Other SOTA sites: SOTAwatch | SOTA Home | Database | Video | Photos | Shop | Mapping | FAQs | Facebook | Contact SOTA

New firmware for MTR-4B and 5-B fixes spurious emissions problems


#1

Dear all:

The LNR versions (and maybe others) of the MTR 4B and MTR 5B have a bug that can lead to emissions on unwanted frequencies, which may be illegal and can harm your finals. This may happen only if the RTC is in use. I am not sure about this, though.

See the post by Steve Webber at

https://groups.yahoo.com/neo/groups/AT_Sprint/conversations/messages/18571

"I need to warn of a potential problem with the LNR 5B with the 2x16 LCD display.
With the clock enabled, it is possible for the DDS chip to get the wrong data and transmit out of band. This happens because the data clocking bit for the LCD and DDS have opposite sense. The DDS clocks in data when the bit goes low, the LCD clocks in data when the bit goes high. Unfortunately, I failed to ensure the clock bit was in the proper state at the beginning of a data transfer to each device.
This problem was not initially apparent because when the RTC is not on, the sequence used to write to the DDS and LCD is such that the clock sense is corrected. However, when the RTC is turned on, a write to the display to update the time leaves the data clock bit in the wrong state for the next write to the DDS, which could occur when you go to transmit. This might also be the cause of lock up problems some users have experienced when the RTC is displayed.
Firmware rev 1.8 and later fixes this problem. To find out what firmware rev you have, turn power on while holding down the Function button.
Otherwise, it would be a good idea to keep a SWR meter in line and if you see the SWR suddenly go high
when transmitting, you know your out of band. Tuning up and back down one step will correct the problem."

As far as I know, the bug has been spotted by Chris, GU3TUX.

The new firmware version 1.9 is available from
https://groups.yahoo.com/neo/groups/AT_Sprint/files/LNR%205B%20FIRMWARE%20(2X16%20LCD)/

It is recommended to upgrade your MTR to this version. I have not done so yet, but the required programmer is inexpensive (15 EUR), available e.g. from Reichelt:

I have not yet found the proper wiring for in-circuit programming of the MCP430, but will post here when found.

73 de Martin, DK3IT


#2

@M1BUU seems to have done this before, see

so he might be able to share instructions on the flashing procedure.

73 de Martin, DK3IT


#3

I have asked Colin @M1BUU whether he could share details on the update procedure. In the meantime, here are a few links:

The 3-pin ISP connector on the MTR-5B PCB is directly connected to the TEST, RST and GND pins of the MCU (see attached image).

mtr5b-isp

So I guess you just need to connect these three wires from the TI LaunchPad to the ISP connector.

What I am still looking for is a command-line tool for OSX for flashing a given binary file.

Martin


#4

This might be what we need:

http://software-dl.ti.com/msp430/msp430_public_sw/mcu/msp430/MSP430Flasher/latest/index_FDS.html

The User’s Guide is here:


#5

Hi all,

I think I am now done on the hardware side; still need to install the software and do the actual firmware update, but this requires a fresh evening.

Here is what I did:

  1. I build an adapter from “pogo pins”; these are contacts used for in-circuit testing and other applications, see https://en.wikipedia.org/wiki/Pogo_pin

You can get them cheaply on eBay; I bought these https://www.ebay.de/itm/282449741005?ViewItem=&item=282449741005

As for the head type, “E2” works best for solder pads with a hole.

I cut four tiny pieces of perfboard and glued them together with superglue; make sure the holes are perfectly aligned and the super-glue does not run into the holes. The two outer pieces should have the copper layer to the outside. This stabilizes the pogopins.

I then added three jumper wires for connection with the LaunchPad.

  1. The (likely) pinout of the PRGM solder pads on the MTR-5B LNR revision is as shown on the attached picture.
    This is untested; use it at your own risk, but I think this is correct, since the rightmost pin is connected to the ground layer, and the leftmost runs via R28 to 3.3V and should hence be “RST”, which leaves the middle pin for “TEST”.

Will share experiences once I will have been successful.

73 de Martin, DK3IT