I've killed my jride

Hardware questions and modifications

I've killed my jride

Postby MikeModified » Sun Apr 27, 2014 7:49 pm

I had surprisingly little trouble assembling the jride (an old dog CAN learn new tricks :lol: ). It's been hanging on the side of my first jr for the last couple of weeks without a case or support. I'm using a Kingston 8gb CF with a startech.com ide2cf adapter.

First I had to learn that to get a bootable hd with the Kingston, I had to use MS-DOS 5.0 fdisk rather than 6.22, that is boot with 5.0, run fdisk for the initial partition, then reboot with 6.22 and set up the rest. Worked as a bootable perfectly for more than a week.

Then, wanting to do everything at once, I waited until I had another printer sidecar and had modified one for bidirectional. That was yesterday evening.

I gently removed the jride from my jr (not for the first time), assembled it into an empty printer sidecar and installed. Before buttoning up, powered up to test, and Drive 0:Not installed!

Pulled the sidecar off, discovering that I needed screws to hold the jride together, I put it aside until this morning. Ran down to the hardware store and bought screws (3/48 x 1/2" btw, could be a little shorter) and assembled it again. Still Drive 0:Not installed.

Pulled it apart and carefully inspected each solder joint, all look "good", no cold joints. No bridges.

Then I noticed that RTC gives me:

"ERROR: Unable to update DOS date
" Sunday April 27, 14814 - (correct time)"
OR
"ERROR: Unable to update DOS date
" Unknown Unknown 24, 1958 - (correct time)"

Several more inspections and even a quick touch of the soldering iron to a couple of joints. Nothing.

Went out and bought another (different brand) CF. No changes.

Pulled a hd that ran a couple days ago in a Tandy 1000 with the xtide and tried it. No changes.

I'm stuck. This is my first assemble job. What/how to troubleshoot?

BTW memory is correct. Does the same thing booting 5.0

A couple things I noticed that I had intended to ask about:

1. The LED only lites when the jr is off and there is 5v to the ide2cf adapter.
2. When I tested my old hd's when setting up two xtides in Tandy 1000s, I had two drives that were bootable, three that the xtide adapter would recognize but wouldn't boot from, and several that weren't recognized. When I tested the three recognized/non-booting drives with the jride, none of them were recognized. I didn't bother with the unrecognized.

Mike
MikeModified
 
Posts: 122
Joined: Sat Mar 15, 2014 5:10 pm
Location: Portland, OR

Re: I've killed my jride

Postby jmetal88 » Sun Apr 27, 2014 8:07 pm

The RTC in mine actually does that periodically. I just have to set the DOS time manually when it happens and re-update the clock. Seems to hold on to the time longer if I use the PCjr regularly. As for the other, I'm not sure. I did have problems getting a couple of the jrIDEs that I assembled recognizing my drive at first, but they always worked after unplugging and re-plugging things a few times.
jmetal88
 
Posts: 797
Joined: Sun Jul 25, 2010 10:22 am

Re: I've killed my jride

Postby Brutman » Sun Apr 27, 2014 8:13 pm

I'm not clear on what you mean when you say something like "memory is correct". Just to be perfectly clear:

The machine will show the 40 column splash screen and count up to 640KB. After that, if the BIOS on the jrIDE is working correctly, you will see a BIOS boot message in 80 column mode with the following text:

Code: Select all
PCjr jrIDE BIOS version 2012-11-12 (http://www.brutman.com/PCjr)
Based on XT-IDE BIOS v0.11 (http://www.vintage-computer.com)

Base memory: 112  Total memory: 736

Drive 0: ...


The BIOS does the memory count to 640KB. Be we know the jrIDE can provide up to 736KB so we tell the BIOS that is what is present.

After that, it will try to detect your hard drives. Any "recent" drive that understands LBA addressing should be recognized by the jrIDE and report in. Any drive over 8GB in size should work. Drives bigger than 1 or 2GB generally work, but there are some cheap and nasty Western Digital drives out there as large as 6.4GB that do not support LBA addressing, so they will report in but not be usable.

If you are not seeing the BIOS messages then you need to reflash the BIOS as a first step. If you can't flash a new BIOS on then you should start looking for bad solder connections, solder bridges, etc.

Alan - thoughts?
Brutman
Site Admin
 
Posts: 970
Joined: Sat Jun 21, 2008 5:03 pm

Re: I've killed my jride

Postby MikeModified » Sun Apr 27, 2014 8:36 pm

Sorry. I get the display of recognized memory as you show it.

Mike
MikeModified
 
Posts: 122
Joined: Sat Mar 15, 2014 5:10 pm
Location: Portland, OR

Re: I've killed my jride

Postby Brutman » Sun Apr 27, 2014 8:48 pm

Nothing to be sorry about - we just need to be very explicit because debugging remotely is hard.

If you are getting the BIOS messages you are not in terrible shape. I'd concentrate on looking for soldering problems, stray metal where it doesn't belong causes bridges, and bent pins on the IDE connector. When connecting a hard drive I use the newer 80 pin style cables because the quality is better. Be sure to let the drive spin up first before powering on the machine; if they both power on at the same time some drives are not fast enough and will not report in when queried by the adapter. This only applies to real drives of course - your CF card setup should always be ready.

I doubt there is a fault in the CPLD chips but those can be flashed too if you are really paranoid. Alan can speak more to their reliability.


Mike
Brutman
Site Admin
 
Posts: 970
Joined: Sat Jun 21, 2008 5:03 pm

Re: I've killed my jride

Postby jmetal88 » Sun Apr 27, 2014 9:16 pm

Brutman wrote:Nothing to be sorry about - we just need to be very explicit because debugging remotely is hard.

If you are getting the BIOS messages you are not in terrible shape. I'd concentrate on looking for soldering problems, stray metal where it doesn't belong causes bridges, and bent pins on the IDE connector. When connecting a hard drive I use the newer 80 pin style cables because the quality is better. Be sure to let the drive spin up first before powering on the machine; if they both power on at the same time some drives are not fast enough and will not report in when queried by the adapter. This only applies to real drives of course - your CF card setup should always be ready.

I doubt there is a fault in the CPLD chips but those can be flashed too if you are really paranoid. Alan can speak more to their reliability.


Mike


I have had the exact opposite experience with the 80-conductor cables. I mean, I don't have very many of them, but with the ones I do have, I have a lot more issues with drives working properly as opposed to the old 40-conductor cables.
jmetal88
 
Posts: 797
Joined: Sun Jul 25, 2010 10:22 am

Re: I've killed my jride

Postby alanh » Sun Apr 27, 2014 10:15 pm

If the memory count up is working, that means all the connections between the side-car connector and the memory and many of the connections between the side-car connector and the main PLD are good.

Do you get a reliable POST display? Are the RTC problems consistent or intermittent? Is the RTC chip seated well? It's a large chip and the socket don't hold it very well I'm afraid. Also check for pushed back pins on the main board. I've done that twice accidentally when inserting a side car.
alanh
 
Posts: 258
Joined: Tue May 10, 2011 6:52 pm
Location: Atlanta, GA

Re: I've killed my jride

Postby MikeModified » Sun Apr 27, 2014 10:51 pm

Alan:

Ah! I tried to put in all pertinent info and forgot: every reported test, except the known-good-to-xtide HD test, were conducted on two Jr's with the same results.

POST, in every observation, was good. One time, out of 50?, jride (I think) gave me an "ERRORH"; a three-fingered salute corrected that.

RTC, since first observed, always gives the error messages, the first example is more common; no difference in hard and soft boots that I can tell.

I did mung one female pin (the lower rightmost) on the jride slightly, but I connected the bride to the printer sidecar and had good continuity from the printer female to the jride male, so I assume that's not a problem. I'll double check the main board pins on both Jr's in the morning, although I (casually) inspected the main Jr numerous times.

More? :)

Mike
MikeModified
 
Posts: 122
Joined: Sat Mar 15, 2014 5:10 pm
Location: Portland, OR

Re: I've killed my jride

Postby jmetal88 » Mon Apr 28, 2014 8:05 am

Have you tried my suggestion for the RTC about setting the clock from DOS and issuing the update command yet? That always fixes my RTC when that happens to me.
jmetal88
 
Posts: 797
Joined: Sun Jul 25, 2010 10:22 am

Re: I've killed my jride

Postby alanh » Mon Apr 28, 2014 1:37 pm

Again, do you get a reliable POST display?
alanh
 
Posts: 258
Joined: Tue May 10, 2011 6:52 pm
Location: Atlanta, GA

Next

Return to PCjr Hardware

Who is online

Users browsing this forum: No registered users and 7 guests