TRX2a Serial Problem

General discussion

Re: TRX2a Serial Problem

Postby AA4PB » 15 Jun 2012 17:55

Thanks Adrian. I just tried it with the Win7 computer and a real RS232 port with the same result. I originally had no serial, found the ST232 IC in backwards and replaced it with a new one. Now the normal serial stuff works okay but not the boot loader. I suspect that the boot loader has been corrupted in the process so I'll have to look into reloading it. All of the 0.1uf caps look okay.

Bob
AA4PB
 
Posts: 17
Joined: 09 Jun 2012 04:46
Location: USA

Re: TRX2a Serial Problem

Postby AA4PB » 30 Jun 2012 01:31

If I have corrupt bootloader code can I reload the bootloader hex file using an ICD2 and then use the bootloader to reload the operating program hex file - or do I need a special first time load hex file that contains both the bootloader and the operating program? If I need the combined file where can I obtain it?

The voltages measured on the ST232 charge pump outputs are + and - 8.6VDC and the RS-232 ouput is +/- 6.5VDC. The waveform coming from the computer is good with nice square waves. All the RS-232 tests at 9600 baud work okay. The radio works great. The bootloader message comes up on the LCD and I see the square waves (from the U characters) coming from the PC but the bootloader doesn't appear to be doing the rising edge detection - it just times out and goes to the radio operating code. I've tried all baud rates including 9600 with the same result. I've also monitored the TXD going to the PC and get nothing until the bootloader times out and the operating software sends the normal sign-on message.

I noticed in the ASM source code for the bootloader that it selects the alternate IO pins but they do not exist on the dsPIC30F6014A so I assume that bit in the register is ignored and the primary IO pins are selected. This is likely the case because others are able to use their bootloaders.

What I am ultimately wanting to do is to upgrade the firmware to Adrian's latest version.

Bob
AA4PB
 
Posts: 17
Joined: 09 Jun 2012 04:46
Location: USA

Re: TRX2a Serial Problem

Postby 5B4AIY » 08 Jul 2012 07:49

For those following this post, after a number of email exchanges between myself and Bob, he discovered that the open-circuit DC voltage on the TX DATA line was only -8V rather than the -9V that it should be. He measured tha DC voltage on pin 16 of IC10 and found it was only about 4V, rather than the 5V it should have been. This indicated that either IC10 was drawing an excessive amount of current, or that R22 was no longer 10 ohm. It turned out to be the latter, R22 was very high. He had originally soldered IC10 backwards, and had subsequently removed the damaged chip and replaced with a new one. R22 was probably damaged as a result of mounting IC10 backwards as in this configuration the IC substrate would have been forward biased and would have been essentialy no more than a forward biased diode junction, effectively shorting the power bus and overloading R22. He is awaiting some replacement resistors, and when R22 is replaced, this should solve this problem.
5B4AIY
 
Posts: 214
Joined: 13 Nov 2011 09:22
Location: Cyprus

Re: TRX2a Serial Problem

Postby AA4PB » 10 Jul 2012 22:18

Success!! Replaced R22 with a new 10 Ohm. Actually I jumpered the defective resistor, now reading about 300K, with the 10 Ohm by placing the 10 Ohm at a more convient spot on the top of the PC board. Otherwise I would have had to remove the front panel and the LCD to get access to R22.

The RS-232 no-load voltage which had measured -8.15V is now reading -9.9V and the bootloader is detecting the baud rate. I assume that the high series resistance and the 2.2uF cap made a nice RC time constant that rounded the edges of the serial signal on the TTL side, thereby preventing the processor from detecting the rising edges. Normal serial data is detected at the center of the bit so the rounded edges didn't matter for the serial data - only the baud rate detection.

Next step will be to load Adrian's new firmware.

My ST232 chip was somewhat difficult to identify pin-1. Both edges are tappered but the side with pin-1 is tappered at a slighly steeper angle. You need to look at the mfgs spec sheet very closely and view the angles from the end of the IC under a magnifying glass to see the angle difference. No notch or other marker on the end of my chip (either the original nor the replacement).

Thanks Adrian for all your assistance.
AA4PB
 
Posts: 17
Joined: 09 Jun 2012 04:46
Location: USA

Previous

Return to Juma Discussion

Who is online

Users browsing this forum: No registered users and 12 guests

cron