I need someone to make me a one off pic.

[cb4ever104]

Hi All .

I'm replacing my channel selector and LED display on my Realistic TRC-449 . The old one has been removed and I've replaced it with one from a SS3900 .

I need a pic to map my new channel selector to my upd858 PLL . At the same time I want to add the "HI" band and have a +10khz option . I have all the binary codes mapped out . Basically I just need someone to do the programming , and make me the circuit .

PM me if you can help and I'll send you the .xls file with the info .

Cheers !

Sammy

[14CS06]

Hello Sammy

The Realistic TRC-449 is very similar to ELECTRONICA 360 - PALOMAR SSB 500- PALOMAR SSB 600- STAG 357- WKS 100-VICE PRESIDENT FRANCK II-UNIVERSE 5500-POL MAR SS 40
But PLL are different
PCB PC-196
Channel SW S 408

@+
Claude

[cb4ever104]

Hi Claude .

Actually it's very similar to the President Adams . Except that it doesn't have the SCAN function . As I mentioned I have already replaced the channel selector and led assembly . The original one is an old style mechanical version with the "stupid" flexible ribbon cable (like the Adams) .

All I need is a pic to map the new channel selector codes over to the upd858 program pins .

P1000904.jpg

P1000905.jpg

P1000906.jpg

Thanks.

Sammy

[lbcomms]

If you can't find anyone to do this, let us know and I'll post a circuit here and put the programming code (the hex file) on my site.
We do this sort of thing a lot here in Oz because in the 80's CB was 18 channels and the radios sold then were the same as their USA versions but with a different (18 position) channel selector switch.

To convert them back to 40 when 40 channels was legalized, all you had to do was buy a matching switch and install it. No chance of buying them now, so these days we use the only switch we can still buy (the Uniden SR-219, used in the PC122, AX144, and dozens of other radios) and adapt them into a PLL02A/MB8719/858 radio. LED segments are just a matter of rewiring, and a chip is made to translate the PLL data lines.

One thing though - it will be an AVR chip, not a PIC. AVR's are a lot easier for these one-off or low volume projects and can be coded in minutes.

The AVR chips will cost about USA$3 and the USB programmer hardware for it (for Windows XP and 7) can be had for about $5 shipped. You'll need to supply it with 5V at about 200mA.

Edit: layout of the adapter we use here for 858 radios. Easily built on a piece of veroboard / perfboard:



It's then just a matter of recoding it to suit your input switch.

NOTE: Not shown for clarity - bypass capacitors in the supply (47uF) and input lines (100nF). Neither value critical, anything close will do.

[163TM847]

Only PIC file i have for the 858 is 120 channel [Low and High] and +10kc but it would only work with the original selector, Why did you swap the original ?

[cb4ever104]

Only PIC file i have for the 858 is 120 channel [Low and High] and +10kc but it would only work with the original selector, Why did you swap the original ?

Cause it was knackered Nick . Stripped it down 3 times , but it still gave me problems . I didn't like the mechanics of it either , with the little metal strip and the ball bearing . The new one is much better .

Sammy

[cb4ever104]

If you can't find anyone to do this, let us know and I'll post a circuit here and put the programming code (the hex file) on my site.
We do this sort of thing a lot here in Oz because in the 80's CB was 18 channels and the radios sold then were the same as their USA versions but with a different (18 position) channel selector switch.

To convert them back to 40 when 40 channels was legalized, all you had to do was buy a matching switch and install it. No chance of buying them now, so these days we use the only switch we can still buy (the Uniden SR-219, used in the PC122, AX144, and dozens of other radios) and adapt them into a PLL02A/MB8719/858 radio. LED segments are just a matter of rewiring, and a chip is made to translate the PLL data lines.

One thing though - it will be an AVR chip, not a PIC. AVR's are a lot easier for these one-off or low volume projects and can be coded in minutes.

The AVR chips will cost about USA$3 and the USB programmer hardware for it (for Windows XP and 7) can be had for about $5 shipped. You'll need to supply it with 5V at about 200mA.

Edit: layout of the adapter we use here for 858 radios. Easily built on a piece of veroboard / perfboard:



It's then just a matter of recoding it to suit your input switch.

NOTE: Not shown for clarity - bypass capacitors in the supply (47uF) and input lines (100nF). Neither value critical, anything close will do.

Hi There Ibcomms .

Many thanks for the information . Just one thing . My channel selector has 8 lines . Generates decimal 91 to 135 . I see yours only has 6 . Would that be a problem ?

Sammy

[lbcomms]

I'm assuming SS3900 = Superstar 3900. Haven't seen one here for years, but from memory they use a MC145106 binary input PLL.

Hard-wire the switch for mid-band and you'll need 8 lines to directly select 91-135 binary, but if you ignore the two most significant lines you should still end up with 40 unique N-codes. The bottom 6 bits should give 64 unique possible combinations (zero to 63 binary inclusive).

Once you have a binary number that corresponds to each of the 40 channels, it's a simple matter of dropping them into a "switch statement" to determine the channel number, from there a second switch statement determines the frequency, and the output "N" code can then be calculated.

It sounds like it's doing it the hard way having 3 sets of code every time you change channel, but it's much easier to maintain the code that way. It executes in about 100 microseconds, which give you a maximum of 10,000 channel changes per second, far more than you could achieve with a mechanical switch.

Post or email me the input binary per channel you've worked out, and I'll come up with some numbers.

It doesn't matter if they are not in sequence, have skips / gaps, or even if its lowest / highest binary value half way through the channel range. As long as each position has a unique output, it'll work.

[cb4ever104]

Hi Ibcomms .

Yeah , the SS3900 is very similar to the mkII Cobra 148 GTL-DX . It uses the MC145106 and 2 binary adders . I'm not too sure what you mean by "hard wire the switch" ? Anyway , I've PM'ed you and I'll send you the Excel file once I hear back from you .

Cheers !

Sammy

[lbcomms]

By "hard wire" I mean just wire the switch as if it was for mid-band only (26.965 to 27.405, the original FCC 40 channels), and don't do the bandswitching through the channel selector using the diode matrix like some of the 80's era sets of that ilk did (Intek 120 was one of these). On the other hand, if it used the pair of binary adders (like the original 120 channel Jackson SSB), then the switch will only have one band and the "banks" will be added by a smaller diode matrix.

If the switch you've transplanted is designed for a radio that uses binary adders, all we need is the N codes that are fed to the input of the adders. I'll check the PM for the excel file now...

[cb4ever104]

The channel selector only operates on one "band" (like the Jackson) . As I mentioned previously it generates binary 91 through to 135 . I got your PM and I'll send you the file later . You can let me know once you've had a chance to look at it .

Sammy

[lbcomms]

OK, got a chance this morning. Supplied data from xlsx file:

Code: Select all

OP GREY (P7) BLACK (P6) GREEN (P5) PURPLE (P4) RED (P3) BROWN (P2) BLUE (P1) WHITE (P0) BINARY DEC
 1 0 1 0 1 1 0 1 1 01011011 91
 2 0 1 0 1 1 1 0 0 01011100 92
 3 0 1 0 1 1 1 0 1 01011101 93
 4 0 1 0 1 1 1 1 1 01011111 95
 5 0 1 1 0 0 0 0 0 01100000 96
 6 0 1 1 0 0 0 0 1 01100001 97
 7 0 1 1 0 0 0 1 0 01100010 98
 8 0 1 1 0 0 1 0 0 01100100 100
 9 0 1 1 0 0 1 0 1 01100101 101
10 0 1 1 0 0 1 1 0 01100110 102
11 0 1 1 0 0 1 1 1 01100111 103
12 0 1 1 0 1 0 0 1 01101001 105
13 0 1 1 0 1 0 1 0 01101010 106
14 0 1 1 0 1 0 1 1 01101011 107
15 0 1 1 0 1 1 0 0 01101100 108
16 0 1 1 0 1 1 1 0 01101110 110
17 0 1 1 0 1 1 1 1 01101111 111
18 0 1 1 1 0 0 0 0 01110000 112
19 0 1 1 1 0 0 0 1 01110001 113
20 0 1 1 1 0 0 1 1 01110011 115
21 0 1 1 1 0 1 0 0 01110100 116
22 0 1 1 1 0 1 0 1 01110101 117
23 0 1 1 1 1 0 0 0 01111000 120
24 0 1 1 1 0 1 1 0 01110110 118
25 0 1 1 1 0 1 1 1 01110111 119
26 0 1 1 1 1 0 0 1 01111001 121
27 0 1 1 1 1 0 1 0 01111010 122
28 0 1 1 1 1 0 1 1 01111011 123
29 0 1 1 1 1 1 0 0 01111100 124
30 0 1 1 1 1 1 0 1 01111101 125
31 0 1 1 1 1 1 1 0 01111110 126
32 0 1 1 1 1 1 1 1 01111111 127
33 0 0 1 0 0 0 0 0 10000000 128  ** ERROR IN SPREADSHEET ??? **
34 1 0 0 0 0 0 0 1 10000001 129
35 1 0 0 0 0 0 1 0 10000010 130
36 1 0 0 0 0 0 1 1 10000011 131
37 1 0 0 0 0 1 0 0 10000100 132
38 1 0 0 0 0 1 0 1 10000101 133
39 1 0 0 0 0 1 1 0 10000110 134
40 1 0 0 0 0 1 1 1 10000111 135

Can you recheck channel 33, the maths for that one doesn't add up.

Now to lose two bits / wires, as only six are needed for 40 channels. Ditch the grey and black.

Bottom 6 bits only, 5V to switch common and pulldown resistors

Code: Select all

CH = Indicated channel number
F = GREEN(P5)
E = PURPLE(P4)
D = RED(P3) 
C = BROWN(P2)
B = BLUE(P1)
A = WHITE(P0) 
KK = Decimal value resulting

CH F E D C B A KK 
 1 0 1 1 0 1 1 27
 2 0 1 1 1 0 0 28
 3 0 1 1 1 0 1 29
 4 0 1 1 1 1 1 31
 5 1 0 0 0 0 0 32
 6 1 0 0 0 0 1 33
 7 1 0 0 0 1 0 34
 8 1 0 0 1 0 0 36
 9 1 0 0 1 0 1 37
10 1 0 0 1 1 0 38
11 1 0 0 1 1 1 39
12 1 0 1 0 0 1 41
13 1 0 1 0 1 0 42
14 1 0 1 0 1 1 43
15 1 0 1 1 0 0 44
16 1 0 1 1 1 0 46
17 1 0 1 1 1 1 47
18 1 1 0 0 0 0 48
19 1 1 0 0 0 1 49
20 1 1 0 0 1 1 51
21 1 1 0 1 0 0 52
22 1 1 0 1 0 1 53
23 1 1 1 0 0 0 56
24 1 1 0 1 1 0 54
25 1 1 0 1 1 1 55
26 1 1 1 0 0 1 57
27 1 1 1 0 1 0 58
28 1 1 1 0 1 1 59
29 1 1 1 1 0 0 60
30 1 1 1 1 0 1 61
31 1 1 1 1 1 0 62
32 1 1 1 1 1 1 63
33 0 0 0 0 0 0 00
34 0 0 0 0 0 1 01
35 0 0 0 0 1 0 02
36 0 0 0 0 1 1 03
37 0 0 0 1 0 0 04
38 0 0 0 1 0 1 05
39 0 0 0 1 1 0 06
40 0 0 0 1 1 1 07

It's easy to set up a switch formula here to save two wires.

Formula: If SelectorSwitch is 27 or greater, then add 64, otherwise add 128

Check a couple:
Channel 19 = 49+64, which is 113.
Channel 38 = 5+128, which is 133.
All good...

With the original AVR circuit I use here, and most microprocessor mechanical inputs such as switches and sensors, one side of the input is connected to ground and the other side to the input. A "pullup" resistor will pull the line up to 5V when the switch is open, and when the switch is closed the input voltage is pulled down to zero volts.

Two advantages here, a short or wiring error in the switch wiring will not mean an over-current situation, and the switching action won't induce noise into the 5V rail if bypass capacitors on the inputs are used. With a 30 year old switch and inside a noise sensitive radio, bypass capacitors are highly recommended.

6 bits, but inverted with 6 x pullup resistors and common of switch to ground

Code: Select all

CH = Indicated channel number
F = GREEN(P5)
E = PURPLE(P4)
D = RED(P3) 
C = BROWN(P2)
B = BLUE(P1)
A = WHITE(P0) 
KK = Decimal value resulting

CH F E D C B A KK  
 1 1 0 0 1 0 0 36
 2 1 0 0 0 1 1 35
 3 1 0 0 0 1 0 34
 4 1 0 0 0 0 0 32
 5 0 1 1 1 1 1 31
 6 0 1 1 1 1 0 30
 7 0 1 1 1 0 1 29
 8 0 1 1 0 1 1 27
 9 0 1 1 0 1 0 26
10 0 1 1 0 0 1 25
11 0 1 1 0 0 0 24
12 0 1 0 1 1 0 22
13 0 1 0 1 0 1 21
14 0 1 0 1 0 0 20
15 0 1 0 0 1 1 19
16 0 1 0 0 0 1 17
17 0 1 0 0 0 0 16
18 0 0 1 1 1 1 15
19 0 0 1 1 1 0 14
20 0 0 1 1 0 0 12
21 0 0 1 0 1 1 11
22 0 0 1 0 1 0 10
23 0 0 0 1 1 1 07
24 0 0 1 0 0 1 09
25 0 0 1 0 0 0 08
26 0 0 0 1 1 0 06
27 0 0 0 1 0 1 05
28 0 0 0 1 0 0 04
29 0 0 0 0 1 1 03
30 0 0 0 0 1 0 02
31 0 0 0 0 0 1 01
32 0 0 0 0 0 0 00
33 1 1 1 1 1 1 63
34 1 1 1 1 1 0 62
35 1 1 1 1 0 1 61
36 1 1 1 1 0 0 60
37 1 1 1 0 1 1 59
38 1 1 1 0 1 0 58
39 1 1 1 0 0 1 57
40 1 1 1 0 0 0 56

Again, easy to use a couple of code lines to calculate the original channel or N-code numbers.

Connect the PLL side of the switch commom pin (that in the original radio would have gone to 5V) to ground.
Add 6 resistors, one side to 5V and the other side to each of the P5 to P0 lines above.

Verify the above numbers and I'll put together a list of parts.

I'm assuming here you are comfortable putting together a basic circuit together on veroboard / perfboard...

One more thing, how do you prefer the channels on the other band to work?

Here in Oz, most folk want the channels without skips or jumps, i.e. high band gives 27.415 to 27.805 in 10KHz steps.
For example, high 15 is 27.555, high 23 is 27.635, and so on. Makes it easy to know what frequency you are on, just add 40.

Or do you guys over there prefer the old "+450KHz" method, that gives 27.415 to 27.855 complete with the gaps and channels out of order at 23/24/25? Makes no difference to the difficulty either way, the only change needed either way is a single code block, the wiring is identical.

[cb4ever104]

Hi Sue .

Thanks very much for all your help . Sorry about channel 33 . Copy and paste error on my part I think the rest are fine . I understand everything you've said . Outputs P0 to P5 generate unique values , so they can be used . I'm comfortable building circuits , as long as it's not SMD , as I don't have the adequate soldering equipment . I'd like the radio to generate our "old" FCC MID band and HI band , with the channel spacings (+450khz) . Can you send me a list of parts that I'll need to buy . I can get the regulator and standard parts locally , but I'll need info on the chip and programmer .

Thanks


Sammy

[Crusader]

Old 449,i modded one of these a few years ago for high band,i used this mod...

http://www.cbtricks.com/pub/secret_cb/v ... 3_pg51.pdf

but of course it is like the Adams with the first 7 channels being duplicated,a great rig

[Crusader]

Some more interesting info on the 858 chassis here...

http://bigstrap.tripod.com/