OpenTracker USB review

Updated 24/10/14.

I purchased an Opentracker USB (OT-USB) after a review of documentation and some screen shots of the configuration tool OTWINCFG which suggested that the product ‘understood’ the challenges.

From the outset, problems were encountered in trying to configure the device and I managed to lock it up after downloading the ‘latest’ firmware from Argent’s site.

Web site download links not current

As it turned out, the link was to an older version of firmware with problems, and the resolution was to let OTWINCFG directly get the undescribed firmware update as it was current. It is highly likely that in trying to solve problems with the USB interface, a well intentioned attempt to make sure that the firmware was the best available actually exacerbated problems in commissioning the device.

I did manage to recover the device by updating firmware via the serial port as the USB port bootloader was crippled.

USB interface very flaky

The USB problems are well known, and continue with the current version of firmware and OTWINCFG. I have tried to use OTWINCFG on three computers, XP, WIn7, and WIn8.1 and the ONLY one that is usable is XP though it takes some careful procedure to coax it to connect. OTWINFG did not display a version, so I can only report that OTWINCFG as downloaded in early Aug 2014 is very flaky and without doubt would be the weakest feature of the whole system instead of a strength.

Open squelch DCD

The OT-USB includes an open squelch DCD (OSDCD) which should detect data modulation rather than simply noise or squelch opening. If this works, it is the best method of operation and was a key reason in selecting the device.

Experience has been that the OT-USB often transmits over one local digi which uses an OT3. VK2RHR-1 is strong and grossly overdriven, but in a way, that is not an excuse for failure of DCD as overdriven signals are the norm on APRS, certainly in this area.

To my mind, the challenge with OSDCD is to minimise the false positives, and minimise false negatives. False positives result in transmitter latency as it holds off waiting for a clear channel, false negatives result in transmitting over other data signals which results in collision and loss data (possibly from both transmitters).

False negatives are to be avoided as a priority, and OT-USB could be improved as it does transmit over even strong data signals from a ‘same family’ digi.

Update: VK2RHR-1’s modulation level has been adjusted down to reasonable level, and it makes little difference to the tendency of OT-USB to transmit over a busy channel. Essentially, Argent’s software DCD is not as reliable as the broadband energy detect DCD in preventing collisions.

Channel access

OT-USB does appear to implement the standard AX.25 ppersistence channel access algorithms, and from the configuration menu, implements its own ‘quiet time’ hold off. Of course, if two stations use the same quiet time (eg default), and both attempt to digi the same packet, they will collide.

OT-USB is not alone in its failure to be AX.25 compliant.

Preamble

OT-USB uses an isochronous preamble followed by two frame codes. I doubt that is AX.25 compliant, X.25 compliant or HDLC compliant, and although benefits may be claimed, it is not fully interoperable with equipment that assumes the preamble is entirely frame codes (eg TNC-X).

OT-USB is not alone in its failure to be AX.25 compliant.

Decoding

On test, OT-USB would not decode Foxtrak with DK7IN, aprstracker v0.11 or TinyTrak v1 firmware. (The same was noticed in evaluation of T3-135, so it appears a family trait.)

OT-USB uses a firmware DSP based demodulation / decoding algorithm which failed to decode the above trackers which were all decoded by four TNCs using different modem chips. The fault may be in the firmwares used on the Foxtrak, but OT-USB was not tolerant of modulation that was reliably decoded by the other TNCs.

 KISS

The KISS implementation appears not to honour KISS commands for PPERSIST, SLOTTIME, TXDELAY, and so does not have configurable AX.25 channel access supported by the KISS command set.

OT-USB is not alone in its failure to be AX.25 compliant.

Failure of documented capability

The remote access capability does not work as documented in the manual, the SECRET command is not implemented and Argent advise that it was temporarily removed due to a shortage of memory… more than a year ago. The product is a fraud, it does not live up to the functional description in the user manual, and it hints a culture at Argent Data.

Conclusions

OT-USB:

  • works in a fashion;
  • the configuration software is very poor due to flaky USB access and the probing scheme implemented;
  • does not decode some popular trackers;
  • OSDCD is poorer than desired, and essentially useless;
  • KISS implementation is partial;
  • Channel access algorithm is not AX.25 compliantl
  • some documented capability is not delivered.

OT-USB if quite function rich, but the very flaky USB interface frustrates exploitation of its features and its lack of tolerance of the signals from some trackers reduces its utility.