Re: [time-nuts] time-nuts Geller 10 volt ref. modification.
Geller 10 volt ref.
Hello WarrenS,
Is it possible to have some details of your 1 transistor 1st order
temperature modification,
on the Geller 10 volt ref. ?
Helge
2008/11/29 time-nuts-request@febo.com
Send time-nuts mailing list submissions to
time-nuts@febo.com
To subscribe or unsubscribe via the World Wide Web, visit
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
or, via email, send a message with subject or body 'help' to
time-nuts-request@febo.com
You can reach the person managing the list at
time-nuts-owner@febo.com
When replying, please edit your Subject line so it is more specific
than "Re: Contents of time-nuts digest..."
Today's Topics:
- Re: Calibration and temperature (Lux, James P)
- Re: Voltage standards (Bruce Griffiths)
- Re: Voltage standards (WarrenS)
- Re: any way to bootstrap a frequency standard, into a,
voltage or resistance standard? (Ed Palmer) - Re: Cesium vs H Maser clocks (Tom Van Baak)
- Re: Cesium vs H Maser clocks (Brian Kirby)
- TBolt TC recs wanted (Richard Moore)
Message: 1
Date: Fri, 28 Nov 2008 19:52:35 -0800
From: "Lux, James P" james.p.lux@jpl.nasa.gov
Subject: Re: [time-nuts] Calibration and temperature
To: Discussion of precise time and frequency measurement
time-nuts@febo.com
Message-ID: <C555FC83.2F5E%James.P.Lux@jpl.nasa.govC555FC83.2F5E%25James.P.Lux@jpl.nasa.gov
Content-Type: text/plain; charset="iso-8859-1"
On 11/28/08 11:27 AM, "Bill Hawkins" bill@iaxs.net wrote:
Has any work been done on temperature compensation of quartz or other
oscillators to avoid the expense, space, and power of ovens? The
oscillating material must have a repeatable temperature curve, of
course.
Look at MCXOs, a very clever technique using the different between third
overtone and fundamental to measure the temperature of the rock.
Jim Lux
Message: 2
Date: Sat, 29 Nov 2008 17:12:40 +1300
From: Bruce Griffiths bruce.griffiths@xtra.co.nz
Subject: Re: [time-nuts] Voltage standards
To: Discussion of precise time and frequency measurement
time-nuts@febo.com
Message-ID: 4930C138.3020105@xtra.co.nz
Content-Type: text/plain; charset=ISO-8859-1
Richard Moore wrote:
What to do? Seems like a possible stable, yet low-cost source would
be to buy 10 or so LM399s and hook them up to a power supply to age
for 6 to 12 months, then use two or four in parallel (with suitable
resistor buffering) driving a chopper amp like the LT1050 to give
gain. Powered by a good, stiff regulated supply, and kept away from
stray air currents, this arrangement wouldn't cost as much as a
single LTZ1000A (which in singles is over $100 USD)
LTZ1000A list price is about $54 (excluding freight and taxes) when
ordering from Linear Technology.
The LTC1050 is a bit marginal unless you bootstrap its power supply the
LTC1151 is a better fit.
and -- dang it,
there's always a catch -- after lab calibration, could be a source
that might need adjustment only once every few years to stay within 5
or 10ppm or so. Paralleling more LM399s would be better, and with an
initial purchase of 10, you could throw away the stinkers and
parallel all the rest. So, short of having a Fluke 732A or those
8-1/2 digit meters, that's my recommendation for a low-cost V
standard. Unless you know someone, it'll cost less to build than to
calibrate...
Dick Moore
Bruce
Message: 3
Date: Fri, 28 Nov 2008 20:51:12 -0800
From: "WarrenS" warrensjmail-one@yahoo.com
Subject: Re: [time-nuts] Voltage standards
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Message-ID: 029701c951de$1adadb70$6401a8c0@WSOffice
Content-Type: text/plain; charset="iso-8859-1"
Just to add my experiences and throw in another two cents worth.
I live in California, and a few years back I did some experiments
with Geller on the East coast shipping back and forth some of his 10V
References in the regular mail that I had added a simple 1 transistor
1st order Temperature compensation circuit to some of his selected aged
units and proved to my satisfaction that better than 2PPM transfer
accuracy was obtained across the country. In house I can consistently
get better than 1 PPM transfer with the same units. Of course there
are more accurate ways, but for us cheap guys It sure beats the $500
I heard Fluke charges to do their underlying Fluke 732B calibration.
WarrenS
RE Message 8 and a couple of earlier posts on standards
Date: Sat, 29 Nov 2008 10:38:53 +1300
From: Bruce Griffiths <bruce.griffiths at xtra.co.nz>
Subject: Re: [time-nuts] any way to bootstrap a frequency standard
into a, voltage or resistance standard?
To: Discussion of precise time and frequency measurement
<time-nuts at febo.com>
Message-ID: <493064ED.1030501 at xtra.co.nz>
Content-Type: text/plain; charset=ISO-8859-1
WB6BNQ wrote:
Ed,
No way in hell does that thing qualify as a voltage standard. It
barely qualifies as a VERY short term transfer device assuming the
temperature does not change.
If you want a real voltage reference then buy, from eBay, a Fluke
731B voltage standard. You will still need to get it calibrated,
but then it will hold under 10ppm for well over a year or
more. Actually, if they are adjusted correctly, you can get less
than 5ppm and it will hold it at a given temperature. These items
are serious devices and their latest versions (very expensive)
are better than a properly maintained and operated group of
standard cells.
The latest Fluke voltage standards use a selected LTZ1000 plus a
precision resistor array.
The earlier versions used selected Motorola zener diode references.
The latest versions allow the LTZ1000 chip temperature to be cycled to
largely eliminate hysteresis which may occur when they lose power
during
shipment.
Bill....WB6BNQ
Bruce
Since we've kinda strayed from the original topic, I'll throw in a
couple of cents. The V references in these 6-1/2 and 7-1/2 digit DMMs
(that I currently own at least one each of): Keithley 196, HP 3456,
3457, and 34401, use selected -- some more than others -- National
LM199/299/399 series references, which have their own heaters. The
Datron 1080 series (that I own two of), as well as the 1070 series
and 1060 series, use selected pairs of zeners connected in parallel,
but not housed in ovens. The Fluke 8502/05/06 (and my Fluke 510A ACV
std) use the aforementioned Motorola transistor+buried zener
reference, also not in ovens.
The HP 3458, as mentioned before, uses the Linear LTZ1000/1000A,
which has its own heater. As to others, the old Fluke 895A, 887A, and
885A 6-digit analog Differential Meters (with K-V bridges for
comparing the reference to the unknown in various ways), all use a
pair of zeners in series, in an oven. A Fluke engineer told me years
ago that those zeners were selected to have slightly different TC
peaks so when one was falling in V the other was rising, so that they
would spread the peak over a larger temperature range. They had a
very elaborate automated system for testing the temp, voltage, and
current values to do the pairing.
My Fluke 732A DC reference standard, which uses the Motorola
transistor+buried zener reference, housed in an oven together with
all of the various fixed and variable resistors and the other active
devices, has an extremely low drift rate -- unmeasurable over a month
in any definitive way -- as measured with a rented HP 3458A. I have
disabled the batteries because I discovered, thanks to the 3458A,
that when turned off, cooled off, and turned back on, within a day,
the output returned within 0.05ppm to it's cal'd value -- I know
there can be hysteresis, I just couldn't see it.
I have one of the little Geller 10V refs built around the AD587
reference chips, and if you keep it enclosed and away from air
currents, it's pretty stable, but would need to be in an oven to
approach the performance of the LM199 series.
All this is to say that if you can get a working HP 3458A or a Datron
1280 series (which is also the Fluke 8805), or a Datron 1270 series
meter, all of which are self-calibrating in various ingenious ways,
it will serve you well as a secondary standard for measurement on
everything that it can measure. I can't, or more properly, won't at
the moment, spend the current freight of around $4,000 USD for a
known good working 3458 or 1281.
What to do? Seems like a possible stable, yet low-cost source would
be to buy 10 or so LM399s and hook them up to a power supply to age
for 6 to 12 months, then use two or four in parallel (with suitable
resistor buffering) driving a chopper amp like the LT1050 to give
gain. Powered by a good, stiff regulated supply, and kept away from
stray air currents, this arrangement wouldn't cost as much as a
single LTZ1000A (which in singles is over $100 USD) and -- dang it,
there's always a catch -- after lab calibration, could be a source
that might need adjustment only once every few years to stay within 5
or 10ppm or so. Paralleling more LM399s would be better, and with an
initial purchase of 10, you could throw away the stinkers and
parallel all the rest. So, short of having a Fluke 732A or those
8-1/2 digit meters, that's my recommendation for a low-cost V
standard. Unless you know someone, it'll cost less to build than to
calibrate...
Dick Moore
Message: 4
Date: Fri, 28 Nov 2008 23:18:39 -0600
From: Ed Palmer ed_palmer@sasktel.net
Subject: Re: [time-nuts] any way to bootstrap a frequency standard,
into a, voltage or resistance standard?
To: time-nuts@febo.com
Message-ID: 4930D0AF.9050708@sasktel.net
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Bill,
I agree with the comments made by you and everyone else. That's why I said
"..if it's good enough..". But there's one thing that kind of slid by
everyone. All the other suggestions conclude with some variation of
"..needs to be calibrated..". Is there any other way to get a
NIST-traceable voltage into your shop for $35? (I'm not challenging you,
I'd really like to know if there is another alternative) Granted, it's only
at the 10 uv level, but for many applications that is good enough.
Let's be honest, as time-nuts we're spoiled rotten. Almost all of us have
got rubidium standards at 10E-11 per day or better and at a ridiculously low
price. Many have got cesium standards and a few have got hydrogen masers.
We can calibrate them against GPS or Loran-C to obsessive-compulsive
levels. So far, none of the other physical standards allow that level of
independence for hobbyists. So when dealing with other measurement units
(e.g. volts, ohms, etc) we either have to ease the requirements by orders of
magnitude or increase the budget by orders of magnitude!
If you're doing this as part of your job and can justify the need and the
expense - great! I'm a hobbyist and I won't be spending $1843 any time soon
to have NIST calibrate a DC solid-state voltage reference. For my needs
Geller Labs provides good value for the very nominal cost.
And just to be clear, I have no relationship with Geller Labs.
Ed
WB6BNQ wrote:
No way in hell does that thing qualify as a voltage standard. It barely
qualifies as a VERY short term transfer device assuming the temperature does
not change.
If you want a real voltage reference then buy, from eBay, a Fluke 731B
voltage standard. You will still need to get it calibrated, but then it
will hold under 10ppm for well over a year or
more. Actually, if they are adjusted correctly, you can get less than
5ppm and it will hold it at a given temperature. These items are serious
devices and their latest versions (very expensive)
are better than a properly maintained and operated group of standard
cells.
I completely disagree with Brian about buying any standard cells.
Whatever voltage value they had is lost upon shipment. Shaking the cells
changes the value and it will not return to the
original value. If you did have a set of cells, you would want at least
4 of them. Then study statistical math all over again because you will need
it to use the cells.
As Brian does suggest, you would be much better off picking up a hp3456A,
3457A, 3458A or getting one of Fluke 8500 series if it has the Ohms and AC
options included. The basic Fluke 8500 series
is DC only mainframe.
Measuring your resistors, using one of the above DVMs, in the 4-wire mode
is about the best you could possibly do. To do any better would require
some very serious effort.
By the way you could buy several of the latest and greatest Fluke super
DVMs for the cost of what it would take to do a Josephson array and still
have money left over to fund that divorce.
Bill....WB6BNQ
Ed Palmer wrote:
It's nowhere near the idea of a Josephson array, but if a NIST-traceable
10V +-10uV reference is good enough to satisfy your voltage-nut urges, you
can buy it from www.gellerlabs.com for $35.
I also have a few standard resistors (e.g. 1.000002 ohms) that I'd be
interested in calibrating, but I can't seem to come up with a practical way
of doing it.
Frustrating, isn't it? :-)
Ed
Message: 5
Date: Fri, 28 Nov 2008 22:30:33 -0800
From: "Tom Van Baak" tvb@LeapSecond.com
Subject: Re: [time-nuts] Cesium vs H Maser clocks
To: "Discussion of precise time and frequency measurement"
time-nuts@febo.com
Message-ID: 9FB14428535B4E0EAE3C34C33EE617F6@pc52
Content-Type: text/plain; format=flowed; charset="Windows-1252";
reply-type=original
Isn't the temperature the only thing to correct for?
No, not at all. Read the links that I provided to see that a real
cesium standard is not quite so simple, at least when you get
down to the ten to the -13, -14, -15 levels. At that level there
are all sorts of cool things that push or pull the frequency and
need to be corrected for.
The definition of the second is "...the duration of 9 192 631 770
periods of the radiation corresponding to the transition between the
two hyperfine levels of the ground state of the cesium 133 atom." (and
affirmed by the CIPM in 1997 that this definition refers to a cesium
atom in its ground state at a temperature of 0 K)
Right. But realize that most of the cesium standards that we use
are not running at 0 Kelvin. So there is a correction for that.
In order for cesium beam standards to even work, one must apply
a slight magnetic field, the so-called C-field, which rather strongly
distorts the shape of the resonance peak. The definition assumes
zero magnetic field, so this too must be modeled and corrected for.
That's why, for example, the hp 5062c runs at 9,192,631,774.3133 Hz,
not the textbook 9,192,631,770 Hz. An internal synthesizer takes
care of this correction.
The NIST papers list a dozen or so of these corrections, each of
which is a nice lesson in atomic physics by itself.
Note also that clocks at NIST run about 1.8e-13 fast due to the high
elevation of Boulder, CO (general relativity), which is yet another
factor that has to be corrected for compared to the official sea-level
definition of the second.
That other factors can change the relative frequency of different Cs
clocks is a problem with the definition, not an indication that any
particular one is better than another. If a magnetic field changes the
relative frequency, but that isn't reflected in the definition, is it
not the definition which is faulty, and not the timepiece? The second
is imprecise in this regard.
The definition is fine -- it applies to the ideal conditions. But if you
decide to build an apparatus to implement the definition, and if for
whatever reason the ideal conditions can't be met in your apparatus,
then is it up to you, the clock builder, to anticipate this and make
corrections for it so that your clock still counts SI seconds at the
output BNC connector.
The other thing to note is that most cesium standards come with a
specification, based on design. I don't have the exact numbers but
a 5061A might be accurate out-of-the-box to 1e-11 while a 5071A
might be accurate to 1e-13. This reflects the difference in design,
manufacturing tolerances, and the number of internal frequency
offsets that are controlled or compensated for in hardware or in
firmware. So the definition of the SI second is fine; it's just that
some clocks can get closer to realizing the definition than others.
/tvb
Message: 6
Date: Sat, 29 Nov 2008 01:25:39 -0600
From: Brian Kirby kirbybq@bellsouth.net
Subject: Re: [time-nuts] Cesium vs H Maser clocks
To: Tom Van Baak tvb@leapsecond.com, Discussion of precise time and
frequency measurement time-nuts@febo.com
Message-ID: 4930EE73.5080903@bellsouth.net
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
The clocks (rubidium and cesium) in the GPS satellites are also run at a
different frequency because of their altitudes....
Also magnetic field are different on the earth and they sometimes adjust
the "C" fields to correct for these differences.
Isn't the temperature the only thing to correct for?
No, not at all. Read the links that I provided to see that a real
cesium standard is not quite so simple, at least when you get
down to the ten to the -13, -14, -15 levels. At that level there
are all sorts of cool things that push or pull the frequency and
need to be corrected for.
The definition of the second is "...the duration of 9 192 631 770
periods of the radiation corresponding to the transition between the
two hyperfine levels of the ground state of the cesium 133 atom." (and
affirmed by the CIPM in 1997 that this definition refers to a cesium
atom in its ground state at a temperature of 0 K)
Message: 7
Date: Sat, 29 Nov 2008 00:14:04 -0800
From: Richard Moore richiem@hughes.net
Subject: [time-nuts] TBolt TC recs wanted
To: time-nuts@febo.com
Message-ID: 604DA6FE-2511-4CBB-8D1B-2758A3FA5C58@hughes.net
Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed
Dear nuts -- What TBolt time constant setting have you found to work
best for best frequency accuracy and-or stability?
Dick Moore
time-nuts mailing list
time-nuts@febo.com
https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
End of time-nuts Digest, Vol 52, Issue 85