The WiNRADiO 1550e is a headless software-controlled receiver that came out about 1999. It covers 150 kHz to 1500 MHz in CW, USB, LSB, FM-narrow, and FM-wide modes. The USA version excludes cellular frequencies by default. It was designed to run on DOS, Windows 98, and MacOS Classic. There were a wide variety of expensive accessory applications available in the early 2000s for scanning, signal analysis, and other purposes. The manufacturer offered support for building APIs to connect various program to the main software. In contrast, the maker provided little technical information about the radio itself, including its control protocol. The WiNRADiO 1550e may be seen as comparable to the ICOM-PCR 1000.
As a receiver, the WR-1550e is a middling performer, with reasonable (but not good) sensitivity, poor adjacent-signal rejection, and is prone to intermodulation. Its manufacturer claims about 0.3uV sensitivity across most of its range for SSB/CW for a 10dB S/N. In side-by-side comparisons, the ICOM PCR-1000 appears to be more sensitive and better at nearby signal rejection. The speaker output is 0.2 watts.
This is a software-controlled radio rather than a software-defined radio. That means it's a standard receiver controlled by software via a serial RS-232 or PCMCIA connection. The software supplied with the radio, WINRADIO.EXE (still available on the manufacturer's website), was apparently designed for Windows 98SE, and has not been updated. It will not install or run on modern Windows systems. The Macintosh software is similarly ancient, and designed to run in the old "Classic" environment (e.g., OS9, pre-OSX). It will not run natively on any modern Mac. There are reports that the Windows software will run on Windows 7, and under Windows 7 in emulation--although Windows XP is probably the latest version for reliable use. I was unable to get the WINRADIO.EXE to install at all on Windows 7. There is an available DOS application, DOSRADIO.EXE, which can be made to work under DOSBOX-X on Windows. I was able to get DOSRADIO.EXE to install and run on DOSBOX-X on a Mac. Unfortunately, due to an apparent bug in DOSBOX-X for Mac, the program could not see the serial port.
It should be noted that DOSRADIO.EXE is a command-line program. However, the control commands listed in the online manual do not correspond to the commands accepted by the program. The actual command protocol is helpfully listed by the program when you attempt to use the commands from the manual. As for ports, only COM1-4 are available in the software, with COM1 an apparent default. To get any of these programs to work on modern systems, you will need a USB-Serial converter and might need to redirect port assignments.
DOSRADIO.EXE seems highly limited, offering only basic functions, such as frequency, volume, and mode control. However, that's basically all that's offered by the main program, WINRADIO.EXE, in its graphical user interface. This reveals the reality of the 1550e. It hardly does anything more, by itself, than receive. There are no built in filters or DSP. No DTMF or CTCSS either. Scanning and everything else seems to be done externally, in software. If there are native scanning, memory, and other functions available in the radio, I was unable to find them. So, for filters, the best you get is about 2.5 kHz in SSB/CW mode. DSP is not even an option. According to the instructions generated by DOSRADIO.EXE itself, you should be able to control the BFO frequency. But, the DOS software doesn't seem to work, defaulting the BFO to preset values.
The manufacturer offers extensive API support for creating links to other applications, such as Visual Basic. In contrast to the extensive information about APIs, there seems to be no published service manual, nothing about the control protocol, and no available schematics. The only clues to the 1550E control protocol are found in the source files for the DOSRADIO program. These reveal a problem for anyone hoping to use the radio: you cannot directly control the 1550E with relatively simple serial commands as you can with the PCR-1000. What should be done with embedded control software--the complex calculations to set up the PLL and other circuits--is done externally. Iterative matrix algebra is used (unnecessarily) to set the frequency. Much simpler math can do the same thing. Another calculation is required to set the BFO frequency. Yet another calculation, this one digital, to pull control bits from parameter byte, is required to set the internal detectors. Basically, you're figuring out the parameters for the PLL, detectors, and filters, and then sending them to the radio encoded (by another set of calculations) into a 14-byte string. Fortunately, the volume, mute, and few other things can be set directly.
Just to give you a feel of the issue, here's the C code (it's all in C) to calculate the "R" and "nTOT" values for the PLL to use to set the operating frequency. (This is just part of the code in just one file. There are 13(!) more C files, with similar content, in the DOS package).
static int GetVcoParams(int hRadio, DWORD fvco, DWORD fref, int minR, int maxR, UINT *R, DWORD *Ntot)
{
double Kn, Kp, af;
MATRIX M, tM, cM;
int i;
UINT tR;
DWORD tN, raf;
Kn = (double)fvco / fref;
M[0][0] = 0;
M[0][1] = 1;
M[1][0] = 1;
M[1][1] = (long)Kn;
Kp = modf(Kn, &Kn);
memcpy(tM, M, sizeof(MATRIX));
while ((Kp > 1e-6) && (M[0][1] < maxR))
{
Kn = 1 / Kp;
tM[1][1] = (long)Kn;
memcpy(cM, M, sizeof(MATRIX));
M[0][0] = cM[0][0] * tM[0][0] + cM[0][1] * tM[1][0]; // M = cM * tM - matrix multiplication
M[0][1] = cM[0][0] * tM[0][1] + cM[0][1] * tM[1][1];
M[1][0] = cM[1][0] * tM[0][0] + cM[1][1] * tM[1][0];
M[1][1] = cM[1][0] * tM[0][1] + cM[1][1] * tM[1][1];
Kp = modf(Kn, &Kn);
}
if (M[0][1] > maxR)
memcpy(M, cM, sizeof(MATRIX));
*R = M[0][1];
while (*R < minR)
*R += M[0][1];
*Ntot = M[1][1] * *R / M[0][1];
Kp = (double)*Ntot / *R;
af = Kp * fref;
raf = (long)(af + 0.5) - fvco;
if (labs(raf) > 50)
for (i = 1; i < 25; i++)
{
if (modf(Kp * i + 1e-10, &Kn) < 1e-6)
{
Kn = 50 / i;
if (raf > 0)
GetVcoParams(hRadio, (long)(af - Kn), fref, minR, maxR, &tR, &tN);
else
GetVcoParams(hRadio, (long)(af + Kn + 1.0), fref, minR, maxR, &tR, &tN);
fvco = (long)(tN / tR * fref + 0.5) - fvco;
if (abs(raf) > labs(fvco))
{
*R = tR;
*Ntot = tN;
raf = fvco;
}
break;
}
}
return raf;
}
A different calculation is used for wide-FM. These values are further processed and encoded. You need to read the PLL from the radio to then calculate the values to set the filters. This is the hex string that tells the radio to tune to 145 MHz and set the right detectors and filter:
6D 2C 6E 00 48 6F 2A 6C 33 F0 4E F1 31 93
Mode is a separate command, as is setting the BFO, if you need it. In CW and SSB, there's an unexplained frequency offset of +13kHz to consider, as well. On the PCR-1000, in contrast, you set the frequency, mode, and filters with a single command with straightforward values which could be sent with a batch file:
K0 0145000000 05 02 00 . (this would be 145MHz FM, 15kHz filter)
You can probably see why there are several non-OEM controllers for the ICOM PCR-1000 and nothing for the WiNRADiO 1550e. HAMLIB supposedly supports it--it's in the list--but not actually. Perhaps there are customers who need the customization of the PLL parameters that this system offers. None of that seems to be offered through the software, however, bringing us back to the question of why these calculations were not embedded.
In sum, in its time, the WR-1550e was an expensive, but feature-limited, computer-controlled wideband radio, inferior in most ways to competitors such as the ICOM PCR-1000. Extensive software packages were available to make up for some its native limitations. None of the software necessary to operate the radio runs on modern systems, except potentially under emulation. I could get only DOSRADIO to work, even with emulation. There seems to be extensive support for the radio on the manufacturer's site. But it is of limited value now. As noted before, the site appears to offer nothing technical about the radio or its control protocol. Multiple inquires made via the manufacturer's site for information have gone unanswered. There is almost nothing useful about it on the internet, either. So, you're pretty-much entirely on your own. That makes this radio something to avoid, except perhaps by tinkerer's who are well-versed in C who want to try to reconstruct the control protocol from the DOSWINDOWS source files. I got mine for $50, which is probably more than it is actually worth. If you need a medium performance wide-band computer controlled receiver, get a modern SDR. |
