fluxengine
fluxengine copied to clipboard
davidgiven
Misbehaving with GreaseWeazle
Greaseweazle 0.31, using an F4
I've two 5.25" drives connected, a 40track as drive 1, and 80 (1.2mb, jumpered for 300rpm) as drive 0.
fluxengine analyse driveresponse --cylinder 0 --min-interval-us=0 --max-interval-us=30 --interval-step-us=.1 --write-img=driveresponse.png --usb.greaseweazle=COM3 -d drive:1
This is a bit hit and miss, sometimes I just get the drive clicking whilst "Measuring rotational speed", sometimes I get this: "Measuring rotational speed... Interval 0.00: failed Interval 0.08: failed Interval 0.17: failed Interval 0.25: failed Interval 0.33: failed Interval 0.50: failed Interval 0.58: failed Interval 0.67: failed Interval 0.75: failed Interval 0.83: failed Interval 1.00: failed Interval 1.08: failed Interval 1.17: failed Interval 1.25: failed Interval 1.33: failed Interval 1.50: failed Interval 1.58: failed Interval 1.67: failed Interval 1.75: failed Interval 1.83: failed Interval 2.00: Error: failed to send command: No error"
Once I got: " Interval 1.75: failed Interval 1.83: failed Interval 2.00: failed Interval 2.08: 8.7368 Interval 2.17: 2.1666 Interval 2.25: 2.2499 Interval 2.33: 2.3333 Interval 2.50: 2.4999 Interval 2.58: 2.5833 Interval 2.67: 2.6667 Interval 2.75: 2.7499 Interval 2.83: 2.8333 Interval 3.00: 3.0001 Interval 3.08: 3.0835 Interval 3.17: 3.1666 Interval 3.25: 3.2498 Interval 3.33: 3.3332 Interval 3.50: Error: GreaseWeazle error: Underflow"
If I try and use drive 0, the activity light goes on but the motor doesn't.
I have an F7 rather than an F4, but it shouldn't make a difference? The last error looks like it's running out of bandwidth, so do make sure it's plugged into a USB 2 port rather than a USB 1 port.
FWIW I do see from the last set of errors that 2.08us pulse intervals are not being reported correctly by the drive (it's getting back 8.7us pulses instead). That probably indicates the minimum pulse width supported by the drive.
No, it shouldn't - actually, mine is an F730 now I look closely :D It's actually plugged into a USB3 port, so there shouldn't be any issues there. Any ideas what's going on with drive 0?
I'm happy to debug etc, etc.
On Fri, 3 Dec 2021 at 13:08, David Given @.***> wrote:
I have an F7 rather than an F4, but it shouldn't make a difference? The last error looks like it's running out of bandwidth, so do make sure it's plugged into a USB 2 port rather than a USB 1 port.
FWIW I do see from the last set of errors that 2.08us pulse intervals are not being reported correctly by the drive (it's getting back 8.7us pulses instead). That probably indicates the minimum pulse width supported by the drive.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/davidgiven/fluxengine/issues/360#issuecomment-985505171, or unsubscribe https://github.com/notifications/unsubscribe-auth/AATU2PAOK7MSUROAVUURBV3UPC6NFANCNFSM5JJGJCRA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
FWIW I can reproduce the driveresponse issue for pulses shorter than 2us on a greaseweazle v4. It is independent of drive used.
Just to be clear, the drives work fine when using GreaseWeazle's own tools?
Also, actually FluxEngine runs out of USB bandwidth writing pulses closer than about 1.8us. (The shorter the pulse intervals, the more data is required.) So there's a hard-coded limit of 2.0us below which it just refuses to do anything. The GreaseWeazle is a USB 3 device and works intermittently down to about 1us, but my drive starts to fail at 1.2us.
I don't have any Victor 9000 disks but I've got an image here which I think has been read in a 300 rpm drive, and the pulse interval of the outer tracks is about 1.45us. I can write that with the GreaseWeazle but it fails miserably on the FluxEngine.
Yes, no problem with the drives at all with the greaseweazle software - drives 0 and 1 happily co-exist. I've just tried a few different drives, I can't get below about 2.2us to read reliably with any of my 80 track 5.25s, which isn't going to be much cop for the victor discs I fear. That's TEACs and Mitsubishis and NECs... I've got a couple more to try tomorrow.
On Fri, 3 Dec 2021 at 23:32, David Given @.***> wrote:
Just to be clear, the drives work fine when using GreaseWeazle's own tools?
Also, actually FluxEngine runs out of USB bandwidth writing pulses closer than about 1.8us. (The shorter the pulse intervals, the more data is required.) So there's a hard-coded limit of 2.0us below which it just refuses to do anything. The GreaseWeazle is a USB 3 device and works intermittently down to about 1us, but my drive starts to fail at 1.2us.
I don't have any Victor 9000 disks but I've got an image here which I think has been read in a 300 rpm drive, and the pulse interval of the outer tracks is about 1.45us. I can write that with the GreaseWeazle but it fails miserably on the FluxEngine.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/davidgiven/fluxengine/issues/360#issuecomment-985909755, or unsubscribe https://github.com/notifications/unsubscribe-auth/AATU2PHQX4TGHSA3OAS4KFTUPFHPDANCNFSM5JJGJCRA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
0.31 currently
On Fri, 3 Dec 2021, 23:42 David Given, @.***> wrote:
What GreaseWeazle firmware do you have, BTW?
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/davidgiven/fluxengine/issues/360#issuecomment-985912996, or unsubscribe https://github.com/notifications/unsubscribe-auth/AATU2PHNSNNR4SVX45VOVT3UPFIWPANCNFSM5JJGJCRA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
The Greaseweazle V4 is still only USB 1.1 and I can't get more than about 8Msps of bandwidth out of it, which could be the problem.
(in my case I'm always writing pulses 2us or longer so I've had no problems introduced by this)
You might also want to check the state of pin 2 during writing, it should be high for HD disks.
FWIW, here's what I get on my 1.2MB drive via a GreaseWeazle. You can see the smear in the bottom right from spurious pulses as the interval gets too big, and this really weird repeated pattern on the left when the pulses are too short. The good range seems to be about 1.1us to about 7us.
If it's helpful, I've produced a bunch of frequency responses for all of my floppy drives on hand. These were all produced with the Greaseweazle V4, so none of them go below 2.00us.
https://docs.google.com/spreadsheets/d/1EG5jYsqBiQIu0XConvSuIooGkA_WonFZVIxuY0oHloE/edit#gid=0