New expansion - UPD858 SSB chassis

[Edd200sx]

I’d have to check the xtals

Those are very rare down here, can only recall records of doing one in 10+ years, and that was for a routine repair, not a modification.

At a minimum, we would need a schematic, the crystal frequencies, and the downmix frequencies (N-codes) for channel 1 and 40 on each band.

On the plus side, no need for broadbanding mods, looks like it was designed in the first place for that bandwidth...

I had a good search online can’t find a schematic for it.
Have to keep digging

[lbcomms]

I had a good search online can’t find a schematic for it.

In most cases it doesn't have to be exact, if the PLL area is identical that will be good enough.

Failing that, that section will need to be reverse engineered, which will add lots of time to the project (well the first one you do)...

[Edd200sx]

IMG_9331.jpeg

Best I can do is this pic for now.

145106 pll with 4008’s

[Techquest]

Look for the Dirland SS3900 its a 12 band two groups of 6

[Otter]

Crystals are 12.66 and 15.36. The frequency at the bottom of the lowest band (12.66MHz) is 24.265, Low band with Hi Xtal (15.36) is 26.965

The varicap tuning for the front end and the output comes from a resistor ladder arrangement which provides a voltage based on the most significant three output lines from the binary adders.

[Edd200sx]

Crystals are 12.66 and 15.36. The frequency at the bottom of the lowest band (12.66MHz) is 24.265, Low band with Hi Xtal (15.36) is 26.965

The varicap tuning for the front end and the output comes from a resistor ladder arrangement which provides a voltage based on the most significant three output lines from the binary adders.

Cheers Tim

[lbcomms]

Best I can do is this pic for now.
145106 pll with 4008’s

That's almost enough information, just need the PLL "N" codes for channels 1 and 40 on each band (Excel will use this to work out channels 2 to 39).

Easiest way is to measure the frequency at pin 2 of the MC145106 PLL and multiply the MHz result by 100.

Example, the Uniden 858 chassis has 0.91 MHz on channel 1 and 1.35 MHz on channel 40 at the equivalent pin on the 858 chip.
0.91 x 100 = 91 and 1.35 x 100 = 135, so the "N" codes on that chassis go from 91 on channel 1 to 135 on channel 40.

This will need to be done for each bank of 40 channels in the radio.

There should be 3 spare pins on the Nano which could be used to control the varicap tuning if you want to do away with the existing band switching scheme.

[Otter]

For the 158, surely the n-codes drop out of this ? (Bottom band Ch.1)
N = 100x( 24.265 -12.66 - 10.695) = 91
You'd add 45 to the ch.1 N-code for the next 5 bands.
91,136,181,226,271,316
Then the Hi band Xtal:
Using the hi Xtal N = 100x( 26.965 -15.36 - 10.695) = 91
91,136,181,226,271,316

Channel 40 N-codes are channel 1 N-code + 44

[Murf]

Hi, I have an old AWA Teleradio AC-5112S Marine 27MHz transceiver with a D8284 PLL.
Has this mod been successfully applied to a radio with this PLL yet please?
Great work by the way, kudos.
Thanks,
Murf. VK4BEM/VK3MP

[lbcomms]

Most of the work on that radio will be to add LSB - marine radios are USB/AM only and nearly all 26/27MHz usage other than marine is LSB. So to make it usable, you'll need to add LSB mode first. It's a standard 2824 chassis with the LSB parts left out, so you'll probably need a junker radio to get the coils etc that you'll require.

Once that is done, it's almost the same as the 858 chassis to modify / expand, click the link below:

viewtopic.php?t=59085

Enjoy...

[RIGeek]

I was hoping to try this modification out. Is there any link to the downloads?

Thank you
73

[EC8AVE]

Hello, good afternoon. I'm interested in trying this channel expansion mod for a Cobra 138XLR to give it a new lease of life. I wanted to know where I could get the files for the mod.
Thanks.

[Antoni]

To continue the thread's resurrection, does anybody know how good these one-small-board programmable oscillators are re phase noise? How clean are they?

Many a chicken box's good reputation relies on a clean local...

[lbcomms]

Many a chicken box's good reputation relies on a clean local...

The programmable oscillators are not clean, they are intended more for clocking digital ICs rather than for use as a local oscillator in a radio. Out of band signals are sometimes only 20dB below the programmed output frequency, so then used as a local oscillator - especially on a radio capable of covering more than a couple of MHz - the spurs will mix with other signals entering the first mixer and some may appear in the IF passband. In non-technical terms, you'll hear signals and noise that are on a different frequency that you are listening to, and on transmit you could potentially interfere with users on a different frequency to the one you are using.

Doesn't apply when using them to the Uniden (and others, except the 80's Cybernets) though. The programmable oscillator is only used as a downmixer to lock the loop, so the 858 / 2824 conversions stay clean. There is a low pass filter between the phase detector and the VCO control voltage with a cutoff of only a few KHz, so the other components of the new downmix signal don't get through to the VCO control voltage and the LO stays clean, no worse than the original circuitry.

This is why the mod won't work on the early 80's cybernets like the PLL02A. The downmix oscillator mixes with the VCO and both outputs are used - the difference frequency (20-17 = approx 3 MHz) locks the loop, and sum becomes the local oscillator (20+17 = 37, minus 10.695 = 27 MHz). The programmable oscillator's spurs happily mix with the VCO and appear on the LO - and the radio performance suffers - badly.

However, they can be used as a local oscillator directly on the old crystal locked SSB 23 channel sets - but only if you want to bring them up to 40 channel operation. Common here in Oz as nearly all 27MHz operation is on 35 LSB, makes an otherwise useless radio usable again. Those old "crystal plex" synthesisers had spurs galore, and the better quality sets had lots of internal filtering to clean them up. They are just wide enough to do a 40 channel conversion OK, but get really ugly really quickly if you try to broadband them.

Hope this explanation helps...