JimDrew wrote:The index pulse sensor is typically a hall sensor with 3.5" disks, and an optical with 5.25" disks. Even with 5.25" disks the optical sensor response time is in tens of nanoseconds and consistent, not microseconds and triggering randomly, so there should never be any type of problem getting the same data start position every time you read a revolution.

Ok, let's dump some tracks to see if practice matches your theory. For each rev, the first four and the last four samples, plus total # of transitions on that rev:

Code: Select all

`Rev 0: 002E 0080 0080 007F ... 0081 007E 0081 007F - Transitions: 42292`

Rev 1: 0080 007F 0080 0080 ... 0080 007F 0080 007F - Transitions: 42293

Rev 2: 0081 007F 0080 0080 ... 007F 0080 0080 0080 - Transitions: 42294

Rev 3: 007F 0080 0080 0080 ... 007E 0080 0080 0080 - Transitions: 42293

Rev 4: 007F 0081 007F 0080 ... 0080 007F 0080 007F - Transitions: 42292

Rev 0: 0011 0088 007D 0081 ... 007F 0081 007E 0088 - Transitions: 38454

Rev 1: 00F1 0087 007E 0081 ... 0081 007E 0088 00F1 - Transitions: 38455

Rev 2: 0088 007E 0080 0080 ... 007F 0081 007E 0088 - Transitions: 38453

Rev 3: 00F1 0088 007E 0080 ... 0080 007F 0087 00F2 - Transitions: 38455

Rev 4: 0087 007E 0081 007F ... 007F 0081 007E 0088 - Transitions: 38453

Rev 0: 004A 009A 009E 00E7 ... 00E7 00ED 00E6 00EA - Transitions: 43175

Rev 1: 009A 009F 00E7 00ED ... 00E8 00ED 00E5 00EB - Transitions: 43174

Rev 2: 009A 009F 00E7 00ED ... 009D 00E7 00EE 00E5 - Transitions: 43173

Rev 3: 00EA 009A 009F 00E7 ... 009D 00E8 00EC 00E6 - Transitions: 43174

Rev 4: 00EB 0099 009F 00E7 ... 00E9 00ED 00E5 00EB - Transitions: 43175

Rev 0: 0099 00EB 009A 009F ... 009E 00E9 00ED 00E5 - Transitions: 42433

Rev 1: 00EB 0099 009F 00E7 ... 0099 009E 00E9 00ED - Transitions: 42431

Rev 2: 00E5 00EA 009A 009F ... 0099 009E 00E9 00ED - Transitions: 42432

Rev 3: 00E6 00EA 0099 00A0 ... 009A 009E 00E9 00ED - Transitions: 42432

Rev 4: 00E6 00EA 0099 009F ... 009A 009E 00E9 00EC - Transitions: 42432

The first transition shows that the jitter sometimes is quite considerable. The last transitions, if you check the range, you can see that sometimes they don't match across revolutions.

This was taken from a handful of images I processed recently. I am posting, obviously, only those tracks that the dump shows jitter. Some tracks were neat, without any apparent jitter, at least not with this simple hex dump. But I didn't have to dump many tracks to find those with jitter. And I didn't code any special analysis routines to search for jitter. Just dumped the first few tracks from a handful of images.

Now let's see track 0 for that image that was making trouble to NewRisingSun:

Code: Select all

`Track: 0/0`

Rev 0: 006D 00A3 00A7 00E4 ... 00E3 00F8 00EB 00F1 - Transitions: 45115

Rev 1: 00A0 00A4 00E8 00F5 ... 00A2 00A5 00E6 00F4 - Transitions: 45112

Rev 2: 00EC 00F2 009F 00A4 ... 00A1 00A2 00E6 00F9 - Transitions: 45112

Rev 3: 00EA 00EF 00A3 00A3 ... 00A3 00E8 00F6 00E9 - Transitions: 45115

Rev 4: 00F1 00A2 00A3 00E8 ... 00A5 00E3 00F6 00E8 - Transitions: 45114

The jitter is obvious, and it is expected that you might have troubles trying to build a circular buffer out of a single rev.