AVR removed = Serial Port hangs not

Zitat

- I removed AVR chip from IEC-ATA V2 board
- I connected IEC-ATA through Serial Cable and on 2nd port of IEC-ATA I connected diskdrive
- yes, IEC-ATA can't work, but connection COMPUTER - SERIAL - IECATA - DISKDRIVE works!

when I have AVR on the board serial port hangs and I can't access diskdrive...

So, here's offer Q: can be AVR damaged and this is the problem why serial port hangs?

If yes, I can buy new AVR and flash it and will not hangs?

Alles anzeigen

This means that the input- (LS244) and output- (LS07) drivers are ok. But as soon as you insert the AVR it pulls the IEC-lines to low, as much as I understand even when JR is closed. Please try this:
- Switch off power
- Connect IEC-ATA with inserted AVR to C128
- Close JR (=> AVR is immediatly in reset-routine when power is switched on)
- Switch on power
- Switch to C64-mode
- Type '10 poke56576,7:printpeek(56576):goto10' <return>
- Tell me the displayed number <--- (1)
- Remove JR
- Tell me the displayed number <--- (2)

It could be possible that the AVR is damaged. Is it possible that you inserted it one time in wrong position (the mark of the AVR is 180 degrees opposite to the mark of the RAM-chip)? I still don't know exactly whether you managed to flash the AVR a single time sucessfully? It could be possible that here the AVR was flashed to an external clock, which doesn't exist...

Zitat

... when I connect complete IEC-ATA (with AVR), C128DCR at start boots (search for boot sector) - and there's main problem, if I don't press STOP+RESTORE, computer hangs on searching boot sector...

What I don't understand is this:
"It could be possible that here the AVR was flashed to an external clock, which doesn't exist..."

I don't have a C128, so I don't know much about it. Isn't it possible to press a key while switching on the C128 so that it goes directly to the C64-mode without searching the boot-sector? If so then please try.
If this is not possible then: If the C128 doesn't hang after having pressed STOP&RESTORE then please try.
I flashed my AVR some hundred times. One single time I had the problem that the syncronisation of C64-sends to AVR-receives was so corrupted that a write-flash-command of the C64 was so corrupted that the AVR understood it as write-fusebyte-command and so wrote a wrong fusebyte which caused an external clock. Newer versions of the flasher try to prevent this, but I don't know whether you have an older or newer version.

Zitat

So, booting is not so big problem (STOP+RESTORE is one solution) and 2nd, better is start directly to C64 mode - holding C= (CBM) key through start.
At monday I get my two ordered new AVRs.
So, I try it...

It's really not easy to help you. Three weeks ago you promised: "I will never again create new thread, if is it not important...". Now you opened again a new one??
It's possible to make the ?peek(56576)-try at you C128, but you don't do it?? I can't understand why you don't do things which could help to understand the problem.

Zitat

I tried your instructions, and results are here:

RESET locked: #71
RESET free: #71

71 means %01000111. Bit 7 (low) ist Data-In of the IEC-line, bit 6 (high) is Clock-In. This means that the AVR pulls Data to active = 0V even when it is in a hardware-reset. This is very strange, because while a reset all lines are inputs and so go to +5V. But this explains why the IEC-bus is locked when you insert the AVR. Can you meassure the voltage directly at pin 9 (reset) of the AVR when JR is closed? But be careful; don't make a short-circuit to neighbour-pins. Or you can wait until the new MEGA162 arrives. Insert it when the power is switched off. It's less dangereous when you connect the IEC-ATA to the C128 when everything is switched off.

Zitat

I used one and flashed it - with software from Dienstagstreff.de - successfully.
HDD is connected with the red label on power side, both diodes lights (through RESET computer red diode blinks one time)
Jumpers are set: JC1-2 and RESET is open

I would use the NLQHD-firmware. It has debugbytes, which could help to solve problems. Please use the lates flasher ( 488 ), which is attached at the troubleshooting-thread. Please flash the latest firmware (V0.20) and tell exactly what happens. The jumpers are ok. There should be the number '1' at the HD. The red wire of the IDE-cable must be at the side with this number.

Zitat

(1) I flashed AVR with NLQ firmware - successfully
Turn off computer, connect HDD, set jumpers: JC1-2 / RESET-free, turn on IEC-ATA & HDD, turn on computer - same problem Serial Pot hangs...
(2) So, I experimented: Turn off all, disconnect IEC-ATA from Serial Cable, turn on computer, load CAL&CON and run it. Before choice of PAL/NTSC I connected IEC-ATA to Serial Cable and turn on power of HDD & IEC-ATA...
(3)(at power on both diodes lights)

It's good that you could flash the AVR correctly two times. It's strange that it doesn't work anyway. Are you shure that the firmware-binfile isn't corrupted? How did you transfer the NH1IEAT0.BIN from your PC to the C64? Do you use a D64-image or the file itself? Is it possible that the transfer-program added any linefeeds or changed anything else?
(2) The C64 isn't made for hot-plugging; this could be dangerous e.g. for the CIAs of the C128; you shouldn't do this. Does your C128 still work with your only external serial device, 64HDD?
(3) What does this mean? Explain exactly what happens when you switch on the power. Do both LEDs go on and never go off... ?
If an uncorrupted firmware is flashed without errormessage while flashing then this will work:
- Remove JR, connect IEC-ATA, connect IDE-HD and switch on power of C128, IEC-ATA and IDE-HD and press the C128-key to jump at once to the C64-mode
- Wait about 10 seconds until the system hangs
- Type poke56576,255 <return>
- Close JR for 1s and remove it again
- Wait about 30s until the red LED blinks (If the red LED doesn't blink then there is any bug with the firmware in the AVR-FLASH)
- Load "sw-debugtrans 14" from your 1571 and run it
- Save the file and send it to me

If an uncorrupted firmware is flashed without errormessage while flashing then this will work:
- Remove JR, connect IEC-ATA, connect IDE-HD and switch on power of C128, IEC-ATA and IDE-HD and press the C128-key to jump at once to the C64-mode
- Wait about 10 seconds until the system hangs
- Type poke56576,255 <return>
- Close JR for 1s and remove it again
- Wait about 30s until the red LED blinks (If the red LED doesn't blink then there is any bug with the firmware in the AVR-FLASH)
- Load "sw-debugtrans 14" from your 1571 and run it
- Follow the instruction
- When the screen goes on again and the program asks you whether you want to save the file to a disk, then you must close JR again (because else the crashed AVR locks the bus and you cannot save. When you close JR all pins of the AVR go to input, so they become 5V, so the IEC-bus is free)
- Save the file and send it to me

Zitat

So, I did it as you wrote - all was correctly.
I attached file DEBUGBYTES.PRG that you wanted.

I think that you flashed the AVR successfully. You flashed V0.20: Load the file which you sent me, start a monitor-program and type "m 6900". The monitor displays the reset-error-message: 73,e,00,20. Track- and sector-number is the version, V0.20, so you already have the correct version. Flashing and reading the debugbytes is a big step to solve the problem.
The AVR writes two debugbyte only into the debugbuffer then it hangs. It tries to acces the HD, but the HD doesn't answer. Don't change anything at the AVR-flash, it is correct. The problem is now the connection from AVR to HD. Try this:
- Is the red-marked wire of the IDE-cable connected to side of the HD-plug where the number '1' is?
- Is the red-marked wire of the IDE-cable connected to side of the IEC-ATA-plug where the floppy-power-supply-connector is (not the side where the RAM is)?
- Is your power supply ok?
- Perhaps you can test the HD at your 64HDD-PC: Connect your IEC-ATA-HD to your IBM-PC, but use the IEC-ATA-power-supply (not the IBM-PC-one) and use the IEC-ATA-IDE-cable (not the IBM-PC-one)
- Do you have another HD for trying at IEC-ATA?
- Perhaps you can use your 64HDD-HD for testing at IEC-ATA. You will receive an error because it isn't FAT32, but at least you should be able to access IEC-ATA. As long as you don't use the program 'nlq-par&for' (partition and fromat) you don't loose datas on this HD

Zitat

I have enlargement of PC-power supply for HDD and FDD. IEC-ATA V2: Black/Red/Black/Yellow from the left of board.

This is strange. My power-supply-cable is red/black/black/yelloworange.
Does your IEC-ATA-HD work at the 64HDD-PC?
Can you connect IEC-ATA to the C128, connect the HD via IDE-cable to IEC-ATA but use the power-supply of the 64HDD-PC?

Zitat

Power supply of IEC-ATA has the same source.
enlargement of cable is only for comfortable using IEC-ATA.
So, what with hanging after removing RESET jumper?

Ok, now I understand. Make a last test, whether your enlargement-cable is ok:
Connect the enlargement-cable to the HD of 64HDD. If 64HDD doesn't work then the Y-cable is defective. If it works then probably the IEC-ATA-hardware is defective. Accessing the external RAM works, seen at the debugfile. The 162 is new, so probably also not defective.
Skern, the Dienstagstreff-man who had repaired your circuit, phoned me. He told that your circuit worked when he tested it before sending and that he had flashed the NLQHD-firmware with devicenumber 8. So everything is very strange.