Schedule February 2009

Wed February 04 - general

• Present were the usual suspects - Ron Williams, Bob Erickson, Frank King, Allen Palmer, Joe Preston, Ed Thelen, Robert Garner. (Robert quickly disappeared into some CHM meeting.)

• Ed helped Joe find a flaky circuit breaker in a CT 729 mod 2 tape drive, the over current kind.
  When set (connected) it had a resistance varying from
  above the ability of the meter to register
  including all points in between down to
  0.4 ohms
  (It took Ron to jiggle the reset button just right to get the resistance down to 0.4 ohms allowing for 1.0 ohms of lead resistance.)

• Frank King - worked on the CT 1403 printer hammers. When he was working on the DE (German) 1403, adjustments worked as expected (after replacing the plastic ribbon that limited the magnetic gap of the print magnet assemblies to about 0.002 inch (the original ribbon had disintegrated). The print hammer adjustments for the Connecituit 1403 would not adjust correctly - so he dissassembled the unit much further than the DE system. And we got to see more :-))

  Starting with basics, this is the electromagnet coil and armature which provides the motion.

  They are mounted so that there are 10 pushing levers per inch. IBM used wide coils so needed four orientations to get 10 to the inch. The orange plastic is "Kapton" - a really tough -

  Each lever strikes (at the proper time) a hammer (a pair shown here)

  The lever pushed the other end, and this end strikes the paper into the ribbon into the print slug (just opposite of a typewriter)

  and an end view

  He found several thousanths of hard crud (maybe compacted paper dust??) on two of the armatures which prevented full movement of the two armatures. Jack knives have many uses, including crud removal ;-))

• Allen Palmer - worked on a DE 729 tape drive

• Joe Preston - worked on a CT 729 tape drive

• Ron Williams and Bob Erickson replaced the bad belt to the CT 1402 card punch dynamic timer, from our big bag of spares :-))) but then found this - I think they said something was wrong. ;-))

  They also logged "Punch clutch does not latch to the poinnt that the keeper can hold it."

Thursday February 05 - Tape Team

from Bob Feretich

TAU Debug Status 2-5-09 (Jeff S., Sam S., Ron W., and me) We spent the first hour discovering something new about a 729 tape drive. Not all 729s have bypass relays. When we turned on the DE 1401 system today both 729s were turned off locally. When we tried to read data from the Emulator, many errors were detected by the TAU. Scoping the Read Data Bus showed that the positive humps of the data waveforms were being clipped at Ground. Since this was similar to an Emulator bug that we shot last week, we thought that somehow the bug had reappeared. However, further debugging showed that 729 #2 was responsible for the clipping. We eventually traced the bug to the fact that the drive was powered off. Turning its power on fixed the bug. 729 #1 does not exhibit that behavior because it has the "dual-port feature" installed and the relays in this feature remove the drive from the bus when it is powered off. We replaced a noisy and weak fan on gate 02b1. We verified the "Rewind & Unload" driver fix. We reran the Tape Print and Tape Copy programs. All source/destination combinations of Tape Copy ran without errors. Even tape copy between the two real 729s. We are officially declaring victory and will be moving on to attempt to get the Tape Autocoder running. Regards, Bob Feretich

Wednesday February 11 - General

• Present were: Ron Williams, Bob Erickson, Bill Flora, Frank King, Allen Palmer, Glenn Lea, Joe Preston, Stan Paddock, Ed Thelen - and the Tape Team: Bob Feretich, Jeff Stutzman, Sam Sjogren. Robert Garner e-mailed Tues evening of a sudden presentation at his work - no bagels today :-((

• San Sjogren and Jeff Stutzman want to port Ron Mak's 1401 software development system to the Apple Mac. Ron Mak's interface is in JAVA, no problem. The AutoCoder assembler from Van Snyder is in FORTRAN 90. Sam found a FORTRAN that runs on the MAC - so hopefully porting will not be a big deal.

  Here is a sample program and output

• A serious specification problem came up in the tape emulation. The maximum record size is 2,000 characters, a PIC memory size limit. No larger is available.

  One Autocoder overlay is about 4,000 characters long. The existing code issues a tape read instruction, and the overlay overlays everything.

  Like what to do now - call Van Snyder for help? He may have another version?

• Lunch time

  Joe Preston went to the 30th RAMAC birthday party some years ago, and received this silver momento

• Ed Thelen spent far too long disassembling and playing with the bad circuit breaker discovered last Wednesday.

  Ron Williams wants Ed to fix SMS cards. Ed claims that if a transistor does not have a hot filament, it just can't work !!

• Joe Preston swapped out a good circuit breaker from the "Organ Donor" 729 tape drive, and continued debugging one of the CE 729 Mod 2's

• Bob Erickson's face says it all - a signal required to punch cards is not being generated, the logic diagrams do not match the machine, circuit tracing got lost in a bundle, and a day was basically wasted. Ron Williams and Bill Flora were not delighted either.

Thursday February 12 - Tape Team

Saturday February 14 - 2nd Saturday

There seems general agreement that folks will come in for the free food at the warehouse on the third Saturday.

Wednesday February 18 - General

• Present were: Ron Williams, Bob Erickson, Bill Flora, Frank King, Glenn Lea, Joe Preston, Judith Haemmerle, Stan Paddock, Ed Thelen, Robert Garner. Allen Palmer is traveling in Italy. Grant Saviers came in about 3:00, due at a ?Volunteer Coordinating Committee? meeting at 4:30

• Today could be called the day of the 729 Tape Drive Doors. You and I are used to rather light weight house doors (especially interior doors) and shower doors. Well - I'm here to tell you that the 729 Tape Drives have Industrial Weight doors. Solid, keep the tapes in and the people out, doors.
  In case you are wondering, I bought a nice little 1 bath 3 bedroom home in Phoenix for what one of those tape drives was selling for in 1963.

  One of the CT drives had a door that was an interferrence fit with the top button assembly. You could give the door a good shove, and it would close - but this did not seem satisfactory to Joe.

  Then Glenn Lea remembered that one of the CT doors had trouble holding the sliding glass window up. There is a spring/cable assembly on each side of the window. One of the cables had de-tached itself from the spring assembly. Again. Time to get a "new" cable from the "Organ Donor" 729.

• Ron Williams, Bob Erickson and Bill Flora are figuring that the 1402 is the reason that: The 1402 punch was not working when the unit went out of service. A belt had been installed before the unit went out of service, and the main shaft was not aligned with the timing unit. They suspect that who ever was servicing the unit in CT did not know how to align the unit correctly The 1402 reader failed about the time the unit went out of service - and was likely the reason the 1401 system was taken out of service.

• Judith Haemmerle worked on cleaning and oiling 1401 ventalation fans that had been removed from service. After re-assembling the second one, she decided to test them. We got the 250 volts from the correct 1401 utility plug. One worked just fine :-)) but the other had no torque at all :-((
  After some fumbling about, and wondering about the brittle insulation, and confusion on circuit tracing (not our best day !!) Grant Saviers came by and straighted us out. There is an open circuit in the fan someplace affecting both phases. :-((

  Resolved: before working on the mechanics, check out the electrics - if the electrics are bad and difficult to repair, don't waste time with the mechanicals. :-))

• Grant explained his planned SMS card tester to Stan, and myself -

Thursday February 19 - Tape Team

from Bob Feretich

1401 TAU Debug Status - Thursday, 2/19/09 (Jeff S., Sam S., and me) Today we loaded and debugged new firmware for the Emulator. The new version has changes to support the transfer of 3000 character records. We also needed to make MAX_XFER_SIZE checking changes to the web application, user mode driver, and kernel driver. The 729 Tape Drive Emulator can now process records up to 3000 characters long!!! We tried to boot the Autocodet.IOCS tape image that I received from Van. The 1401 read the entire tape, over 4000 records. But then the 1401 crashed showing several errors. Symptom: Check light on. B-reg VRC error on with value of 0x00. Instruction state machine indicated the an opcode was being fetched. Instruction Addr Reg = 1F75 C C C 8 4 4 4 2 2 1 1 1 1 We rebooted the tape a few times. This type of error occurred every time. I'm not sure if the I-Addr-Reg was identical each time. Regards, Bob Feretich

Saturday February 21 - 3rd Saturday

Report from Stan Paddock

Ron Williams, Bob Erickson and Stan Paddock decided to come into the CHM to do some useful repair work. Ron and Bob worked on the punch section of the CT IBM 1402. They found a bank of heavy duty cam driven contacts that were not only oily, they looked more like the teeth of a Great White shark. They pulled the bank of contacts out of the machine; And filed each set one by one. Until they were all good for another 20 years. While the punch is still not fully operational, it is further along that it was last Wednesday. While Ron and Bob were doing BIG work on the CT machine, Stan replaced the same fan that Frank replaced last Wednesday. Seems to have lost the grease and started to howl. This task took awhile as we had around 50 visitors and Stan got to do his pitch to all of them. The docents would bring a group in and leave them with me. Lots of fun. It is surprising how many visitors either worked on 1401s, ran 1401s or programmed 1401s. Stan Paddock San Jose, CA

Wednesday February 25 - General

Thursday February 26 - Tape Team Canceled

Due to the lack of actionable tasks to perform and other pressing needs, 
I am canceling this Thursday's 1401 debug session at the CHM.

We have two Autocoder tapes. One has been modified and contains a record 
that is too large to read. We loaded the other tape last Thursday and 
after the entire tape was read, the 1401 check stopped. We have no 
documentation on what the program was trying to do when it crashed. (The 
crash was documented in last weeks status report.) We also know that 
this tape requires a 1407 console, which we don't have.

I have tried to communicate with Paul Pierce, who has IBM 1401 Autocoder 
tape images on his web site. These images are of the1401/1410  IBM 
Autocoder SW product. They will run in 4000 characters of memory and 
should not require a console unit. However, the format of the images is 
undetermined and Paul hasn't responded to my queries.

Since proceeding with our plan to bring-up Autocoder on the 1401 isn't 
actionable, since we don't have other pressing TAU or tape issues, and 
since I have other things that need to be accomplished, I am canceling 
Thursday's session.

Regards,
Bob Feretich

Saturday February 28 - 4th Saturday

Jim Somers is expecting three big groups of students - e-mail from Jim Somers to Bill Worthington -
"I was notified yesterday that we have 100 college students from the Monterey Bay area visiting on Feb 28th from 11 am to 3 pm. Unfortunately the professor did not check with us prior to scheduling the visit. If he had he would have found out that we already have two group tours booked for the day. So I am now trying to figure out what we will do with all of these students.

"The reason I am writing you is to figure out if it would be possible to schedule a second data process tour/1401 demo for the 11 am time frame."