Well, I received maczrool's chip and replaced it. Still getting the NHDFCS. It looks like the chip may be good and something else may be wrong as MLord eluded to.

Here is the output just for kicks:
If there is anyone present who wants to upgrade the flash, let them speak now,
or forever hold their peace...it seems not.

e000 v1.04
Copying kernel...
Calling linux kernel...
Uncompressing Linux..................................... done, booting the kerne
l.
Linux version 2.2.17-rmk5-np17-empeg52-hijack-v361 (root@ibbm) (gcc version 2.95
.3 20010315 (release)) #2 Sun Jan 25 22:34:59 EST 2004
Processor: Intel StrongARM-1100 revision 11
NetWinder Floating Point Emulator V0.94.1 (c) 1998 Corel Computer Corp.
empeg-car player (hardware revision 9, serial number 10101551)
Command line: mem=16m
Calibrating delay loop... 207.67 BogoMIPS
Memory: 15016k/16M available (980k code, 20k reserved, 364k data, 4k init)
Dentry hash table entries: 2048 (order 2, 16k)
Buffer cache hash table entries: 16384 (order 4, 64k)
Page cache hash table entries: 4096 (order 2, 16k)
POSIX conformance testing by UNIFIX
Linux NET4.0 for Linux 2.2
Based upon Swansea University Computer Society NET3.039
NET4: Linux TCP/IP 1.0 for NET4.0
IP Protocols: ICMP, UDP, TCP
TCP: Hash tables configured (ehash 16384 bhash 16384)
IrDA (tm) Protocols for Linux-2.2 (Dag Brattli)
Starting kswapd v 1.5
SA1100 serial driver version 4.27 with no serial options enabled
ttyS00 at 0xf8010000 (irq = 15) is a SA1100 UART
ttyS01 at 0xf8050000 (irq = 17) is a SA1100 UART
ttyS02 at 0xf8030000 (irq = 16) is a SA1100 UART
Signature is 636f6972 'rioc'
Found custom animation at offset 0x9b790
Tuner: loopback=0, ID=-1
Scheduling custom logo.
empeg display initialised.
empeg dsp audio initialised
empeg dsp mixer initialised
empeg dsp initialised
Could not find CS4231A (version=80)
empeg remote control/panel button initialised.
empeg usb initialised, PDIUSBD12 id 1012
empeg state support initialised 0089/88c1 (save to d0004a80).
empeg RDS driver initialised
empeg power-pic driver initialised (first boot)
RAM disk driver initialized: 16 RAM disks of 4096K size
empeg single channel IDE
Probing primary interface...
Probing primary interface...
empeg-flash driver initialized
smc chip id/revision 0x0049
smc_init couldn't find card
RAMDISK: ext2 filesystem found at block 0
RAMDISK: Loading 320 blocks [1 disk] into ram disk... done.
VFS: Mounted root (ext2 filesystem).
empeg-pump v0.03 (19980601)
Press Ctrl-A to enter pump...el panic: VFS: Unable to mount root fs on 03:05

Sine no one has replied, I'll make a wild-ass guess and suggest that maybe there's a trace on the board that's gone bad that's preventing the chip from working properly.

I would also recommend keeping the thread bumped up until Monday. A lot of people might be out for the weekend. For the record, I have no idea either...

Hi,

I just finished a nasty IDE header repair, that required me to trace some of the IDE signals back to source in order to reconnect them to the badly damaged header.

Lacking a schematic for the player makes repairs very tricky, but I think I now understand some more of how it is wired up internally.

The SA-1100 has a built-in PCMCIA controller -- which basically presents an ISA bus (IBM PC AT expansion bus) interface for the CPU and peripherals. The IDE drives (which by definition have built-in controller electronics) simply attach directly to this bus, as does the CS4231A, the USB controller, and other stuff (?).

If both the CS4231A and the IDE are not working, it could be due to:

(1) the logic that manages chip selects for them, if there is any such logic (they could just be using separate address lines for each device). To check this, one will need to customize a kernel to loop repeatedly, toggling the chip select(s) high for one second, then low for one second, ad infinitem, so that you can then use a logic probe / scope / voltmeter to test for the signal(s) getting to the CS4231A pins. Then repeat this test for the IDE header CSEL lines.

(2) the problem could be the IDE header, but if you disconnect all drives and the CS4231A still fails, then the header is likely innocent.

Anyway, the way to get this player working again is to use the approach described in (1) above -- custom code here and there to toggle appropriate signals, and see which ones are getting to the chip and which ones are not.

The code would be like this:

while (1) {

   unsigned long timestamp = jiffies;

   unsigned char data = 0xff;

   while (jiffies < (timestamp + HZ))

       outb_p(data, port_address);  // or inb() 

   while (jiffies < (timestamp + (HZ+HZ)))

       outb_p(0x00, port_address);  // or inb()

}

Note to picky programmers: the jiffie logic is not 100% correct, but is simple and works well enough for this isolated situation.

Cheers

WOW!!

That is more complicated than I am probably able to do myself. I am able to disconnect the hard drive and test the serial port to see what happens which I will do tonight when I get home.

After that, I don't think I am able to do any of that other troubleshooting.

Is there someone I could send my board to and have them diagnose the problem and then pay them for their services (up to a reasonable amount)?

I cannot promise anything, but so far I've been having good fortune with extremely tricky repairs. You could send it to me (no hard drives, please), and when I have time and inclination I'll pick away at it. I think $50 to poke at it, maybe more if I can actually fix it -- to be agreed upon between us at that eventuality.

Email (not PMs) works best.

Cheers

That sounds great.

What do you need? Just the motherboard? The case? The IDE cable? The display? Etc?

Just the mainboard, I think. I'll mostly be poking new kernels into it, and observing results on the serial link.

Cheers

You have a PM.

...so far I've been having good fortune with extremely tricky repairs.
Mark, with all the repairs you seem to be doing recently, you could start a new business: Empeg revival by The Lord.

Speaking of which.. your player arrived here today, Russ, and I've already repaired the display. Looks great!

Still gotta assemble a pair of drive cables before I can send it back to you by 14-day post.

Cheers

Awesome! Just curious, were the fuses blown or was it only the connectors?

Fuses were fine, hard drive header looked okay, though I did touch up the soldering a bit there just to make sure. So it was really just the display cable not making proper contact -- no suprise to anyone on that.

More by email once I mail the unit back to you.

Cheers

Well, I'm sitting here listening to music playing from this "dead" RioCar now!

The unit arrived in the mail earlier today, and I have spent a few hours cutting and resoldering traces to isolate the fault.

The problem was a stuck "D5" line (data bit 5) on the peripheral (ISA) bus.

I traced this signal down to the IDE header, the USB chip, the CS4231A, and the Ethernet port. As well as to a 3V/5V line transceiver connecting to the processor/memory bus.

The line transceiver was first up, and checked out okay on both sides.

Next, the USB chip. Again okay. Damn.

Next, the CS4231A -- I lifted the D5 pin, and connected it directly to the line transceiver in an isolated fashion -- the chip passed self test first try. Good resoldering job there, whoever did the chip replacement!

Next, as everything so far was innocent, I lifted D5 from the ethernet chip. Bingo.. system boots with audio etc.. Reconnect the pin again, and all dies. So, I'm going to leave it at that.

Cheers

REALLY close-up macro shots of my hack job are now visible here:

http://rtr.ca/NHDFCS_empeg/

Keep in mind that the features in the photos are nearly microscopic -- not as ugly in real life as they first appear. Some cleanup is needed, for sure.

Cheers

That's truly evil. Wow.

-brendan

Wow Mark, excellent detective work there. You must be very proud of yourself. Truly impressive!

Stu

AMAZING!!!

Definitely worth every penny!!! I will finally have my RIO Back. I have been missing it.

I don't know how the problem arised but I am assuming it came from static of some sort. My room has rug and is relatively dry. I try to touch something before touching the RIO to remove the static but maybe someone else touched it. I am going to look for replacement ethernet chips and transeiver chips.

If someone knows where to get those, that would be very helpful.

Hey,
I appreciate all the work you did MLord. Please don't take any of the following as insults as they are all able to be worked through.

I received the Rio Board yesterday as well as the LAN91C96 Chips. Unfortunately it is the wrong chip (it is rectangular, not square).

I had the person that is going to solder it look at the board and she brought up a couple of problems.
1.) The pad on the LVX4245 D5 pin is gone. That leaves nothing to solder to. She is going to create a jumper from there to the D5 line that you created. A more permanent solution than what you created.
2.) The pad on the SMSC LAN91C96 D5 pin is also gone. She wants to create a jumper but needs to know where to jumper to. We looked at the board and couldn't figure out where it went. Do you know where to connect the other end of the jumper to?

>The pad on the LVX4245 D5 pin is gone. That leaves nothing to solder to.

Occupational hazard! The pad was actually still under the lifted pin -- pressing the pin down made contact when I last tried it, but the jumper wire can be used just fine as well.

>The pad on the SMSC LAN91C96 D5 pin is also gone.
It connects to the nearby via (hole) that has a wire sticking out of it that I left in place as a marker for it -- also shown in the photo I took of that chip.

Cheers