Re: [time-nuts] How are non-Trimble oscillators disciplined?
hp_cisco...@yahoo.com said: One thing I am wondering about is disciplining - how much of this is HW and how much is SW? How are non-Trimble oscillators disciplined? It it common practice to provide a GPS antenna input? I think you are missing the big picture. There are two different types of boxes. There are oscillators. They come in all types of quality. Many of the good ones include an oven to keep the crystal at a constant temperature. OCXO is a common term: Oven Controlled Crystal Oscillator. There are GPSDOs: GPS Disciplined Oscillators, for example the Trimble Thunderbolt. They generally start with a good crystal, then add a GPS unit, some hardware to compare the crystal output with the GPS output, and a microprocessor and software to control everything. Crystals can be tuned slightly by changing the capacitance in parallel with the crystal. You can do that by using a diode for the capacitor and changing the back-bias across the diode. Usually, that voltage comes from a DAC. Discipline just means making it do what you want. Usually, that's put out the right time and/or frequency. GPS is a handy way to get both time and frequency. GPS and a good crystal are a good fit. GPS has lots of short term noise but very very good long term accuracy. Crystals have good short term stability but lots of long term drift. How are non-Trimble oscillators disciplined? Any way you want. Whatever fits your application. One approach is to adjust something with a screwdriver. You have to do that frequently enough so that it meets your needs. Usually, crystals come with specs, things like max drift over a month or year. So you can figure out if you need to calibrate it monthly or annually. It's easier to get better accuracy with GPS. You might have fun browsing data sheets. Feed OCXO to google and see what you get. There is lots of info available on the HP Z3801A. http://www.realhamradio.com/GPS_Frequency_Standard.htm (Time sink warning.) -- These are my opinions. I hate spam. ___ 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] How are non-Trimble oscillators disciplined?
I would like to experiment with some other OCXO's, but am not sure about how they might be disciplined. Almost all of them have just a just pins on their connector, ground, power (one or two voltages) and control with is a variable voltage to apply and the ooutput frequency varies over a small range that is proportional to the input voltage. The exact ratio of Hz/V is device dependance. Typically to discipline an OCXO you use a DAC to drive the control voltages. You would use a micro controller to look at some kind of phase detector and adjust the DAC. The T-Bolt is nice because all this is done for you in a low cost self contained unit. But you don't learn much that way, except how to connect cables. Chris Albertson Redondo Beach, California ___ 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] How are non-Trimble oscillators disciplined?
On 10/01/2012 02:11 AM, Frank Hughes wrote: Hi, Thanks for all the good advice to get me started in this fun technology! Last weekend finished putting up the antenna that was obtained from the suggested China source, powered up a new to me Trimble Thunderbolt, obtained an old used laptop w/ DB-9 serial HW, and have been experimenting w/ the Trimble and LH software. I ran the 10Mhz signal into an HP 5087A DA to feed all the other things in the racks that accept a 10Mhz external reference, works great! One thing I am wondering about is disciplining - how much of this is HW and how much is SW? How are non-Trimble oscillators disciplined? It it common practice to provide a GPS antenna input? You have to realize that the Thunderbolt is an amazingly open design. Just the fact that there is connectors on the PCB to hook up external oscillators makes it sane. Using a thunderbolt to control an off-board oscillator boils down to neither HW or SW details, but rather configuration details (once the oscillator is hooked up). You need to adjust the loop gain to match the EFC sensitivity of your oscillator. The loop bandwidth is another parameter to look for. Is this a default in HW on some oscillator controllers, and/or is there some industry standard command set or protocol to activate disciplining? Industry standard? I'd love to see one. Just look at the HP family of SmartClocks (Z3801A etc) and the Thunderbolt series you will realize they are different in many ways. The different design teams have taken somewhat different approaches. As long as they do that, their preferred way of doing things will differ. Learn your tool and how you can adapt it to your clocks. I would like to experiment with some other OCXO's, but am not sure about how they might be disciplined. If they have an EFC/VC input, check the voltage range and the scale of that range. It can vary a lot. Prepare to do adapter-boards doing scaling and level shifting. Toss in an offset trimmer. Using a voltage reference is highly recommended. Cheers, Magnus ___ 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.
