Just finished a board to play with crystal oscillators

[hamster_nz]

I've just sent off a board to play around with crystal oscillators and tuning them, loosely based on Figure 6 of this app-note
https://www.analog.com/media/en/technical-documentation/application-notes/an12fa.pdf

The cutouts are so I can make a small insulated enclosure for the center section out of EPP or PS foam, and use the 1/4W resistors to heat it up.

It has a mishmash of SMD and THP is because I'm using mostly junk-box parts.

Anyhow, just thought I'ld share - is anybody else out there working on any experiments, just for the heck of it?

[iMo]

It depends what is your goal..
You will not get something even close to the cheapest tcxo's performance with that design, imho. The performance will be something like the cheapo canned oscillators from old motherboards.
7805 is noisy and not stable enough, you need a special oscillator circuit and crystal to get stability and noise, the LM319 is rather slow, heating with those resistors has no sense unless you control the temperature..

[hamster_nz]

It depends what is your goal..

It's just for experimenting/learning, rather than watching TV

The temperature will be monitored & controlled by a small micro, or more likely a spare Raspberry Pi Zero.

[iMo]

It depends what is your goal..

It's just for experimenting/learning, rather than watching TV

The temperature will be monitored & controlled by a small micro, or more likely a spare Raspberry Pi Zero.

You will experiment, then you will spend $25 for a double ovenized ocxo from ebay, and then you will again sit and watch TV

[hamster_nz]

You will experiment, then you will spend $25 for a double ovenized ocxo from ebay, and then you will again sit and watch TV

Sounds like all of hobby electronics :-)

[iMo]

Indeed, almost all of us went through the same exercise you plan.. Spent months and $$ and finally bought a few ocxos off ebay.

PS: not to discourage you from your experiments - the electronics itself is doable (low noise, high stability, low tempco design), the mechanics is doable (box, heater, etc).
What is difficult is to get the Crystals - the famous vendors use special cut crystals (usually their know-how), with pretty specific params, especially with aging, temperature vs. stability, oscillation modes, phase noise, power levels, eq. LCR params, packaging, etc. Also, you must be able to measure the output with required precision/resolution..