• Please review our updated Terms and Rules here

Kalok Model KL 320 - Copy Data

SpeedStor

SpeedStor

I will probably add the following SpeedStor information to my web site, also expanding on it.


Controller Test
-----------------

Navigation: [Diagnostics] [Controller]

SpeedStor asks the controller to do a self-test, and displays the response.

Note that it is not possible for the controller to COMPLETELY test itself. The passing of this test simply provides confidence.

Some people have reported that this test fails for their controller, even though their controller/drive combination appears to be fully functional. Maybe their controller is slightly faulty (e.g. bad buffer RAM, if any), but maybe their controller simply does not support the self-test function.



Seek Test
-----------------

Navigation: [Diagnostics] [SeekTest]

SpeedStor simply gets the controller to step (seek) the drive's heads to various cylinders.

You will hear the heads being stepped (moved). In the case where the drive has a stepper arm/flag/interrupter (example [here]), you will see the arm move. Often, if you put your hand on the drve, you will feel the stepping.

In the times that I have tried, disconnecting the data cable does NOT cause this test to fail. That tells me that the controller is not reading the disk surface, and from that, no true verification that the heads are where they are meant to be.
So, im my opinion, the passing of this test is no garantee that the drive is seeking (stepping) properly.


Read Test
-----------------

Navigation: [Diagnostics] [ReadTest]

Requirement: That the low-level format (LLF) on the drive is one that the controller can read. If there is a LLF mismatch between controller and drive, then all tracks will read as bad (note that other problems can also cause that symptom).

SpeedStor gets the controller to read sectors on the tracks (track = cylinder/head combination) that you specify.
Bad reads get listed below the purple banner.
 
Hey Again,

Does anyone know if there is a switch I can use with Speedstor to set the display to monochrome?

Even though there's a color graphics card in the computer, the display in monochrome (black/white only). The television is connected to the computer via a single cable plugged into the RCA jack on the card. I am not sure I'll be able to read everything on the Speedstor screens with the current display output.

I searched online and couldn't find any help. All of the information I read said the program is compatible with CGA, EGA and monochrome displays.
Thanks,

Jeff
 
... 1. The display is terribly hard to read (some of the information is not readable). I am using a small flat panel television connected to the RCA jack on the CGA card. I did some Googling on the subject and found information that says that it is compatible with CGA. I was hoping there might be a command line switch that I could set; but, I can't find any documentation that provides this information.
I get good results using my particular 32" Panasonic TV (although, no colour). It sounds like your TV is the issue.

You could try changing from '80 column by 25 line' mode to '40 column by 25 line', which is what IBM intended for operation via the RCA connector (due to limitations of TVs at the time). However, programs such as SpeedStor, designed only for 80x25, will display oddly in 40x25 mode.

If you want to try that, then if MODE.COM is on your boot disk, type "MODE CO40" at the DOS prompt. Otherwise, change the setting of the two video switches on the 5160 motherboard.

Does anyone know if there is a switch I can use with Speedstor to set the display to monochrome?
sstor /nocolor
 
Good Morning,

First, thanks modem7 for the information on the /nocolor switch. It worked great.

Now for the test results:

Controller Test: Passed

Seek Test: There was a lot of drive activity that ended when there was a message that stated that the test failed on cylinder 263

Read Test: The read test continued up to the point at which 31 entries were listed as defects. At that time there was another message that there was an error 0X20 indicating a drive failure. Not sure if it means anything; but, the last head listed in the defect report was 62 (of 63).

Following the read test I ran another controller test just to see what would happen. That test passed.

Are we at the end of the line?

I did a little looking into data recovery services. I'm not sure what is actually on the drive, so I'm not really sure it's worth pursuing this option. (Not to mention the fact that some of those companies have less than admired reputations.)

Thanks,

Jeff
 
Looks like the drive has a hardware failure.

I don't know if anyone else has any ideas but I have never gotten data back from a drive that failed both read and seek tests.
 
Hi Again Guys,

Well, my optimistic thoughts about this drive are pretty much gone. We sure gave it a good try.

I really appreciate all the help you have all provided. You have a great group here. It's great that someone new like me can come here and be treated so well. Thank you all.

After all this I'm thinking about listing the 5150 for sale. I don't know that I'll ever want to go to this extreme with it in the future. Is your classified area a pretty good place to do that?

Take Care,

Jeff
 
.. the last head listed in the defect report was 62 (of 63).
Head 62 !!! The Kolac KL320 has only 4 heads. It sounds like the controller is reporting the wrong drive geometry to SpeedStor, and SpeedStor is thus testing cylinders and heads that do not exist (resulting in error).

On an XT, when SpeedStor is run, and navigation to the Diagnostics screen done, SpeedStor will show the parameters of the hard drive, as reported by the controller.
For example, below is what SpeedStor shows for my ST-412 drive.

kl826415.jpg


For the Kolac KL320, expected to be shown is 615 cylinders, 4 heads, 17 sectors per track. I bet that you are not seeing that.

As a guess, your dynamic controller (see [here]) is blindly accepting 'bad' geometry stored on the reserved cylinder.

Is data still recoverable? Possibly. You can probably read various sectors on the hard drive by using Norton Disk Editor in 'physical' mode. You specify the cylinder/head/sector combination that you want to read. Recovery at that level is quite laborious, and you would need to do research (e.g. workings of the FAT table) in order to piece things together.
 
Head 62 !!! The Kolac KL320 has only 4 heads. It sounds like the controller is reporting the wrong drive geometry to SpeedStor, and SpeedStor is thus testing cylinders and heads that do not exist (resulting in error).
Don't you think he meant cylinder 62 and not head 62? :) :) :)
 
Hi Stone,

Don't you think he meant cylinder 62 and not head 62? :) :) :)

I'll go back an run the test one more time. The results I posted were from running the test in 40 x 25 display mode (before modem7 shared the /nocolor switch).

There were 31 errors reported. The first was head 0, cylinder 2372. The second was head 6, cylinder 2332 and the very last was head 62, cylinder 2316. There were various heads listed, some with multiple cylinder errors. (The list of heads includes 0, 6, 9, 13, 20, 22, 27, 28, 30, 31, 32, 33, 35, 37, 43, 46, 51, 54, 55, 59, 60, 61 and 62.)

I'll run the test again this morning and report back.

Unless the display mode caused the quirky data to be generated, I don't really have any explanation. I can assure you guys, though, that the error information that I stated here was what Speedstor generated.

Take Care,

Jeff
 
Mornin',

Well, here are the drive parameters from the controller screen.... lol, there is a problem.

Mfg/Mod: WD SuperBIOS
Cyls: 4096
Heads: 64
Sec: 17
Precomp: 0
Lzone: ?
Total Bytes: 2,281,700,864

And yet the controller passed the controller test.

I have checked the cabling. Not sure what more can be done. I suppose the cable connectors on controller card could be oriented 180 degrees different...

Thanks,

Jeff
 
Hi Stone,

Have you tried entering drive parameters manually? If not...

Sounds like a good plan... I just want to be careful not to make things worse.

Are you suggesting to input data for cylinders (64) and heads (4)? Do I also need to input the data for the drive capacity, or will that work out from the other parameters?

Thanks again.

Jeff
 
Hey Again,

I decided to take a look, thinking I might be able to figure out which parameters need to be changed.

Once I started Speedstor, I selected Manual Setup, then Type, then Drive 1. At this point there were a number of hard drives listed, and a selection for Manual Parameters. The Kaylok drive was not listed.

So, I chose Manual Parameters. All of the entries were "0". My thought was that I would populate the entries for the number of cylinders, number of heads and sectors. There was also a note at the top stating that only the parameters required different numbers should be changed. The concern that I have is that the other entries were all "0", and that might not be correct.

I'm thinking that there may be another drive type that's listed that might be close; but, it would be a guess on my part.

The following are the parameters that are being displayed:

Number of Cylinders (I know that this should be 615)
Number of Heads (I know that this should be 4)
Starting Cylinder for Pre-Compensation: ?
Control Byte: ?
Sectors per Track (AT): ?
Landing Zone Cylinder: ?
Starting Cylinder for Reduced Write Content (XT): ?
ECC Burst Byte (XT): ?

Any suggestions for either a similar drive type, or the parameters to input?

Thanks,

Jeff
 
Well, here are the drive parameters from the controller screen.... lol, there is a problem....
It suggests to me that the drive geometry information that the controller earlier wrote to its reserved cylinder, has been lost.

A Seagate ST-225 would probably work as well.
I agree. For what you are doing, you can specify the Seagate ST-225. It also has 615 cylinders, 4 heads, and 17 sectors per track. All we are doing is making SpeedStor aware of the drive's geometry, since the controller was unable to supply the information.

When you do run the read test, note that is not unusual for drives of this age to have bad areas on them. If I was to run SpeedStor's (or equivalent) read test on the MFM drive in my IBM 5170, that would reveal about five faulty tracks (which DOS compensated for when I high-level formatted the partition).

It was a few years ago since I last ran a read test over the entire drive. Maybe if I was to do it again, there would be many more bad areas due to deterioration.

Maybe your drive has deteriorated such that parts involved in the boot sequence are affected. But if some parts (tracks) can still be read by SpeedStor and other tools, then some recovery is possible. The bad news is that the recovery process would be laborious.
 
When you do run the read test, note that is not unusual for drives of this age to have bad areas on them.
Not only that but when these drives were new they came out of the factory with bad areas on them. They all have lists glued to them with the (factory created) bad sectors enumerated. It was very rare for a drive to be manufactured completely error free.


Maybe your drive has deteriorated such that parts involved in the boot sequence are affected. But if some parts (tracks) can still be read by SpeedStor and other tools, then some recovery is possible. The bad news is that the recovery process would be laborious.
Yes, if Track Zero is hosed it would require extraordinary methods to get any data off that (or any) drive.
 
Back
Top