uPD2824 PLL expander concept / first prototype

A forum for the discussion of matters of a technical nature. All such activities are undertaken at the readers discretion and own risk. If you don't know what you are doing, don't blame us if it all goes wrong!
User avatar
Admiral
Legend
Legend
Posts: 10000
Joined: 08 Mar 2011, 21:20
Call Sign: 26TM157
Location: MK-UK

Re: uPD2824 PLL expander concept / first prototype

Post by Admiral » 23 Dec 2015, 11:26

My software will only run on DOS and a B&W Toshiba laptop circa 1996. I've tried all the tricks of throttling the processor and using virtual machines, absolutely won't have it.

But the radio is only three years old.
Winner of the 2017 IBTL 'Summer Sizzler' competition

lbcomms
Super Member
Super Member
Posts: 263
Joined: 04 Oct 2015, 08:10
Location: Sydney, Australia

Re: uPD2824 PLL expander concept / first prototype

Post by lbcomms » 23 Dec 2015, 15:13

TM86 wrote:If you decide to turn this into a product, consider this a request to buy one.
The plan is that it will be eventually, just need to find the time to hand make a couple more prototypes, and if that goes well then sit down and design an SMD board with all 3 sections on it.

It's a part time / spare time project for us, so it might take us a month or it might take three or longer, it depends on how busy we get at work...

TM86
Regular
Regular
Posts: 31
Joined: 13 Jul 2015, 02:04
Call Sign: 2TM86

Re: uPD2824 PLL expander concept / first prototype

Post by TM86 » 24 Dec 2015, 00:31

It's a part time / spare time project for us, so it might take us a month or it might take three or longer, it depends on how busy we get at work...
I understand. I call it the interrupt driven day. You can do it IF people will stop interrupting you. But, busy is better than not if you want to pay the bills.

Best of luck to you and the rest of your shop.

User avatar
Ashtec
Legend
Legend
Posts: 9157
Joined: 30 Dec 2009, 23:12
Call Sign: M6LTD
Location: Uk

Re: uPD2824 PLL expander concept / first prototype

Post by Ashtec » 28 Dec 2015, 09:43

Interesting post there lbcomms keep up the good work..................!!!

lbcomms
Super Member
Super Member
Posts: 263
Joined: 04 Oct 2015, 08:10
Location: Sydney, Australia

Re: uPD2824 PLL expander concept / first prototype

Post by lbcomms » 29 Dec 2015, 23:52

Using the same circuit minus the added PLL, another radio was converted in a similar manner.
Took this one on because I want to get a few prototypes installed in different radios done before getting a proper board designed and fabricated, and CB repairs are a bit infrequent these days.

Radio was an early Cybernet SSB radio, a "Super Panther". First generation PLL02A SSB chassis.

It was originally a 23 channel, but had been converted to 18 channels, an old Oz allocation from the late 1970's. 16 of the channels were in the 23 channel allocation, but the numbers didn't line up. Channel 1 was FCC channel 5 and channel 18 was FCC channel 22. Channel 7 was USA 11A (27.095), and channel 16 was 19A (27.195). This must have caused a ton of confusion in the day.

Stage 1: Repair the radio

The radio was in very good physical condition for its age, but not working electrically, lights up but no Tx or Rx.

There was some minor damage to the traces from the ancient 23 to 18 conversion where they presumably were a bit rough with the desoldering. A bit ugly visually, but electrically OK, so no repairs were needed or performed on the trace damage.

The radio was not stock, but had been slightly modified around 1978, reversing the function of the variable dimmer and the Local/DX switch to make it a bright/dim selector and an RF gain control. Going by the documentation that came with the radio, this was an "option" offered by the retailer that sold the radio new. The owner wanted to retain this alteration.

The other mod was a simple rear panel added switch wired to a trace cut on the 320KHz pin of the PLL, giving 27.335 to 27.545 MHz. This switch and wiring was removed.

The owner wanted a straightforward 80 channel conversion, using only the existing channel selector switch. All other front panel controls had to be factory standard, except for the old dimmer / DX-Local role reversal mod. Not even an LED conversion for the meter light allowable, this owner really wanted to keep the thing looking original...

Image
The radio as received. Amazing condition for a 38 year old piece of kit.

Image
Clean inside too.

Image
Clean on the print side, looking good. No hack jobs present, this makes life so much easier.

Image
The faulty component. This thing was 4 MHz off frequency and not responding to the tuning voltage from the PLL.

Image
Epoxy removed from the print side with a hot air pencil and a dental pick, allowing the faulty component to be isolated.

Image
Trace cut to isolate the bad varactor. A new one was then soldered in on the print side of the tiny VCO board.

Image
Even a thousandth of an inch movement (relative to the board and the other components) from car vibration will mess this up.
Conformal coating - a special type of hard varnish for electronics - was added, then dried for an hour on a hot air blower.
It was then returned to the radio, and adjusted for 2V on channel 1. PLL was now locked and radio transmits and receives on all 18 channels.

Sort of. Another couple of problems show up. Tx on SSB is intermittent and knocking the radio makes it come and go.
A similar fault can be seen on AM, but it's only the meter that's intermittent, an external meter shows the wattage is fine.
On the other hand, the external meter showed that SSB transmit was really failing.

Both were easy to find. Two trimpots had gone very dark and multicolored instead of silver.
The two trimpots - Tx meter adjustment and SSB Tx audio gain adjustment...

Image
The two darkened trimpots.

Image
Close up of the corrosion on the SSB Tx audio trimmer.

Image
Both were replaced, but we only had 1K, not the 500 ohm needed in the SSB Tx audio circuit. This formed part of a voltage divider circuit, so a 1K chip resistor was added to board to keep the math correct with a 1K pot in place.

A quick dose of switch cleaner in the front panel pots and switches and the radio worked as good as new. Checked a couple dozen random electrolytics for ESR and leakage current, all were 100%, so the radio was not recapped.

All working as good as new. Next step: Mods :thumbup:

lbcomms
Super Member
Super Member
Posts: 263
Joined: 04 Oct 2015, 08:10
Location: Sydney, Australia

Re: uPD2824 PLL expander concept / first prototype

Post by lbcomms » 29 Dec 2015, 23:54

With the radio now working as new, it was time to update it for 2015, whilst keeping it looking original, with the exception of the channel numbers it displays and operates on.

Owner only needed / wanted operation up to about 27.805 MHz, so we decided to retain the original PLL circuitry in this installation.

Step 1: Expand the bandwidth:

Image
Jury rig pot to manually tune the VCO. Quicker than temp mods to the PLL.

Image
Jury rig pot wired to the VCO. The pot now tunes the radio over the whole VCO range, about 26.2 to 28.2 MHz.

One thing that was apparent straight away - the Tx in this Cybernet was very narrow, good for about 30 channels before it rolled off quite quickly. Only a mixer could be that tight, so time to open it up...

Image
Tx mixer standard. Both of these transformers were quite narrow, but served only to block the 10MHz and 38MHz signals. With the unwanted signals at least 11MHz away, there is a lot of room to play with.

Image
Tx mixer modified. The tap has been changed on the first can, and the second one bypassed altogether. The worst spur was the 38MHz first LO, it measured -36dBC, plenty good enough for a CB set.

Image
Top view of the modified Tx mixer stage.

Image
This shield is needed to prevent the stage from oscillating, so it was put back and the bandwidth measured.

Before: -1dB TX B/W wad 350KHz.

Changes:
C42 > 5p6 (Tx mixer)
C43 > wire link
T4 > change tap from center to outer
T5 > Isolate non-grounded pins and bypass with wire link
C14 > 18p (1st Rx LO)
C61 > 5p6 (sniffer)
C63 > 3u3 (sniffer)
RV3 > 100K (sniffer)

After: -1dB TX B/W = 1170KHz

(1dB loss = reduction in PEP power, i.e. from 15W down to 12W)

We only needed 800KHz B/W, so that finished step 1. The jury rig pot was removed and the PLL IC output pin reconnected.

Step 2: Turn the original 18 channel switch into an 80 channel switch:

Image
The second prototype of the EXPANDA circuit. This was much neater than the first one used in the Hornet.

Image
Wiring added.

Image
The board installed into the radio.

Image
Wiring to the channel display LEDs. Like the last 2, user definable pinning makes for a nice neat wiring job.

Image
The owner wanted an internal link or switch so they could easily "convert" it to and from a standard 40 channel mode, without using a computer. A dual internal presettable switch was added, giving as choice of 40, 80, or 99 channel operation.

Image
Board secured in the radio.

Image
Wiring to the PLL side of the channel switch. Only one wire is used (Violet), this is the one that changes state in between channel switch positions. The other 7 inputs (used to determine the direction of rotation) come from the units digit pins of the original channel switch. It's done this way with binary channel switches, otherwise some steps around the position of channel 23 go in the wrong direction (because clockwise from 23 to 24 would go down, not up as expected, as the PLL frequency code would be going down but the channel number is going up).

The wiring was secured against vibration failures with a small amount of RTV silicone adhesive.

Image
The completed radio, ready for return to its owner...

User avatar
prettypolly
Radio Addict
Radio Addict
Posts: 920
Joined: 09 Jan 2013, 01:57
Call Sign: 26FB185

Re: uPD2824 PLL expander concept / first prototype

Post by prettypolly » 30 Dec 2015, 01:29

Very impressive

User avatar
Admiral
Legend
Legend
Posts: 10000
Joined: 08 Mar 2011, 21:20
Call Sign: 26TM157
Location: MK-UK

Re: uPD2824 PLL expander concept / first prototype

Post by Admiral » 30 Dec 2015, 08:31

Excellent work as ever, but can I say from a simpletons point of view, with respect, that expander board seems to be completely over engineering to get 80 channels from a PLL02A board.
Winner of the 2017 IBTL 'Summer Sizzler' competition

User avatar
Ashtec
Legend
Legend
Posts: 9157
Joined: 30 Dec 2009, 23:12
Call Sign: M6LTD
Location: Uk

Re: uPD2824 PLL expander concept / first prototype

Post by Ashtec » 30 Dec 2015, 08:53

I've got to say i'm very impressed.......

User avatar
cjay
Top Poster
Top Poster
Posts: 1775
Joined: 06 Jun 2013, 21:21
Call Sign: cjay
Location: Manchester
Contact:

Re: uPD2824 PLL expander concept / first prototype

Post by cjay » 30 Dec 2015, 09:42

Hmm, lots of over engineering but as proof of concept it's all good, that's the sort of fun stuff I like to do.

Wasn't there a comment about 'chips with numbers filed off' in the first message though?

lbcomms
Super Member
Super Member
Posts: 263
Joined: 04 Oct 2015, 08:10
Location: Sydney, Australia

Re: uPD2824 PLL expander concept / first prototype

Post by lbcomms » 30 Dec 2015, 15:30

cjay wrote:Wasn't there a comment about 'chips with numbers filed off' in the first message though?
This was deliberate in this case, we heard someone (presumably a friend of the owner) bragging on air how he was going to claim the design as his own, so it was out with the Dremel before the radio was returned :D

The final version will be open source (GPL) for both the hardware (schematic and PCB design) and the firmware, once that's done anybody will be able to make their own and add their own changes if they like - but they won't be able to claim copyright or prevent others from making it for themselves. We had someone try that on us a few years back with another product we were involved with, this time we are not going to give anyone the chance...

lbcomms
Super Member
Super Member
Posts: 263
Joined: 04 Oct 2015, 08:10
Location: Sydney, Australia

Re: uPD2824 PLL expander concept / first prototype

Post by lbcomms » 30 Dec 2015, 15:42

Admiral wrote:seems to be completely over engineering to get 80 channels from a PLL02A board
In this case it was, most people over here are using the Uniden made MB87xx and uPD2824 radios, we just did the 02A to verify the design on a parallel expandable PLL rig. Didn't want to spend too much time broadbanding it for 3MHz+ coverage, adding FM, or anything else too complex.

The owner only wanted 80 channels coverage, so it was perfect for us to test things like the software switch debouncing. The final version when used in an 02A radio will most likely be something similar to the Hornet, maybe a selectable 40 channel and continuous gapless frequency mode covering all of 26 / 27 MHz.

User avatar
cjay
Top Poster
Top Poster
Posts: 1775
Joined: 06 Jun 2013, 21:21
Call Sign: cjay
Location: Manchester
Contact:

Re: uPD2824 PLL expander concept / first prototype

Post by cjay » 30 Dec 2015, 16:39

lbcomms wrote:
cjay wrote:Wasn't there a comment about 'chips with numbers filed off' in the first message though?
This was deliberate in this case, we heard someone (presumably a friend of the owner) bragging on air how he was going to claim the design as his own, so it was out with the Dremel before the radio was returned :D
Ah, fair play then. I've had stuff 'ripped off' before, it's not fun.

Maybe set the code protect too, at least until you're ready to release into the wild.

Mattylad
Top Poster
Top Poster
Posts: 1518
Joined: 03 May 2014, 20:09
Call Sign: RDX64
Location: Lancashire

Re: uPD2824 PLL expander concept / first prototype

Post by Mattylad » 30 Dec 2015, 18:04

Can I ask what software you intend to use for schematic\pcb layout?

TM86
Regular
Regular
Posts: 31
Joined: 13 Jul 2015, 02:04
Call Sign: 2TM86

Re: uPD2824 PLL expander concept / first prototype

Post by TM86 » 31 Dec 2015, 03:36

lbcomms wrote:The final version will be open source (GPL) for both the hardware (schematic and PCB design) and the firmware, once that's done anybody will be able to make their own and add their own changes if they like - but they won't be able to claim copyright or prevent others from making it for themselves. We had someone try that on us a few years back with another product we were involved with, this time we are not going to give anyone the chance...
Thank you for that.

Full disclosure, I've been working on something similar for Cobra 2000 off and on, mostly off, for a while now, based on earlier work I did to revive an SBE Console V. Your solution is far more elegant than the brute force methods I've been using.

That being said, I still want to buy a kit and drop it in my PC244. Just because I can.

Post Reply

Who is online

Users browsing this forum: No registered users and 7 guests