Hello Gerry, A lot of time thinking about to buy a frequency counter to about 1 GHz. Today, after a lot of reading, viewing and thinking, not without your help I made the decision and bought RACAL - DANA 1998 Thank you
On this particular OCXO that pin seems to be not connected, I get 0v out. Looking at the data sheet this could be as an option a reference output but in my particular case it is not. Now I have other OCXO's that do seem to have a reference output, but when I have tested this the voltage changes when I drive it into a 10K load so on balance I think generating a separate reference voltage is a more universal approach. Thanks for your comments. Gerry
I installed a Morion MV85A OCXO on my HP5316A frequency counter which was plagued with a standard crystal oscillator time base not even a TCXO. I like that it generates a 4.5 volt reference for trimming the frequency internally. Tie a 20K (value specified by manufacturer) quality multi turn pot between the Vref pin and GND and tie the wiper to the Adjust pin. I'm pretty sure the designers are using the ovens constant internal temperature to ensure the thermal stability of the reference output. I thought pretty much all OCXOs had an internal reference voltage output interesting that Isotemp unit didn't it certainly had the unused pin for it.
Thanks for the video Gerry. I have the Racal 1992 of military vintage, equipped with an OCXO at 5MHz with a frequency doubling kludge. It has the same five pin connector to the main board that you reference in your sketch. Since the adjustment of the oscillator will not accurately calibrate, this look like a worthwhile project for me. My counter dates from 1991, and has about 3500 hours on it according to an hours gauge hidden on the left inside near the front panel power switch.
No worries, I updated the blog page to include a link to the adaptors on the site too. Should have done that before so thanks for asking and prompting me. Gerry
My guess is 720P on youtube would also look better, since it is encoded from the original which is 1080i, if it was all encoded and edited in progressive. Now it looks much softer than it should be, and 1080P on youtube reveals the problem.
Just had a look and you are right, the default shooting mode is 60i, I will try the next video on 30p and see how that works. Let me know on the next video :)
That may help, I am not sure if that would make the OCXO oven work less well though. I have had the counter on for about 7 hours now and its still rock solid so I think its probably working OK. Gerry
After following the instructions in this video I got rug burns on my butt, but now my frequency counter is deadly accurate. I don't know whether to thank you or cuss you out, lol.
If that OCXO is anything like the ones I've played around with, the pin you've labeled N/C is actually a stable reference that CAN be used with a pot to trim the oscillator (unlike the +5v as you point out..).
Hi Paul, sure it would work, but you will almost certainly need to add a buffer so you can drive multiple 50R loads and transmission lines at the same time. A simple video amp (see the Rubidium LAB standard I done a few weeks ago) is a good choice. Gerry
Still having that Interlaced/deinterlaced issue, WHY? Set your camera to 25P/30P NOT 50i/60i. You don't need to do any interlacing/deinterlacing after that. Do you like this effect or something?
Just noticed if I set the video quality to 720P it doesn't show that lack of deinterlacing. 1080P (which I have my youtube automatically set to) shows lack of deinterlacing like in this picture: imgur(dot)com [remove spaces and this] /6lGsUWA My guess is 30 is progressive and 60 is interlaced, and you shoot at 60i or add deinterlacing which causes it to look like that. Wish I could set only your videos to 720P and watch everything else 1080p which has that option.
Unfortunately my camera is a Sony NX5U (US version) and does not shoot in 25fps mode, only 30/60. However, I have had a look at that I I don't see the problem on Windows or MAC, I asked some other people and they also do not see the problem so I am a bitt miffed I am afraid.
I'm not really qualified to speak but wouldn't it help to wrap the entire subassembly with an insulating material? (Assuming that the crystal oven has its own temperature control) My video is excellent, btw.
I know it's old video but I have a question. Is there any reason for CMOS Isotemp OCXO's to cost way cheaper than sine wave variants? Just got a 131-100 CMOS one of the eBay for $12. I plan to stick it inside my Fluke PM6666. I probed the original oscillator (low cost crystal) and it outputs square wave.
Hi Gerry, I've watched your videos for awhile, thanks. I realize this is thread is old but I just designed a really great precision osc. so I don't have to run my rubidium constantly. All I do is check it every six months or so. If you want a board, I'll run one out to you.
@@gerrysweeney Hi, It's based on OCXO's. What I did was set it up for OCXO's at 5 or 10M and 12 or 5 volts, the board allow you to switch in several different ones or run an external. It's been running in my lab for a few years now with my favorite OCXO the Milliren 260 (MR sells these for $1k each!). You can get them on eBay but I see the price is going up now but still at ~$35 (Dec 2018). MR rates them at ~17 years MTBF so watch the dates although they are so good that you get plenty of life after MTBF. The circuit uses an Atmel uC using a Rotary Encoder to run a 12 bit DAC to a thevenin equivalent circuit. Doubling is from a Cypress CY22800. I usually turn on the Rubidium every few months, tweak the OCXO and then shut off the Rubidium. The MTI 260 is near Rubidium, look it up if you don't know about it. I'll send you the circuit if you like, you can contact me through my web site www.dsgnspec.com (also the board with uC and programmed CY22800 if you want it). I haven't sent it out to any of my publishers yet so don't post it until I tell you. Rob
Great job. But at 930 you discuss thermal drift of the ten turn pot. Who cares? You're using it as a voltage divider. Both sections will drift at the same rate. The ratio is virtually immune from drift. All that matters is the drift of the source feeding the divider and its load regulation.
How did your mod behave itself since this video?, i have a dana 1998 and its internal reference is a rather old metal box with a crusty pcb and half dead foam. Do you think your mod might work for my old dana 1998?
Hi Paul, there is no easy answer to this one, all PCB programs have their little differences and none are easy to use - you have to put the time in and learn them. I use a commercial product called DipTrace which is about the best of a bunch but of course you have to pay for it. I would have a look at KiCad which is open source and free, when I last looked at it I could not get on with it but it is actively developed and many people say its decent. Gerry
Sorry, I didn't have my hearing aid in. I thought you said take off your pants and slide your bum across the floor like a dog. I guess what you said was to solder an oscillator in. Oops, it could happen to anyone!
Hello Gerry,
A lot of time thinking about to buy a frequency counter to about 1 GHz.
Today, after a lot of reading, viewing and thinking, not without your help I made the decision and bought RACAL - DANA 1998
Thank you
On this particular OCXO that pin seems to be not connected, I get 0v out. Looking at the data sheet this could be as an option a reference output but in my particular case it is not. Now I have other OCXO's that do seem to have a reference output, but when I have tested this the voltage changes when I drive it into a 10K load so on balance I think generating a separate reference voltage is a more universal approach. Thanks for your comments. Gerry
I installed a Morion MV85A OCXO on my HP5316A frequency counter which was plagued with a standard crystal oscillator time base not even a TCXO.
I like that it generates a 4.5 volt reference for trimming the frequency internally. Tie a 20K (value specified by manufacturer) quality multi turn pot between the Vref pin and GND and tie the wiper to the Adjust pin.
I'm pretty sure the designers are using the ovens constant internal temperature to ensure the thermal stability of the reference output. I thought pretty much all OCXOs had an internal reference voltage output interesting that Isotemp unit didn't it certainly had the unused pin for it.
Thanks for the video Gerry. I have the Racal 1992 of military vintage, equipped with an OCXO at 5MHz with a frequency doubling kludge. It has the same five pin connector to the main board that you reference in your sketch. Since the adjustment of the oscillator will not accurately calibrate, this look like a worthwhile project for me. My counter dates from 1991, and has about 3500 hours on it according to an hours gauge hidden on the left inside near the front panel power switch.
No worries, I updated the blog page to include a link to the adaptors on the site too. Should have done that before so thanks for asking and prompting me. Gerry
My guess is 720P on youtube would also look better, since it is encoded from the original which is 1080i, if it was all encoded and edited in progressive. Now it looks much softer than it should be, and 1080P on youtube reveals the problem.
Just had a look and you are right, the default shooting mode is 60i, I will try the next video on 30p and see how that works. Let me know on the next video :)
That may help, I am not sure if that would make the OCXO oven work less well though. I have had the counter on for about 7 hours now and its still rock solid so I think its probably working OK. Gerry
After following the instructions in this video I got rug burns on my butt, but now my frequency counter is deadly accurate. I don't know whether to thank you or cuss you out, lol.
If that OCXO is anything like the ones I've played around with, the pin you've labeled N/C is actually a stable reference that CAN be used with a pot to trim the oscillator (unlike the +5v as you point out..).
Hi Paul, sure it would work, but you will almost certainly need to add a buffer so you can drive multiple 50R loads and transmission lines at the same time. A simple video amp (see the Rubidium LAB standard I done a few weeks ago) is a good choice. Gerry
Still having that Interlaced/deinterlaced issue, WHY? Set your camera to 25P/30P NOT 50i/60i. You don't need to do any interlacing/deinterlacing after that. Do you like this effect or something?
Great addition to a nice counter. Might do a similar mod to my HP5315A which only has a TCXO and is quite temperature sensitive.
Yeah I did think that but I am not sure there would be such a demand for them.
Just noticed if I set the video quality to 720P it doesn't show that lack of deinterlacing. 1080P (which I have my youtube automatically set to) shows lack of deinterlacing like in this picture: imgur(dot)com [remove spaces and this] /6lGsUWA
My guess is 30 is progressive and 60 is interlaced, and you shoot at 60i or add deinterlacing which causes it to look like that.
Wish I could set only your videos to 720P and watch everything else 1080p which has that option.
Unfortunately my camera is a Sony NX5U (US version) and does not shoot in 25fps mode, only 30/60. However, I have had a look at that I I don't see the problem on Windows or MAC, I asked some other people and they also do not see the problem so I am a bitt miffed I am afraid.
I'm not really qualified to speak but wouldn't it help to wrap the entire subassembly with an insulating material? (Assuming that the crystal oven has its own temperature control)
My video is excellent, btw.
Hi Mike, its a simple and low cost mod well worth doing to any counter IMHO, Glad you found it useful. Gerry
I know it's old video but I have a question. Is there any reason for CMOS Isotemp OCXO's to cost way cheaper than sine wave variants? Just got a 131-100 CMOS one of the eBay for $12. I plan to stick it inside my Fluke PM6666. I probed the original oscillator (low cost crystal) and it outputs square wave.
Hi Gerry,
I've watched your videos for awhile, thanks. I realize this is thread is old but I just designed a really great precision osc. so I don't have to run my rubidium constantly. All I do is check it every six months or so. If you want a board, I'll run one out to you.
Hi There, what is the board based on, an OCXO? Got any info published anywhere?
@@gerrysweeney
Hi,
It's based on OCXO's. What I did was set it up for OCXO's at 5 or 10M and 12 or 5 volts, the board allow you to switch in several different ones or run an external.
It's been running in my lab for a few years now with my favorite OCXO the Milliren 260 (MR sells these for $1k each!). You can get them on eBay but I see the price is going up now but still at ~$35 (Dec 2018). MR rates them at ~17 years MTBF so watch the dates although they are so good that you get plenty of life after MTBF.
The circuit uses an Atmel uC using a Rotary Encoder to run a 12 bit DAC to a thevenin equivalent circuit. Doubling is from a Cypress CY22800. I usually turn on the Rubidium every few months, tweak the OCXO and then shut off the Rubidium.
The MTI 260 is near Rubidium, look it up if you don't know about it.
I'll send you the circuit if you like, you can contact me through my web site www.dsgnspec.com (also the board with uC and programmed CY22800 if you want it). I haven't sent it out to any of my publishers yet so don't post it until I tell you.
Rob
Hi Tommy, you are welcome, good luck with the hack. Gerry
How did you get rug burns on your but? Didn't happen to me!
Great job. But at 930 you discuss thermal drift of the ten turn pot. Who cares? You're using it as a voltage divider. Both sections will drift at the same rate. The ratio is virtually immune from drift. All that matters is the drift of the source feeding the divider and its load regulation.
What is name of site for so8 adapters?
Super project! Kudos!
How did your mod behave itself since this video?, i have a dana 1998 and its internal reference is a rather old metal box with a crusty pcb and half dead foam.
Do you think your mod might work for my old dana 1998?
zx8401ztv It continues to perform very well, and yes I am sure the mod would also work in the 1998. Gerry
Gerry Sweeney Thankyou Gerry :D, ive took a copy of your diagram just in case i have one of those slow brain days lol :D
zx8401ztv You are most welcome, good luck with the mod. Gerry
Hi Paul, there is no easy answer to this one, all PCB programs have their little differences and none are easy to use - you have to put the time in and learn them. I use a commercial product called DipTrace which is about the best of a bunch but of course you have to pay for it. I would have a look at KiCad which is open source and free, when I last looked at it I could not get on with it but it is actively developed and many people say its decent. Gerry
Thank you. I tried al combination of tin, teen, dy, dee, de and di possible and in the end is "die" :)
...shouldn't "somethink" be "something" ...?
Yet another detailed and informative Vid Gerry.
Just one point. Shouldn't that have been .... improves with age, like a fine wife :)
Hello
Hello Danny, did you have any idea what I was talking about :)
@@gerrysweeney Nope not a clue
Err.....ok...!
Gerry, stop these Racal-Dana video's please!........you'll have me spending more money on mine.....:)
Ah yes, goes without saying....:)
Sorry, I didn't have my hearing aid in. I thought you said take off your pants and slide your bum across the floor like a dog. I guess what you said was to solder an oscillator in. Oops, it could happen to anyone!
Haha, you need an RbS first so you can calibrate it :) More money still....
Visit tindie.com and search for "SMD Adapters - Set #1"