OCXO Stable Reference and Control Voltages?

[FlyingHacker]

Hi,

What do you folks think is best practice for the reference and control voltages for fine tuning an OCXO? Is a passive divider network off VCC sufficient, or should I be looking at a reference IC, and then a passive divider off the reference? Or something else entirely?

The oscillator itself is quite stable, but if the voltages drift (relative to one another) then any tuning is not so useful.

I'd like to maximize long term accuracy, and have to recal only a couple times a year.

The OCXO is this guy:

http://www.ctscorp.com/wp-content/uploads/2015/12/OCXO196.pdf

Thanks in advance.

[AdamHi]

I've always assumed that you'd want to divide off a stable separate voltage source - but you got me thinking that maybe i'm wrong! If you find the answer elsewhere, please keep this post updated.

BTW - where are you sourcing this OCXO? eBay?

--adam

[grumpydoc]

Depends of the accuracy that you need (or can "calibrate" to).

The tuning range is 7x10^-7, therefore 5% change in your V_FC will be 3.5x10^-8 (35ppb).

Only if you can measure to better than that is worring about V_FC to better than 5% going to do you any good.

[NivagSwerdna]

Maybe you could watch

That's the design I plan to use if I ever get around to adding an OCXO to my Racal Dana 1999.

[suicidaleggroll]

OCXOs typically provide their own reference designed for this purpose.  The pdf you linked shows it's on pin 2.  Use that reference with a voltage divider or similar, and feed the control voltage back into pin 1.

[Cerebus]

The hint is in the datasheet for the part. There's an option - option R - to get this OCXO with a reference voltage output. If they are going to bother to make a part that provides a reference voltage output I'd take that as a hint that it might be a good idea. If you do the calculations on the tuning  range (+/- 0.7ppm) versus the Allan deviation (0.01 ppb @ 1.0 sec) you'll get an idea of the voltage stability needed - 1 part in 140,000 (~7 ppm) would introduce an error equivalent to the Allan deviation.

Moreover, I'd hope that they arranged the drift of the vref output to follow the oscillator drift (with respect to tempco, power etc) in a compensatory fashion - making it a more attractive option than providing your own reference.

[FlyingHacker]

OCXOs typically provide their own reference designed for this purpose.  The pdf you linked shows it's on pin 2.  Use that reference with a voltage divider or similar, and feed the control voltage back into pin 1.

Ahhhhh! I misunderstood how it worked. I thought I was supposed to *apply* 4.0V to that VRef pin, and then apply the correction voltage to the VC pin. This makes much more sense...

Hopefully I did not damage the unit, but it doesn't look like I did on the scope (and in comparison to the other OCXO unit).

Thanks.

[Vgkid]

The internal ocxo voltage is generally rather stable. I have a few that are of similiar size to your CTS unit.
If you want, I can power one up(NOT A CTS, a C-MAC), and record its stability for you. Will not be the greatest, as will have only been running for a few hours before measure>.

[FlyingHacker]

The internal ocxo voltage is generally rather stable. I have a few that are of similiar size to your CTS unit.
If you want, I can power one up(NOT A CTS, a C-MAC), and record its stability for you. Will not be the greatest, as will have only been running for a few hours before measure>.

I appreciate the offer. I am trying to learn about these things. You are referring to recording the change in voltage of the Vref over time? I could do that with my bench meter as well.

Or are you talking about recording the frequency drift? I don't believe I personally have any tools to do that. The bench meter (which does logging) can only go to about 300kHZ (and does not have an external clock), and my frequency counters cannot record things (and they are not even oven controlled, which is why I am messing with these OCXOs.)

[NivagSwerdna]

The hint is in the datasheet for the part. There's an option - option R - to get this OCXO with a reference voltage output.

Yes but the OP didn't state they had option R.  In the absence of option R then the solution might depend on your budget...

.. if you are not price sensitive then a voltage ref ic would work (that's what I'm planning to do) (I think that's what was done in the video I referenced)

[Vgkid]

@ Flying hacker: correct. Still will post anyway.
if you want to seethe long term stability check my 'Allen Deviation' thread in the beginners section.
Somthing is not going good with my setup, if you look at the aldev graphs.
@ NivagSwerdna: I agree. A rather decent 5v ref chip will work out good.

[FlyingHacker]

The hint is in the datasheet for the part. There's an option - option R - to get this OCXO with a reference voltage output.

Yes but the OP didn't state they had option R.  In the absence of option R then the solution might depend on your budget...

.. if you are not price sensitive then a voltage ref ic would work (that's what I'm planning to do) (I think that's what was done in the video I referenced)

Yes, as the OP I do indeed have option R. So that will work great. Thanks. I misunderstood the application.

Vgkid: I will check that thread. Thanks.

[uncle_bob]

The OCXO is this guy:

http://www.ctscorp.com/wp-content/uploads/2015/12/OCXO196.pdf

Thanks in advance.

Hi

If you are looking at that specific CTS OCXO on eBay right now ... beware ... the current drop is defective. They have an internal spurious oscillation. The suspicion is that they sent them to the dump as scrap and they wound up on eBay instead.

Bob

[FlyingHacker]

The OCXO is this guy:

http://www.ctscorp.com/wp-content/uploads/2015/12/OCXO196.pdf

Thanks in advance.

Hi

If you are looking at that specific CTS OCXO on eBay right now ... beware ... the current drop is defective. They have an internal spurious oscillation. The suspicion is that they sent them to the dump as scrap and they wound up on eBay instead.

Hmm. The two I got seem pretty stable. with 20MHz BW limit it looks like a perfect sine. At 300MHz it has a little distortion, but rock solid. How would one identify this problem? I don't have a spectrum analyzer/FFT. My scopes are old school Tek gear.

[Theboel]

The OCXO is this guy:

http://www.ctscorp.com/wp-content/uploads/2015/12/OCXO196.pdf

Thanks in advance.

Hi

If you are looking at that specific CTS OCXO on eBay right now ... beware ... the current drop is defective. They have an internal spurious oscillation. The suspicion is that they sent them to the dump as scrap and they wound up on eBay instead.

Bob

Can You more detail about your information Sir. I need a few of this ocxo.

[Vgkid]

Vgkid: I will check that thread. Thanks.

If you read through it and find something that interests you, I can post the image of the graph(you most likely don't have TimeLab)or anything else that you want, as I still have the files, or atleast I could download them.
-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_
In some documents they suggest using trimmers with tempco's better than 50PPM, but that can get expensive, just use a string of nice resistors, with a pot for trimming.

[FlyingHacker]

Vgkid: I will check that thread. Thanks.

If you read through it and find something that interests you, I can post the image of the graph(you most likely don't have TimeLab)or anything else that you want, as I still have the files, or atleast I could download them.
-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_
In some documents they suggest using trimmers with tempco's better than 50PPM, but that can get expensive, just use a string of nice resistors, with a pot for trimming.

TimeLab looks interesting. Do you then have a TimePod to connect to? This is all stuff beyond my hobby budget for now (and I need a Spectrum Analyzer first anyway).

I checked the thread. It may be a little too academic for my uses (read that as, I don't have the time to dedicate to figuring most of it out ) It is interesting to read, though. Thanks.

[Cerebus]

The hint is in the datasheet for the part. There's an option - option R - to get this OCXO with a reference voltage output.

Yes but the OP didn't state they had option R.  In the absence of option R then the solution might depend on your budget...

Yes, but that's kind of irrelevant in the context of the availability of the option suggesting that such a thing might be useful in conjunction with this part.

[uncle_bob]

The OCXO is this guy:

http://www.ctscorp.com/wp-content/uploads/2015/12/OCXO196.pdf

Thanks in advance.

Hi

If you are looking at that specific CTS OCXO on eBay right now ... beware ... the current drop is defective. They have an internal spurious oscillation. The suspicion is that they sent them to the dump as scrap and they wound up on eBay instead.

Bob

Can You more detail about your information Sir. I need a few of this ocxo.

Hi

Multiple people who have purchased these OCXO's on eBay over the last few weeks have all reported a spurious signal on the output at a level high enough to make the useless for their needs. The signal is broadband and show up as a series of spurs all at about -100 dbc spaced at a uniform interval of 40 to 80 Hz (depending on the exact unit tested).

Bob

[FlyingHacker]

Multiple people who have purchased these OCXO's on eBay over the last few weeks have all reported a spurious signal on the output at a level high enough to make the useless for their needs. The signal is broadband and show up as a series of spurs all at about -100 dbc spaced at a uniform interval of 40 to 80 Hz (depending on the exact unit tested).

Thanks. I don't have a spectrum analyzer, but these look solid on a 300MHz scope. I will slow the scope down and see if I see anything. Does this look like AM modulation, what?

They also seem to work fine as an input to at least two frequency counters. I ordered mine about a month ago. They were bought from a seller selling them in pairs.

[FlyingHacker]

I can trigger off some kind of harmonic at around 333 Hz. Not sure what that is.

[uncle_bob]

Multiple people who have purchased these OCXO's on eBay over the last few weeks have all reported a spurious signal on the output at a level high enough to make the useless for their needs. The signal is broadband and show up as a series of spurs all at about -100 dbc spaced at a uniform interval of 40 to 80 Hz (depending on the exact unit tested).

Thanks. I don't have a spectrum analyzer, but these look solid on a 300MHz scope. I will slow the scope down and see if I see anything. Does this look like AM modulation, what?

They also seem to work fine as an input to at least two frequency counters. I ordered mine about a month ago. They were bought from a seller selling them in pairs.

Hi

It shows up as phase modulation. With a *lot* of care, you can catch it on a scope. You sort of need to know the modulation frequency first and then adjust the scope's trigger to duplicate it.

Bob

[uncle_bob]

I can trigger off some kind of harmonic at around 333 Hz. Not sure what that is.

Hi

You may be triggering off of a spur that is at 333/N where N could be 4,5,6,7 ... who knows.

Bob

[FlyingHacker]

Yeah, I don't think I would be able to tell if these are affected without a spectrum analyzer.

Anyone have one of the bad ones on a scope?

[uncle_bob]

Yeah, I don't think I would be able to tell if these are affected without a spectrum analyzer.

Anyone have one of the bad ones on a scope?

Hi

If you bought them in the last month or so, it's just about certain they are out of the defective lot. They show up in singes, pairs, and trays of 40 or so. There are a variety of seller names. I doubt there are very many real sellers behind all those names.

Bob