10.7 Mhz FM Deviation Meter Schematic.

[LaserSteve]

Posting for a friend.

Schematic of FM Deviation meter for 10.7 Mhz based on Quartz Crystal and FM IF Chip from 4th Edition RSGB Handbook


At 1 KHz sine input to the TEST TX input, the RX output will be 100 mV per KHz of Deviation with components as shown.

Pic Attached, apologies for the poor scan.

AC voltmeter or other load on circuit must have 200 K-Ohm or greater input impedance.

Requires you to tap the IF of the receiver.

R5 is 50 or 53 Ohms, so will load the IF.

Link to Datasheet:

http://www.intersil.com/content/dam/Intersil/documents/ca30/ca3089.pdf

Steve 

[Richard Crowley]

Nice. Now we just need a time machine to go back to the era when you could still find CA3089.
Not clear how they can be so confident about the calibration?

[LaserSteve]

There are NOS CA3089 and TDA1200  all over Ebay.

In this case, back in the late 1980s when the circuit was developed, they built a large  batch and optimized the design.

I'm aware component tolerances  are an issue. I wish I had the original publication the RSGB lifted this from, but I don't.

Considering most of the Deviation Meter designs on the web are even cruder, this is a winner.

A hint is that the simple directions insist that the receiver be in full limiting. So I imagine the designer was depending on chip tolerances for the calibration.

Do consider the ultra high Quality Factor  and Temperature Stability  of the Crystal Resonator as opposed to the normal Quadrature Coil.

With a good MMIC added on the front end it may not even load the rig that much.

I'm sure the gentleman in question will eventually use Bessel Nulls, but he asked for a hardware specific design. In his case, he needs consistency as he tunes a few older HTs. This will provide it. He does not have access to a Service Monitor or high end SpecAn.

Also take a look at the New Japan Radio web site. Plenty of FM demod chips in the pipeline. Same for On Semi.

Steve

[T3sl4co1l]

So build your own.  They provide the internal schematic.

Looks to be a self contained FM receiver set, sans front end and mixer, so; substitute with whatever you like, and wire up the deviation meter part just the same.

Once you have IF with limiting, you can toss on a phase or ratio detector just fine and read the deviation (and detected audio) directly, or do it with a fancy self contained chip.

Tim

[LaserSteve]

The fun in rolling your own breadboarded clone chips is figuring out how to deal with the multi-emitter transistors and excellent transistor matching available to chip designers.   Note Q80 in the CA3089 data sheet. I'm quite sure there are more then a few of those in the actual silicon.

Steve

[T3sl4co1l]

Indeed.  Quite a lot can be substituted at marginal cost to performance though, at least with a strong enough understanding of what's going on.

Multiple emitters aren't as bad as they look; just take N transistors and connect their B's and C's in parallel.  Whether the Vbe's need to be matched, depends on the circuit (certainly wouldn't hurt, but might not be a big deal here).  What's harder is current mirrors (Vbe's in parallel) and diff pairs (Vbe's in series, opposing), which can be addressed in some cases (say, emitter resistors for good mirror matching at high current).

Tim

[Solder_Junkie]

While I am in favour of making all manner of radio related stuff, you can often find Racal 9009 modulation meters for sale between 30 and 50 GBP on eBay. I used one for years in the telecoms industry and they work very well.

[Solder_Junkie]

A circuit you might consider is this one:
http://users.tpg.com.au/ldbutler/DeviationMeter.htm
The XR215 chips are listed on eBay UK.

[dannyf]

A PLL should do -> essentially by detecting the voltage fed tot he VCO.

Alternatively, you can run the IF + AD9850's signal through an XOR gate -> steady output means in sync -> precision frequency readout.