DEMO9S08QD4: programming through BDM port header

[sauliuz]

Hello,

Using DEMO9S08QD4 board to program QD4 processors for a while and it works fine. Up till now I was using QD4 DIP socket. Now decided to switch to BDM port header which si located on a side for optional use (have to install pin header). But it does not work. I monitored BDM signals with scope both on pin header and on dip socket and they seem to be ~identical. It just does not want to connect to a chip. I atteched demoboard schematics with red lines which shows that BDM wires to pin header and to dip socket comes from different programmer chip pins and I don't understand why it is not the same one. Maybe someone has a solution, becouse it is much more confortable to use pin header.

Saulius

DEMO9S08QD4_SCH.pdf

Message Edited by t.dowe on 2009-10-27 01:46 PM

[Ake]

Hi,

I don't know if this helps, but I suspect that if you want control the MCU on another board with the BDM cable, you must remove the MCU from your DEMO board first. 

At least that was something that was the rule for the HC908s.

Regards,

Ake

[sauliuz]

>> you must remove the MCU from your DEMO board first.

Well, if debugger will try to communicate to target chip through BKGD and one more chip on the way will start answering the same messages obviously data packet conflict should occure... So removing existing QD4 chip on demo board was the first thing I have done.

I tried variuous pull-ups on both reset and BKGD wires befors starting this thread too. Even tried additional capacitance on reset pin.

As I see the problem might be in debugger configuration (I checked it and I couldn't find anything related), or maybe some additional changes must be done to the board, not just pin header soldering...

I solved this problem in simplest way - I soldered additional wire to connect normal BGDN pin in DIP socket and GDM_PORT header pin nr. 5 (which was not used). Adjusted my cables accordingly and using programmer normally. But that still leaves the question why they put BKGD_IN/EBGND signal to header, which is not working???

Saulius

[peg]

Hi Saulius,

I don't know anything about this board other than what can be gleaned from the schematic, but it looks very much like a cut down version of a QG one. Interestingly the QG does not have a seperate connection to the P&E black box for the BDM header, but shares one with the chip socket.

One thing I notice is that the BKGD signal going to the header does not have a pullup. Perhaps you could try applying a 3k3 pullup resistor onto this line.

I have programmed/debugged hundreds of GB/GT's without this pullup but others have found it necessary.