Data recovery and disk repair questions and discussions related to old-fashioned SATA, SAS, SCSI, IDE, MFM hard drives - any type of storage device that has moving parts
April 16th, 2018, 23:06
ISTM = It Seems To me
I havent followed this thread well..
try looking for a manual for your motherboard. check what the POST codes are. 3 beeps may be RAM. in the manual check the configuration of RAM and capacity etc to make sure you are using the right amount/type etc.
BTW, when posting, best include pictures of both sides of an item, and also when taliking about hardware, model numbers of everything, not just the bit you think is the issue. saves many back and forth posts with questions and answers to build a picture of your setup/issues.
as an aside, are you able to format a floppy succesfully, or do you still get the "Format Complete, Unable to Write BOOT"
April 17th, 2018, 9:55
Are those really just three beeps, equally spaced, or are them a longer one, then silence, then two short ones in rapid succession ?
Or are they more than three , like one beep repeating continuously ?
Have you tried the new vga card in all slots ?
Can you post pics of the hard disk you are using, with special interest in the connector region ?
April 17th, 2018, 15:15
If the new motherboard has one, try installing the "old" VGA card in an 8-bit ISA slot. Otherwise use a scalpel, or similar, to cut the trace to data bit 14 at the appropriate connector pin on the card.
As for the row of 3 pins on the replacement card, I suspect they configure the card to run in 8-bit mode or to autodetect the presence of a 16-bit slot.
April 17th, 2018, 17:38
Woudln´t it be better to try to determine if his "new" vga card is working or defective first ? Cutting traces on the board of his only working card could make things worse ( more so if the wrong trace(s) end up cut ) ...
IMHO, it would be easier to try changing the jumpers in the new card to see if that is the cause. I never had to fiddle with jumpers in these kind of cards, but that doesn´t mean someone hasn´t moved jumpers just to be in a prettier position ( like people seem to have a need to to when they see a crooked capacitor in a board ) .
And if something in the card was causing problems with one of the bits in the bus, wouldn´t it cause errors in the display ( or wrong characters ) or other errors ?
April 17th, 2018, 18:10
Yes, by all means get the new card working, if possible. As for cutting the trace to DB14, that is on the 16-bit end of the card. When the card is working in 8-bit mode, that end of the card would be ignored.
Maybe the OP could insert a very thin insulator between DB14 and the slot, say a thin piece of paper, or place tape over the connector pin.
April 17th, 2018, 19:00
fzabkar wrote:Maybe the OP could insert a very thin insulator between DB14 and the slot, say a thin piece of paper, or place tape over the connector pin.
I agree with that, and think that it would be more prudent. Cutting traces is always dangereous when one can "slip" and cut the wrong trace, or count from the wrong position ...
Maybe it is something I do not remember or never knew, but could you explain your reasoning for considering that the vga could be causing problems with that bit in the bus ? I can only imagine that if something was wrong with the card, it would not work okay and this would be seen, in artifacts on the screen or wrong colors....
That thing with one bit of the ide traffic being wrong seems so much like those cases with a broken pin in the drive connector that I can´t say the hdd is free of guilt.
I need to see if I have some small IDE disk here. If I cut the relevant pin from it "for science", could we try to reproduce what the OP is experiencing ?
April 18th, 2018, 4:33
ISTR that the OP has already tried 2 HDDs and 2 IDE cards, both with the same symptom.
As for the VGA card, many (most?) can run in an 8-bit ISA slot, if properly jumpered. If the OP's card is jumpered for 8-bit mode, then it ignores all the C and D signals. That includes data bits 8 - 15 and address bits 16 - 23. Remember that the OP also has an XT IDE (16-bit) controller card that runs in an 8-bit slot.
April 18th, 2018, 8:25
We didn´t exactly saw pictures of all of those parts. Maybe what the OP consider as trying "another cable" is not the same as you and I imagine.
As he now has changed motherboards, that would rule that out also, so clues would indicate the vga card , and like mr. Spock could say, "curious" to see a vga card cause that bit flipping. Of course, I´m not saying it is impossible. We use to have some interesting discussion with customers here about the difference between possible and probable.
Also, maybe something initially damaged the hdd/cables/superio so that now that carries over to new boards ?
As for the XT controller, I think that would present some more problems when configuring and getting it to work with a modern ( ha! ) board and hdd ?
April 18th, 2018, 15:13
XT BIOS has no support for HDDs, so the card has its own ROM. I don't know if, or how, the XT ROM will get along with an AT BIOS which has its own IDE support. I merely mentioned the XT card since it demonstrates that it is possible to get a 16-bit peripheral to communicate via an 8-bit bus.
The OP did measure the voltages at the C pins when writing an all-ones bit pattern to the drive. He stated that the voltage on DB14 was dead on zero for the whole time, so ISTM that the problem must be right there, not further downstream. That said, a DMM is a poor tool for observing bus activity.
April 18th, 2018, 20:03
What where the voltages in the other pins ( C16, C15 for example ) during the test ?
Pics of the hard disk board, and closes of the connector region, please ?
Also, the OP is using just the 4-pin connector to power the drive, right ?
To Frank : If "something" is keeping the C17 line down, a contraption with a led connecting that pin to some other pin that is always high should indicate something, wouldn´t it ?
Also, trying with the minimum number of ram sticks to see it the behaviour is the same ? If the OP could check the disk in another computer, I would suggest formatting it blind ( without the vga card inserted ) to see if now the problem would disappear.
April 19th, 2018, 4:32
rogfanther wrote:What where the voltages in the other pins ( C16, C15 for example ) during the test ?
The OP's measurements are a little unclear but ...http://forum.hddguru.com/viewtopic.php?t=36610&p=257605#p257605
rogfanther wrote:To Frank : If "something" is keeping the C17 line down, a contraption with a led connecting that pin to some other pin that is always high should indicate something, wouldn´t it ?
ISTM that would not provide any more information than what has already been observed.
Bill1962 wrote:And it is 0 volt, it is not like it is 0.7 volt, or 1.3 volts or whatever.
April 19th, 2018, 8:16
The measurements are unclear, because he didn´t say what was measured where.
Give the lack of information, the fact that the computers are an ocean away, and that he doesn´t want to test with another disk, it would probably be be easier if he buys a complete ( and confirmed working ) system. Then he can proceed with installing his software without scavenging for problem causes.
May 8th, 2018, 17:13
Has anyone checked for a boot sector write protect option in BIOS yet?
Also, I have an old 386 box laying around that also had this exact drive before replacing it with ST31276A with an add-on IDE card. I would suggest looking in BIOS for a Type 47 "User Defined" type, My 1991 AMI had it, yours should too.
Powered by phpBB © phpBB Group.