Log of July 2007 events
Back to Home Page
Tuesday July 03,
Our RAMAC signal tracks from Joe Feng, July 23 Signal interface to Control Box from Dick Oswald, July 24, 225 KByte .pdf
Tuesday July 03
Drawings - sent July 01, 2007
Box Lay Out
.pdf - 372 KB
Control Box Schematics
.pdf - 223 KB
Power Supply Schematics
.pdf - 54 KB
. . .
- Present were Dave Bennet, Joe Feng, Al Hoagland, Dick Oswald, Ed Thelen
- Al asked when can we expect some hardware, reference to Dick Oswald's drawings above
- Manual interface
- - an LED will light when head settled, seek complete
- - two decimal thumb switches for disks (1-50?)
- - Three decimal thumb switches for tracks - access 0 to 119
- - If high digit is 3, negative track, track -1 = 319
- - two sets of the above disk and track switches for from/to repeat seeks
- - How to select Home postion?
- - a pot to limit the repetition rate of back and forth seeking above, 10 ms to 2 seconds
- - a 12 position selector switch for something
- - I think I can get into two of the Xilinks programable arrays
- Servo system
- - +- 7.5 volts when a long distance away
- - Classical parabolic trajectory, to null
- - when close to null, then set the detent
- - when detent set, will coherst it to center of track
- - test point for offset, window comparitor, plus/minus 1/4 track
- - null accuracy better than original?
- - I think we are better than 1/4 track
- - adjustment pots?? procdure in CE manual for pot adjustments
- - we should get Pat Connolly to help align pots?
- - no contact with Pat Connolly for over a year??
- What about safety circuits?
- - I have a watch dog that is a pot that is an adjustable timer, will turn off drive
- -What about crash stop detectors? Safety circuits? - Jack knows a lot about that safety wise.
- - What about crash stop detectors? Safety circuits?
- Power supply discussion - see schematics -
- Dave - is there going to be a place set up for (Dick) to work around here?
- - Joe volunteered his table top 19 inch rack :-))
- Considerable discussion of lead free edicts, and its effect on parts for us.
- - most dual-in-line parts will not be available, not manufactured
- - mostly surface mount parts will be available :-(( major pain
- - Dick has a stock of parts he plans to use
- - plated through hole problems with lead free solder
- Joe Feng wondered about driving write currents to switch fast enough
- - Is there a value for head inductance?
- - [SeniorDesignProjectReport-Phase2.pdf page 19-20 says:
- - - bit rate = 80 KHz
- - resistance is 2.5 ohm
- - they used 160 to 200 ma.
- - no comment about inductance.
- - They used 20 volts with a pot to limit current.]
- - Al says the student were required to make precise measurements of inductance.
- - [I can't find it, easier to do it again?
- PC interface -
- - 8 line data bus (Unidirectional) mulitplexed between all of the switches and LED readout
- - Ein is 3 lines, 8 states for the multiplexor
- - Control lines for the single shots
- Action Items
- - Remind Joe to bring in table top 19 inch rack, going to Hawaii next week
- - Al to check through Santa Clara for Pat Connolly
- - how far (resistance?) is home position from end of the pot?
- - What voltage/current does it take to drive the head L*di/dt
- resistance of head? (2.5 ohms?), inductance of head?
- Things we forgot to ask
- - plug compatable with existing RAMAC connections?
- - connections with read/write circuits - - on-track, busy,
- Things to be worked out, Dick and Ed
- - connections? with Ed Thelen's National Instrument
- - -NI USB-6008 multifunction board board http://sine.ni.com/nips/cds/view/p/lang/en/nid/14604
- - http://www.ed-thelen.org/RAMAC/Log2006-11.html#Human%20Interface
Our RAMAC signal tracks - received from Joe Feng, July 23
-- "Albert Hoagland" wrote: > Unless the coating was oriented during spin coating the disks > should be unmagnetized unless recorded upon. I cannot remember > your info showed that all tracks were formatted for 5 records > per track. Were "empty" records magnetized (i.e., erased) in the > beginning or just left unmagnetized? The analyses on the coating > could clarify if there is any doubt. > Al Al, From memory, there were 28 tracks that were not properly formatted (plus the one track that had been overwritten by the SCU students). They did not have sync bursts followed by gaps at the index times. Most (26) of these tracks had recorded transitions, but with low and variable amplitude, about 10-25% compared to nearby formatted tracks. The transitions were constant frequncy at a lower density than allowed in formatted data, and the envelopes were heavily modulated. Since most of these had adjacent tracks that were formatted with much higher and "constant" amplitudes, I interpreted these as unsaturated due to reduced write current rather than due to a circumferential disk defect. Two of the tracks appeared to have no transitions. I am not sure that I could reliably distinguish between a virgin, unrecorded track and a dc-erased track. I am inclined to think they were virgin tracks because I do not recall seeing a transient glitch that could be attributed to turning the write current off or on. However, I also did not make an extra effort to look carefully for such a glitch. The Expo67 organization was probably formed before the first RAMAC was ever built. Joe
-- "Albert Hoagland"
wrote: > My last question is were all the tracks you mentioned on the same > disk or same side of a disk? I ask since the total number of > tracks on the stack is 10,000. Al, They were not confined to a single surface or disk. They were found on three surfaces: there were 7 on the bottom of disk 29, 14 on the bottom of disk 33, and 7 on the bottom of disk 36. Except for two near the ID of disks 29 and 36 and one near the MD of disk 36, they were all on the OD third of the disks. Joe