A better frequency comparison than using the counter is to just run the GPS ref into 1 ch of the scope and the Xtal oven ref into the other channel of the scope and compare the drift.
I agree with this approach. Additionally, I would set up the scope to display a Lissajous pattern to visualize the frequency and phase difference of the two signals. Thanks John!
@@rfburns3472 I would rather use a phase detector IC. If the output is almost constant (ie. almost perfect frequency) stick an ADC on it, to measure phase drift.
Yes, for the final fine tune: "normal trigger" on a 1PPS on one channel View the 10MHz on another channel Adjust the time base so that you see about 10 cycles of the 10MHz updating once per second. Adjust so the 10MHz holds still. Any you will be within about 0.1PPM easily
I made a similar mod to my unit and it too stays in agreement with my GPS stabilized frequency counter. A 74hc14 on the SQ output cleaned it up nicely. Happy New Year to you and yours. John k0ebc
Hello John. That is an excellent suggestion using the 74hc14. I should have shown that inserting a 7 MHz low pass filter also did a great job cleaning up the square wave.
A few thoughts: 1) I like the idea of the box but I think a green LED for "good" is better than blue. 20 years ago, blue LEDs were the cool new thing. Today they don't look "movie prop for the future" any more. 2) I think I would have added another buffer to make one or perhaps two 50 Ohm outputs that could drive a cable but not pass current in the ground of the oscillator PCB. 3) For a little more fun, you could switch the extra buffer between 3V, 5V and perhaps 10V
I use those, I use the tinyPFA to adjust them and have one of them modified to work with a 25Mhz oscillator. They work well within their specifications. Some of the units use junk oscillators taken from some old boards but the majority I have seen are around 0.5ppm and better. The tiny PFA is specifically designed to permit long time measurements of 2 oscillators and compare the, they are basically a nanoVNA H4 with special software. One can buy the unit as is or if you have an H4 just load the software.
Novel channel name, reminds me of when I was very young, I had one of those electronics experimenters box's that could be linked up to make 20 or more circuits. I set it up as a transmitter, but it only went a few feet, so I got some extra transistors and tried making an amplifier for better range, now it went clear across the road. Then I got an 'RF burn' from it and figured I better stop before I get into trouble.
An instrument maker would put that board inside another insulated and temperature-controlled oven. Double-oven oscillators are a thing and cheap to implement. As for SMAs coming loose, they don't if you torque them properly, which is, yanno, how they're meant to be used. :P So presumably this thing is designed to have 50 ohm impedance. So as not to distort the waveform, we'd want to ensure the BNCs are actually impedance controlled units - many cheaper units are quite random!
Great comment! I agree with the double oven approach. That would be an improvement. Also, the device should be in a metal enclosure for better shielding. Yes, the SMA's need to be torqued down but this interferes with my desire to make quick connect/disconnects and BNC's work better for that.
New Guy here. RF Burns, Nice channel name. I dont know if everyone gets it. I`ve had a lot of those because i`m a curious Ham. I even have a few 6o hz high voltage burns along the way.
Al, you got me laughing! For those of you wondering, Al and I have suffered many RF (radio frequency) burns from playing around with our amateur radio transmitters!
Unfortunately the AliExpress item is "no longer available" and for some reason searches, even their suggested searches" for things like "10mhz oxco" are turning up nothing. I found a nearly-identical board on Amazon for $10, though.
Right now on AliExpress....I see it for 3.00 and Free Shipping....Hurry! www.aliexpress.us/item/3256806828290652.html?spm=a2g0o.cart.0.0.70ea38daLVNPhV&mp=1&_gl=1*1mrnseb*_gcl_aw*R0NMLjE3MzU3NjkxOTAuQ2owS0NRaUE3Tk83QmhEc0FSSXNBRGdfaElZX3RVQ2xMNmJlckplQVFHZ1A5N3BuWW1kUEpZOXhhVkJPY0l6OTlUbnZpWFZpZkFDWEJjQWFBb1NLRUFMd193Y0I.*_gcl_au*NjczODA3OTU1LjE3MjgyNjIxNzE.*_ga*NWUwYzJlNDItODdiYi00MDc1LWE2YjItZGMxYzVlNThjMGJjLjE3MDA4NTc5NTQyODE.*_ga_VED1YSGNC7*MTczNTc2ODQ5Ny40My4xLjE3MzU3NjkxOTAuMzAuMC4w&gatewayAdapt=glo2usa
Many of these PCB's, with SMA connectors, are not soldered underneath the board. This is good. Use a very sharp pointed wire-cutter type tool and snip the two ground and center SMA connections from the board. The SMA connector should fall right off. Then use a soldering gun or iron to remove the left-over pieces of the SMA. They will fall right off the board. So easy to do.
2 วันที่ผ่านมา +2
My two units had the same issue. I complained and got a partial refund from AliExpress. I did change a couple of resistors to re-center the frequency adjustment. My review and the mods are shown in this video: th-cam.com/video/uJwIGjTJsKM/w-d-xo.html
A better frequency comparison than using the counter is to just run the GPS ref into 1 ch of the scope and the Xtal oven ref into the other channel of the scope and compare the drift.
I agree with this approach. Additionally, I would set up the scope to display a Lissajous pattern to visualize the frequency and phase difference of the two signals. Thanks John!
@@rfburns3472 I would rather use a phase detector IC. If the output is almost constant (ie. almost perfect frequency) stick an ADC on it, to measure phase drift.
Yes, for the final fine tune: "normal trigger" on a 1PPS on one channel
View the 10MHz on another channel
Adjust the time base so that you see about 10 cycles of the 10MHz updating once per second.
Adjust so the 10MHz holds still. Any you will be within about 0.1PPM easily
I made a similar mod to my unit and it too stays in agreement with my GPS stabilized frequency counter. A 74hc14 on the SQ output cleaned it up nicely. Happy New Year to you and yours. John k0ebc
Hello John. That is an excellent suggestion using the 74hc14. I should have shown that inserting a 7 MHz low pass filter also did a great job cleaning up the square wave.
Among the thousands of products invented and produced by HP, I think one more will be launched!
A few thoughts:
1) I like the idea of the box but I think a green LED for "good" is better than blue. 20 years ago, blue LEDs were the cool new thing. Today they don't look "movie prop for the future" any more.
2) I think I would have added another buffer to make one or perhaps two 50 Ohm outputs that could drive a cable but not pass current in the ground of the oscillator PCB.
3) For a little more fun, you could switch the extra buffer between 3V, 5V and perhaps 10V
Oh wow, nice suggestions to combat any loss of interest in the project!
I use those, I use the tinyPFA to adjust them and have one of them modified to work with a 25Mhz oscillator. They work well within their specifications. Some of the units use junk oscillators taken from some old boards but the majority I have seen are around 0.5ppm and better. The tiny PFA is specifically designed to permit long time measurements of 2 oscillators and compare the, they are basically a nanoVNA H4 with special software. One can buy the unit as is or if you have an H4 just load the software.
Thanks GermanJohn for your idea!
Novel channel name, reminds me of when I was very young, I had one of those electronics experimenters box's that could be linked up to make 20 or more circuits.
I set it up as a transmitter, but it only went a few feet, so I got some extra transistors and tried making an amplifier for better range, now it went clear across the road.
Then I got an 'RF burn' from it and figured I better stop before I get into trouble.
What a learning experience that was Paul!
An instrument maker would put that board inside another insulated and temperature-controlled oven. Double-oven oscillators are a thing and cheap to implement.
As for SMAs coming loose, they don't if you torque them properly, which is, yanno, how they're meant to be used. :P
So presumably this thing is designed to have 50 ohm impedance. So as not to distort the waveform, we'd want to ensure the BNCs are actually impedance controlled units - many cheaper units are quite random!
Great comment! I agree with the double oven approach. That would be an improvement. Also, the device should be in a metal enclosure for better shielding. Yes, the SMA's need to be torqued down but this interferes with my desire to make quick connect/disconnects and BNC's work better for that.
Nice idea using an old wall-wart as a project case. Gotta remember that.
As the new power adapter cases become smaller, the older cases get harder to find. Yes indeed, save them for your future projects.
New Guy here. RF Burns, Nice channel name. I dont know if everyone gets it. I`ve had a lot of those because i`m a curious Ham. I even have a few 6o hz high voltage burns along the way.
Al, you got me laughing! For those of you wondering, Al and I have suffered many RF (radio frequency) burns from playing around with our amateur radio transmitters!
@@rfburns3472 Im KD6CFE Orange County CA. 73`s HiHi
Unfortunately the AliExpress item is "no longer available" and for some reason searches, even their suggested searches" for things like "10mhz oxco" are turning up nothing. I found a nearly-identical board on Amazon for $10, though.
Right now on AliExpress....I see it for 3.00 and Free Shipping....Hurry!
www.aliexpress.us/item/3256806828290652.html?spm=a2g0o.cart.0.0.70ea38daLVNPhV&mp=1&_gl=1*1mrnseb*_gcl_aw*R0NMLjE3MzU3NjkxOTAuQ2owS0NRaUE3Tk83QmhEc0FSSXNBRGdfaElZX3RVQ2xMNmJlckplQVFHZ1A5N3BuWW1kUEpZOXhhVkJPY0l6OTlUbnZpWFZpZkFDWEJjQWFBb1NLRUFMd193Y0I.*_gcl_au*NjczODA3OTU1LjE3MjgyNjIxNzE.*_ga*NWUwYzJlNDItODdiYi00MDc1LWE2YjItZGMxYzVlNThjMGJjLjE3MDA4NTc5NTQyODE.*_ga_VED1YSGNC7*MTczNTc2ODQ5Ny40My4xLjE3MzU3NjkxOTAuMzAuMC4w&gatewayAdapt=glo2usa
@@rfburns3472 the SMA connectors it uses cost more than the entire board.
I bought a couple of the ones with the green board they were both on freq. $16
Almost bought one. Glad they worked out for you.
Doesn't the square wave form appear similar to that of an un-calibrated probe? I just want to adjust the capacitance to square it up.
I would say "similar" but not quite as atrocious looking.
just curious, how difficult was it to unsolder those sma connectors? I like bnc connectors more too. Maybe its an age thing.
Many of these PCB's, with SMA connectors, are not soldered underneath the board. This is good.
Use a very sharp pointed wire-cutter type tool and snip the two ground and center SMA connections from the board. The SMA connector should fall right off.
Then use a soldering gun or iron to remove the left-over pieces of the SMA. They will fall right off the board. So easy to do.
My two units had the same issue. I complained and got a partial refund from AliExpress. I did change a couple of resistors to re-center the frequency adjustment. My review and the mods are shown in this video:
th-cam.com/video/uJwIGjTJsKM/w-d-xo.html
Excellente vidéo. Très bon détail.
Well they are cheap for a reason. Production run rejects.
overpriced at $6
Problem now is the overpriced ones are no longer available. You have to go with 2x overpriced to buy one now.
@@Hank_W6IR Keep looking on AliExpress for a "flash sale". They come and go quickly Hank.
able/baker