 6/3/17 9:56 PM, Donald E. Pauly wrote:
It was only in the early 70s that Analog Devices invented the AD590
solid state temperature sensor. It made thermister bridges obsolete.
There is a difference between something like a platinum resistance
thermometer (PRT or RTD) and a thermistor, but they both are "measure
resistance to measure temperature" devices.
Yes, the AD590 is a useful part (I've got some in a device being
launched in August), but PRTs,thermistors, and thermocouples are still
widely used.
I don't know that the inherent precision (at room temperature)of the
various techniques is wildly different. A 1mV/K signal (AD590 into a 1k
resistor) has to be measured to 0.1mV for 0.1 degree accuracy. That's
out of 300mV, so 1 part in 3000
A type E thermocouple is 1.495 mV at 25C and 1.801 at 30C, so about 0.06
mV/K slope. Measure 0.006mV for 0.1 degree (plus the "cold junction"
issue). 1 part in 250 measurement.
Modern RTDs all are 0.00385 ohm/ohm/degree at 25C. Typically, you have
a 100 ohm device (although there are Pt1000s), so it's changing 0.385
ohm/degree. 1 part in 3000
Checking the Omega catalog.. A 44007 has nominal 5k at 25C, and is 4787
at 26C, so 1 part in 24.
Especially these days, with computers to deal with nonlinear calibration
curves, there's an awful lot of TCs and Thermistors in use. The big
advantage of the AD590 and PRT is that they are basically linear over a
convenient temperature range.
In a variety applications, other aspects of the measurement device are
important - ESD sensitivity, tolerance to wildly out of spec temperature
without damage, radiation effects etc. Not an issue here, but I'll note
that the thermistor, PRT, and thermocouple are essentially ESD immune.
The AD590 most certainly is not.
If you go out and buy cheap industrial PID temperature controller it
will have input modes for various thermocouples and PRTs. I suppose
there's probably some that take 1uA/K, but it's not something I would
expect.
So I wouldn't say thermistor bridges (or other temperature measurements)
are obsolete.
In message 3ca81847-63c4-f803-994d-8e07c9973ba0@earthlink.net, jimlux writes:
Modern RTDs all are 0.00385 ohm/ohm/degree at 25C. Typically, you have
a 100 ohm device (although there are Pt1000s), so it's changing 0.385
ohm/degree. 1 part in 3000
Depending how much money you want to spend, you can also get pt10k
and even pt100k RTD's, to satisfy particular needs for resolution,
self-heating, inductance, mass and the many and varied noises.
And if course, we cannot talk PT100 and fail to repeat the old pun:
"PT100 is the gold standard for temperature measurement"
:-)
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.
Hi
Have you ever tried to actually buy a crystal built to a specification? There is a
tolerance on them. That has a profound impact on what you can buy.
Bob
On Jun 4, 2017, at 12:56 AM, Donald E. Pauly trojancowboy@gmail.com wrote:
You have a fundamental misunderstanding of the AT curve family. See
my QBASIC plot at
http://gonascent.com/papers/hp/hp5061/photos/newxtl.jpg . The
commonly described AT cut is shown as the largest sine wave in the
blue rectangle. The left side of the rectangle is -55°C, the center
is 25° C and the right side is 105° C. The bottom of the rectangle is
-16 ppm and the top is +16 ppm.
Main Cut
Temp Freq
-55° C -16 ppm
-15° C +16 ppm
+25° C ±0 ppm
+65° C -16 ppm
105° C +16 ppm
You can get a lower turnover point of 24° C and an upper turnover
point of 26° C. Their amplitude would be °±0.250 ppb. As the turnover
points approach each other, their amplitude approaches zero. The line
joining all the turnover points is y= -8·x^3. The zero temperature
for 25° is y=4·x^3. Practical tolerance these days is on the order of
0.1 minutes of arc. This is within the width of the traces in the
graph.
You are way off on your 0° to 50° C crystal.
["Umm …. errr … it’s quite easy to get a +/- 2 ppm 0-50C AT cut
including the tolerance on the cut angle."]
Temp Freq
0° C -0.488 ppb (lower limit)
12.5° C +0.488 ppb (lower turning point)
25° C ±0
37.5° C -0.488 ppb (upper turning point)
50° C +0.488 ppb (upper limit)
As I claimed, a Thermal Electric Cooler has never been used to build a
crystal oscillator. In the 50s, TEC efficiencies were on the order of
1% and were useless. The Soviets made coolers more practical in the
70s with better materials. I saw one used at Telemation that was able
to measure dew point by condensing water vapor on a mirror. It looks
like efficiencies have now improved to 33% or so.
It was only in the early 70s that Analog Devices invented the AD590
solid state temperature sensor. It made thermister bridges obsolete.
Switching amplifiers are required to drive thermal coolers if you want
to preserve efficiency.
πθ°μΩω±√·Γλ
WB0KVV
---------- Forwarded message ----------
From: Bob kb8tq kb8tq@n1k.org
Date: Fri, Jun 2, 2017 at 12:22 PM
Subject: Re: HP5061B Versus HP5071 Cesium Line Frequencies
To: "Donald E. Pauly" trojancowboy@gmail.com
Cc: "rward0@aol.com" rward0@aol.com, time-nuts time-nuts@febo.com
Hi
Any real crystal you buy will have a tolerance on the angle. In the
case of a crystal cut for turn the temperature will be a bit different
and you will match your oven to it. If you attempt a zero angle cut,
you will never really hit it and there is no way to compensate for the
problem.
Bob
On Jun 2, 2017, at 3:19 PM, Donald E. Pauly trojancowboy@gmail.com wrote:
A cut at that angle has no turn over temperature. The zero temperature
coefficient point is 25°. Its temperature coefficient everywhere else
is positive.
On Friday, June 2, 2017, Bob kb8tq kb8tq@n1k.org wrote:
Hi
If you are going to use an oven, it’s better to run it at the turn temperature of
the crystal. That would put you above 50C for an AT and a bit higher still for an SC.
Bob
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
I've bought dozens of them over the years and talked to crystal
engineers for tens of hours. I watched them plated and tuned at a
crystal filter company in Phoenix. I own Virgil Bottom's book on the
subject and understood half of it.
πθ°μΩω±√·Γλ
WB0KVV
---------- Forwarded message ----------
From: Bob kb8tq kb8tq@n1k.org
Date: Sun, Jun 4, 2017 at 5:15 AM
Subject: Re: [time-nuts] HP5061B Versus HP5071 Cesium Line Frequencies
To: Discussion of precise time and frequency measurement time-nuts@febo.com
Cc: "rward0@aol.com" rward0@aol.com, "Donald E. Pauly"
trojancowboy@gmail.com
Hi
Have you ever tried to actually buy a crystal built to a
specification? There is a
tolerance on them. That has a profound impact on what you can buy.
Bob
I stand by my remark that thermistors have been obsolete for over 40
years. The only exception that I know of is cesium beam tubes that
must withstand a 350° C bakeout. Thermistors are unstable and
manufactured with a witches brew straight out of MacBeth. Their
output voltages are tiny and are they inconvenient to use at different
temperatures.
Where did you get the idea to use a 1 k load for an AD590? If you run
it from a -5 V supply you can use a 15 k load to a +5V supply. This
gives 15 V/C° output. If you drive it from a 10 Meg impedance current
source, you get 30,000 V/ C°. If I remember correctly, I drove a
power MOSFET heater gate directly in my prototype oven 20 years ago.
It would go from full off to full on in 1/15 ° C. Noise is 1/25,000 °
C in a 1 cycle bandwidth.
The room temperature coefficient of an AT crystal is -100 ppb per
reference cut angle in minutes. (-600 ppb/C° for standard crystal)
The practical limit in a crystal designed for room temperature is
about 0.1' cut accuracy or ±10 ppb/C°. If you have access to an
atomic standard, you can use feed forward to get ±1 ppb/C°. If the
temperature can be held to ±0.001° C, this is ±1 part per trillion.
This kind of accuracy has never been heard of. Feed forward also
allows you to incorporate the components of the oscillator into the
thermal behavior. It does no good to have a perfect crystal if the
oscillator components drift.
πθ°μΩω±√·Γλ
WB0KVV
---------- Forwarded message ----------
From: jimlux jimlux@earthlink.net
Date: Sun, Jun 4, 2017 at 4:47 AM
Subject: Re: [time-nuts] Fwd: HP5061B Versus HP5071 Cesium Line Frequencies
To: time-nuts@febo.com
On 6/3/17 9:56 PM, Donald E. Pauly wrote:
It was only in the early 70s that Analog Devices invented the AD590
solid state temperature sensor. It made thermister bridges obsolete.
There is a difference between something like a platinum resistance
thermometer (PRT or RTD) and a thermistor, but they both are "measure
resistance to measure temperature" devices.
Yes, the AD590 is a useful part (I've got some in a device being
launched in August), but PRTs,thermistors, and thermocouples are still
widely used.
I don't know that the inherent precision (at room temperature)of the
various techniques is wildly different. A 1mV/K signal (AD590 into a
1k resistor) has to be measured to 0.1mV for 0.1 degree accuracy.
That's out of 300mV, so 1 part in 3000
A type E thermocouple is 1.495 mV at 25C and 1.801 at 30C, so about
0.06 mV/K slope. Measure 0.006mV for 0.1 degree (plus the "cold
junction" issue). 1 part in 250 measurement.
Modern RTDs all are 0.00385 ohm/ohm/degree at 25C. Typically, you
have a 100 ohm device (although there are Pt1000s), so it's changing
0.385 ohm/degree. 1 part in 3000
Checking the Omega catalog.. A 44007 has nominal 5k at 25C, and is
4787 at 26C, so 1 part in 24.
Especially these days, with computers to deal with nonlinear
calibration curves, there's an awful lot of TCs and Thermistors in
use. The big advantage of the AD590 and PRT is that they are basically
linear over a convenient temperature range.
In a variety applications, other aspects of the measurement device are
important - ESD sensitivity, tolerance to wildly out of spec
temperature without damage, radiation effects etc. Not an issue here,
but I'll note that the thermistor, PRT, and thermocouple are
essentially ESD immune. The AD590 most certainly is not.
If you go out and buy cheap industrial PID temperature controller it
will have input modes for various thermocouples and PRTs. I suppose
there's probably some that take 1uA/K, but it's not something I would
expect.
So I wouldn't say thermistor bridges (or other temperature
measurements) are obsolete.
Hi
Ok, when you wrote the specification for your crystals what was the tolerance on the angle
for those crystals? What did the suppliers who quoted to your spec say about the angle tolerance
you specified? When they shipped against your volume requirements how did they do against the
specification? When your incoming QA tested the crystals what did they find? When you put the
crystals into production oscillators and tested the result how did they perform?
Bob
On Jun 4, 2017, at 11:09 AM, Donald E. Pauly trojancowboy@gmail.com wrote:
I've bought dozens of them over the years and talked to crystal
engineers for tens of hours. I watched them plated and tuned at a
crystal filter company in Phoenix. I own Virgil Bottom's book on the
subject and understood half of it.
πθ°μΩω±√·Γλ
WB0KVV
---------- Forwarded message ----------
From: Bob kb8tq kb8tq@n1k.org
Date: Sun, Jun 4, 2017 at 5:15 AM
Subject: Re: [time-nuts] HP5061B Versus HP5071 Cesium Line Frequencies
To: Discussion of precise time and frequency measurement time-nuts@febo.com
Cc: "rward0@aol.com" rward0@aol.com, "Donald E. Pauly"
trojancowboy@gmail.com
Hi
Have you ever tried to actually buy a crystal built to a
specification? There is a
tolerance on them. That has a profound impact on what you can buy.
Bob
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Hi
I think you have thermistors and thermocouples a bit mixed up. You can get
quite substantial output voltages from a thermistor bridge….
Bob
On Jun 4, 2017, at 11:44 AM, Donald E. Pauly trojancowboy@gmail.com wrote:
I stand by my remark that thermistors have been obsolete for over 40
years. The only exception that I know of is cesium beam tubes that
must withstand a 350° C bakeout. Thermistors are unstable and
manufactured with a witches brew straight out of MacBeth. Their
output voltages are tiny and are they inconvenient to use at different
temperatures.
Where did you get the idea to use a 1 k load for an AD590? If you run
it from a -5 V supply you can use a 15 k load to a +5V supply. This
gives 15 V/C° output. If you drive it from a 10 Meg impedance current
source, you get 30,000 V/ C°. If I remember correctly, I drove a
power MOSFET heater gate directly in my prototype oven 20 years ago.
It would go from full off to full on in 1/15 ° C. Noise is 1/25,000 °
C in a 1 cycle bandwidth.
The room temperature coefficient of an AT crystal is -100 ppb per
reference cut angle in minutes. (-600 ppb/C° for standard crystal)
The practical limit in a crystal designed for room temperature is
about 0.1' cut accuracy or ±10 ppb/C°. If you have access to an
atomic standard, you can use feed forward to get ±1 ppb/C°. If the
temperature can be held to ±0.001° C, this is ±1 part per trillion.
This kind of accuracy has never been heard of. Feed forward also
allows you to incorporate the components of the oscillator into the
thermal behavior. It does no good to have a perfect crystal if the
oscillator components drift.
πθ°μΩω±√·Γλ
WB0KVV
---------- Forwarded message ----------
From: jimlux jimlux@earthlink.net
Date: Sun, Jun 4, 2017 at 4:47 AM
Subject: Re: [time-nuts] Fwd: HP5061B Versus HP5071 Cesium Line Frequencies
To: time-nuts@febo.com
On 6/3/17 9:56 PM, Donald E. Pauly wrote:
It was only in the early 70s that Analog Devices invented the AD590
solid state temperature sensor. It made thermister bridges obsolete.
There is a difference between something like a platinum resistance
thermometer (PRT or RTD) and a thermistor, but they both are "measure
resistance to measure temperature" devices.
Yes, the AD590 is a useful part (I've got some in a device being
launched in August), but PRTs,thermistors, and thermocouples are still
widely used.
I don't know that the inherent precision (at room temperature)of the
various techniques is wildly different. A 1mV/K signal (AD590 into a
1k resistor) has to be measured to 0.1mV for 0.1 degree accuracy.
That's out of 300mV, so 1 part in 3000
A type E thermocouple is 1.495 mV at 25C and 1.801 at 30C, so about
0.06 mV/K slope. Measure 0.006mV for 0.1 degree (plus the "cold
junction" issue). 1 part in 250 measurement.
Modern RTDs all are 0.00385 ohm/ohm/degree at 25C. Typically, you
have a 100 ohm device (although there are Pt1000s), so it's changing
0.385 ohm/degree. 1 part in 3000
Checking the Omega catalog.. A 44007 has nominal 5k at 25C, and is
4787 at 26C, so 1 part in 24.
Especially these days, with computers to deal with nonlinear
calibration curves, there's an awful lot of TCs and Thermistors in
use. The big advantage of the AD590 and PRT is that they are basically
linear over a convenient temperature range.
In a variety applications, other aspects of the measurement device are
important - ESD sensitivity, tolerance to wildly out of spec
temperature without damage, radiation effects etc. Not an issue here,
but I'll note that the thermistor, PRT, and thermocouple are
essentially ESD immune. The AD590 most certainly is not.
If you go out and buy cheap industrial PID temperature controller it
will have input modes for various thermocouples and PRTs. I suppose
there's probably some that take 1uA/K, but it's not something I would
expect.
So I wouldn't say thermistor bridges (or other temperature
measurements) are obsolete.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Moin,
This discussion is kind of getting heated.
Let's put some facts in, to steer it away from
opinion based discussion.
On Sun, 4 Jun 2017 08:44:33 -0700
"Donald E. Pauly" trojancowboy@gmail.com wrote:
I stand by my remark that thermistors have been obsolete for over 40
years. The only exception that I know of is cesium beam tubes that
must withstand a 350° C bakeout. Thermistors are unstable and
manufactured with a witches brew straight out of MacBeth. Their
output voltages are tiny and are they inconvenient to use at different
temperatures.
If you really mean thermistors, and not, as Bob suggested thermocouples,
then I have to disagree. The most stable temperature sensors are
platinum wire sensors. The standards class PRT's are the gold standard
when it comes to temperature measurement, for a quite wide range
(-260°C to +960°C) and are considered very stable. They offer (absolute)
accuracies in the order of 10mK in the temperature range below 400°C.
Even industrial grade PRT sensors give you an absolute accuracy better
than 0.1K up to 200-300°C. The "cheap" PT100 are more of the order of 1-10°C
accuracy... all numbers just using a two-point calibration.
For more information on this see [1] chapter 6 and [2] for industrial sensors.
NTC sensors have a higher variablity of their parameters in production
and are usually specified in % of temperature relative to their reference
point, which is usually 25°C. Typical values are 0.1% to 5%. Additionally
there is a deviation from the reference point, specified in °C, which
is usually in the order of 0.1°C to 1°C.
The NTC sensors are less accurate than PT sensors, but offer the advantage
of higher resistance (thus lower self-heating), higher slope (thus better
precision). Biggest disadvantage is their non-linear curve. Their price
is also a fraction of PT sensors and due to that you can have them in
many different forms, from the 0201 SMD resistor, to a large stainless
steal pipe that goes into a chemical tank. NTCs are the workhorse in
todays temperature measurement and control designs.
The next category are band-gap sensors like the AD590. Their biggest
advantage is that their 0 point is fix at 0K (and very accurately so).
Ie they can be used with single point calibration and achieve 1°C accuracy
this way. Their biggest drawback their large thermal mass and large
insulating case, because they are basically an standard, analog IC.
Ie their main use is in devices where there is a lot of convection and
slow temperature change. Due to their simple and and quite linear
characteristics, they are often used in purely analog temperature
control circuits, or where a linearization is not feasible.
But only if price isn't an issue (they cost 10-1000 times as
much as an PTC). Their biggest disadvantage, beside their slow
thermal raction time, is their large noise uncorrelated to the
supply voltage, and thus cannot be compensated by ratiometric measurement.
They are also more suceptible to mechanical stress than NTC's and PT's,
due to their construction. Similar to voltage references (which they
actually are), their aging is quite substantial and cannot be neglected
in precision application.
With a 3 point calibration, better than 0.5°C accuracy can be achieved
(modulo aging) within their operating temperature range, which is
rather limited, compared to the other sensor types.
I don't know enough about thermocouples to say much about them, beside
that they are cumbersome to work with (e.g. the cold contact) and
produce a low voltage (several µV) output with quite high impedance,
which makes the analog electronics difficult to design as well.
With todays electronics, the easiest sensors to work with are NTC and
PT100/PT1000 as most high resolution delta-sigma ADCs have direct support
for 3 and/or 4 wire measurement of those, including compensation for
reference voltage/current variation. Using a uC as control element
also opens up the possibility to linearize the curve of NTCs without
loss of accuracy. Usually measurement precision, with a state-of-the-art
circuit, is limited by noise coupling into the leads of the sensor
and noise in and around the ADC. (see [3-5])
Where did you get the idea to use a 1 k load for an AD590?
Jim was refering to a circuit he used in a satellite. Not to your circuit.
The room temperature coefficient of an AT crystal is -cd 100 ppb per
reference cut angle in minutes. (-600 ppb/C° for standard crystal)
The practical limit in a crystal designed for room temperature is
about 0.1' cut accuracy or ±10 ppb/C°. If you have access to an
atomic standard, you can use feed forward to get ±1 ppb/C°. If the
temperature can be held to ±0.001° C, this is ±1 part per trillion.
This kind of accuracy has never been heard of.
It has been heard of. The 8607 was spec'ed to <2e-10 p-p deviation
over temperature range (-30°C to 60°C). Also, to hold the temperature
stable to 0.001K in a room temperature environment (let's say 10K variation),
you need a thermal gain of >10k. That's quite a bit and needs considerable
design effort. Most OCXO design's I am aware of are in the order of 100
(the DIL14 designs) to a few 1000 for single ovens, to a few 10k for
double ovens. The only exception is the E1938 which achieves >1M.
But that design is not for the faint hearted. I don't remember seeing
any number, but i would guess the 8607 has a thermal gain in the
order of 100k to 1M as well, considering it being a double oven in
a dewar flask.
Also, what do you mean by atomic standard and feed forward?
If you have an atomic standard you don't need to temperature
stabilize your quartz. You can just simply use a PLL to lock
it to your reference and achieve higher stability than any oven
design.
Feed forward also
allows you to incorporate the components of the oscillator into the
thermal behavior. It does no good to have a perfect crystal if the
oscillator components drift.
Beyond tau=100s, the temperature and moisture sensitivity of the
electronics, combined with the aging of the electronics and the
crystal will be the limit of stability. Of course, this is under
the assumption that you achieved a thermal noise limited design
and thus the 1/f^a noise of the oscillator is negligible in the
time range considered.
Attila Kinali
[1] "Traceable Temperatures - An Introduction to Temperature Measurement
and Calibration", 2nd edition, by Nicholas and White, 2001
[2] "Thin-film platinum resistance thermometer for use at low temperatures
and in high magnetic fields", Haruyama, Yoshizaki, 1986
[3] "Completely Integrated 4-Wire RTD Measurement System Using a Low Power,
Precision, 24-Bit, Sigma-Delta ADC", Analog Circuit Note CN-0381
http://www.analog.com/CN0381
[4] "Completely Integrated 3-Wire RTD Measurement System Using a Low Power,
Precision, 24-Bit, Sigma-Delta ADC", Analog Circuit Note CN-0383
http://www.analog.com/CN0383
[5] "2- 3- 4- Wire RDT (Pt100 to PT1000)Temperature Measurement"
Ti Presentation
http://www.ti.com/europe/downloads/2-%203-%204-Wire%20RTD%20Measurement.pdf
--
You know, the very powerful and the very stupid have one thing in common.
They don't alters their views to fit the facts, they alter the facts to
fit the views, which can be uncomfortable if you happen to be one of the
facts that needs altering. -- The Doctor
Where do digital sensors (e.g. ds1820 and some more recent parts from TI)
fit into this ?
On Mon, Jun 5, 2017 at 12:59 AM, Attila Kinali attila@kinali.ch wrote:
Moin,
This discussion is kind of getting heated.
Let's put some facts in, to steer it away from
opinion based discussion.
On Sun, 4 Jun 2017 08:44:33 -0700
"Donald E. Pauly" trojancowboy@gmail.com wrote:
I stand by my remark that thermistors have been obsolete for over 40
years. The only exception that I know of is cesium beam tubes that
must withstand a 350° C bakeout. Thermistors are unstable and
manufactured with a witches brew straight out of MacBeth. Their
output voltages are tiny and are they inconvenient to use at different
temperatures.
If you really mean thermistors, and not, as Bob suggested thermocouples,
then I have to disagree. The most stable temperature sensors are
platinum wire sensors. The standards class PRT's are the gold standard
when it comes to temperature measurement, for a quite wide range
(-260°C to +960°C) and are considered very stable. They offer (absolute)
accuracies in the order of 10mK in the temperature range below 400°C.
Even industrial grade PRT sensors give you an absolute accuracy better
than 0.1K up to 200-300°C. The "cheap" PT100 are more of the order of
1-10°C
accuracy... all numbers just using a two-point calibration.
For more information on this see [1] chapter 6 and [2] for industrial
sensors.
NTC sensors have a higher variablity of their parameters in production
and are usually specified in % of temperature relative to their reference
point, which is usually 25°C. Typical values are 0.1% to 5%. Additionally
there is a deviation from the reference point, specified in °C, which
is usually in the order of 0.1°C to 1°C.
The NTC sensors are less accurate than PT sensors, but offer the advantage
of higher resistance (thus lower self-heating), higher slope (thus better
precision). Biggest disadvantage is their non-linear curve. Their price
is also a fraction of PT sensors and due to that you can have them in
many different forms, from the 0201 SMD resistor, to a large stainless
steal pipe that goes into a chemical tank. NTCs are the workhorse in
todays temperature measurement and control designs.
The next category are band-gap sensors like the AD590. Their biggest
advantage is that their 0 point is fix at 0K (and very accurately so).
Ie they can be used with single point calibration and achieve 1°C accuracy
this way. Their biggest drawback their large thermal mass and large
insulating case, because they are basically an standard, analog IC.
Ie their main use is in devices where there is a lot of convection and
slow temperature change. Due to their simple and and quite linear
characteristics, they are often used in purely analog temperature
control circuits, or where a linearization is not feasible.
But only if price isn't an issue (they cost 10-1000 times as
much as an PTC). Their biggest disadvantage, beside their slow
thermal raction time, is their large noise uncorrelated to the
supply voltage, and thus cannot be compensated by ratiometric measurement.
They are also more suceptible to mechanical stress than NTC's and PT's,
due to their construction. Similar to voltage references (which they
actually are), their aging is quite substantial and cannot be neglected
in precision application.
With a 3 point calibration, better than 0.5°C accuracy can be achieved
(modulo aging) within their operating temperature range, which is
rather limited, compared to the other sensor types.
I don't know enough about thermocouples to say much about them, beside
that they are cumbersome to work with (e.g. the cold contact) and
produce a low voltage (several µV) output with quite high impedance,
which makes the analog electronics difficult to design as well.
With todays electronics, the easiest sensors to work with are NTC and
PT100/PT1000 as most high resolution delta-sigma ADCs have direct support
for 3 and/or 4 wire measurement of those, including compensation for
reference voltage/current variation. Using a uC as control element
also opens up the possibility to linearize the curve of NTCs without
loss of accuracy. Usually measurement precision, with a state-of-the-art
circuit, is limited by noise coupling into the leads of the sensor
and noise in and around the ADC. (see [3-5])
Where did you get the idea to use a 1 k load for an AD590?
Jim was refering to a circuit he used in a satellite. Not to your
circuit.
The room temperature coefficient of an AT crystal is -cd 100 ppb per
reference cut angle in minutes. (-600 ppb/C° for standard crystal)
The practical limit in a crystal designed for room temperature is
about 0.1' cut accuracy or ±10 ppb/C°. If you have access to an
atomic standard, you can use feed forward to get ±1 ppb/C°. If the
temperature can be held to ±0.001° C, this is ±1 part per trillion.
This kind of accuracy has never been heard of.
It has been heard of. The 8607 was spec'ed to <2e-10 p-p deviation
over temperature range (-30°C to 60°C). Also, to hold the temperature
stable to 0.001K in a room temperature environment (let's say 10K
variation),
you need a thermal gain of >10k. That's quite a bit and needs considerable
design effort. Most OCXO design's I am aware of are in the order of 100
(the DIL14 designs) to a few 1000 for single ovens, to a few 10k for
double ovens. The only exception is the E1938 which achieves >1M.
But that design is not for the faint hearted. I don't remember seeing
any number, but i would guess the 8607 has a thermal gain in the
order of 100k to 1M as well, considering it being a double oven in
a dewar flask.
Also, what do you mean by atomic standard and feed forward?
If you have an atomic standard you don't need to temperature
stabilize your quartz. You can just simply use a PLL to lock
it to your reference and achieve higher stability than any oven
design.
Feed forward also
allows you to incorporate the components of the oscillator into the
thermal behavior. It does no good to have a perfect crystal if the
oscillator components drift.
Beyond tau=100s, the temperature and moisture sensitivity of the
electronics, combined with the aging of the electronics and the
crystal will be the limit of stability. Of course, this is under
the assumption that you achieved a thermal noise limited design
and thus the 1/f^a noise of the oscillator is negligible in the
time range considered.
Attila Kinali
[1] "Traceable Temperatures - An Introduction to Temperature Measurement
and Calibration", 2nd edition, by Nicholas and White, 2001
[2] "Thin-film platinum resistance thermometer for use at low temperatures
and in high magnetic fields", Haruyama, Yoshizaki, 1986
[3] "Completely Integrated 4-Wire RTD Measurement System Using a Low Power,
Precision, 24-Bit, Sigma-Delta ADC", Analog Circuit Note CN-0381
http://www.analog.com/CN0381
[4] "Completely Integrated 3-Wire RTD Measurement System Using a Low Power,
Precision, 24-Bit, Sigma-Delta ADC", Analog Circuit Note CN-0383
http://www.analog.com/CN0383
[5] "2- 3- 4- Wire RDT (Pt100 to PT1000)Temperature Measurement"
Ti Presentation
http://www.ti.com/europe/downloads/2-%203-%204-Wire%
20RTD%20Measurement.pdf
--
You know, the very powerful and the very stupid have one thing in common.
They don't alters their views to fit the facts, they alter the facts to
fit the views, which can be uncomfortable if you happen to be one of the
facts that needs altering. -- The Doctor
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/
mailman/listinfo/time-nuts
and follow the instructions there.
The other issue that needs to be considered is the drift in temperature sensor characteristics when operated at a constant temperature (as is typical in a continuously operated crystal oven). High quality thermistors can achieve drifts of around 1mK/month. Its unlikely that something as complex as an AD590 will achieve a similar drift (1nA/month in a operating current of 300uA or so at 25C). High quality PRT sensors drift even less than thermistors when operating at constant temperature.
Bruce
.On 05 June 2017 at 11:59 Attila Kinali <attila@kinali.ch> wrote:
Moin,
This discussion is kind of getting heated.
Let's put some facts in, to steer it away from
opinion based discussion.
On Sun, 4 Jun 2017 08:44:33 -0700
"Donald E. Pauly" <trojancowboy@gmail.com> wrote:
I stand by my remark that thermistors have been obsolete for over 40
years. The only exception that I know of is cesium beam tubes that
must withstand a 350° C bakeout. Thermistors are unstable and
manufactured with a witches brew straight out of MacBeth. Their
output voltages are tiny and are they inconvenient to use at different
temperatures.
If you really mean thermistors, and not, as Bob suggested thermocouples,
then I have to disagree. The most stable temperature sensors are
platinum wire sensors. The standards class PRT's are the gold standard
when it comes to temperature measurement, for a quite wide range
(-260°C to +960°C) and are considered very stable. They offer (absolute)
accuracies in the order of 10mK in the temperature range below 400°C.
Even industrial grade PRT sensors give you an absolute accuracy better
than 0.1K up to 200-300°C. The "cheap" PT100 are more of the order of 1-10°C
accuracy... all numbers just using a two-point calibration.
For more information on this see [1] chapter 6 and [2] for industrial sensors.
NTC sensors have a higher variablity of their parameters in production
and are usually specified in % of temperature relative to their reference
point, which is usually 25°C. Typical values are 0.1% to 5%. Additionally
there is a deviation from the reference point, specified in °C, which
is usually in the order of 0.1°C to 1°C.
The NTC sensors are less accurate than PT sensors, but offer the advantage
of higher resistance (thus lower self-heating), higher slope (thus better
precision). Biggest disadvantage is their non-linear curve. Their price
is also a fraction of PT sensors and due to that you can have them in
many different forms, from the 0201 SMD resistor, to a large stainless
steal pipe that goes into a chemical tank. NTCs are the workhorse in
todays temperature measurement and control designs.
The next category are band-gap sensors like the AD590. Their biggest
advantage is that their 0 point is fix at 0K (and very accurately so).
Ie they can be used with single point calibration and achieve 1°C accuracy
this way. Their biggest drawback their large thermal mass and large
insulating case, because they are basically an standard, analog IC.
Ie their main use is in devices where there is a lot of convection and
slow temperature change. Due to their simple and and quite linear
characteristics, they are often used in purely analog temperature
control circuits, or where a linearization is not feasible.
But only if price isn't an issue (they cost 10-1000 times as
much as an PTC). Their biggest disadvantage, beside their slow
thermal raction time, is their large noise uncorrelated to the
supply voltage, and thus cannot be compensated by ratiometric measurement.
They are also more suceptible to mechanical stress than NTC's and PT's,
due to their construction. Similar to voltage references (which they
actually are), their aging is quite substantial and cannot be neglected
in precision application.
With a 3 point calibration, better than 0.5°C accuracy can be achieved
(modulo aging) within their operating temperature range, which is
rather limited, compared to the other sensor types.
I don't know enough about thermocouples to say much about them, beside
that they are cumbersome to work with (e.g. the cold contact) and
produce a low voltage (several µV) output with quite high impedance,
which makes the analog electronics difficult to design as well.
With todays electronics, the easiest sensors to work with are NTC and
PT100/PT1000 as most high resolution delta-sigma ADCs have direct support
for 3 and/or 4 wire measurement of those, including compensation for
reference voltage/current variation. Using a uC as control element
also opens up the possibility to linearize the curve of NTCs without
loss of accuracy. Usually measurement precision, with a state-of-the-art
circuit, is limited by noise coupling into the leads of the sensor
and noise in and around the ADC. (see [3-5])
Where did you get the idea to use a 1 k load for an AD590?
Jim was refering to a circuit _he_ used in a satellite. Not to your circuit.
The room temperature coefficient of an AT crystal is -cd 100 ppb per
reference cut angle in minutes. (-600 ppb/C° for standard crystal)
The practical limit in a crystal designed for room temperature is
about 0.1' cut accuracy or ±10 ppb/C°. If you have access to an
atomic standard, you can use feed forward to get ±1 ppb/C°. If the
temperature can be held to ±0.001° C, this is ±1 part per trillion.
This kind of accuracy has never been heard of.
It has been heard of. The 8607 was spec'ed to <2e-10 p-p deviation over temperature range (-30°C to 60°C). Also, to hold the temperature stable to 0.001K in a room temperature environment (let's say 10K variation), you need a thermal gain of >10k. That's quite a bit and needs considerable
design effort. Most OCXO design's I am aware of are in the order of 100
(the DIL14 designs) to a few 1000 for single ovens, to a few 10k for
double ovens. The only exception is the E1938 which achieves >1M.
But that design is not for the faint hearted. I don't remember seeing
any number, but i would guess the 8607 has a thermal gain in the
order of 100k to 1M as well, considering it being a double oven in
a dewar flask.
Also, what do you mean by atomic standard and feed forward?
If you have an atomic standard you don't need to temperature
stabilize your quartz. You can just simply use a PLL to lock
it to your reference and achieve higher stability than any oven
design.
Feed forward also
allows you to incorporate the components of the oscillator into the
thermal behavior. It does no good to have a perfect crystal if the
oscillator components drift.
Beyond tau=100s, the temperature and moisture sensitivity of the
electronics, combined with the aging of the electronics and the
crystal will be the limit of stability. Of course, this is under
the assumption that you achieved a thermal noise limited design
and thus the 1/f^a noise of the oscillator is negligible in the
time range considered.
Attila Kinali
[1] "Traceable Temperatures - An Introduction to Temperature Measurement
and Calibration", 2nd edition, by Nicholas and White, 2001
[2] "Thin-film platinum resistance thermometer for use at low temperatures
and in high magnetic fields", Haruyama, Yoshizaki, 1986
[3] "Completely Integrated 4-Wire RTD Measurement System Using a Low Power,
Precision, 24-Bit, Sigma-Delta ADC", Analog Circuit Note CN-0381
http://www.analog.com/CN0381
[4] "Completely Integrated 3-Wire RTD Measurement System Using a Low Power,
Precision, 24-Bit, Sigma-Delta ADC", Analog Circuit Note CN-0383
http://www.analog.com/CN0383
[5] "2- 3- 4- Wire RDT (Pt100 to PT1000)Temperature Measurement"
Ti Presentation
http://www.ti.com/europe/downloads/2-%203-%204-Wire%20RTD%20Measurement.pdf
--
You know, the very powerful and the very stupid have one thing in common.
They don't alters their views to fit the facts, they alter the facts to
fit the views, which can be uncomfortable if you happen to be one of the
facts that needs altering. -- The Doctor
_______________________________________________
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
On 6/4/17 4:59 PM, Attila Kinali wrote:
Where did you get the idea to use a 1 k load for an AD590?
Jim was refering to a circuit he used in a satellite. Not to your circuit.
We've also used 3k. It's more about supply voltage, expected
temperature range, and the ADC you're using (if any). 1k is handy if
you're running off 5V and are feeding a 1 volt full scale ADC - room
temp is 0.3 V. Note that the minimum voltage across an AD590 is 4V,
so if you've got a 3V supply, you're out of luck.
10k gives you 3V at room temp, and is quite ok into a 5V ADC, as long as
your supply is at least 7-8 volts.
There is self heating to worry about if you have a high supply voltage
(12V @ 0.3 mA is 3.6 mW), but realistically, all sensors have that
problem (unless you are using a PRT in some sort of bridge that nulls
the current)
Hi
If your objective is a resolution of < 0.001 C at something < 1 second, the current crop of
digital sensors don’t quite do what you need to do. They are a terrific way to do wide range
measurements that might feed into some sort of correction algorithm. A conventional
thermistor bridge falls apart if you try to run it -55 to +125. The range of resistances
involved results in significantly lowered resolution at the end(s) of the range.
Bob
On Jun 4, 2017, at 8:18 PM, Adrian Godwin artgodwin@gmail.com wrote:
Where do digital sensors (e.g. ds1820 and some more recent parts from TI)
fit into this ?
On Mon, Jun 5, 2017 at 12:59 AM, Attila Kinali attila@kinali.ch wrote:
Moin,
This discussion is kind of getting heated.
Let's put some facts in, to steer it away from
opinion based discussion.
On Sun, 4 Jun 2017 08:44:33 -0700
"Donald E. Pauly" trojancowboy@gmail.com wrote:
I stand by my remark that thermistors have been obsolete for over 40
years. The only exception that I know of is cesium beam tubes that
must withstand a 350° C bakeout. Thermistors are unstable and
manufactured with a witches brew straight out of MacBeth. Their
output voltages are tiny and are they inconvenient to use at different
temperatures.
If you really mean thermistors, and not, as Bob suggested thermocouples,
then I have to disagree. The most stable temperature sensors are
platinum wire sensors. The standards class PRT's are the gold standard
when it comes to temperature measurement, for a quite wide range
(-260°C to +960°C) and are considered very stable. They offer (absolute)
accuracies in the order of 10mK in the temperature range below 400°C.
Even industrial grade PRT sensors give you an absolute accuracy better
than 0.1K up to 200-300°C. The "cheap" PT100 are more of the order of
1-10°C
accuracy... all numbers just using a two-point calibration.
For more information on this see [1] chapter 6 and [2] for industrial
sensors.
NTC sensors have a higher variablity of their parameters in production
and are usually specified in % of temperature relative to their reference
point, which is usually 25°C. Typical values are 0.1% to 5%. Additionally
there is a deviation from the reference point, specified in °C, which
is usually in the order of 0.1°C to 1°C.
The NTC sensors are less accurate than PT sensors, but offer the advantage
of higher resistance (thus lower self-heating), higher slope (thus better
precision). Biggest disadvantage is their non-linear curve. Their price
is also a fraction of PT sensors and due to that you can have them in
many different forms, from the 0201 SMD resistor, to a large stainless
steal pipe that goes into a chemical tank. NTCs are the workhorse in
todays temperature measurement and control designs.
The next category are band-gap sensors like the AD590. Their biggest
advantage is that their 0 point is fix at 0K (and very accurately so).
Ie they can be used with single point calibration and achieve 1°C accuracy
this way. Their biggest drawback their large thermal mass and large
insulating case, because they are basically an standard, analog IC.
Ie their main use is in devices where there is a lot of convection and
slow temperature change. Due to their simple and and quite linear
characteristics, they are often used in purely analog temperature
control circuits, or where a linearization is not feasible.
But only if price isn't an issue (they cost 10-1000 times as
much as an PTC). Their biggest disadvantage, beside their slow
thermal raction time, is their large noise uncorrelated to the
supply voltage, and thus cannot be compensated by ratiometric measurement.
They are also more suceptible to mechanical stress than NTC's and PT's,
due to their construction. Similar to voltage references (which they
actually are), their aging is quite substantial and cannot be neglected
in precision application.
With a 3 point calibration, better than 0.5°C accuracy can be achieved
(modulo aging) within their operating temperature range, which is
rather limited, compared to the other sensor types.
I don't know enough about thermocouples to say much about them, beside
that they are cumbersome to work with (e.g. the cold contact) and
produce a low voltage (several µV) output with quite high impedance,
which makes the analog electronics difficult to design as well.
With todays electronics, the easiest sensors to work with are NTC and
PT100/PT1000 as most high resolution delta-sigma ADCs have direct support
for 3 and/or 4 wire measurement of those, including compensation for
reference voltage/current variation. Using a uC as control element
also opens up the possibility to linearize the curve of NTCs without
loss of accuracy. Usually measurement precision, with a state-of-the-art
circuit, is limited by noise coupling into the leads of the sensor
and noise in and around the ADC. (see [3-5])
Where did you get the idea to use a 1 k load for an AD590?
Jim was refering to a circuit he used in a satellite. Not to your
circuit.
The room temperature coefficient of an AT crystal is -cd 100 ppb per
reference cut angle in minutes. (-600 ppb/C° for standard crystal)
The practical limit in a crystal designed for room temperature is
about 0.1' cut accuracy or ±10 ppb/C°. If you have access to an
atomic standard, you can use feed forward to get ±1 ppb/C°. If the
temperature can be held to ±0.001° C, this is ±1 part per trillion.
This kind of accuracy has never been heard of.
It has been heard of. The 8607 was spec'ed to <2e-10 p-p deviation
over temperature range (-30°C to 60°C). Also, to hold the temperature
stable to 0.001K in a room temperature environment (let's say 10K
variation),
you need a thermal gain of >10k. That's quite a bit and needs considerable
design effort. Most OCXO design's I am aware of are in the order of 100
(the DIL14 designs) to a few 1000 for single ovens, to a few 10k for
double ovens. The only exception is the E1938 which achieves >1M.
But that design is not for the faint hearted. I don't remember seeing
any number, but i would guess the 8607 has a thermal gain in the
order of 100k to 1M as well, considering it being a double oven in
a dewar flask.
Also, what do you mean by atomic standard and feed forward?
If you have an atomic standard you don't need to temperature
stabilize your quartz. You can just simply use a PLL to lock
it to your reference and achieve higher stability than any oven
design.
Feed forward also
allows you to incorporate the components of the oscillator into the
thermal behavior. It does no good to have a perfect crystal if the
oscillator components drift.
Beyond tau=100s, the temperature and moisture sensitivity of the
electronics, combined with the aging of the electronics and the
crystal will be the limit of stability. Of course, this is under
the assumption that you achieved a thermal noise limited design
and thus the 1/f^a noise of the oscillator is negligible in the
time range considered.
Attila Kinali
[1] "Traceable Temperatures - An Introduction to Temperature Measurement
and Calibration", 2nd edition, by Nicholas and White, 2001
[2] "Thin-film platinum resistance thermometer for use at low temperatures
and in high magnetic fields", Haruyama, Yoshizaki, 1986
[3] "Completely Integrated 4-Wire RTD Measurement System Using a Low Power,
Precision, 24-Bit, Sigma-Delta ADC", Analog Circuit Note CN-0381
http://www.analog.com/CN0381
[4] "Completely Integrated 3-Wire RTD Measurement System Using a Low Power,
Precision, 24-Bit, Sigma-Delta ADC", Analog Circuit Note CN-0383
http://www.analog.com/CN0383
[5] "2- 3- 4- Wire RDT (Pt100 to PT1000)Temperature Measurement"
Ti Presentation
http://www.ti.com/europe/downloads/2-%203-%204-Wire%
20RTD%20Measurement.pdf
--
You know, the very powerful and the very stupid have one thing in common.
They don't alters their views to fit the facts, they alter the facts to
fit the views, which can be uncomfortable if you happen to be one of the
facts that needs altering. -- The Doctor
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/
mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
On Mon, 5 Jun 2017 01:18:59 +0100
Adrian Godwin artgodwin@gmail.com wrote:
Where do digital sensors (e.g. ds1820 and some more recent parts from TI)
fit into this ?
AFAIK, these are all band-gap temperature sensors. But unlike a discrete
sensor, you have the problem that they only contain a low resolution
ADC on die (somewhere between 8 and 14 bit). If your goal is to measure
temperature and report it with an accuracy of about 1°C, then these are
the easiest to use sensors you can buy. Sensor noise doesn't really matter
with them, as it is dominated by the low ADC resolution. I don't have any
long term stability data on those, but given their use-case I do not think
that they are very stable. Although long term stability might not be an
issue at all, again due to low ADC resolution.
If you need better precision, accuracy, or stability, then choosing one
of the modern delta-sigma ADCs that directly support thermistors
(e.g. like AD7124) is not much more difficult, though a bit more expensive
(around 10USD instead of 5USD like for an TMP107). Additionally you need
to calbirate the system, which means you need a reference temperature sensor
and a setup with which you can produce different temperatures. Though for
an oven kind of temperature control, one can live without calibration.
Attila Kinali
--
You know, the very powerful and the very stupid have one thing in common.
They don't alters their views to fit the facts, they alter the facts to
fit the views, which can be uncomfortable if you happen to be one of the
facts that needs altering. -- The Doctor
Hi
On Jun 5, 2017, at 7:30 AM, Attila Kinali attila@kinali.ch wrote:
On Mon, 5 Jun 2017 01:18:59 +0100
Adrian Godwin artgodwin@gmail.com wrote:
Where do digital sensors (e.g. ds1820 and some more recent parts from TI)
fit into this ?
AFAIK, these are all band-gap temperature sensors. But unlike a discrete
sensor, you have the problem that they only contain a low resolution
ADC on die (somewhere between 8 and 14 bit). If your goal is to measure
temperature and report it with an accuracy of about 1°C, then these are
the easiest to use sensors you can buy. Sensor noise doesn't really matter
with them, as it is dominated by the low ADC resolution. I don't have any
long term stability data on those, but given their use-case I do not think
that they are very stable.
Based on using them in a lot of designs, they are indeed quite stable. They are not
going to rival a thermistor or an RTD, but compared to their resolution they are stable.
Put another way, if they read out at the (say) 0.5 C level, you can come back a year later
and the temperature repeats at < the 0.5 C level.
None of this is simple or straightforward. All temperature sensors have a sensitivity
to strain. They all exhibit some level of hysteresis. That can make aging measurements
a bit challenging.
Bob
Although long term stability might not be an
issue at all, again due to low ADC resolution.
If you need better precision, accuracy, or stability, then choosing one
of the modern delta-sigma ADCs that directly support thermistors
(e.g. like AD7124) is not much more difficult, though a bit more expensive
(around 10USD instead of 5USD like for an TMP107). Additionally you need
to calbirate the system, which means you need a reference temperature sensor
and a setup with which you can produce different temperatures. Though for
an oven kind of temperature control, one can live without calibration.
Attila Kinali
--
You know, the very powerful and the very stupid have one thing in common.
They don't alters their views to fit the facts, they alter the facts to
fit the views, which can be uncomfortable if you happen to be one of the
facts that needs altering. -- The Doctor
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Hi, guys
I have been following time nuts and volt nuts for some time out of interest and fascination. Although my personal backyard hobby is more along a volt nuts line, the two worlds often collide - like in this discussion of temperature sensors, and in particular their long term stability. NTC thermistors appear to be very commonly used in ovens used to stabilize voltage references (solid state as well as chemical) . I have long wondered about their stability. If, as Bruce asserts, "high quality thermistors can achieve drifts of around 1mK/month" then it appears that this level of drift is a significant factor in the "apparent" aging of, say, a bank of Weston cells (which is still my best backyard shot at a voltage reference).
I have had no luck with Google; Bruce's statement is the first quantified allusion that I have seen to this subject. Is there any actual data available on the long term performance of NTC sensors?
Roman
On 5 Jun 2017, at 9:53 AM, Bruce Griffiths bruce.griffiths@xtra.co.nz wrote:
The other issue that needs to be considered is the drift in temperature sensor characteristics when operated at a constant temperature (as is typical in a continuously operated crystal oven). High quality thermistors can achieve drifts of around 1mK/month. Its unlikely that something as complex as an AD590 will achieve a similar drift (1nA/month in a operating current of 300uA or so at 25C). High quality PRT sensors drift even less than thermistors when operating at constant temperature.
Bruce
.On 05 June 2017 at 11:59 Attila Kinali <attila@kinali.ch> wrote:
Moin,
This discussion is kind of getting heated.
Let's put some facts in, to steer it away from
opinion based discussion.
On Sun, 4 Jun 2017 08:44:33 -0700
"Donald E. Pauly" <trojancowboy@gmail.com> wrote:
I stand by my remark that thermistors have been obsolete for over 40
years. The only exception that I know of is cesium beam tubes that
must withstand a 350° C bakeout. Thermistors are unstable and
manufactured with a witches brew straight out of MacBeth. Their
output voltages are tiny and are they inconvenient to use at different
temperatures.
If you really mean thermistors, and not, as Bob suggested thermocouples,
then I have to disagree. The most stable temperature sensors are
platinum wire sensors. The standards class PRT's are the gold standard
when it comes to temperature measurement, for a quite wide range
(-260°C to +960°C) and are considered very stable. They offer (absolute)
accuracies in the order of 10mK in the temperature range below 400°C.
Even industrial grade PRT sensors give you an absolute accuracy better
than 0.1K up to 200-300°C. The "cheap" PT100 are more of the order of 1-10°C
accuracy... all numbers just using a two-point calibration.
For more information on this see [1] chapter 6 and [2] for industrial sensors.
NTC sensors have a higher variablity of their parameters in production
and are usually specified in % of temperature relative to their reference
point, which is usually 25°C. Typical values are 0.1% to 5%. Additionally
there is a deviation from the reference point, specified in °C, which
is usually in the order of 0.1°C to 1°C.
The NTC sensors are less accurate than PT sensors, but offer the advantage
of higher resistance (thus lower self-heating), higher slope (thus better
precision). Biggest disadvantage is their non-linear curve. Their price
is also a fraction of PT sensors and due to that you can have them in
many different forms, from the 0201 SMD resistor, to a large stainless
steal pipe that goes into a chemical tank. NTCs are the workhorse in
todays temperature measurement and control designs.
The next category are band-gap sensors like the AD590. Their biggest
advantage is that their 0 point is fix at 0K (and very accurately so).
Ie they can be used with single point calibration and achieve 1°C accuracy
this way. Their biggest drawback their large thermal mass and large
insulating case, because they are basically an standard, analog IC.
Ie their main use is in devices where there is a lot of convection and
slow temperature change. Due to their simple and and quite linear
characteristics, they are often used in purely analog temperature
control circuits, or where a linearization is not feasible.
But only if price isn't an issue (they cost 10-1000 times as
much as an PTC). Their biggest disadvantage, beside their slow
thermal raction time, is their large noise uncorrelated to the
supply voltage, and thus cannot be compensated by ratiometric measurement.
They are also more suceptible to mechanical stress than NTC's and PT's,
due to their construction. Similar to voltage references (which they
actually are), their aging is quite substantial and cannot be neglected
in precision application.
With a 3 point calibration, better than 0.5°C accuracy can be achieved
(modulo aging) within their operating temperature range, which is
rather limited, compared to the other sensor types.
I don't know enough about thermocouples to say much about them, beside
that they are cumbersome to work with (e.g. the cold contact) and
produce a low voltage (several µV) output with quite high impedance,
which makes the analog electronics difficult to design as well.
With todays electronics, the easiest sensors to work with are NTC and
PT100/PT1000 as most high resolution delta-sigma ADCs have direct support
for 3 and/or 4 wire measurement of those, including compensation for
reference voltage/current variation. Using a uC as control element
also opens up the possibility to linearize the curve of NTCs without
loss of accuracy. Usually measurement precision, with a state-of-the-art
circuit, is limited by noise coupling into the leads of the sensor
and noise in and around the ADC. (see [3-5])
Where did you get the idea to use a 1 k load for an AD590?
Jim was refering to a circuit _he_ used in a satellite. Not to your circuit.
The room temperature coefficient of an AT crystal is -cd 100 ppb per
reference cut angle in minutes. (-600 ppb/C° for standard crystal)
The practical limit in a crystal designed for room temperature is
about 0.1' cut accuracy or ±10 ppb/C°. If you have access to an
atomic standard, you can use feed forward to get ±1 ppb/C°. If the
temperature can be held to ±0.001° C, this is ±1 part per trillion.
This kind of accuracy has never been heard of.
It has been heard of. The 8607 was spec'ed to <2e-10 p-p deviation over temperature range (-30°C to 60°C). Also, to hold the temperature stable to 0.001K in a room temperature environment (let's say 10K variation), you need a thermal gain of >10k. That's quite a bit and needs considerable
design effort. Most OCXO design's I am aware of are in the order of 100
(the DIL14 designs) to a few 1000 for single ovens, to a few 10k for
double ovens. The only exception is the E1938 which achieves >1M.
But that design is not for the faint hearted. I don't remember seeing
any number, but i would guess the 8607 has a thermal gain in the
order of 100k to 1M as well, considering it being a double oven in
a dewar flask.
Also, what do you mean by atomic standard and feed forward?
If you have an atomic standard you don't need to temperature
stabilize your quartz. You can just simply use a PLL to lock
it to your reference and achieve higher stability than any oven
design.
Feed forward also
allows you to incorporate the components of the oscillator into the
thermal behavior. It does no good to have a perfect crystal if the
oscillator components drift.
Beyond tau=100s, the temperature and moisture sensitivity of the
electronics, combined with the aging of the electronics and the
crystal will be the limit of stability. Of course, this is under
the assumption that you achieved a thermal noise limited design
and thus the 1/f^a noise of the oscillator is negligible in the
time range considered.
Attila Kinali
[1] "Traceable Temperatures - An Introduction to Temperature Measurement
and Calibration", 2nd edition, by Nicholas and White, 2001
[2] "Thin-film platinum resistance thermometer for use at low temperatures
and in high magnetic fields", Haruyama, Yoshizaki, 1986
[3] "Completely Integrated 4-Wire RTD Measurement System Using a Low Power,
Precision, 24-Bit, Sigma-Delta ADC", Analog Circuit Note CN-0381
http://www.analog.com/CN0381
[4] "Completely Integrated 3-Wire RTD Measurement System Using a Low Power,
Precision, 24-Bit, Sigma-Delta ADC", Analog Circuit Note CN-0383
http://www.analog.com/CN0383
[5] "2- 3- 4- Wire RDT (Pt100 to PT1000)Temperature Measurement"
Ti Presentation
http://www.ti.com/europe/downloads/2-%203-%204-Wire%20RTD%20Measurement.pdf
--
You know, the very powerful and the very stupid have one thing in common.
They don't alters their views to fit the facts, they alter the facts to
fit the views, which can be uncomfortable if you happen to be one of the
facts that needs altering. -- The Doctor
_______________________________________________
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
Hi
Well, as part of the process of designing them into OCXO’s you do indeed check their long term stability.
The test is done in an indirect fashion so you only come up with a “it’s below the limit” sort of number. The
typical process involves running a group of OCXO’s on turn to check the frequency and then shifting them
off turn to make a sort of thermometer. After a few months of frequency readings you take them back to turn
for a while. Relative frequency shift math gives you a stability number for the thermistor and the rest of the
circuitry. You may repeat the run for months / shift process a couple of times. If the answer isn’t “I can’t see
a difference” you look for a new thermistor. Since it’s a long drawn out test, the tendency is to stick with a
vendor’s part for quite a while. The parts also tend to be design specific so what works in my (say SMT)
design may not work well in your (say chip and wire) design.
Bob
On Jun 5, 2017, at 9:20 AM, romeo987 romeo987@westnet.com.au wrote:
Hi, guys
I have been following time nuts and volt nuts for some time out of interest and fascination. Although my personal backyard hobby is more along a volt nuts line, the two worlds often collide - like in this discussion of temperature sensors, and in particular their long term stability. NTC thermistors appear to be very commonly used in ovens used to stabilize voltage references (solid state as well as chemical) . I have long wondered about their stability. If, as Bruce asserts, "high quality thermistors can achieve drifts of around 1mK/month" then it appears that this level of drift is a significant factor in the "apparent" aging of, say, a bank of Weston cells (which is still my best backyard shot at a voltage reference).
I have had no luck with Google; Bruce's statement is the first quantified allusion that I have seen to this subject. Is there any actual data available on the long term performance of NTC sensors?
Roman
On 5 Jun 2017, at 9:53 AM, Bruce Griffiths bruce.griffiths@xtra.co.nz wrote:
The other issue that needs to be considered is the drift in temperature sensor characteristics when operated at a constant temperature (as is typical in a continuously operated crystal oven). High quality thermistors can achieve drifts of around 1mK/month. Its unlikely that something as complex as an AD590 will achieve a similar drift (1nA/month in a operating current of 300uA or so at 25C). High quality PRT sensors drift even less than thermistors when operating at constant temperature.
Bruce
.On 05 June 2017 at 11:59 Attila Kinali attila@kinali.ch wrote:
Moin,
This discussion is kind of getting heated.
Let's put some facts in, to steer it away from
opinion based discussion.
On Sun, 4 Jun 2017 08:44:33 -0700
"Donald E. Pauly" trojancowboy@gmail.com wrote:
I stand by my remark that thermistors have been obsolete for over 40
years. The only exception that I know of is cesium beam tubes that
must withstand a 350° C bakeout. Thermistors are unstable and
manufactured with a witches brew straight out of MacBeth. Their
output voltages are tiny and are they inconvenient to use at different
temperatures.
If you really mean thermistors, and not, as Bob suggested thermocouples,
then I have to disagree. The most stable temperature sensors are
platinum wire sensors. The standards class PRT's are the gold standard
when it comes to temperature measurement, for a quite wide range
(-260°C to +960°C) and are considered very stable. They offer (absolute)
accuracies in the order of 10mK in the temperature range below 400°C.
Even industrial grade PRT sensors give you an absolute accuracy better
than 0.1K up to 200-300°C. The "cheap" PT100 are more of the order of 1-10°C
accuracy... all numbers just using a two-point calibration.
For more information on this see [1] chapter 6 and [2] for industrial sensors.
NTC sensors have a higher variablity of their parameters in production
and are usually specified in % of temperature relative to their reference
point, which is usually 25°C. Typical values are 0.1% to 5%. Additionally
there is a deviation from the reference point, specified in °C, which
is usually in the order of 0.1°C to 1°C.
The NTC sensors are less accurate than PT sensors, but offer the advantage
of higher resistance (thus lower self-heating), higher slope (thus better
precision). Biggest disadvantage is their non-linear curve. Their price
is also a fraction of PT sensors and due to that you can have them in
many different forms, from the 0201 SMD resistor, to a large stainless
steal pipe that goes into a chemical tank. NTCs are the workhorse in
todays temperature measurement and control designs.
The next category are band-gap sensors like the AD590. Their biggest
advantage is that their 0 point is fix at 0K (and very accurately so).
Ie they can be used with single point calibration and achieve 1°C accuracy
this way. Their biggest drawback their large thermal mass and large
insulating case, because they are basically an standard, analog IC.
Ie their main use is in devices where there is a lot of convection and
slow temperature change. Due to their simple and and quite linear
characteristics, they are often used in purely analog temperature
control circuits, or where a linearization is not feasible.
But only if price isn't an issue (they cost 10-1000 times as
much as an PTC). Their biggest disadvantage, beside their slow
thermal raction time, is their large noise uncorrelated to the
supply voltage, and thus cannot be compensated by ratiometric measurement.
They are also more suceptible to mechanical stress than NTC's and PT's,
due to their construction. Similar to voltage references (which they
actually are), their aging is quite substantial and cannot be neglected
in precision application.
With a 3 point calibration, better than 0.5°C accuracy can be achieved
(modulo aging) within their operating temperature range, which is
rather limited, compared to the other sensor types.
I don't know enough about thermocouples to say much about them, beside
that they are cumbersome to work with (e.g. the cold contact) and
produce a low voltage (several µV) output with quite high impedance,
which makes the analog electronics difficult to design as well.
With todays electronics, the easiest sensors to work with are NTC and
PT100/PT1000 as most high resolution delta-sigma ADCs have direct support
for 3 and/or 4 wire measurement of those, including compensation for
reference voltage/current variation. Using a uC as control element
also opens up the possibility to linearize the curve of NTCs without
loss of accuracy. Usually measurement precision, with a state-of-the-art
circuit, is limited by noise coupling into the leads of the sensor
and noise in and around the ADC. (see [3-5])
Where did you get the idea to use a 1 k load for an AD590?
Jim was refering to a circuit he used in a satellite. Not to your circuit.
The room temperature coefficient of an AT crystal is -cd 100 ppb per
reference cut angle in minutes. (-600 ppb/C° for standard crystal)
The practical limit in a crystal designed for room temperature is
about 0.1' cut accuracy or ±10 ppb/C°. If you have access to an
atomic standard, you can use feed forward to get ±1 ppb/C°. If the
temperature can be held to ±0.001° C, this is ±1 part per trillion.
This kind of accuracy has never been heard of.
It has been heard of. The 8607 was spec'ed to <2e-10 p-p deviation over temperature range (-30°C to 60°C). Also, to hold the temperature stable to 0.001K in a room temperature environment (let's say 10K variation), you need a thermal gain of >10k. That's quite a bit and needs considerable
design effort. Most OCXO design's I am aware of are in the order of 100
(the DIL14 designs) to a few 1000 for single ovens, to a few 10k for
double ovens. The only exception is the E1938 which achieves >1M.
But that design is not for the faint hearted. I don't remember seeing
any number, but i would guess the 8607 has a thermal gain in the
order of 100k to 1M as well, considering it being a double oven in
a dewar flask.
Also, what do you mean by atomic standard and feed forward?
If you have an atomic standard you don't need to temperature
stabilize your quartz. You can just simply use a PLL to lock
it to your reference and achieve higher stability than any oven
design.
Feed forward also
allows you to incorporate the components of the oscillator into the
thermal behavior. It does no good to have a perfect crystal if the
oscillator components drift.
Beyond tau=100s, the temperature and moisture sensitivity of the
electronics, combined with the aging of the electronics and the
crystal will be the limit of stability. Of course, this is under
the assumption that you achieved a thermal noise limited design
and thus the 1/f^a noise of the oscillator is negligible in the
time range considered.
Attila Kinali
[1] "Traceable Temperatures - An Introduction to Temperature Measurement
and Calibration", 2nd edition, by Nicholas and White, 2001
[2] "Thin-film platinum resistance thermometer for use at low temperatures
and in high magnetic fields", Haruyama, Yoshizaki, 1986
[3] "Completely Integrated 4-Wire RTD Measurement System Using a Low Power,
Precision, 24-Bit, Sigma-Delta ADC", Analog Circuit Note CN-0381
http://www.analog.com/CN0381
[4] "Completely Integrated 3-Wire RTD Measurement System Using a Low Power,
Precision, 24-Bit, Sigma-Delta ADC", Analog Circuit Note CN-0383
http://www.analog.com/CN0383
[5] "2- 3- 4- Wire RDT (Pt100 to PT1000)Temperature Measurement"
Ti Presentation
http://www.ti.com/europe/downloads/2-%203-%204-Wire%20RTD%20Measurement.pdf
--
You know, the very powerful and the very stupid have one thing in common.
They don't alters their views to fit the facts, they alter the facts to
fit the views, which can be uncomfortable if you happen to be one of the
facts that needs altering. -- The Doctor
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
time-nuts mailing list -- time-nuts@febo.com
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.
In message 20170605133013.526e8505158e68b6a8091e05@kinali.ch, Attila Kinali w
rites:
Where do digital sensors (e.g. ds1820 and some more recent parts from TI)
fit into this ?
AFAIK, these are all band-gap temperature sensors.
The Ds1820 is based on the frequency difference between two
free-running silicon oscillators with different physical design.
--
Poul-Henning Kamp | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG | TCP/IP since RFC 956
FreeBSD committer | BSD since 4.3-tahoe
Never attribute to malice what can adequately be explained by incompetence.