If I wanted to try to compensate for temperature variation of my OCXO... where would I attach a thermistor? What would be a good method of attachment to best respresent the OCXO temp without opening the OCXO and without an undue influence of ambient temperature? The OCXO in question is Datum 1111C... was thinking of using thermal epoxy to hold it down and then putting some insulating material (suggestions) on top of it. -- Doc Bill Dailey KXØO

You should sense the ambient temperature, the OCXO temperature is supposed to be constant. Sensing the ambient you can "anticipate" the move of the OCXO tem,perature controller trying to contrast the variation. Indeed I have seen in the Morion MV201 that the EFC doesn't go only to its pin but also to the temperature controller. It seems that this type of temperature compensation is already used. On Wed, Oct 17, 2012 at 11:16 PM, Bill Dailey <docdailey@gmail.com> wrote: > If I wanted to try to compensate for temperature variation of my OCXO... > where would I attach a thermistor? What would be a good method of > attachment to best respresent the OCXO temp without opening the OCXO and > without an undue influence of ambient temperature? The OCXO in question is > Datum 1111C... was thinking of using thermal epoxy to hold it down and then > putting some insulating material (suggestions) on top of it. > > > > -- > Doc > > Bill Dailey > KXØO > _______________________________________________ > 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. >

Bill Dailey wrote: > If I wanted to try to compensate for temperature variation of my OCXO... > where would I attach a thermistor? What would be a good method of > attachment to best respresent the OCXO temp without opening the OCXO and > without an undue influence of ambient temperature? The OCXO in question > is > Datum 1111C... was thinking of using thermal epoxy to hold it down and > then > putting some insulating material (suggestions) on top of it. > > > > -- > Doc > > Bill Dailey > KXØO Probably easier to monitor oven current and use it to tweak the set point of the oven or tweak the EFC line of the oscillator. As I explained in my 1997 FCS paper on zero gradient ovens, the gradient is the main thing limiting thermal gain. If you have a serious gradient problem, then locating the thermistor far away from the crystal (probably much farther away than the internal one) isn't likely to be productive. The other issue to be aware of is that at some point, the tempco of the electronics comes in play, as explained in my 1997 FCS paper on bridge stabilized oscillators. Increasing the thermal gain of the 10811 beyond 1,000 doesn't improve the tempco because of oscillator circuit pulling. You could possibly improve the tempco by using oven current to tweak the set point. This might cancel out circuit pulling drift, regardless of what was going on with the crystal. I've never actually tried this, but it would be good research project well within the ability of any time nut. Rick Karlquist N6RK

Yes, usually increasing temperature increases frequency, so if you see an oven current increase then slightly take down the EFC (and the other way around if the current decreases). On Thu, Oct 18, 2012 at 12:47 AM, Rick Karlquist <richard@karlquist.com>wrote: > Bill Dailey wrote: > > If I wanted to try to compensate for temperature variation of my OCXO... > > where would I attach a thermistor? What would be a good method of > > attachment to best respresent the OCXO temp without opening the OCXO and > > without an undue influence of ambient temperature? The OCXO in question > > is > > Datum 1111C... was thinking of using thermal epoxy to hold it down and > > then > > putting some insulating material (suggestions) on top of it. > > > > > > > > -- > > Doc > > > > Bill Dailey > > KXØO > > Probably easier to monitor oven current and use it to tweak the > set point of the oven or tweak the EFC line of the oscillator. > As I explained in my 1997 FCS paper on zero gradient ovens, > the gradient is the main thing limiting thermal gain. If you > have a serious gradient problem, then locating the thermistor > far away from the crystal (probably much farther away than the > internal one) isn't likely to be productive. > > The other issue to be aware of is that at some point, the > tempco of the electronics comes in play, as explained in my > 1997 FCS paper on bridge stabilized oscillators. Increasing > the thermal gain of the 10811 beyond 1,000 doesn't improve > the tempco because of oscillator circuit pulling. You could > possibly improve the tempco by using oven current to tweak > the set point. This might cancel out circuit pulling drift, > regardless of what was going on with the crystal. I've never > actually tried this, but it would be good research project > well within the ability of any time nut. > > Rick Karlquist N6RK > > > _______________________________________________ > 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 am using a fury board with on off board oscillator. So a thermistor is representing ocxo current. Right now without tempco I have 1.7 ns sd in ti. If I can eliminate "bumps" when the temperature changes slightly I can back the efc gain down below my current setpoint of 0.5. Right now if I go any lower on efc gain the thermal changes perturb the system a bit too much. When the temp is constant I can bring the efc gain way down...and therefore the sd to close to 1 ns. What I am trying to achieve is a much more stable and narrow time interval. Current performance is pretty close to as good as it gets for a gpsdo but I think I can do a little better. Sent from my iPad On Oct 17, 2012, at 5:55 PM, Azelio Boriani <azelio.boriani@screen.it> wrote: > Yes, usually increasing temperature increases frequency, so if you see an > oven current increase then slightly take down the EFC (and the other way > around if the current decreases). > > On Thu, Oct 18, 2012 at 12:47 AM, Rick Karlquist <richard@karlquist.com>wrote: > >> Bill Dailey wrote: >>> If I wanted to try to compensate for temperature variation of my OCXO... >>> where would I attach a thermistor? What would be a good method of >>> attachment to best respresent the OCXO temp without opening the OCXO and >>> without an undue influence of ambient temperature? The OCXO in question >>> is >>> Datum 1111C... was thinking of using thermal epoxy to hold it down and >>> then >>> putting some insulating material (suggestions) on top of it. >>> >>> >>> >>> -- >>> Doc >>> >>> Bill Dailey >>> KXØO >> >> Probably easier to monitor oven current and use it to tweak the >> set point of the oven or tweak the EFC line of the oscillator. >> As I explained in my 1997 FCS paper on zero gradient ovens, >> the gradient is the main thing limiting thermal gain. If you >> have a serious gradient problem, then locating the thermistor >> far away from the crystal (probably much farther away than the >> internal one) isn't likely to be productive. >> >> The other issue to be aware of is that at some point, the >> tempco of the electronics comes in play, as explained in my >> 1997 FCS paper on bridge stabilized oscillators. Increasing >> the thermal gain of the 10811 beyond 1,000 doesn't improve >> the tempco because of oscillator circuit pulling. You could >> possibly improve the tempco by using oven current to tweak >> the set point. This might cancel out circuit pulling drift, >> regardless of what was going on with the crystal. I've never >> actually tried this, but it would be good research project >> well within the ability of any time nut. >> >> Rick Karlquist N6RK >> >> >> _______________________________________________ >> 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.