Dual Diversity HRO Receiver

[w5hro]

Hi All

One of my upcoming projects is going to be building a homebrew receiver. I was out in the garage and I have two scrap HRO-50T receivers and I have attached a photo of one below. Well, they are not fully scrap and they have all most all of the parts but I'm going to use them to build a Dual Diversity HRO receiver in one big cabinet.

What I'm wondering is should I just stick with the 455kc IF's or should I expand both sides and make them double conversion? If so what frequency IF would be best and why? I have a near perfect HRO60 and HRO50T, but to be honest the extra stage of conversion the HRO-60 has doesn’t really provide that much more benefit in receiver performance. It does to a certain extent, but I’m not sure it’s worth the added effort.

[k4kyv]

I would just use the two 455 kc/s IFs, since you already have them.  Use one oscillator to feed two separate mixers.

If you really feel into diving into precise mechanical construction, combine the two variable capacitor units into one, using one of the gear boxes.  You will need a 7 section capacitor, one section for the oscillator and three each for each one of the separate rf sections: two rf stages and one mixer in each.

Trying to combine the plug-in coil sections would be cumbersome, but it would be feasible due to the modular construction of the coil units.  You would have a pretty long coil drawer, with 7 coils per section instead of four.

Maybe it would be easier just to keep the two receivers separate, so you tune in each one individually.  Tie the AGCs together and combine the audio outputs from the two detectors to feed one audio amplifier.

Diversity reception works with DSB AM, since the phase of the audio output from the receiver detector is independent of the phase of the rf carrier or sidebands.  It doesn't work very well with CW or SSB, since the phase of the audio from the detector is a product of the phase of the BFO and the phase of the rf signals.  Even if you fed both detectors with the same BFO, the random phase variations between signals arriving in the two antennas would cause cancellations and additions, re-introducing just what you are trying to eliminate with diversity reception - fading of the signal.

[w3jn]

The main advantage to double conversion is better image rejection at higher frequencies.  If that's important to you, go for it.  If most of your work will be at 40 and below, it's probably not worth it.

The second reason you might want to do this is, as you know, it's best for dynamic range to determine the final receiver selectivity right after the first mixer (rather than developing selectivity gradually via cascaded IF stages).  You can do this with a high frequency xtal filter, then downconvert to 455 right after that.  

[w3jn]

You need a common VFO if you want diversity - the LO needs to be EXACTLY in phase sync on each receiver chain, otherwise you'll end up getting beat notes in the audio channel.  You can do what you want with two different LOs with a switching arrangement - LO1 or LO2 as the master, or separate.

First IFs are usually fairly carefully chosen to avoid spurious responses between the harmonics of the LO and signals sneaking in the passband.  In the old Collins SSB manual there was a (almost unreadable and indecipherable) chart plotting LO vs IF vs spurious response, as I recall.  Were it me, I'd choose a first IF according to the availability of a crystal filter and/or IF transformers, and just deal with the spurious responses.

[w3jn]

Mechanically connecting won't work.  The LOs need to be in electrical phase, if you're going to combine the audio channel.  If you're not going to do that, and you're not going to tie the AVC channels together, it's just tantamount to using 2 completely separate receivers.

There's nothing wrong at all with just using 2 receivers on the same frequency.  However, unless the LOs are in phase and the IFs are pretty much identical, the effect can be irritating rather than cool.  Even so, I did this using a Cubic R-3030 dual receiver (each reciever was identical, but completely independent) - fed the audio from each receiver into a stereo amp.  It kinda sucked because selective fading would wipe out one channel (and the noise would rise) and there was an irritating phase beat between the two receivers until I tied the timebases together.

If you tie the AVC channels together, the stronger signal's AVC will automagically reduce the gain on the receiver with the weaker signal, thereby knocking its noise down.

The old skool diversity setups would select the detector with the strongest audio and feed that to the amp (or RTTY detector, as was common).

[k4kyv]

You need a common VFO if you want diversity - the LO needs to be EXACTLY in phase sync on each receiver chain, otherwise you'll end up getting beat notes in the audio channel.  You can do what you want with two different LOs with a switching arrangement - LO1 or LO2 as the master, or separate.

That is true only for SSB/CW reception.  With AM envelope (or synchronous) detection there is no beat note because the audio phase is dependent only on the intermodulation between the AM carrier and sidebands (AKA coherent detection), and is not a function of the phase of the LO or BFO.  The only advantage of a common LO is that both sections will precisely track frequency.

The R-390 series has provisions (per the user manual) for diversity reception by simply tuning both receivers to the same dial setting and tying the AGCs and audio outputs together. I once set up two R-390As with separate antennas, and listened to SWBC stations without tying anything together.  I could hear the dominant signal shift, sometimes very rapidly, from one to the other.  With the speakers separated about like you would set up stereo speakers, it caused a trippy effect with the audio randomly shifting all over the room. Sometimes the audio would fade out completely on one receiver while full quieting on the other and vice versa.

With CW and SSB, there will be be no beat notes from the oscillators if  the receiver is working properly with no rf leakage to the audio output.  However, the audio outputs from the two detectors will themselves beat with each other if the LO and BFO are not exactly in synch.  With both in synch, we still have random phase variations from the separate antennas that will add and subtract randomly, generating fading, nullifying any advantage of the diversity effect.  I have heard of some kind of signal processing with CW that restores the diversity effect, but I'm not  sure anything can be done with SSB.  However, maybe a solution has been developed via DSP technology.

[k4kyv]

The R-390 series have a set of terminals on the strip in back for setting them up in diversity mode. Not sure if the audio is combined at the low level or high level.

[w3jn]

Before you build it, Brian, why not JS those two HRO-50 carcasses into operation and experiment with it a bit?  See which method works the best.

Remember for best diversity effect the two antennas need to be separated by several wavelengths, or use polarization diversity.

[W3SLK]

Don said:

With the speakers separated about like you would set up stereo speakers, it caused a trippy effect with the audio randomly shifting all over the room.

What do you know about this Don? Too much 'Windowpane' during the late 60's early 70's?          

[ke7trp]

I run an R390 and an SP600 in poor mans diversity.  SP600 is on a big L and the R390 is on the Horiz Zep.  The AGC is tied together. I tune each with sperate VFO and digital Freq counters hooked to the oscilators so I can make sure they are dead on.  Its pretty neat.  I can listen to the AMI net here and one station blasts in on the SP600 and its speaker and the next might come in on the R390 and its speaker. Sometimes on a fading band, the station goes from left to right. But its always 100% copy.

C

[ke7trp]

I got the idea as we got a Northern 159 radio that was used in tripple Diverstity.  I never used that radio. Its for sale. But after doing all this research we decided to try it on the other rigs. It is really neat. I find I dont use it as much as I thought after 6 months. I used it last night on the AMI net and it was helpfull. 

On SSB, I would think this would be a different story..

C

[KM1H]

It would seem that a SLM such as the HP 3586B could be used for this. Lots of inputs and outputs on the back panel and they are a lot cheaper than a pair of HRO-500's or FRR-59's with the rest of the package. Only shortcoming is the 3.1 KHz maximum BW filter but Im sure JN has that already figured out   Im still learning on mine.

From what Ive heard and read about the DD-1 is they easily get out of adjustment.

Carl