Black*star Meteor frequency counter

[Andrewausfa]

Originally Posted by Nymrod121

Andrew, your earlier reply in the thread indicates that your unit (" 'X' series ") has the 'Ext Ref Osc' facility - does it have a high-stability oscillator built-in or is the option a socket and switch, similar to what I've been working on?

Best wishes
Guy

Hi Guy, the TCXO is built in on a seperate PCB as below

Regards - Andrew

[Nymrod121]

Hi Andrew & thanks for taking the lid off/photographing/uploading the picture
Interesting that the original crystal is still in place - that clarifies another line of thought I'd been following.
Being as how your unit receives power from a wall-wart that would leave plenty of 'real-estate' inside the box to retro-fit any spare 10MHz TCXO one might have - the one I've been considering is a Carpenter unit on a PCB recovered from a skip at work some while back, the only downside being that it needs + 18v DC @ up to 500mA. Nothing I can't resolve in due course but I need to prove the concept before doing any further re-engineering

Best wishes
Guy

[karesz*]

Guy,
what kind of TCXO have you exactly?
Sorry, but isn`t an ovenized XO these?
rgds, Karl

[Nymrod121]

Karl,

(Temperature-Controlled Crystal Oscillator - I'd call it ovenised )

The model is 'Ovenair - Audio - Carpenter' type OSC 73-52, 10.000MHz.

Best wishes
Guy

[mhennessy]

Some impressive work going on here

Just to clarify:

TCXO = temperature controlled crystal oscillator. These don't have an oven, but essentially use a temperature sensor to apply a correction voltage to a varicap diode.

OCXO = oven-controlled crystal oscillator. As the name suggests

While I didn't quickly spot a datasheet, the OSC 73-52 looks like an OCXO.

This thread is close to my heart, as I've been thinking about this for many years. I have lots of devices that would take a 10MHz reference, and they all have references of varying qualities. My "best" frequency counter - a Philips PM6680 with the 2.7GHz option - comes with the basic XO option - not even TCXO - and while I rarely need much accuracy (who does ), it just "feels" wrong to have such an advanced bit of gear with a low-end reference.

I have a lovely old HP OCXO module that I plan to build into a box one day, but I guess I keep putting it off because the GPS-disciplined option is so very tempting (if a bit spendy). I was really impressed with Jeremy's build. But we have a GPS reference at work, so really, I should just build the HP unit up and bring it to work to check every year or so...

They need a beefy PSU to cope with warm-up, but once up to temperature, OCXOs generally have a pretty low power dissipation, so can be left on 24/7. I plan to stick mine in the IT cabinet in the cellar, which has a pretty constant temperature (and dry waste heat in a cellar is always welcome).

[Nymrod121]

<chuckle> Hi Mark
Thanks for the clarification. What's the HP module you have?
Cheers
G.

[mhennessy]

It's at home, so I'll let you know later tonight - remind me if/when I forget

[Nymrod121]

OK, will do

By way of a temporary bodge, I connected a 'three-days-since-switched-on' 5MHz OCXO (G4XWD Jim, take a bow) to the external reference of my HP8640B, set the latter to 1215.00kHz in phase-lock mode then adjusted the 5MHz OCXO to give minimal beat with a certain nearby broadcast station . That got the 5MHz OCXO to +/-0.1Hz ... I'm happy with that (and no doubt my ex-Crowsley chum Ian-G4JQT will join me in recollecting that MF group synch tolerances were +/-0.05Hz back in our day - 1979-82).

I then took a second feed of 5MHz to the Meteor's front panel input and adjusted its on-board crystal trimmer to read as near 5.0000000 MHz as I could get. The reasoning behind all of the above is that 1215kHz being a synch'ed group, the drives are referenced to GPSDO (and many thanks to another ex-Beeb chum [Steve Gwilliam] for his helpful advice/confirmation of this).

Cheers
Guy

[karesz*]

Guy,
I asked for exact model because TCXOs have very similar power need as an XTAL oscillator, also dont need +/- voltages and usual is +5/12V or even 3V3-- to give them.
Here is an annoncement over your oscillator and is to read-"Ovenaire OSC73-53 also has HP part# A-0960-0465-1."
For small GPSDO projects is often interessant these old HP OCXO, HP 10811-60111 Precision Oscillator from HP5328A Counter, but in EU are the prices often more as 100-150$ (alone for the ovenized XO)...
Karl

[Nymrod121]

Hello Karl,
Thanks for the links - the HP OCXOs are very tempting! - but I must be self-disciplined (if not GPS) and hold back on spending any more money at the moment ... we have bad weather here and our heating bill next month may be higher than usual
I'll look more at 'Simon's dialogs' after we have had our supper.
Best wishes
Guy

[MrBungle]

I have considered one of these as an OCXO option: http://www.qrp-labs.com/ocxokit

You don't have to follow the instructions and use the synthesizer bit

Annoyingly I had an HP 5315A counter with OCXO but the thing didn't have 10MHz in or out as it was the portable version so I shifted it. That was solid as a rock.

[mhennessy]

Originally Posted by Nymrod121

What's the HP module you have?

It's a HP 10811-60111. An oldie, but good by all accounts. It's a bit of a lump, at around 3 by 2.5 by 2 inches.

I can't remember how I came by it (I've had it for over a decade), so can't comment about price/availability, etc.

The main supply is 12V at 30mA, and the heater supply is 20-30V at 0.5A when cold, and about 120mA when warm. A small toroid and a couple of linear regulators should do the job admirably. The absolute value of the heater supply voltage isn't critical, but it needs to be stable and low noise. I haven't decided what to do about a DA yet. I do have a load of PAL video DAs that will do 10MHz with no problem, but then I have a draw full of EL2020 video op-amps crying out for a job. I imagine I'll put the DA on my bench behind all the gear, leaving the oscillator in the IT bay. But the way things are going, that's at least a decade away

When I build the oscillator, I'll try to make it easy to change to an alternative module in the future - I think it's best to keep options open with this sort of project.

But enough of my rambling

[Nymrod121]

@MrB - thanks for the link; I recall reading that article recently after a similar prompt elsewhere in VRF ... ?

@markhennessy - the HP OCXO is impressive-looking, even if it is a 'lump' !

http://literature.cdn.keysight.com/l...0811-90002.pdf

Jeremy's article looks really nice, too. So much to do, so little time ...

Off to brave the cold ... back to the Meteor!

AtB
Guy

[Nymrod121]

Right then. It seems that to reliably drive the Meteor in 'external reference' mode the incoming signal has to be 4 volts p-p. However, the OAC 10MHz OCXO doesn't provide anywhere near this (less than 2v p-p) and its waveform seems to be rather sensitive to capacitive loading effects, so this indicates to me that a buffer amplifier is needed to match the OCXO's preferred load impedance (whatever that is) and also provide the 4v p-p needed by the Meteor when in 'ext ref' mode.
More work tomorrow, subject to snow clearance duties

Guy

[karesz*]

Guy,
your ovenized OAC-XO is the reference oscillator in HPs 3585A 40MHz SA from 1979 too_as find it please in my attachment, but they didnt gave a detailed schema in Band-III as wroten...
rgds, Karl

[MrBungle]

Very common HP OCXO module mentioned above. Works in the 5328 counters as well if you can get over the fan noise. I had two of them for a bit, in the living room! The module can be picked up for not much if you search for the HP part number. If it has OCXO in the title then it’ll go for a fortune.

Teardown and repair here: http://www.kerrywong.com/2014/07/01/...b-ocxo-repair/

I have ordered the QRP Labs kits. I’ll post about it when I receive them. The plan is to lock their OCXO module to the GPS 1pps with a PLL. This will require adding VCO support to the OCXO and repurposing it which should be fun. I’ve only built one PLL before and it didn’t work very well so no promises.

[Nymrod121]

Hi Karl/MrB,

This is the one I've been using - turns out it's from a HP3325 function generator. Interestingly, it has its own on-board rectifier/smoothing because in its OEM state the supply was from the mains transformer - as an Option (001). The idea being that if you kept the mains supply present, the reference oscillator always had power even if the instrument itself was turned off. That makes perfect sense to me and hence it would seem logical to extend that long-term stability to its upcycled use within the Meteor.

Regarding last night's work, I'm thinking of adapting Jeremy's 74AS04/74AS541 solution (Mark's post #25 refers) in order to bring the drive level up and provide at least one buffered output for other use (via the rear panel SMA socket I fitted the other evening).
AtB
Guy

[MrBungle]

Looks like a good solution. Worth looking at 74lvc as well. Much faster edges than the 74ac and cheaper!

[Pamphonica]

Guy asked me for a bit more info on my simple buffer amp. I've sketched it from memory. It was so simple I never drew a circuit for it.
Good point about 74LVC but I had a few AC chips spare so used those.
IC1 can be 74AC04 or 74AC14, IC2 was 74AC541. Please note, they are 74AC, not 74AS. The diagram incorrectly shows 74AS.
Jeremy

[Nymrod121]

Many thanks, Jeremy!