Reverse engineering an AKD2001

Per previous post, my unusal/1st gen AKD

Some notes:

 

Processor: MC6802

PLL NJ88C30, PLL Reference Oscillator 10MHz, Ref Osc divisor = 400, channel spacing = 25KHz

On RX VCO runs at RX Frequency - 1st IF (10.7MHz)

TX VCO runs at TX Frequency. 

EPROM 27C128 Mapped to $E000-$FFFF (mostly unused, No code before FC00, only 1K used)  

It seems to be a very simple design where the various 'peripheral' parts are memory mapped using a 74LS138 3-8 Decoder, it's not fully decoded so any write which addresses A13, A14, A15 or any combination will enable an output on the 'LS138


 

CP is the clock pulse for the 4024 Binary counter which generates the 1750 Toneburst 

CLK is the PLL Clock

Latch is the PLL Latch signal

FPSEL and FP14 are related to the front panel, not sure how that works yet

OE is the EPROM Select 

And some NOR logic:

 

'FP' signals are related to the front panel, the front panel connector:

 

Amazingly, the entire operating code for the radio takes only 336 bytes!

I don't think I could even persuade a modern processor to start up in that few bytes.

Memory map of EPROM: 

Reset Vector at $FFFE-$FFFF is $FC00 and yes, the code begins at $FC00

IRQ Vector is $FC71 and there's an IRQ handler routine there. 

IRQ seems to be from the front panel, suspect it's generated on keypress

Channel table seems to start at $FE00-$FFC7

The entire channel table: 

0F 00 16 A8 15 14  
1F 10 16 A9 15 15 
2F 20 16 AA 15 16 
3F 30 16 AB 15 17 
4F 40 16 AC 15 18 
5F 50 16 AD 15 19 
6F 60 16 AE 15 1A 
7F 70 16 AF 15 1B 
00 80 16 B0 15 04 
00 90 16 B1 15 05 
00 01 16 B2 15 06 
00 11 16 B3 15 07 
00 21 16 B4 15 08 
00 31 16 B5 15 09 
00 41 16 B6 15 0A 
00 51 16 B7 15 0B 
00 61 16 B8 15 0C 
00 71 16 B9 15 0D 
00 81 16 BA 15 0E 
00 91 16 BB 15 0F 
00 02 16 BC 15 10 
00 12 16 BD 15 11 
00 22 16 BE 15 12 
00 32 16 BF 15 13 
00 42 16 94 14 E8 
00 52 16 95 14 E9 
00 62 16 96 14 EA 
00 72 16 97 14 EB 
00 82 16 98 14 EC 
00 92 16 99 14 ED 
00 03 16 9A 14 EE 
00 13 16 9B 14 EF 
00 23 16 9C 14 F0 
00 33 16 9D 14 F1 
00 43 16 9E 14 F2 
00 53 16 9F 14 F3 
00 63 16 A0 14 F4 
00 73 16 A1 14 F5 
00 83 16 A2 14 F6 
00 93 16 A3 14 F7 
00 04 16 A4 14 F8 
00 14 16 A5 14 F9 
00 24 16 A6 14 FA 
00 34 16 A7 14 FB 
00 44 16 C8 15 1C 
00 54 16 C9 15 1D 
00 64 16 CA 15 1E 
00 74 16 CB 15 1F 
00 84 16 CC 15 20 
00 94 16 CD 15 21 
00 05 16 CE 15 22 
00 15 16 CF 15 23 
00 06 16 A8 14 FC 
00 16 16 A9 14 FD 
00 26 16 AA 14 FE 
00 36 16 AB 14 FF 
00 46 16 AC 15 00 
00 56 16 AD 15 01 
00 66 16 AE 15 02 
00 76 16 AF 15 03 
00 07 16 C0 15 14 
00 17 16 C1 15 15 
00 27 16 C2 15 16 
00 37 16 C3 15 17 
00 47 16 C4 15 18 
00 57 16 C5 15 19 
00 67 16 C6 15 1A 
00 77 16 C7 15 1B 
00 09 16 C0 14 FC 
00 19 16 C1 14 FD 
00 29 16 C2 14 FE 
00 39 16 C3 14 FF 
00 49 16 C4 15 00 
00 59 16 C5 15 01 
00 69 16 C6 15 02 
00 79 16 C7 15 03 

 

Each channel seems to be 6 bytes of data,  

First channel: 

"0F 00 16 A8 15 14" 

 0F, not sure yet but the first byte only appears on repeater channels where the tone burst can be enabled.

00, The channel number for display on the LEDs

16 A8 is the TX VCO divisor

15 14 is the RX VCO divisor

Our Reference frequency is 25KHz

TX Divisor $16 $A8 equals 5800 Decimal, 

5800 * .025 = 145MHz

RX Divisor $15 $14 = 5396

Same as before, 5396 * .025 = 134.9MHz 

RX Frequency = RX VCO + 10.7MHz IF = 145.6 

So the first channel is set up as a repeater, 

The 4024 chip in the radio seems to be a 1750Hz Toneburst generator, pretty useless these days but it must be switched on or off, not sure if it's in the EPROM or not, yet. 

 

 More later.

Location: Greater Manchester, UK

Comments

Post a Comment

Popular posts from this blog

3D printing the things.

Image
I bought a very cheap Yaesu VX1-R dual band handy last year, it's a cute little thing that covers 2 and 70 but unfortunately it came without the battery. No problem, if it works I'll just buy a replacement, simple. But WOW, the FNB-52 battery is *expensive*, more than three times what I paid for the radio and I could not find one in stock anywhere (I have since found them)... I did find out that the battery pack is a standard 14650 Li-Ion cell in a molded plastic case (presumably including a battery management circuit) and a quick search of everyone's favourite tat bazaar eBay found a pack of two for a not expensive £9 so I ordered some and waited. When they arrived I gave them a few hours on charge and then tried one in the handy, it worked but of course it rattles about and runs the risk of being damaged or, worse, exploding if charged incorrectly as the cell isn't 'protected'. Now I do have a charger but it meant I'd have to remove th...
Read more

MiniPA 70 build, almost there.

Image
A quick update, I needed a heatsink so I found one from an old PC which is more than adequate, weighing in at almost 650 grammes I reckon it should be able to handle the heat. Problem though, it's not flat. Which makes it kind of difficult to mount the board. Now, I could use some standoffs, PCB pillars cut to length but I'm a little light on machine tools to make an accurate job of it and it'd annoy me that it was 'wobbly'. As it happens, I recently bought myself a new toy. A Flashforge 3D printer so this seemed an ideal opportunity to learn to use a CAD program and find out how dimensionally accurate the printer is. It turns out, it's pretty good. And the PLA takes a thread really nicely. So, my Mini PA board now has mounts, I just need to drill and tap some holes in the heatsink.
Read more

Opto-isolated RS-232 for Icom CI-V

Image
First thing I should say, this is my 'design' and I offer no guarantee that it will work for you, be careful if you build one, do your own testing to be certain it's not going to blow up your laptop and or radio, cause a rift in space time and the instant destruction of the entire universe as we know it. OK, guff out of the way, the Opto Isolated CI-V Interface I mentioned earlier, And the schematic! It's rough, hand drawn and there have been a couple of changes to component values, mainly because I found a document that suggests the CI-V interface port on the radio shouldn't have to sink more than 2.5mA, so one resistor has been changed (the 680R on pin 2 of the right hand side opto is now 2K) and another removed completely (the pull up on pin 5 of the left hand side opto) to comply with that. If I short the CI-V line with my ammeter, I get this: 1.6 milliamps, I'm comfortable with that. Currently (yeah yeah) it works reliably as a serial loop...
Read more