• Please review our updated Terms and Rules here

XTIDE tech support thread

hargle said:
you likely have some that were SDP'ed by my eeprom burner and some that are not
Click to expand...

I received the two new XTIDE kits today (good timing). I just tried one of the new EEPROMs on the old XTIDE card.

I first tried flashing with Software Data Protection set to N. I got polling timeout error since SDP was working properly (write cycle never began because of SDP so polling couldn't work either).

Next i tried flashing with Software Data Protection set to Y. Flashing was successful just like on the old Atmel EEPROM i've been using. Your EEPROM burner certainly works fine since SDP was successfully enabled.

I tried flashing on clone XT with 8088@4.77MHz (it has underclocked 8MHz CPU), 640kB, CGA.
It has BIOS that displays:
(C) MICROWAY TURBO version 2.0 1985
International Business System
 
Hi! Thanks! Does that explain all the issues k* is seeing? It sounds like there are differences between his boards. I suspect if he sent me his EEPROMs, I could send them through the programmer and wipe them back to "like new" state and reprogram them from there -- SDP or not. It would be interesting to see what would happen using EEPROMs programmed and verified using the programmer versus the flashing software.

I've been running the 27C64's since the beginning and I've never had any trouble at all. The XTIDE has been solid as a rock since day 1 for me even with the wirewrap prototype units. Although I'll admit I've been in and out as time permits so I may have missed some things. Speaking of, I need to reassemble my XT clone and put it away since the development phase of this project is long over. It is still sitting next to the work bench in the basement!

Thanks and have a nice day!

Andrew Lynch
 
Thank you, i'll try that build out on my second card and see what happens, also i'm going to try the second card in my tandy, maybe my zenith and see what happens with it, i'm confident that with your help i can get this sorted out. Downloaded that build of idecfg, thats tommorrows project :), And, i'm not trying to drive anyone away, this is a VERY great project, i just seem to be the one that has run into a problem and can't figure it out. Consider me the guinea pig for this revision. And guys, the good card formats a 2gb hard drive much FASTER than my pentium machine does. I'm just frustrated is all, i admit i tend to get a bit feisty when frustrated.

Hargle, i'm sure it's NOT how you flashed the chips, as i put the 5 other chips on the good card and they flashed to rc2 properly, and the kit flashes properly, and half my solder joints look like absolute crap. Funny thing is the bad cad will read the chips, it just refuses to write to them, and when it does it borks the chip. All the solder joints are good, a friend of mine is double checking for me, i'll have that back tommorrow, is it possible one of my IC'S are bad? Is there a way to test

Hargle, i got the keyboard xt-at interface, any ideas where i can get the female to female plug to bridge the gap? can i use a 5-pin din midi cable? i can get those a dime a dozen in female to female

once again. THANK YOU for doing this project
 
Last edited:
carlsson said:
If one gets the boot menu, but no hard drives are detected (I tried 2-3 different ones + various cables), could it still be a soldering problem? I have assembled both my cards, and neither works. :-( One gets to the menu, one is not even detected. I even made sure to not have any I/O conflicts in my PC.

Apart from visual inspection and carefully resolder what looks bad, is there any meaningful ways to measure connectivity across spots? I saw there exists some diagnostic tools requiring Hargle's BIOS but I'm not sure if they will help me at the moment.
Click to expand...

Did you get this figured out? I recieved my card today and soldered it up. The BIOS comes up but no drives are detected.
 
k2x4b524[ said:
Hargle, i got the keyboard xt-at interface, any ideas where i can get the female to female plug to bridge the gap? can i use a 5-pin din midi cable? i can get those a dime a dozen in female to female
Click to expand...
Yes, 5 pin MIDI cables work perfectly, provided all 5 pins are wired through. If yours don't work, i can pick up 3 footers here for $0.99 that I know work.
 
Unknown_K said:
Did you get this figured out? I received my card today and soldered it up. The BIOS comes up but no drives are detected.
Click to expand...

I have a card here with the exact same problem that I am debugging too. Funny that I've built 30+ cards now and perhaps got a bit too cocky with my builds-this one just isn't working. I've been trying to debug it in my spare time, except I have no spare time.:D

Check the joints on the dip switch. the upper row of pins often takes a lot of heat to get the solder to stick since there is such a huge chunk of metal underneath them. If just one connection on the IO side (left 4 switches) isn't connected right, your card will be decoding somewhere else, and the BIOS won't see it, and no drives will be found.

That didn't fix my card, but it is something common that can happen. My next task is to attempt to clean up some flux that may be interfering, as well as re-flowing all the joints. We'll see, and I'll keep posting what I find with mine.
 
hargle said:
I have a card here with the exact same problem that I am debugging too. Funny that I've built 30+ cards now and perhaps got a bit too cocky with my builds-this one just isn't working. I've been trying to debug it in my spare time, except I have no spare time.:D

Check the joints on the dip switch. the upper row of pins often takes a lot of heat to get the solder to stick since there is such a huge chunk of metal underneath them. If just one connection on the IO side (left 4 switches) isn't connected right, your card will be decoding somewhere else, and the BIOS won't see it, and no drives will be found.

That didn't fix my card, but it is something common that can happen. My next task is to attempt to clean up some flux that may be interfering, as well as re-flowing all the joints. We'll see, and I'll keep posting what I find with mine.
Click to expand...

I will check it, but the BIOS is searching at address 300 which is where the jumpers were set to as per the instructions.
 
On my initial boot, there was a (silent) I/O address conflict with my network card. I removed the NIC with no difference. Then I reconfigured the NIC to use another I/O address, but no difference. Perhaps I should reflash the XT-IDE in case the unintended conflict caused some problems?
 
Hi! If we get to the point of truly borked XT-IDE boards I would like to see them. There may be some design improvements we can roll into a future generation.

The soldering seems to be giving people problems due to the lack of VCC and ground pad thermal relief but that is an intentional design feature. Adding thermal relief pads are easier to solder but they limit the chips from using the ground fill zones as heatsinks. Hot chips mean low reliability and short life spans. Nice cool chips will last a lot longer and perform more reliably.

If the sockets are properly soldered to the thick non thermal relief ground pads, you should have a solid low impedance good thermally conducting path to ground. Solid ground paths are important for a long term reliable PCB. Of course if it is preventing people from assembling the board properly then it is self defeating. It's a trade off but my preference is for solid supply rails whenever possible.

Thanks and have a nice day!

Andrew Lynch
 
Unknown_K said:
I will check it, but the BIOS is searching at address 300 which is where the jumpers were set to as per the instructions.
Click to expand...

Well, the jumpers don't have anything to do with the IO space it's using. 300h is selected by the dipswitch.

What I'm saying is that if the dipswitch isn't soldered in perfectly, one or more of the lines of it will be floating, meaning the address where the card is actually decoding might not be at 300h. I caught one of these in a build I did where one of the pins on the dipswitch wasn't joined correctly on the memory address range (switches 5-8 ). So even when the dipswitches were set to D000h as the instructions said, because one pin wasn't soldered in right, the card was actually decoding at D400h.

For the Memory range, this is not as critical, since the card will still work no matter where the memory range is decoding, provided there isn't a conflict with another card.
For the IO range, it is *crucial* to get the decoding correct, because the BIOS is saying "look at 300h for the card" but the card might be sitting at 320h, so they miss each other.
 
Last edited:
Sorry for the trouble, But i'm haveing troubles with my xt-ide

Sorry for the trouble, But i'm haveing troubles with my xt-ide

I seem to be haveing troubles getting my xt-ide card to run.

When my friend first brought it too me, i added jumpers for jp1 and 2 and tried to fire it up uinder the default settings and the default bios (IDE_XT.bin)

Unfortunly, it kept detecting the drives as slaves, and would not boot from them

Useing the MFM controller to boot to dos proved to be a loong process, and even booting from a floppy netted the same results: Even if the xt-ide bios detected a drive, dos was unable to see it at all.

Ive removed all of the carsd in the system except for the ega card (Paradise or clone), tried re-flashing the bios with all 3 versions (ide_xt, ide_at, and ide_xtp) on another system, all to no avail.

The closest ive gotten is the card to de6tect the drive, find a bootsector....and then be presented with a "non system disk or disk error" in which case ive booted from the floppy or mfm drive, entered fdisk, and found the drive absent (IE, only sees one hard disk when both drive controllers are plugged in, dosent see any when the xt-ide is plugged in alone

I seem to be faceing a similar prolbem that a previous poster encountered. 3 of the drives ive used with the card no longer register when jumper for master, and jumpering cable select only brings them up as slave. Ive yet to bootup the drive diagnostics for the drives yet, however

The drives I've tried so far are:

Quantum pro drive lps (50meg) - jumered for master
Western digital Cavier 26400 (6.4 gig)
western digital 2200 (2gig)
Seagate Medalist st32122a (2.1G)

In all except the quantum, ive tried different jumper positions on the drive...Most do not register in xt-ide.

Ive noticed that when booting from the menu, the boot process goes slightly faster if i use a later revision of the card/bios (2.1 as opposed to 1.7) though its still a extremely longer process then booting off of the controller alone.

Ive tried 4 differant i/o ranges and memory addresses. Most will fail to bring up any drive, no matter what. I've also tried flasing with differant command ports, but it all results in the same.

The furthest ive gotten was the controller plugged into the motherboard ive used for flashing (a p-2 with 2 isa slots) and the Quantum. It booted, from which i imeatly transfered the drive and controller to the XT machine (8088 samsung clone running at 4.7 mhz) and at first the machine reported "no bootsector found" and then, on retry, "boot sector found" but then "non-system disk or disk error" and, booting into dos from the mfm drive, the drive fails to register in dos.

Ive currently tried with both the bios enabled and disabled. I have no jumpers installed on any of the IRQ pins. The only modification ive had to make is when i switched over to trying a 80-pin cable was to snip one of the pins to enable the cable to attach.

HELP!
 
Have you tried removing all the jumpers on the WD drives? As mentioned before, WD has a peculiar need to be told when a drive is a master in a single drive configuration vs master in a 2 drive setup.
__
Trevor
 
salamontagne said:
When my friend first brought it too me, i added jumpers for jp1 and 2 and tried to fire it up uinder the default settings and the default bios (IDE_XT.bin)

Unfortunly, it kept detecting the drives as slaves, and would not boot from them
Click to expand...

Did you add a jumper to K in the L position?
You need THREE jumpers, as referenced on page 1 of this thread.

That's the cable select pin, and if that baby is floating, you would very easily have drives detected as slaves.
 
hargle said:
I have a card here with the exact same problem that I am debugging too. Funny that I've built 30+ cards now and perhaps got a bit too cocky with my builds-this one just isn't working. I've been trying to debug it in my spare time, except I have no spare time.:D

Check the joints on the dip switch. the upper row of pins often takes a lot of heat to get the solder to stick since there is such a huge chunk of metal underneath them. If just one connection on the IO side (left 4 switches) isn't connected right, your card will be decoding somewhere else, and the BIOS won't see it, and no drives will be found.

That didn't fix my card, but it is something common that can happen. My next task is to attempt to clean up some flux that may be interfering, as well as re-flowing all the joints. We'll see, and I'll keep posting what I find with mine.
Click to expand...

I have fixed my card. I ended up doing 2 things to it: I hit every single pin and re-flowed the solder. All you really need to do is just hit the pin with the iron and move it a little bit. the solder will flow and seat itself in a different position.
I also changed out the entire IC set from another card. The card now works as expected. I'll be keeping an eye out on this one and do some extra testing on it though.

The old set of ICs I used on a different card, and that new card works fine, so the only explanation to the fix is that it was at least one pin somewhere that was keeping the card from seeing any drives. Unfortunately, this wasn't a great debugging experience-it was pretty simple.
 
Yes, I've added a jumper for K. In a bit, I'll swap out the jumpers with others. I've rarely seen jumpers go bad, but it could be the prolbem.
 
Today I successfully flashed 8 different cards from RC1 to RC2. No problems with any of them at any time.

I used a zenith 8088 XT clone machine, had a VGA card in it and nothing else. 640k, 4.77MHz, single 8 Gig hard drive attached to the XTIDE, two 360k floppies attached to the on-motherboard controller.

To do the work, I copied idecfg and xt_bios.bin onto the hard drive, booted to the drive and used idecfg off the hard drive that I was about to flash the controller on. I used only the defaults that the idecfg program has for flashing. I'm certain the atmel parts I flashed were of the same stock that was shipped out with everyone else's card. That is, pre-flashed with software data protection already enabled.

Everything worked exactly as suspected.
 
hargle, aitotat, no dice, my zenith and tandy keep returning data verification errors, with rc1 of the program and rc2 of the program on the second pre-fab card. Lynch, i may take u up on your offer to send you the board so you can give it a once-over, maybe there is something i missed, but in the mean time, one more thing to try....

*update*

I have a REALLY REALLY finicky card.
I finally got it to flash on the tandy, using the defaults in the IDECFG program
BUT i recommend using the version of idecfg that matches the bios revision

Anyway a little guide perhaps.
Should flashing fail repeatedly, try this

1. Remove card, check for bad solder joints, loose IC's, switches and jumpers not set right.
From here on down is how i finally got it to work *still finicky*
2. remove bios enable jumper, install card, turn on computer
3. reinstall bios enable jumper while machine is on
4. enter idecfg *matching bios revision if possible*, flash to the defaults
5. if the flash fails, hopefully it won't at this point, in IDECFG, turn SDP off, and change another random setting, reflash
6. flashing WILL fail *did for me*
7. repeat 2,3 and 4
7.5 warm reboot *reset button if you have it*
8. upon successful flash, in idecfg load the settings from the rom and alter as needed, reflash from the rom.
9. reboot, have fun with functional card.

I recommend this be done on AT LEAST a 286-10, or an 8086-8
You may NOT have a bad card, just one that is being really, really, REALLY anal about life :)
 
Last edited:
Got much further along....and I apoligize for not reading the first page of this thread.... Could have saved alot of headaches

Re-flashed my card to RC 2, Xt_IDE.bin. Default settings. Attached a seagate ST364218 6.4 Gig, formatted with win98 ms-dos, swapped
back onto xt-ide card

Boot sector found...but froze.

Booted from floppy (Dos 5.0), fdisk, delete partition, create new, reboot from floppy

tried to format, no drive found, entered fdisk, no partitions defined, exit, fdisk /mbr, reboot

fdisk again, create partition, reboot....xt-ide failed to detect drive, reboot, found drive

create partition, on reboot no partiton existed.

Anyone got a clue what could be going on? This time, i swear, i read from the first page of this thread...
 
I think i know what is the problem with flashing.

I assembled the two new XTIDE kits. Both of the cards worked on first try except that i couldn't flash the other with SDP enabled.

The other card had Atmel AT28C64X instead of AT28C64B. I replaced the X EEPROM with another B EEPROM and flashing was successful. The Atmel EEPROM ending with X does not appear to support Software Data Protection.
 
You must log in or register to reply here.
Back
Top