From dimitri at dotp.com Wed Oct 7 17:00:24 2015 From: dimitri at dotp.com (Dimitri.p) Date: Wed, 07 Oct 2015 14:00:24 -0700 Subject: [GPS_Standard] EFC voltage "cleanliness" Message-ID: <52.F0.15149.CE785165@cdptpa-oedge01> Has anyone ever looked at their EFC output with a scope and if you have, how "clean" is your EFC ? I'm using the controller with a 10811-60111 oscillator and I seem to have a 200mV Pk-Pk jungle of signals when using wide bandwidth on my scope (150 MHz) that calm down to about 20-40mv when I limit the scope bandwidth to 20MHz . Either way it doesn't seem right so I wanted to compare notes. Thanks, Dimitri From dimitri at dotp.com Sun Oct 11 19:42:08 2015 From: dimitri at dotp.com (Dimitri.p) Date: Sun, 11 Oct 2015 16:42:08 -0700 Subject: [GPS_Standard] EFC voltage "cleanliness" In-Reply-To: <56160AFA.7020206@e2000.gdynia.pl> References: <52.F0.15149.CE785165@cdptpa-oedge01> <56160AFA.7020206@e2000.gdynia.pl> Message-ID: <41.74.31992.5D3FA165@cdptpa-oedge01> I also discoverd a a 0mv to 2.5mv to 23mv potential difference between different GND points. So there is "something going on". There is some investigating to be done, for sure. I did use a 15k resistor and 0.1uF cap right at the oscillator EFC input that doesn't "seem" to make a difference. Dimitri At 11:19 PM 10/7/2015, Michal wrote: >Hi Dimitri, > >What you have observed is most often a matter of (relatively) long >ground probe wire. >To verify it you should just connect a hot point of a probe to your >ground and you will notice the same high level of (digital) noise. >Use as short as possible contact between your ground and grounding >ring contact of a scope. >Many demo boards are already equipped with two contacts for such >short contact and it really minimize this effect. > >You may also add some extra filtering on EFC pin with 100nF-1uF >capacitor just on the OCXO pins and 1k series resistor. > >Best 73, >Michal >sp2iqw > > >W dniu 2015-10-07 o 23:00, Dimitri.p pisze: >>Has anyone ever looked at their EFC output with a scope and if you >>have, how "clean" is your EFC ? >> >>I'm using the controller with a 10811-60111 oscillator and I seem >>to have a 200mV Pk-Pk jungle of signals when using wide bandwidth >>on my scope (150 MHz) that calm down to about 20-40mv when I limit >>the scope bandwidth to 20MHz . >> >>Either way it doesn't seem right so I wanted to compare notes. >> >>Thanks, >>Dimitri >> >> >> >> >>______________________________________________________________ >>GPS_Standard mailing list >>Home: http://mailman.qth.net/mailman/listinfo/gps_standard >>Help: http://mailman.qth.net/mmfaq.htm >>Post: mailto:GPS_Standard at mailman.qth.net >> >>This list hosted by: http://www.qsl.net >>Please help support this email list: http://www.qsl.net/donate.html From dimitri at dotp.com Sun Oct 11 19:52:13 2015 From: dimitri at dotp.com (Dimitri.p) Date: Sun, 11 Oct 2015 16:52:13 -0700 Subject: [GPS_Standard] EFC voltage "cleanliness" In-Reply-To: References: <52.F0.15149.CE785165@cdptpa-oedge01> Message-ID: <27.0A.31992.536FA165@cdptpa-oedge01> Which oscillator are you using and what's the largest sample size (S) you've tried ? Dimitri At 09:14 PM 10/10/2015, Dogma wrote: >You are so right about the noise on the EFC line, it's so bad. I am >looking at making a window comparator and isolator to solve this >problem, as mine never gets better that like 3.x-8 From dimitri at dotp.com Sun Oct 11 22:00:07 2015 From: dimitri at dotp.com (Dimitri.p) Date: Sun, 11 Oct 2015 19:00:07 -0700 Subject: [GPS_Standard] EFC voltage "cleanliness" In-Reply-To: References: <52.F0.15149.CE785165@cdptpa-oedge01> <27.0A.31992.536FA165@cdptpa-oedge01> Message-ID: <54.B3.30767.C241B165@cdptpa-oedge03> If you built the controller on a breadboard, I'll say all bets are off as far as power supply issue go. Depending on my power supply arrangement, I get -3.90e-09 by the 4th sample. Try powering up the controller circuit from a separate power supply. For testing you could even use a 9V battery feeding the 7805 regulator for a short time. See what kind of difference it makes. The general idea is to keep the power supplies separate so you can figure out which circuit is having the problem. Dimitri At 04:56 PM 10/11/2015, Dogma wrote: >I should add, that this was just made on a >breadboard, chips are bypassed with .1uf caps >and power rail too, but it is noisy.? > > >Jim > >On Sun, Oct 11, 2015 at 4:54 PM, Dogma ><dogma2u at gmail.com> wrote: >I am using a HP10811 and a sample size of 100, >it does better if i just ground the EFC line. > > >Jim > >On Sun, Oct 11, 2015 at 4:52 PM, Dimitri.p ><dimitri at dotp.com> wrote: >Which oscillator are you using and what's the >largest sample size (S) you've tried ? > >Dimitri > > > >At 09:14 PM 10/10/2015, Dogma wrote: >You are so right about the noise on the EFC >line, it's so bad. I am looking at making a >window comparator and isolator to solve this >problem, as mine never gets better that like 3.x-8 > > > From dimitri at dotp.com Thu Oct 15 16:20:41 2015 From: dimitri at dotp.com (Dimitri.p) Date: Thu, 15 Oct 2015 13:20:41 -0700 Subject: [GPS_Standard] GPS_Std_MonTrol plot Error Message-ID: <56.0A.31790.29A00265@cdptpa-oedge01> Sometimes when trying to plot the EFC changes from the port monitor log file I get the following error: '$F01|' is not a valid integer value. The information in the log file looks "clean". It's running on an XP SP3 machine with montrol being the only application open. Rebooting the machine seems to cure the symptom, but it reappears sooner or later. Dimitri