HA12411 FM IF chip - who used it?

[SteveCG]

I've a Trio KT-50L AM/FM tuner (early 1980s) for which I can find no service information. However I've looked up the ICs that are in it and found that it uses an HA12411 chip for its FM IF.

The data sheet for this is rather sparse but it alludes to its use to provide multi-path output (but this Trio tuner does not use this capability). This multi-path possibility is something I'm interested in so I was wondering whether any Forum FM tuner enthusiasts could let me know of tuners that used this chip so that I can see if I can get service information on them and so work out how to use this capability in my Trio tuner.

It is rather a round about way of getting info - but the Trio tuner is rather complicated to circuit trace and that still would not show me how to use the multi-path capability.

BTW: according to the chip datasheet this multi-path capability seems to be to display the signal level voltage output (chip pin 13) on an oscilloscope.

[Radio Wrangler]

Fuzzy memories. I think Cirkit did some FM tuner designs and I seem to recall coming across the Hitachi FM IF device in their pages.

Might be worth looking

David

[Terry_VK5TM]

HA12411, HA12418, HA1137 and CA3089 are all the same chip (the first two being more updated versions but pinouts are the same).

The only thing I could find re the multipath output is in the application note for the NEC uPC1167C2 (another version of the above).

Application circuit below, multipath output is in upper left corner.

[SteveCG]

Terry, yes, that NEC circuit is close to the Hitachi HA12411 datasheet I have - the exception is the Hitachi circuit has pin 16 connected directly to chassis.

[SteveCG]

Thanks Lawrence - I'll look that one up.

[ms660]

The Mitsubishi DA-F20 has H and V multipath outputs it uses a similar IC (HA1137) vertical from pin 13 and horizontal from the audio output (pin 6) via some coupling capacitors.

Link to schematic here:

https://elektrotanya.com/mitsubishi_...wnload.html#dl

This is just to assist in tracing the circuit in the above schematic....In that schematic there are two IF paths for FM, wide and narrow, the HA1137 is configured for narrow band, for wide band the detector uses a familiar discrete components circuit, the narrow/wide selector switch for the two different bandwidths is S1b in the schematic, the left hand section of S1b doing the selection between the two audio outputs from the two separate detectors.

EDIT: I scrubbed my original post due to an error I made, the above is the re-written replacement.

Lawrence.

[SteveCG]

Thanks for the link Lawrence - I've downloaded the .pdf and will study it closely !

[Jon_G4MDC]

Is this multi path just another name for diversity reception?

The idea being two antennas (maybe Vertical and Horizontal polarisation) feed 2 RF front ends and IFs. The audio is taken from which ever path is producing the best signal level with automatic switch over in the event of fading? They would share the same local oscillator.

[Nuvistor]

Multipath is the same as ghosting on analog television. The aerial picking up the main signal plus reflections from other objects.

[Synchrodyne]

As I understand it, multipath reception of FM signals results in amplitude modulation of the resultant signal. Multipath detectors/indicators effectively compare in some way the AM and FM components of the resultant, for example by orthogonal presentation on an oscilloscope, usually vertical for AM and horizontal for FM. With the CA3089 family of FM subsystem ICs, the signal strength output (before heavy filtering) serves as a proxy for the AM part, with the discriminator output providing the FM part.


Cheers,

[Radio Wrangler]

remember that as FM signals are frequency modulated and a multipath path difference can be many wavelengths, the created nulls will be sharp enough that the frequency variation due to modulation will move the resultant signal up and down the slopes of nulls, creating AM on the signal, possibly (modulation) frequency doubled AM.

If the reflected signal is weak, the nulls/peaks are less developed and the amount of AM is less.

Less than perfect flatness in the transmitter or mistuning of the receiver RF stages can also create some AM without there being multipath conditions.

The reflected signal path length will change if its from a moving object, giving varying AM effect... Aircraft flutter.

Oh and all this AM can get back into the received audio via AM to PM conversion in an mperfect limiter in the IF.

Once you dig into it, there's quite a mess!

David

[SteveCG]

As Lawrence posted earlier the Mitsubishi indeed has a vertical output which is connected straight to the HA chip and a horizontal output which is taken from the stereo decoder input signal. This is the situation that Synchrodyne described.

[G8HQP Dave]

Yes, a typical good FM tuner with no multipath and a loud signal will show a smooth curve with a droop at either side. Essentially you are viewing the top of the IF filter response. You need a loud signal so that you see a wide selection of radio frequencies - BBC Radio 3 will just show you the centre of the curve!

Add multipath and the curve becomes wiggly. If the wigggles move then the reflector is moving.

[SteveCG]

Indeed G8HQP_Dave. I was intending to use the BBC Radio 1 signal as it is compressed to within an inch of its life - and not a trace of slight time gap in the 'programme content' either...

[Julesomega]

Ca. 1977 the Pioneer SX-1980 tuner-amp and doubtless a few other high-end models had multipath detection. This was provided by the PA3001-A demodulator and was switched in the same way as Terry's schematic above. Switching it to Multipath switched the audio to AM which you assessed by ear.
Pin 25 went to the Multipathe switch. I can only assume that reducing the load impedance forced it to compromise the limiting though I can't imagine how.

[G8HQP Dave]

Pin 25 went to the Multipathe switch. I can only assume that reducing the load impedance forced it to compromise the limiting though I can't imagine how.

The 'multipath' output from most chips is normally just a signal strength indicator (SSI), which will get some of its voltage from before the limiting takes place. What is often done is that there are three or four IF amp stages, possibly identical, each of which has a diode detector. The three or four detected voltages are just added together to get SSI. For small signals the summed detector output is dominated by the final stage. For slightly bigger signals the last stage limits and just puts out a constant voltage, but the penultimate stage now outputs a significant voltage so the SSI still indicates something useful. And so on. The result is an SSI which is a very crude approximation of a logarithmic response.

[SteveCG]

I tried my first multi-path experiment using the Trio tuner over the weekend. Using a dual beam 'scope I connected the output from pin 13 (tuning meter/multi-path) of the HA12411 chip to one channel of the 'scope and the input to pin2 (raw FM demodulator signal) of the AN115 stereo decoder chip to the other, and tuned to the local BBC Radio 1 freq. Altering the tuner input signal with an attenuator I was seeing what G8HQP_Dave describes: strong signal - constant voltage on pin13, weaker signal - fluctuations in the signal which approximated to the audio content.

I was intending to try the X-Y plot display style however the control to select this on my scope is jammed solid (so I discovered). Mind you since the display would collapse to nearly a spot with any silence in the audio (just leaving the 19 kc/s stereo pilot tone present), that could risk burning the 'scope's phosphor.

I'm going to think now about my next moves...

[G8HQP Dave]

Pilot tone is typically 10% deviation so enough to avoid burning.