Re: [time-nuts] AD5791 for EFC control ?
In message f356n894epicao60crhh56g7itkofer...@4ax.com, David writes: The DAC1220 span and offset drift versus temperature on the other hand are 20 to 50 times worse than that of the AD5791 unless you want to spend the $30 difference in price for deglitching so you can use the AD5791 autocalibration feature. Uhm, you lost me there, autocalibration feature ? -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@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. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
On Sun, 21 Apr 2013 06:41:06 +, Poul-Henning Kamp p...@phk.freebsd.dk wrote: In message f356n894epicao60crhh56g7itkofer...@4ax.com, David writes: The DAC1220 span and offset drift versus temperature on the other hand are 20 to 50 times worse than that of the AD5791 unless you want to spend the $30 difference in price for deglitching so you can use the AD5791 autocalibration feature. Uhm, you lost me there, autocalibration feature ? Sorry, I meant the DAC1220 manual self-calibration feature which corrects for gain and offset errors. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
In message lf47n81l7t8su75pimf6it1pfm74n3p...@4ax.com, David writes: On Sun, 21 Apr 2013 06:41:06 +, Poul-Henning Kamp Uhm, you lost me there, autocalibration feature ? Sorry, I meant the DAC1220 manual self-calibration feature which corrects for gain and offset errors. ahh. But those are not really important in a PLL context... -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@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. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
On Sun, 21 Apr 2013 09:20:28 +, Poul-Henning Kamp p...@phk.freebsd.dk wrote: In message lf47n81l7t8su75pimf6it1pfm74n3p...@4ax.com, David writes: On Sun, 21 Apr 2013 06:41:06 +, Poul-Henning Kamp Uhm, you lost me there, autocalibration feature ? Sorry, I meant the DAC1220 manual self-calibration feature which corrects for gain and offset errors. ahh. But those are not really important in a PLL context... How long is the PLL time constant compared to the rate of temperature changes? Maybe I misunderstood the application. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
In message u6k7n815ugngmc185k08m7578v1kq8d...@4ax.com, David writes: On Sun, 21 Apr 2013 09:20:28 +, Poul-Henning Kamp How long is the PLL time constant compared to the rate of temperature changes? Maybe I misunderstood the application. I'm working on driving the EFC input of a double-oven OCXO -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@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. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
What drives the EFC? Bert Kehren In a message dated 4/21/2013 7:59:23 A.M. Eastern Daylight Time, p...@phk.freebsd.dk writes: In message u6k7n815ugngmc185k08m7578v1kq8d...@4ax.com, David writes: On Sun, 21 Apr 2013 09:20:28 +, Poul-Henning Kamp How long is the PLL time constant compared to the rate of temperature changes? Maybe I misunderstood the application. I'm working on driving the EFC input of a double-oven OCXO -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@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. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
In message 7601e.32c7850c.3ea53...@aol.com, ewkeh...@aol.com writes: What drives the EFC? A computer implemented PLL controls the DAC which drives the EFC -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@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. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
the DAC1220 is definitely worth a second look. Two wire opto coupling makes it even better. On my next PCB run there will be a test circuit. Thank you for the hint. Bert Kehren In a message dated 4/19/2013 9:44:32 A.M. Eastern Daylight Time, herb...@13thfloor.at writes: On Fri, Apr 19, 2013 at 09:20:05AM -0400, ewkeh...@aol.com wrote: As one of our digital loop projects we took a very, very close look at it, but thought once you consider the application, requirements of external references, its influence on temperature performance and cost, a dithered 20 bit LTC 1655 held at 0.1 C is as good a solution. Texas Instruments has a number of precision DACs for a very reasonable price, for example the DAC1220 (20bit low power delta-sigma, ~8 USD) and precision voltage references to go with (e.g. REF5050, ~2 USD). best, Herbert Have not actually tested a 5791 but I know some one that has an evaluation board. I will ask him to contact you off list Bert Kehren In a message dated 4/19/2013 7:22:16 A.M. Eastern Daylight Time, p...@phk.freebsd.dk writes: Has anybody tried using the AD5791 20bit DAC for EFC control ? -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@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. ___ 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. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
Looked up the DAC1220 and thats a very nice chip for ~$8. Amazing 20 bit resolution and easy to hook up. Regards Paul WB8TSL On Sat, Apr 20, 2013 at 6:00 AM, ewkeh...@aol.com wrote: the DAC1220 is definitely worth a second look. Two wire opto coupling makes it even better. On my next PCB run there will be a test circuit. Thank you for the hint. Bert Kehren In a message dated 4/19/2013 9:44:32 A.M. Eastern Daylight Time, herb...@13thfloor.at writes: On Fri, Apr 19, 2013 at 09:20:05AM -0400, ewkeh...@aol.com wrote: As one of our digital loop projects we took a very, very close look at it, but thought once you consider the application, requirements of external references, its influence on temperature performance and cost, a dithered 20 bit LTC 1655 held at 0.1 C is as good a solution. Texas Instruments has a number of precision DACs for a very reasonable price, for example the DAC1220 (20bit low power delta-sigma, ~8 USD) and precision voltage references to go with (e.g. REF5050, ~2 USD). best, Herbert Have not actually tested a 5791 but I know some one that has an evaluation board. I will ask him to contact you off list Bert Kehren In a message dated 4/19/2013 7:22:16 A.M. Eastern Daylight Time, p...@phk.freebsd.dk writes: Has anybody tried using the AD5791 20bit DAC for EFC control ? -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@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. ___ 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. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
where did you find it for $ 8? Bert In a message dated 4/20/2013 3:48:08 P.M. Eastern Daylight Time, paulsw...@gmail.com writes: Looked up the DAC1220 and thats a very nice chip for ~$8. Amazing 20 bit resolution and easy to hook up. Regards Paul WB8TSL On Sat, Apr 20, 2013 at 6:00 AM, ewkeh...@aol.com wrote: the DAC1220 is definitely worth a second look. Two wire opto coupling makes it even better. On my next PCB run there will be a test circuit. Thank you for the hint. Bert Kehren In a message dated 4/19/2013 9:44:32 A.M. Eastern Daylight Time, herb...@13thfloor.at writes: On Fri, Apr 19, 2013 at 09:20:05AM -0400, ewkeh...@aol.com wrote: As one of our digital loop projects we took a very, very close look at it, but thought once you consider the application, requirements of external references, its influence on temperature performance and cost, a dithered 20 bit LTC 1655 held at 0.1 C is as good a solution. Texas Instruments has a number of precision DACs for a very reasonable price, for example the DAC1220 (20bit low power delta-sigma, ~8 USD) and precision voltage references to go with (e.g. REF5050, ~2 USD). best, Herbert Have not actually tested a 5791 but I know some one that has an evaluation board. I will ask him to contact you off list Bert Kehren In a message dated 4/19/2013 7:22:16 A.M. Eastern Daylight Time, p...@phk.freebsd.dk writes: Has anybody tried using the AD5791 20bit DAC for EFC control ? -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@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. ___ 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. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
I looked at the DAC1220 first, but in an EFC application it worries me to no end that it is a Sigma-Delta DAC. Even if I feed it a clock divided down from the OCXO you're EFC'ing, to avoid beatfrequency effects, I would still worry about harmonic effects and noise. Also, the noise-spec is pretty horrid, almost a microvolt/sqrt(Hz) all the way to one kHz. For comparison the AD5791 is purely static and its noise is below 10 nv/sqrt(Hz) from 5Hz and up. -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@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. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
On Sat, 20 Apr 2013 20:14:06 +, Poul-Henning Kamp p...@phk.freebsd.dk wrote: I looked at the DAC1220 first, but in an EFC application it worries me to no end that it is a Sigma-Delta DAC. Even if I feed it a clock divided down from the OCXO you're EFC'ing, to avoid beatfrequency effects, I would still worry about harmonic effects and noise. Also, the noise-spec is pretty horrid, almost a microvolt/sqrt(Hz) all the way to one kHz. For comparison the AD5791 is purely static and its noise is below 10 nv/sqrt(Hz) from 5Hz and up. I wondered about that also but the low frequency noise is only a couple times higher (RMS conversion factor of 6.6) than that of a chopper or FET input precision operational amplifier which themselves only have twice the noise of an AD5791. The DAC1220 span and offset drift versus temperature on the other hand are 20 to 50 times worse than that of the AD5791 unless you want to spend the $30 difference in price for deglitching so you can use the AD5791 autocalibration feature. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
On Sat, Apr 20, 2013 at 08:14:06PM +, Poul-Henning Kamp wrote: I looked at the DAC1220 first, but in an EFC application it worries me to no end that it is a Sigma-Delta DAC. Even if I feed it a clock divided down from the OCXO you're EFC'ing, to avoid beatfrequency effects, I would still worry about harmonic effects and noise. Also, the noise-spec is pretty horrid, almost a microvolt/sqrt(Hz) all the way to one kHz. For comparison the AD5791 is purely static and its noise is below 10 nv/sqrt(Hz) from 5Hz and up. At a price of ~100 USD (100pcs for ~50 USD each) it's also in a different price range as the DAC1220 which can be bought for less than 20 USD (~8 USD when you buy 100). best, Herbert -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@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. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
Bert I was looking at your price in the leading email. You had ~$8. So thats what I thought it was. I may have misread it sorry. Regards Paul WB8TSL On Sat, Apr 20, 2013 at 7:59 PM, Herbert Poetzl herb...@13thfloor.atwrote: On Sat, Apr 20, 2013 at 08:14:06PM +, Poul-Henning Kamp wrote: I looked at the DAC1220 first, but in an EFC application it worries me to no end that it is a Sigma-Delta DAC. Even if I feed it a clock divided down from the OCXO you're EFC'ing, to avoid beatfrequency effects, I would still worry about harmonic effects and noise. Also, the noise-spec is pretty horrid, almost a microvolt/sqrt(Hz) all the way to one kHz. For comparison the AD5791 is purely static and its noise is below 10 nv/sqrt(Hz) from 5Hz and up. At a price of ~100 USD (100pcs for ~50 USD each) it's also in a different price range as the DAC1220 which can be bought for less than 20 USD (~8 USD when you buy 100). best, Herbert -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@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. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
As one of our digital loop projects we took a very, very close look at it, but thought once you consider the application, requirements of external references, its influence on temperature performance and cost, a dithered 20 bit LTC 1655 held at 0.1 C is as good a solution. Have not actually tested a 5791 but I know some one that has an evaluation board. I will ask him to contact you off list Bert Kehren In a message dated 4/19/2013 7:22:16 A.M. Eastern Daylight Time, p...@phk.freebsd.dk writes: Has anybody tried using the AD5791 20bit DAC for EFC control ? -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@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. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
On Fri, Apr 19, 2013 at 09:20:05AM -0400, ewkeh...@aol.com wrote: As one of our digital loop projects we took a very, very close look at it, but thought once you consider the application, requirements of external references, its influence on temperature performance and cost, a dithered 20 bit LTC 1655 held at 0.1 C is as good a solution. Texas Instruments has a number of precision DACs for a very reasonable price, for example the DAC1220 (20bit low power delta-sigma, ~8 USD) and precision voltage references to go with (e.g. REF5050, ~2 USD). best, Herbert Have not actually tested a 5791 but I know some one that has an evaluation board. I will ask him to contact you off list Bert Kehren In a message dated 4/19/2013 7:22:16 A.M. Eastern Daylight Time, p...@phk.freebsd.dk writes: Has anybody tried using the AD5791 20bit DAC for EFC control ? -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@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. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
In message 20130419134411.gb29...@mail.13thfloor.at, Herbert Poetzl writes: On Fri, Apr 19, 2013 at 09:20:05AM -0400, ewkeh...@aol.com wrote: Texas Instruments has a number of precision DACs for a very reasonable price, for example the DAC1220 (20bit low power delta-sigma, ~8 USD) and precision voltage references to go with (e.g. REF5050, ~2 USD). That reminds me of another thing: I wonder if one is better off using a external precision reference or the VREF output from the OCXO ? It's probably one of those sometimes things... -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@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. ___ 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.
Re: [time-nuts] AD5791 for EFC control ?
Thats a great question and not to distract the thread. The modern references seem far superior to the old ones that might be as simple as a zener diode. However the internal references are almost always in a semi temperature controlled environment near the oven. Kind of interesting trade offs. Regards Paul WB8TSL On Fri, Apr 19, 2013 at 10:40 AM, Poul-Henning Kamp p...@phk.freebsd.dkwrote: In message 20130419134411.gb29...@mail.13thfloor.at, Herbert Poetzl writes: On Fri, Apr 19, 2013 at 09:20:05AM -0400, ewkeh...@aol.com wrote: Texas Instruments has a number of precision DACs for a very reasonable price, for example the DAC1220 (20bit low power delta-sigma, ~8 USD) and precision voltage references to go with (e.g. REF5050, ~2 USD). That reminds me of another thing: I wonder if one is better off using a external precision reference or the VREF output from the OCXO ? It's probably one of those sometimes things... -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 p...@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. ___ 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.