Question about disk drive motor control

[ebiguy]

I am not a hardware guru and I would like to understand how the motor is controlled in the 810 or the 1050.

I looked at the original 810 rom and found that this program deals with PHASE1 to PHASE4 through the pins 2,3,4,5 of PORT B

Am I right if I say that the purpose of setting these PORT B pins is to move the head over the desired track ?

I am confused because I thought that a single type 1 command sent to the floppy controller (for example Restore, Step In, Step Out,...) would be enough to move the head.

So what is the real use of these pins ?

Thank you for your explanations.

 

[Bryan]

Stepper motors:

 

https://en.wikipedia.org/wiki/Stepper_motor

 

Unlike a typical motor where you put power in and it begins turning, a stepper motor allows for exact positioning by requiring a sequence of signals to move it a number of degrees. You can think of it like the main gear in a pendulum clock, that can only advance one tooth per swing. In the 1050, the PIA provides that pendulum signal.

 

The drive ROM doesn't give the user access to all the stepper phases because Atari didn't consider it to be important to stop the head between tracks (I think the 1050 can do a half-step in diagnostic mode), but the ROM uses them internally to move the motor.

 

[Bryan]

Follow up since I just figured out exactly what you're asking:

 

The floppy controller chip is designed to interface with a drive mechanism with onboard electronics like you'd use in a PC. The 1050 uses a bare mechanism and has the electronics on the board (cheaper, I guess). So it doesn't use all the functions of the controller.

 

The XF551 has a more standard configuration of microcontroller->FD controller->PC mechanism.

[ebiguy]

Thank you for your answers.

Don't wory, the first answer is also interesting.

 

Does it mean that a different use of the PHASEs in the 1050 is the trick to a noiseless drive (as seen -heard- with the Happy upgrade) ?

[Bryan]

The simplest way of controlling a stepper is with square pulses which cause the motor to move in noisy/violent steps. This makes the head carriage rattle on each step which leads to that seek noise. The Happy doesn't fix this, but a little bit of grease on the head rails does. I recommend a plastic compatible grease like SuperLube.

 

http://www.amazon.com/Super-Lube-21030-Synthetic-Grease/dp/B000XBH9HI

 

Hobby shops usually have suitable greases as well.

[ebiguy]

Ok. I have one drive with the Happy 1050 which is almost silent.

I always thought that the firmware was responsible for that !

I didn't guess that it could be a problem of grease...

I will try that.

Thank you

Edited April 24, 2016 by ebiguy

[1050]

Nice find Bryan, that one slipped under my radar completely. Translucent white means silicone grease which will not creep away as oil based greases would and it has PTFE added (read teflon) too. They have entire line of stuff I'd like to get as well with reasonable pricing. I'm liking the dunce hat wearing smiley face guy too, he reminds me of somebody quite close to me.

 

Noisy 1050 head rails are pretty much the deal where you gots what you got - good lube like this can make a night and day difference where nothing else can really be done about it anyway. First I've ever heard that there may be a software/hardware solution - if only that were the case as some will howl quite loudly and others just barely purr loud enough to hear it.

 

So now you've got me actually thinking one may be able to reduce the power supply voltage and current to the stepper to make it a tad bit quieter while still moving the head reliably. It is a possible, but not something I'm keen to jump into the middle of with both feet at the moment.

Edited April 24, 2016 by 1050

[Bryan]

The problem with reducing the voltage is that the drive's firmware expects the head to settle within a specific time frame. If the stepper is weaker, it might not always make it.

 

The real way to silence it is to ramp the stepper voltages so it glides more than it cogs (see 1:52 in the video). This could be done with a circuit between the PIA and the stepper connector. Personally, it doesn't bother me much.