If you do get a correct 10MHz Rb standard you don't have to worry about keeping it powered up continuously. I put one in a old SCSI hard drive case and whenever I need to do a critical measurement I just power it up and a couple of minutes later all the stuff on the bench has the reference available. Once done it goes off. Good enough for my lab. My Z3801 GPSDO just sits there off and lonely but there if I need it if time isn't critical and can wait for it to warm up, sync and be ready. Late note, you have to be careful using 75Ω BNCs as they're different from 50Ω connectors and using one in the other can deform them. 73 - Dino KLØS
The definition of time is actually TAI (International Atomic time), it is calculated with an weighted average of hundreds of different devices. A national standard such as NIST is trying to be as close as possible to TAI, but will be always a bit off, since TAI is only calculated, there is no transmitter sending this. It is only the master reference for all other clocks.
There might be another problem with GPS clocks, I am not sure, if this is the case for all. In the process Industrie, where I worked before, we had very strict timing requirements and we used GPS receiver for this. They did (by purpose) only very slowly correct the time and frequency, since fast changes might confuse dependent devices. It might also be possible, that you can set this in software.
This is also why I went with the Rb solution. It really let me see how much the GPDSO was getting yanked around. I was able to see with with both statistics measurement of my frequency counter as well as drift on the scope trace triggered off the Rb. I need to get a more modern frequency counter to get some better data acquisition for long term testing.
@@ZenwizardStudios look like that is what you were using, I just wanted to make sure it wouldn't be susceptible to rf and have noise on the line. I know standard bnc cables are junk!
Normal BNC cables are ok as well however the is a HUGE difference in quality BNC and cheap BNC. Also if you order on the less expensive side of things there are sellers that will send you 50 Ohm ends on 70 Ohm coax. I have also seen 50 ohm ends on 75 ohm coax for CCTV systems. Yep there is a subtle but definite connector difference as well.
@@ZenwizardStudios haven't had much luck with keeping a clean ref signal using any when I am working/ testing 1.5+ kw rf amps, ldmofet are crazy bad with odd harmonics. I'll Just go with using the RG318
very interesting video thank you. I'm wanting to get into scope calibration and I was wondering does the Tektronix type 184 have an external input so that I can link it to a GPS 10 MHz reference? or does it just have an internal oven oscillator?
The 184 does not have an external Ref input but it does have an oven. This is the cause of the 3 hour standby +30 minutes on warmup. The oven is on all the time the unit is plugged in. I allow mine to go cold so it is a LONG warm up time. The TG501 on the other hand can be externally referenced to 1Mhz, 5Mhz or 10Mhz. This does however bypass the internal oscillator so if you set a TG501 up for external ref it will not run standalone any more.
If you do get a correct 10MHz Rb standard you don't have to worry about keeping it powered up continuously. I put one in a old SCSI hard drive case and whenever I need to do a critical measurement I just power it up and a couple of minutes later all the stuff on the bench has the reference available. Once done it goes off. Good enough for my lab. My Z3801 GPSDO just sits there off and lonely but there if I need it if time isn't critical and can wait for it to warm up, sync and be ready. Late note, you have to be careful using 75Ω BNCs as they're different from 50Ω connectors and using one in the other can deform them. 73 - Dino KLØS
Some of the GPS locked oscillators were found to lose sync when a leap second was added. Older firmware was confused at the extra pulse
This is kind of where I am leaning at the moment. I have another 1PPS source that will arrive today for some testing.
The definition of time is actually TAI (International Atomic time), it is calculated with an weighted average of hundreds of different devices. A national standard such as NIST is trying to be as close as possible to TAI, but will be always a bit off, since TAI is only calculated, there is no transmitter sending this. It is only the master reference for all other clocks.
There might be another problem with GPS clocks, I am not sure, if this is the case for all. In the process Industrie, where I worked before, we had very strict timing requirements and we used GPS receiver for this. They did (by purpose) only very slowly correct the time and frequency, since fast changes might confuse dependent devices. It might also be possible, that you can set this in software.
This is also why I went with the Rb solution. It really let me see how much the GPDSO was getting yanked around. I was able to see with with both statistics measurement of my frequency counter as well as drift on the scope trace triggered off the Rb. I need to get a more modern frequency counter to get some better data acquisition for long term testing.
What coax are you using for the distribution? I'm in a rf heavy environment and curious if you have had any issues in that regard. Tia.
The coax is RG318. No issues as of yet. The RG318 is also nice and flexible and can get between things easier.
@@ZenwizardStudios look like that is what you were using, I just wanted to make sure it wouldn't be susceptible to rf and have noise on the line. I know standard bnc cables are junk!
@@ZenwizardStudios Thank you!
Normal BNC cables are ok as well however the is a HUGE difference in quality BNC and cheap BNC. Also if you order on the less expensive side of things there are sellers that will send you 50 Ohm ends on 70 Ohm coax. I have also seen 50 ohm ends on 75 ohm coax for CCTV systems. Yep there is a subtle but definite connector difference as well.
@@ZenwizardStudios haven't had much luck with keeping a clean ref signal using any when I am working/ testing 1.5+ kw rf amps, ldmofet are crazy bad with odd harmonics. I'll Just go with using the RG318
very interesting video thank you. I'm wanting to get into scope calibration and I was wondering does the Tektronix type 184 have an external input so that I can link it to a GPS 10 MHz reference? or does it just have an internal oven oscillator?
The 184 does not have an external Ref input but it does have an oven. This is the cause of the 3 hour standby +30 minutes on warmup. The oven is on all the time the unit is plugged in. I allow mine to go cold so it is a LONG warm up time. The TG501 on the other hand can be externally referenced to 1Mhz, 5Mhz or 10Mhz. This does however bypass the internal oscillator so if you set a TG501 up for external ref it will not run standalone any more.
Want one? Of course! Need one? Noooooooope. But who doesn't want a 10MHz square wave sync'ed to GPS?
Yes. You don't need it... at all. But if you want it, start a UA-cam channel, and get your viewers to pay for it.
@@willthecat3861 😄 Wish I'd thought of that.
It depends on your needs. Aligning a 85 year old radio? Nah, you don't need one.