Mike's PCjr Page

Hi, I decided to post a new topic rather than continue replying in the "LBA too old" thread, as so much time has passed and I've found that my problems aren't only with SD.

I'll give an overview of my situation.

I got a functional jr earlier this year (thank you, forum member, for hooking me up; I don't know if you want me to say your name). It is a standard 128KB model, and it works well.

I replaced its CPU with a V20 CPU. I added TexElec's adapters (serial/CGA and keyboard/gameport). I've added a Gonbes GBS-8220 + Bit-C128's CGA-to-digitalCGA adapter, and I'm getting pretty good output to a VGA monitor (some artifacts that remind me of interlaced video, but overall, it's great; much clearer 80 column display than composite video, of course).

And I have a jr-IDE.

I'm very excited about the jr-IDE. I took a trip recently to visit my parents (10 hour drive) and went on an excavation of their old hoarder house (really an excavation; the floor has caved in due to termite damage, and the 2nd floor is covered with bird manure due to broken upstairs windows; had to wear a gas mask while I was in the house, as my brother got cave disease from breathing the air during a previous trip). I found our old (dead) jr! So now I have some spare, untested, parts (probably most work), and more importantly for me, I got 2 sidecars. The 512k memory expansion sidecar donated its shell for the jr-IDE. Parallel port sidecar will be added to my jr after I get the jr-IDE working the way I want.

Now with that said, I have a problem that I have not been able to resolve: I cannot boot from any IDE device attached to the jr-IDE. I spent a lot of time trying an SD card (through an SD-to-IDE adapter). I went through the suggestions in the "LBA too old" thread but haven't had success. I followed the debug instructions to the letter. Nothing. Debug removed the partition, and I recreated it with fdisk and did a "format /s". No dice. I went ahead and tried "sys c:", just to be sure. Nope. And of course I ran "fdisk /mbr". Nothing worked.

I decided to leave it alone and wait for the slow boat to bring me the Kingspec 8GB DOM that I'd ordered a month ago. This is the second one that I ordered, as the unscrupulous ebay seller for the first one sent me some other brand of DOM instead, and it gives the "LBA too old" error.

Anyway, this DOM seems to be the real thing -- Kingspec 8GB. No "LBA too old" error from the jr-IDE; it is identified as Kingspec 8GB, and it shows up in fdisk (with a non-DOS partition, which I deleted then fdisk'd and "format /s"').

However, I cannot boot from the DOM. Same symptoms as the SD -- it works fine if I boot from floppy, but if I press "C" at the jr-IDE Boot Menu to boot from the IDE device (DOM or SD), I will see a green LED on the jr-IDE flash 3 times, then it will just hang. The system will not boot, and it will continue to display the jr-IDE Boot Menu on the screen. It just sits there until I turn off the jr or press Ctrl+Alt+Del.

So.. it can be rebooted with Ctrl+Alt+Del; it's not so locked up that it can't read keypresses. But it just sits there.

I tried the debug steps on the DOM, just like I did on the SD. No dice. I re-fdisk'd, re-"format /s", reran "fdisk /mbr". Nothing. Still won't boot from the DOM.

I'm thinking that this might be the point at which there is something obvious that I missed, because I'm too focused on the problem and can't see the big picture, just like a problem at work that you can't solve but a coworker quickly solves it because his mind is not fixated on one part of the problem.

Any ideas? Could there be a problem with the jr-IDE itself, given that the same thing happens when I use an SD or a DOM? Is there some jumper or DIP switch that I should be checking? I did try swapping the jumper for +5V on pin 20 on and off; no difference.

The slow boat also brought me a 44-pin-to-40-pin IDE adapter (to enable me to use a laptop IDE hard drive in a desktop PC). I have an 80GB laptop IDE hard drive and an external power supply for it; I might just try it, to see if I get the same result.

I am stumped, though, and I'm starting to wonder if it's a problem with the jr-IDE itself.

Thanks for any tips you can give.

Well, that was weird.

I went ahead and hooked up my 80GB IDE laptop HDD to the jr, using the 44-pin-to-40-pin adapter, and... the jr couldn't boot, not even from floppy.

The jr-IDE displayed the name of the hard drive during POST as the first device, but when attempting to boot from that hard drive, it would hang (no surprise, really, since I haven't setup that drive with a DOS partition or anything).

However, I couldn't even boot from floppy! When using a DOS 5 disk, the disk started spinning, and I could hear the drive head seeking a few times, then... just spinning disk, no more drive head movement.

When using a DOS 3.3 disk, I couldn't even start the boot process. The jr-IDE said that it could not find a boot device!

Unplugging the laptop HDD resulted in both the DOS 3.3 and DOS 5.5 disks booting just fine.

So that's weird -- an IDE device being hooked up to the jr-IDE resulted in the jr's floppy drive failing to boot.

I dunno if that would job anyone's brain, but I thought I'd mention my attempt to use a laptop HDD with the jr.

I finally got it to work. I can now boot from the DOM and laptop hard drive. Probably same results for the SD, but I think I'll stick with the DOM, since I finally got it and it fits so well in the sidecar.

I ended up hooking up the DOM, SD-to-IDE adapter, laptop hard drive with 40-to-44pin adapter, and a CF-to-IDE adapter (which I'd tried before but got the "LBA too old" error) to a modern laptop with a USB-to-IDE cable.

I made sure that each worked okay in Windows 7 then used Diskpart in Windows 7 to create a 2GB FAT partition on each device, formatted, then connected to the jr.

When doing this, I removed the NTFS partition from the laptop hard drive and replaced it with a 2GB FAT partition. The jr no longer had that weird problem in which is failed to boot from floppy with that laptop hard drive attached.

The CF-to-IDE adapter still results in "LBA too old"; no real surprise there.

At some point, pressing "C" on the 2nd jr-IDE Boot Menu caused it to actually use the attached IDE device; only pressing "C" on the 1st Boot Menu caused it to hang. That's when I reread the documentation about patching DOS 5 and went through the Debug steps to patch C: for DOS 5... and it worked. At last. First jr-IDE Boot Menu will now boot from the DOM, rather than hanging.

Now to work on the Gotek.

Glad you got it working. Makes the jr so much easier to use. It is OK to say that I sold you the PCjr, I will eventually have a couple more to sell, so having people know that he one I sold you is working should help. For the parallel port side car, doing the bidirectional communication mod really helps if you hook up parallel devices like CF readers or 3.5" disk drives. I have both and only use them when transferring things I've downloaded using a modern machine, but they make that process much easier.

I regret that I wasn't able to get to this thread earlier - weekdays are just too busy and I wind up trying to catch-up on the weekend.

I really need to write a complete how-to to do this; DOS has so many gotchas, never mind the weird combinations of hardware.

The basic steps are:

• Ensure the drive IDs correctly when the jrIDE BIOS scans it. If you see data corruption here there is no point in continuing.
• If you get the device too old for LBA stop; jrIDE can't use CHS. Something is not compatible and needs to be replaced.
• Wipe the first sector (the master boot record) of the drive to zeros. This can be done on the PCjr using debug.com
• Use Fdisk to partition the disk
• The partition that you are going to boot from must be marked as "Active" or "Bootable"
• Use Format to format the disk
• If using DOS 5 or later stop and patch the boot sector. You also need stacks=0,0 in your config.sys and you need to setup jrconfig.sys.

Mike

Add "FDISK /MBR" in there somewhere as well

If you get the device too old for LBA stop; jrIDE can't use CHS. Something is not compatible and needs to be replaced.

99.9% of the time, this message simply means that the data being returned from the drive is all zeros or garbage, meaning there is a hardware incompatibility between the drive in use and the jrIDE.

If I were to go back and work on the BIOS again, I'd scan the incoming data packet from the IDentify command and try to determine if the data returned was valid or not before determining if the single CHS/LBA bit was set before throwing that message.

That's the reason for step 1 - "ensure the drive IDs correctly." If you see "M1x1o1" instead of "Maxtor" you are going to have a bad time.

I think a small stand-alone utility to ident the drive and runs some basic checks might be helpful. I wouldn't want a complicated BIOS, but a standalone util would be great for this.