Lock-up problems

Lock-up problems

Postby nojoopa » Mon Oct 09, 2006 7:28 pm

My unit (prototype from jsaily) locks up during transmissions.

I can LOAD"$",8 most of the time. The following LOAD"$n",8 works about 70%.

Every attempt to load a file locks up the unit, except a 1 block file I created for testing purposes. The next test file in size was 11 blocks and did not load even once.

When the unit locks up, the LED is glowing a contant level which does not seem to be the brightest level (although I may be mistaken).

I tested with two different SD cards. They both behaved the same way.

I can't tell much more details because I don't own a oscilloscope.

Any ideas?
nojoopa
 
Posts: 60
Joined: Mon Oct 09, 2006 7:16 pm

Postby jsaily » Tue Oct 10, 2006 7:40 pm

Hi, have you been able to load any real program from the device? Try loading D64 images or PRG files from root.

Are there any FAT-directories in the card? Those can cause trouble, believe me!
jsaily
 
Posts: 110
Joined: Wed Sep 27, 2006 2:33 pm
Location: Finland

Postby nojoopa » Wed Oct 11, 2006 7:09 am

I have succesfully loaded (and executed) 1,2 and 3 block files from a d64. I attempted to load larger files (4, 11, 57...) many times, but it always got stuck at the "LOADING" phase. Sometimes the LOAD"$",8 or LOAD"$n",8 gets stuck the same way.

There are no FAT-directories in the card, assuming you mean (sub)directories accessable with "cd" on PC.

Here's more info on one of the cards I use:
Code: Select all
amdxp ~ # fsck.msdos -v /dev/sda1
dosfsck 2.11 (12 Mar 2005)
dosfsck 2.11, 12 Mar 2005, FAT32, LFN
Checking we can access the last sector of the filesystem
Boot sector contents:
System ID "mkdosfs"
Media byte 0xf8 (hard disk)
       512 bytes per logical sector
      8192 bytes per cluster
         1 reserved sector
First FAT starts at byte 512 (sector 1)
         2 FATs, 16 bit entries
    124928 bytes per FAT (= 244 sectors)
Root directory starts at byte 250368 (sector 489)
       512 root directory entries
Data area starts at byte 266752 (sector 521)
     62459 data clusters (511664128 bytes)
62 sectors/track, 16 heads
         0 hidden sectors
    999874 sectors total
Checking for unused clusters.
/dev/sda1: 5 files, 110/62459 clusters

There are 5 (d64) files, no subdirectories and no hidden sectors. The fs is FAT16 with 512 bytes/sector.

I don't think the problem is with these cards, but if anyone sends a link to a dd dump of a tested and working card then I can try that.

The DTV was tested with a 1541-II earlier. It didn't lock up, but there were some problems loading some programs (which can be blamed on 15 year old magnetic media, I guess).
nojoopa
 
Posts: 60
Joined: Mon Oct 09, 2006 7:16 pm

Postby jsaily » Wed Oct 11, 2006 8:41 am

I assume you have updated the firmware already? Try to update the firmware again, and if not working you can send it back to me and I will see if there is something wrong. I have one extra PCB here so I can rebuild it if the PCB is broken or something.

I think I have also observed some more lockups with the more recent firmware versions. The lockup can be recovered by resetting the DTV and then pressing the IEC-RESET button. Did you remember to always press the reset button after resetting or powering up the DTV? If not, then you will certainly observe a lockup. After IEC-RESET (if you try to load something before reset) you will see 'DEVICE NOT PRESENT ERROR'. Then try to load again.

Maybe there need to be more delays in the firmware or something, maybe Jan has some ideas.
jsaily
 
Posts: 110
Joined: Wed Sep 27, 2006 2:33 pm
Location: Finland

Postby nojoopa » Wed Oct 11, 2006 9:42 am

I haven't upgraded the firmware yet because my cable is not ready. I guess I have to finish it and dust off my Win98SE partition...

I did use the IEC-RESET. The lock-up can also be recovered by pressing Esc+Pause (RUN/STOP+RESTORE?) and then pressing the IEC-RESET button.

I'll post the results here after the firmware upgrade.

I'll make a firmware upgrade tool for Linux if someone gives me a link to the protocol documentation.
nojoopa
 
Posts: 60
Joined: Mon Oct 09, 2006 7:16 pm

Postby MadModder » Wed Oct 11, 2006 5:32 pm

I have absolutely no problem loading large files, but the thing do locks up sometimes when I load $. It just freezes, and the LED keeps fading. Since I haven't installed a reset for the "drive" I power cycle. And it NEVER works if I don't wait like at least 5 seconds before turning the DTV on again.

I haven't updated to the latest firmware.
MadModder
 
Posts: 39
Joined: Fri Sep 29, 2006 8:22 pm

Postby jsaily » Thu Oct 12, 2006 7:14 am

The device should work fine with large and multipart files... I played Times of Lore and Jumpman just this morning...
jsaily
 
Posts: 110
Joined: Wed Sep 27, 2006 2:33 pm
Location: Finland

Postby nojoopa » Thu Oct 12, 2006 8:20 pm

Firmware upgraded to v.6.10.2. The "detected PCB" was "1541-III" on the terminal, which seems odd. I'll test the detection routine a couple of times tomorrow.

I managed to load a 11 block program once and a 4 block program 6 times in a row. Loading a 59 block (or larger) program always failed. Therefore the upgrade did help, but only a little.

While trying to LOAD"$3",8 I observed the following errors:
- the usual "LOADING"-lockup
- "?DEVICE NOT PRESENT"
- "?FILE NOT FOUND"
- "LOAD ERROR"
The LED was blinking (on/off) after the actual error messages. In the lockup situation the LED stays on.

I'll measure some voltages (at least the IEC lines) during the lockup and post the results tomorrow. I'll try to get access to an oscilloscope next week and see what is going on.
nojoopa
 
Posts: 60
Joined: Mon Oct 09, 2006 7:16 pm

Postby jsaily » Fri Oct 13, 2006 4:21 am

Hmm... this is very odd. If the PCB is not detected correctly then the button pin (RA0, PIC pin 2) is not floating as it should. Jan's pcb detection routine measures the time of capacitor discharge on this pin.

Please check that this pin is not connected to anything! Again, sorry for the trouble. I tested this device with my special firmware which does not do the pcb checking as this was not available until a week ago or so.

I also remember loading and playing large programs from this unit, although the access delay was long as I had two 1541-III's on the same IEC bus.
jsaily
 
Posts: 110
Joined: Wed Sep 27, 2006 2:33 pm
Location: Finland

Postby Jan D » Fri Oct 13, 2006 10:41 am

Hello all,
although I'm very interested in the status of the button input pin (which should be floating (i.o.w. left open), here another thing te keep in mind that may be of interest to other users.

When connecting another device to the IEC-bus (a second drive for instance) make sure that that device is also switched ON. If it is switched off, then the 1541-III seems to work but then somewhere during loading the 1541-III resets. The exact reason is not known yet, but is it caused by the other device that is switched off. So in every case no other devices on the IEC-bus may be switched off otherwise erratic behaviour shall occur. Practically, when a device is switched off, why should you connect it to the bus anyway.
Jan D
 
Posts: 21
Joined: Sun Oct 01, 2006 9:05 am
Location: Netherlands

Postby nojoopa » Fri Oct 13, 2006 1:29 pm

I tested the detection routine some 10 times and it always reported "1541-III", so it was not a random glitch. RA0 seems to be disconnected, but I'll desolder/cut it later and try again.

I measured some voltages at a lockup situation:
IEC-A = 3,3V
IEC-D = 3,3V
IEC-C = 0,1V (?)
Vin of 1541-III-DTV's regulator = 5,3V ( = regulated 6V - 1N4001's dropout)

Hmm...should I connect the Vin directly at the 6V? Could the 1N4001 diode mess things up?

(Edit)

I disconnected the pin from the PCB. The detection routine still shows "Detected PCB: 1541-III". (One small feature request: add a "\r\n" at the end of the last line to be printed.)

So...what's next?
Last edited by nojoopa on Fri Oct 13, 2006 7:01 pm, edited 1 time in total.
nojoopa
 
Posts: 60
Joined: Mon Oct 09, 2006 7:16 pm

Postby MadModder » Fri Oct 13, 2006 6:56 pm

What's your purpose with the diode?
The 78M33C regulator can handle an input of up to 35V. :)
MadModder
 
Posts: 39
Joined: Fri Sep 29, 2006 8:22 pm

Postby nojoopa » Fri Oct 13, 2006 7:10 pm

The regulator might, but I think my keyboard wouldn't :)

My power arrangement:
6V comes from a (cheap) "switching mode universal power supply"
6V is connected to
- DTVs standard regulator input (which powers the DTV)
- a 1N4001 which drops the voltage to 5V-5,3V
The 5V-5,3V is connected to
- PS/2 keyboard
- joystick ports
- 1541-III's regulator input

Using the 6V power source and the 1N4001 was a decision based on my very low budget; I just happened to have those already :)
nojoopa
 
Posts: 60
Joined: Mon Oct 09, 2006 7:16 pm

Postby jpilkinton » Sat Oct 14, 2006 1:28 am

Hello!
I am having the same problem. Although I can't load or run anything.
I am always getting the "LOADING" freeze also. LED bright all the time.
I even tried to load a PRG file and get the same results.
I did the load "$0",8 and it loaded the root directory just fine.
I then tried to load the directory of the d64 image by load "$1",8 and it just says "LOADING" and doesn't do anything else.
All of my SD cards are FAT, but I tried to format it under WinXP and it will only let me choose FAT or FAT32.
Can the SD card be FAT or does it require FAT16 in order to work?
jpilkinton
 
Posts: 67
Joined: Tue Sep 26, 2006 4:48 pm
Location: Jacksonville, AR

Postby jpilkinton » Sat Oct 14, 2006 2:12 am

I just tried the other kit I put together and it does the same exact thing.
Any clues?
I have not made the upgrade cable yet. But it should be working without upgrading the firmware.
jpilkinton
 
Posts: 67
Joined: Tue Sep 26, 2006 4:48 pm
Location: Jacksonville, AR

Next

Return to 1541-III-DTV Software

Who is online

Users browsing this forum: No registered users and 1 guest

cron