Transmission1

Anyone in this forum know about the specs on the crystals used in the cybernet chassis like the GE super base or the Colt Excalibur or the midland 78-999 radios the 10.0525 crystal used in these radios.

Im looking for the specs to have some crystals made and need the

l know its doubled in the circuit
need the frequency tolerance pp
load capacitance xxPF
and other parameters to have crystals made to get 40 up and 40 down on the 02A ssb radios


On the GE Superbase and the Colt Excalibur and the Midland 78.999 ssb radios and most other 02A ssb radios im looking for some xtal information.
the stock is 10.0525 regular 40
the high 40 is10.165
the low 40 is 9.940.
If i was to have some crystals made what would be the specs for the crystal to be cut I cant find any info on the crystals used in these radios and schematics are hard to find with a parts list.

I had a crystal missing from my superstar 2000 and replacement were non existent so a pic chip was used to replace the original pll.

Im looking for the specs on the xtal to have some cut but very hard to find info on.
I did do this for the mb8719/mb8734 chassis and have kits for them 40up/down

I do have a company that will make the xtals but need the specs

They are parallel mode HC49/U case types with a 20pF capacitance rating.

The frequencies you stated will not work optimally though, the oscillator in the radio goes through a frequency doubler to produce a signal in the 20 MHz range that mixes with the 17 MHz VCO. The sum (38 MHz range) becomes the local oscillator and the difference (2 to 3 MHz) is fed to the PLL chip to lock the loop.

Problem: doublers only work optimally at one frequency. Not a problem if you only have one, as in a standard 40 channel radio. Add more and the doubler will become an attenuator. This is why export Cybernets with band crystals use ones that operate directly at the ~20MHz region and omit the doubler:



A much better solution these days would be to put a micro like an Arduino Mini between the channel selector and the PLL02A chip. Would give you super low, low, mid, high, and super high with just the original crystal. One component that costs less than $10 and a bit of wiring / code. It could also incorporate a 10 KHz shifter if you want one (one more wire and one more line of code)... with parts you can buy from eBay / Ali Express / Temu.

Why not try a progrock2 it can have 3 independant frequencies and can be GPS locked but thats only if you need 2 frequencies.
https://qrp-labs.com/progrock2.html

Progrock 2 won't work in this application - 3 reasons:



1) No analog input

This is needed in order for the clarifier to work. The clarifier varactor voltage will have no effect on the Progrock2 module, which means the clarifier control(s) will no longer function.

2) Not enough frequencies selectable

Progrock2 can be configured with either 2 or 3 selectable frequencies. He wanted at least Lo-Mid-Hi and there is an offset needed for LSB to operate, so a minimum of 6 frequencies will need to be selectable. For Super Low / Low / Mid / High / Super High you will need 10 frequencies selectable. If he wants a +10KHz shifter too, he will need 20.

3) Degraded radio performance

The output of the Progrock2 is noisy, 20dB to 30dB higher than a crystal oscillator, with low level spurs a meg each side. Won't affect a digital circuit... but because the oscillator mixes up to become the local oscillator in the radio, the noise and spurs will appear in the receiver. The user will notice the radio picks up more background noise and that it is more susceptible to "splatter" interference from strong stations on other channels due to them mixing with the spurs and appearing in the IF passband (and hence the received audio) of the radio it is installed into.

We used Progrocks and SiLabs synths in work we've done in the last few years - but only in Uniden and RCI chassis (not Cybernets) - in those chassis, the added oscillators are only used to lock the loop (i.e. digital), their local oscillators separated from the PLL downmix by a low pass filter with a cutoff of only a few KHz, so no noise or spurs enters the radios analog circuitry.

I thought getting bespoke crystals made would be expensive, unless you order bulk so unit costs are lower.

In the past I've made my own crystal boards copying the later Cybernet circuit and used the 20Mhz crystals, which are still available btw, and fitted to an old Concorde 1 (10Mhz xtals 059 board). The only issue I had was getting the bandwidth expanded to cover extra frequencies, I put some diode switched trimmer caps on the tuning/mixer coils (like later chassis) which helped, the output stage struggled a bit though but I managed 200 channels out of it.

You could (if you can find one) use a crystal board from a scrap Hygain/Tristar (121 board) radio and just fit that.

Cybernet crystals: https://www.ebay.co.uk/itm/284209692808

If it was me then I'd probably take the easy route and get one of the Microchipz modules that do 120 channels from a single crystal, you could also have the +10/-5khz too. https://www.microchipz.net/cybernet

The only problem I can see is the bandwidth one, 40 channel radios are made with components designed for 40 channels and may not take well to many more channels, multi band radios are made with some differences in component values (lower Q etc.) to increase bandwidth.

You could just buy a classic 'export' radio with it all done of course.

Edit: FYI The older Cybernet export radios (Ham International & Major) that used the 10Mhz crystals actually switched the crystals and trimmers with wires straight to the band switch (is there a worse way?) so poor stability, later (20Mhz) uses voltage switching (lots better). The Colt 485DX Black Shadow (one of my faves) had a crystal board with two oscillator/doublers and two 10Mhz crystals (good enough).

The mixers are the only real bottleneck, the front end and output stage are fairly broad and will do 1.5MHz without too much loss.

The mixers can be made tunable, look at some of the exports with 5 bands, they use PIN diodes at the mixers to switch in some pF of extra capacitance on the lower bands, this method can be added to the 40 channel versions. Just use switching to bias them on for the lower frequencies and experimentation to work out the values - tune the mixers for max at 75% of the frequency range, put the radio down to the 25% mark, and select the capacitance value that gives the highest output. It won't need much, 5 to 20pF should do it.

Couldn't find a schematic for the 485DX, and I've never seen one here in Oz. From what you say they use two doublers, one for each crystal. That will work but would be a bit much for 3 or 5 bands - but still doing it the hard way by modern standards. That microchipz solution is a far better way than crystal switching - same outcome but with none of the stability issues and RF switching methods needed for a multi-crystal conversion.

@lbcomms The Colt 485DX Black Shadow is exactly the same as this (not a very good copy I'm afraid):

When I used AEL crystals in Plymouth (long gone now) they would accept an existing crystal and take measurements from it to derive the specifications.