Re: [time-nuts] Brand new OCXO/TCXO reccomenations
- In an unconditioned/semi-conditioned space. The ambient temperature can
slew 20*C easily over the holdover period. (Think sun hitting an outdoor
enclosure first thing in the morning - the ambient temperature tends to rise
quickly).
It would be interesting to get some data on that. Clouds blocking the sun or
holes in the cloud cover would be a variation.
If the sun hits the enclosure, how long does it take the air inside to heat up
and then how long does it take the gear inside to warm up?
One of the classic tools for problems like this is to attach a big blob of
stuff to the crystal to dampen thermal changes. When you design your test
setup, can you leave room for a brick and collect data with and without when
you toss it into the freezer?
--
These are my opinions. I hate spam.
Some of these sites already have a internet-connected temperature sensor on
a PCB inside a unvented plastic (electronics) enclosure inside the outdoor
weatherpoof metal enclosure. With just a quick look, I found one which
slewed from 21C to 65C over 2 hours within the last couple weeks. I don't
have enough data from that day to tell how much of that 44C was during a
short period - I only have the two samples 2 hours apart.
It's amazing how quickly though a bit of sun on a grey metal box will heat
the box insides up. Cooling is a lot slower though.
I will say though that the idea of adding some thermal mass/insulation
around the oscillator to reduce the temperature change during holdover is
something I'm going to investigate.
On Fri, Sep 25, 2020 at 3:50 AM Hal Murray hmurray@megapathdsl.net wrote:
- In an unconditioned/semi-conditioned space. The ambient temperature
can
slew 20*C easily over the holdover period. (Think sun hitting an
outdoor
enclosure first thing in the morning - the ambient temperature tends to
rise
quickly).
It would be interesting to get some data on that. Clouds blocking the sun
or
holes in the cloud cover would be a variation.
If the sun hits the enclosure, how long does it take the air inside to
heat up
and then how long does it take the gear inside to warm up?
One of the classic tools for problems like this is to attach a big blob of
stuff to the crystal to dampen thermal changes. When you design your test
setup, can you leave room for a brick and collect data with and without
when
you toss it into the freezer?
--
These are my opinions. I hate spam.
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.
--
- Forrest
Hi
On Sep 25, 2020, at 5:44 AM, Hal Murray hmurray@megapathdsl.net wrote:
- In an unconditioned/semi-conditioned space. The ambient temperature can
slew 20*C easily over the holdover period. (Think sun hitting an outdoor
enclosure first thing in the morning - the ambient temperature tends to rise
quickly).
It would be interesting to get some data on that. Clouds blocking the sun or
holes in the cloud cover would be a variation.
A lot of OEM’s have data on that sort of thing. The only time you see 20C in 10
minutes is when the building catches fire :)
Bob
If the sun hits the enclosure, how long does it take the air inside to heat up
and then how long does it take the gear inside to warm up?
One of the classic tools for problems like this is to attach a big blob of
stuff to the crystal to dampen thermal changes. When you design your test
setup, can you leave room for a brick and collect data with and without when
you toss it into the freezer?
--
These are my opinions. I hate spam.
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.
Note that many (maybe all) TCXOs exhibit a hysteresis effect when the
temperature cycles. The stability reported on the datasheets is only
tested with a temperature change in one direction. Cycling temperature
up and down will result in more variation. I've seen two or three
times the datasheet variation when cycling some TCXOs between 0 and 50
C and have had to super select for this. I've also seen non monotonic
jumps. Aging is also very significant in the first few months after
soldering.
Mark
W7MLG
On 9/25/20 8:55 AM, Mark Goldberg wrote:
Note that many (maybe all) TCXOs exhibit a hysteresis effect when the
temperature cycles. The stability reported on the datasheets is only
tested with a temperature change in one direction. Cycling temperature
up and down will result in more variation. I've seen two or three
times the datasheet variation when cycling some TCXOs between 0 and 50
C and have had to super select for this. I've also seen non monotonic
jumps. Aging is also very significant in the first few months after
soldering.
I've attached a couple plots showing the hysteresis of two different
TCXOs (fairly good ones) - so you can get an idea of the amount of
variation.
10 ppm easy, 1 ppm hard.
Those look pretty good compared to some I got. See attached.
I'm purchasing a small, custom frequency run and it is cheaper for me to
super select than for the manufacturer to do so. In fact, they won't set up
that kind of testing for such a low volume. I'm lucky to get them at all.
About 70-80% looked like your examples, but there were a number of
outliers. I would expect that if you buy a common frequency they would have
the process optimized.
Regards,
Mark
W7MLG
On Fri, Sep 25, 2020, 11:51 AM jimlux jimlux@earthlink.net wrote:
On 9/25/20 8:55 AM, Mark Goldberg wrote:
Note that many (maybe all) TCXOs exhibit a hysteresis effect when the
temperature cycles. The stability reported on the datasheets is only
tested with a temperature change in one direction. Cycling temperature
up and down will result in more variation. I've seen two or three
times the datasheet variation when cycling some TCXOs between 0 and 50
C and have had to super select for this. I've also seen non monotonic
jumps. Aging is also very significant in the first few months after
soldering.
I've attached a couple plots showing the hysteresis of two different
TCXOs (fairly good ones) - so you can get an idea of the amount of
variation.
10 ppm easy, 1 ppm hard.
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.
On 9/25/20 5:50 PM, Mark Goldberg wrote:
Those look pretty good compared to some I got. See attached.
I'm purchasing a small, custom frequency run and it is cheaper for me to
super select than for the manufacturer to do so. In fact, they won't set up
that kind of testing for such a low volume. I'm lucky to get them at all.
About 70-80% looked like your examples, but there were a number of
outliers. I would expect that if you buy a common frequency they would have
the process optimized.
interesting.. those were for 49.244792 MHz, which is hardly a standard
frequency. But they were for a space application..
On 9/25/20 8:55 AM, Mark Goldberg wrote:
Note that many (maybe all) TCXOs exhibit a hysteresis effect when the
temperature cycles. The stability reported on the datasheets is only
tested with a temperature change in one direction. Cycling temperature
up and down will result in more variation. I've seen two or three
times the datasheet variation when cycling some TCXOs between 0 and 50
C and have had to super select for this. I've also seen non monotonic
jumps. Aging is also very significant in the first few months after
soldering.
I've attached a couple plots showing the hysteresis of two different
TCXOs (fairly good ones) - so you can get an idea of the amount of
variation.
10 ppm easy, 1 ppm hard.
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to
http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.
time-nuts mailing list -- time-nuts@lists.febo.com
To unsubscribe, go to http://lists.febo.com/mailman/listinfo/time-nuts_lists.febo.com
and follow the instructions there.