This repository is for investigating how floppy disks wear down so that the hobbyists, curators, and retro-computer archaeologists can keep these beasts from the dawn of the computer-age alive as best they can with what materials are now available.

Currently there is a lack of information about the correct replacement for the felt pressure pad seen in single-sided disk drives used in early microcomputers. The pad pushes the rotating diskette against the read/write head and there is a possibility that using the wrong material can lead to a disk wearing out prematurely, especially if they are used as "flippy" disks.¹

At the time the drives were made (late 1970s to early 1980s), professional research into the wear properties was limited because floppy disks were abundant and one could just buy replacement pads from the manufacturer. (Shugart recommendeded doing so every 10,000 hours, if I recall correctly).

There are plenty of anecdotal stories of substitute materials people have used that "work" to allow the drive to read and write, but no information about which materials cause the least damage to magnetic media.

The hope is that by releasing a program which can make testing easier, those who care about such things will be inspired to run tests and share the results. As the most commonly available single-sided drive was the Drive II used in the Apple II line of computers, this program is written in Applesoft BASIC for a system running Apple's DOS 3.3.

In order to test a component, use the TRIBOLOGY TRIAL program which prompts for a track to move the head to and then cycles around on it forever, perhaps scraping off the magnetic material and etching a ring into the diskette.

An ideal test will run for 7 days straight. (Shugart says there should be no damage to the media from the pad or head after 3 × 10⁶ revolutions). However, visible evidence of damage has been appearing in less than 24 hours.

• The program attempts to calculate elapsed wall time but it is not very accurate, so it's better to use an external clock when possible. For reasons not clear to me, the speed of my timing loop depends upon whether it is run automatically as the Apple II "HELLO" program or from the prompt.

• The best metric for evaluation is not yet clear. Perhaps the ideal would be for the test rig to automatically compare the signal level on the track before and after the test. However, the disk controller does not make that information available to the Apple II and so multimeter or oscilloscope may be necessary. For now, visual estimates are being used along with minimal verification that the data has not been corrupted.

Results: No noticeable difference. However, concentric rings at every track are visible under a microscope, so perhaps this floppy has already been worn down?

Results: Track 0 has a black ring on the side facing the felt. That ring has data as the other side of the disk was formatted before the test, but it does not appear to have been affected at all.

This is the same felt pad as in Trial 1-A.

There do appear to be light etch marks visible only when held to the light, but they are on side A, which is touching the head, not the felt. These did not occur in Trial 1-A. Could the felt be pushing too hard since it is too tall? Could it be a density issue? Or is it a problem with this drive's head?

No marks on felt side. No apparent new marks on head side.

Results: Track 30 has a faint, light ring on the side facing the felt. On the reverse (head) side, there is a prominent groove at track 30 which appears dark or light depending upon the angle of the light.

Results: No marks whatsoever. This is a very strong indication that it is indeed the replacement felt pad that is causing the marks in tests 1 & 2 despite them being most prominent on the head side of the disk.