Back to Home
900 MHz MaxTrac
VCO Range Switching
By Robert W. Meister WA1MIK
The 900 MHz conventional MaxTrac radio VCO has two ranges: one covers receive plus normal transmit (896-902 MHz) and the other covers talk-around transmit (935-941 MHz). A signal from the logic board selects the range based on the transmit frequency. This switch occurs around 903 MHz. Anything lower (receive plus normal transmit) uses the low range; anything higher (talk-around) uses the high range. This is determined completely by the radio; RSS has no control over it. From previous experimentation and info I found on the web, I figured that the MaxTrac switches VCO ranges at 903 MHz, so I tried to find the code that selects the proper range. It was straight-forward after I figured out how the radio deals with frequency information.
The information presented here applies to firmware product FVN4019A (conventional 900 MHz MaxTrac).
Most 900 MHz repeater operation uses the 902-903 MHz range for inputs, but there are times when it could be useful to have the radio transmit properly at frequencies just above 903 MHz.
Unfortunately the 900 MHz trunking MaxTrac radio, which uses different firmware, switches VCO ranges at 902.5 MHz. This same fix, implemented at different locations, should also work on radios with trunking firmware. It has not been tried. (Similar code was found around memory address C255h.)
Transmit (and receive) frequencies in the MaxTrac are stored as a 16-bit value which is the number of steps above a base frequency. In the 900 MHz radio, the step size is 12.5 kHz and the base frequency is 894,400 kHz. (It's easier to deal with frequencies in kHz rather than MHz.) To convert 903,000 kHz to the step number, we do the following calculations. Throughout this article, hexadecimal values will have an "h" at the end.
I opened a copy of the 900 MHz firmware (FVN4019A) image file with Hex Workshop. I then searched for the hex value determined above: 02B0h. I found exactly one occurrence, shown in red below, as part of the following instruction sequence. Note that the "File" column lists the addresses in the image file, while the "ROM" column lists the addresses that the code is executing at, when in the radio's EPROM. All of the addresses and data are hexadecimal.
|2E21||AE21||DC 4E||LDD $004E||Get data from RAM|
|2E23||AE23||1A 83 02 B0||CPD #$02B0||Compare to memory value|
|2E27||AE27||25 0E||BCS $AE37||Branch if lower (carry-set)|
|2E29||AE29||1D 00 40||BCLR 0,x,$40||Code to select HI range follows|
That sure has a familiar ring to it. If we change the value 02B0h to something more useful, we should be able to get the VCO to switch ranges at a more convenient frequency. For a test, I chose 915 MHz, because it's the middle of the 902-928 MHz amateur band. Running through the calculations above, I now get a new value of 0670h, which goes into the file at hex addresses 2E25h and 2E26h. Could it be this easy?
Seemed Simple Enough:
To test the operation, I programmed some transmit frequencies into a radio at 902, 903, 904, 914, 915, and 916 MHz and measured the PIN SHIFT signal with a DC voltmeter on pin 14 of the connector that goes between the logic board and the RF board. This pin is low when the VCO's high range (talk-around) is selected and high otherwise (RX and normal TX). At each frequency, a quick press of the PTT button causes the radio to transmit (into a dummy load, of course) and the PIN SHIFT signal can be observed. The data is recorded in the table shown later.
After burning a new EPROM, I installed it into the radio and turned it on. I got a continuous steady beep sound, indicating a problem, which I correctly figured was a checksum error. So I had to find and fix that issue before continuing. See, I knew it couldn't be that easy.
After searching through a partial disassembly of the firmware, I found code where the program verifies that the EPROM's data is correct when the radio is turned on. It does this using a simple 8-bit checksum routine (which is used in a lot of other places in the radio). This table shows the upper half of the radio's memory.
The two resulting checksum bytes of the EPROM, when added together, must equal zero. With the original firmware, the 8-bit checksums for these areas were as shown above; with the new VCO switch frequency value discussed above, these changed to 3Fh and 85h. I changed file location 35FFh (EPROM location B5FFh) from FFh to 3Bh and that restored the original 7Bh checksum value for that half of the EPROM. I erased the previous EPROM and burned a new version. This time the radio powered up and worked perfectly.
With the new firmware, I repeated this test including frequencies around 915 MHz, the "new" magic frequency. As expected, the PIN SHIFT line switched at the new frequency. Here are the tabulated results showing the channel number, transmit frequency, and voltage level on pin 14 (H = ~5VDC, L = ~0VDC):
If you want to use a switching frequency that's different from the 915 MHz I chose, you must calculate and fix the checksum for the lower half of the EPROM yourself. To do that, change location B5FFh to 00h, generate the checksum for 8000h-B5FFh, subtract that value from 17Bh, and put the low two digits into location B5FFh. Generate the checksum again and verify that the result is 7Bh.
What You'll Need:
To do this job, you'll need a hex editor, an EPROM burner, and either an erased EPROM or an EPROM eraser. I had all of this, including a copy of the firmware already in a disk file that I read from another EPROM. The steps you might need to do are:
I use Hex Workshop, available from BPSoft. I use the True-USB PRO GQ-4X Willem Programmer, which I bought for about $100, available from MCUMall. The eraser was an inexpensive ($50) ultra-violet unit that can handle four ICs at a time, available from Circuit Specialists.
As usual, performing such a modification will void the warranty. I cannot be held responsible for any damage you incur during this process.
The author can be contacted at: his-callsign [ at ] comcast [ dot ] net.
Back to the top of the page
Up one level (MaxTrac index)
Up two levels (Motorola index)
Back to Home
This page originally posted on Friday 05-Aug-2011
Article text, artistic layout, and hand-coded HTML © Copyright 2011 by Robert W. Meister WA1MIK.
This web page, this web site, the information presented in and on its pages and in these modifications and conversions is © Copyrighted 1995 and (date of last update) by Kevin Custer W3KKC and multiple originating authors. All Rights Reserved, including that of paper and web publication elsewhere.