An AEC/NIM subrack filled with upgrade NIM modules is shown in figure the-rack.eps . These module locations still leave space for a 3/12 units wide inchworm driver module.
Please make sure that:
hdcal.snp'' for head stack
calibration, documented in FS manual ``Narrow Track Head
Calibration.'' If changes in headstack parameter values in
``head.ctl'' seem dramatic, please try to find a reason why the
headstack(s) got misaligned before proceeding with the actual
realignment procedure.
It is a good idea to check that the basic reproduce electronics, head positioner including reproduce level detector for peaking, and the Mark~III decoder work satisfactorily.
"Load a Mark III test tape using STOP button.
"Set up 'peak' and 'parity' commands.
peak=
parity=600,12,ab,off,g1,g2,g3,g4
"Select arbitrary tracks for playback reproduce.
repro=raw,15,17
stack=,nnn,,f nnn== 0.0, -350.0, -110.0, +110.0 depending on tape
st=for,135,off
"See oscilloscope connected to read reproduce for eye pattern.
peak
"Verify eye pattern in scope.
"Get actual stack position.
lvdt
stack
decode=a,syn should produce 'ff7fbfdf', the sync word pattern in hex.
decode
decode=a,time should give 'ydddhhmm sssssccc' (ccc=checksum?)
decode
decode=a,aux see below for interpretation of the aux field
decode
"Change 'repro=' to see other tracks; watch eye pattern in scope.
repro=raw,1,2
repro=raw,3,4
...
"Check parity errors.
parity
If the test tape has been written using a VLBA formatter, its Mark~III emulation has written VLBA track numbers embedded as two hex digits in the auxiliary field. The total length of the auxiliary field is 8 bytes, 16 hex digits. In principle the track number should be found in two first hex digits, 04 meaning Mark~III track 01, 05 meaning 02, and so on. In practice, added parity bits and little-endian/big-endian interpretation of integers may make recognizing the track number a little bit more difficult. Decoding track numbers from Mark~IV formatter-generated aux fields is a little bit more tedious, as the track numbers is located at hex digits 13 and 14 of total of 16 (again, without considering parity bits!). An example diagram to help in removing parity bits from decoder display:
Decoder display shows
in hex: 1 5 2 3 7 f 5
t t t t t t t t p x x x x y y y y p z z z z z z z z p x ...
0 0 0 1 0 1 0 1 0 0 1 0 0 0 1 1 0 1 1 1 1 1 1 1 0 1 0 1
remove p: X X X
0 0 0 1 0 1 0 1 0 1 0 0 0 1 1 0 1 1 1 1 1 1 0 1
in hex: 1 5 4 6 f d
the "real" hex digits after removing the parity bits.
We recommend that you follow the step-by-step instructions in
``Narrow-Track Head Calibration'' manual of FS9 to perform a standard
Mark~III ``schedule=hdcal,#1'' head positioner calibration
and verify that the values of ``/usr2/control/head.ctl'' and
``/usr2/control/head.new'' do not differ by much. It is
important to check the prerequisites of successful calibration:
lvdt=-9,-9'', ``lvdt=9,9'').stack=0,0,u,u'', ``worm=write'', ``worm'',
``worm=read''``worm''); ``fast'' speeds 100--200μm,
``slow'' speeds 10--30μm.peak'' command response field is between 0.5--2.0 volts when
on peaked track and below 0.2 volts when playing unrecorded tape.
You can also use the the ``hdata'' SNAP command to get the
voltage readout while playing back correctly positioned tape. The
voltage is shown in the sixth response field of this command.The default installation procedure described in the Haystack reference drawing ``4712K003'' assumes that there is currently an AEC/NIM subrack just below the tape transport door and and that the only module currently occupying space in this subrack is a 3/12~units wide AEC/NIM module called the ``inchworm driver'' (the device with many LEDs and pushbuttons in the front panel). This leaves enough space for the new upgrade NIM modules and the locations shown in drawing ``4712K003'' (available in the documentation binder) can be used. (This is the third ``Haystack-compliant'' installation order option presented in figure nim-order.eps .)
Unfortunately, many EVN stations have their inchworm drivers enclosed in a wider AEC/NIM module, 4/12 units or more. Also, at most stations the inchworm controller is a large (6/12 units or more) AEC/NIM module instead of the (typically blue) 19" wide and 5~cm high rack-mountable box used in the original Haystack recorders. The two inchworm-related boxes fill the complete subrack.
The location just below the tape transport door is quite ideal for the AEC/NIM subrack that will hold the R/W modules. Therefore:
The ``new location'' for the largish inchworm module(s) is created by installing the new AEC/NIM subrack provided at the rear of the recorder, at any vertical position that can accept it. We recommend that you move both the inchworm controller and the driver modules to the rear. If you do this, you will be able to put the additional write modules (WRMs) you might eventually acquire in the front rack, just side-by-side with the existing WRM~1 and 2.
(You can also leave the inchworm controller with its LEDs and pushbuttons in the front subrack---but be prepared to relocate it if you eventually install additional write modules.)
The new modules are installed in the order described in the drawing ``4712K003'' observing the options presented in figure nim-order.eps .
It is perhaps most appropriate to take care of power supplies first. This involves connecting three cables from the rear panel of PWM to the various existing power supplies in the recorder.
The first of these three cables, cable ``PWMHWU'' connects J7 of PWM to three AMP Faston-style lugs at the Honeywell/Metrum Unregulated Power Supply. Locate this supply at the bottom of the recorder by referring to the documentation that came with the recorder. Locate the row of lugs called ``TB3'' and connect the wires according to the following table:
PWM rear -- 1830mm/72" cable -- Honeywell
-------------------------------------------
J7-1 -- white -- TB3-1 (+20V)
J7-2 -- polarizing plug AMP 200821-1
J7-3 -- red -- TB3-5 (+14V; in practice less, about +12.3V)
J7-4 -- black -- TB3-4 (VHRET)
Without inserting the plug into J7, apply AC power to the recorder and please verify with a voltmeter that approximately +20V is present at socket J7-1 (when measured against rack ground), that approx. +14V is present at J7-3, and that J7-4 is really at ground potential.
Now remove AC power and connect the plug to J7.
The cables ``PWMMKPJ1'' and ``PWMMKPJ2'' have been supplied with an open end at the Mark~III power supply side. The standard connectors used by Haystack (large AMP CPC connectors) are provided unassembled, just in case you happen to have similar connectors.
The cables ``PWMMKPJ1'' and ``PWMMKPJ2'' have been supplied with AMP CPC connectors mounted according to the pinouts shown in Haystack drawings. Please check the voltages resulting in use of these cables, as the lack of Mark~III equipment has prevented Metsähovi from verifying the correct pin order.
The next step is to locate the -5V connector in the local implementation of ``Mark~III Power Supply.'' In the original Haystack design this is the connector J2.
Attach a mating connector to the cable ``PWMMKPJ2'' following the wire colors presented in the form ``Station-specific PWM Power Supply Cables'' which is located in the section ``Station-specific information'' of the documentation binder. Please write down the connector type and the pin-outs you used on this form.
Locate the +/-12V connector in the local implementation of ``Mark~III Power Supply.'' In the original Haystack design this is the connector J1. Again, follow the wire colors presented in the form and write the pin-outs down on the form.
Connect the cables ``PWMMKPJ1'' and ``PWMMKPJ2'' to the Mark~III Power Supply, put AC power on, and please measure the voltages present at plugs J9 (PWMMKPJ1) and J8 (PWMMKPJ2) according to the following table:
J8-1 -- black -- (-5V RETURN)
J8-3 -- red -- (-5V)
J8-4 -- white -- (-5V RETURN)
J8-6 -- green -- (-5V)
J9-3 -- green -- (-12V)
J9-4 -- red -- (+12V)
J9-5 -- black -- (RETURN)
Switch AC power off and connect the cables to PWM J8 and J9.
Now while you have verified that the PWM will receive all the voltages it will need and you have connected all three input cables to PWM, it is time to verify that PWM operates correctly and provides all the voltages it should.
Turn AC power on. Notice that all LEDs on the PWM front panel should be lit when there are no connectors at the front panel. An unlit +/-12V LED indicates a missing or misplaced input pin at J9.
Measure +5V at front panel output jacks: this will be IOM operating supply voltage.
Measure write voltages present at jacks located to the left of each corresponding WRM power connector. Adjust the voltage to a value suitable for your headstack(s). Customary values lie within 9--11~volts. New headstacks require a larger value than already worn ones. Please check and adjust all four Write outputs, so that the values lie in the ``sensible'' range also in the case you forget to adjust the voltages when adding WRM~3 and 4.
Add the I/O board power cable ``PWMIOM'' to the front panel connector J6. Test the functionality of disabling write voltages by connecting +5V (from front panel jack) with a test cord to a pin of the orange AMP connector:
When you have ensured that your PWM works correctly, you can connect the cable ``PWMRDM'' and two instances of cable ``PWMWRM''. Switch off AC power. Connect the output J5 to the RDM, and J1 to WRM~1 and J2 to WRM~2. Restore AC power and verify that all LEDs are still lit. Remove AC power.
Locate the current wire-wrapped Mark~III I/O board at the side panel of the recorder. Unmount all ribbon cable connectors (verify that they are labeled), unscrew its mounting screws leaving the spacer pillars in the recorder, and finally mount the new IOB with the same screws onto the same spacer pillars.
Plug the orange-colored AMP power connector of cable ``PWMIOM'' into its receptable on the board. Do not connect other cables yet, but switch AC power on. All the PWM LEDs should still be lit. Remove AC power.
Now we are going to add signal cables at the front panels of all three modules, one RDM and two WRMs. It is perhaps easiest to plug all cables into the front panel connectors first and after this thread all the other ends through the rack to their final destinations.
Locate the cables ``RDMWRMIOM'', ``WRMIOMMON'', and four cables ``WRMnIOMSIGn'' where ``n'' is replaced by 1 or 2. (These four write module signal cables are identical.) The ``*MON'' cables are grey, ordinary ribbon cable and the ``*SIG'' cables are color-coded twisted pair ribbon cable.
Connect:
From RDM Front Panel, J4 (Monitor) to WRM1 Front Panel, J4 (Monitor) to IOM, J3 (Control Out); grey with three connectors.
From WRM2 Front Panel, J4 (Monitor) to IOM, J4 (Control Out); grey with two connectors.
From WRM1 Front Panel, J6 (Odd Input) to IOM, J10 (Odd Output); color-coded.
From WRM1 Front Panel, J5 (Even Input) to IOM, J8 (Even Output); color-coded.
From WRM2 Front Panel, J6 (Odd Input) to MkIV Formatter (Odd Output); color-coded.
From WRM2 Front Panel, J5 (Even Input) to MkIV Formatter (Even Output); color-coded.
Connect also the two primary write outputs of the Mark~IV formatter to the I/O board connectors that correspond to J8 and J10. Use the cables supplied with the formatter (or possibly the old cables that were used with Mark~III formatter).
You can keep the Mark~III formatter installed for the initial checkout described in section checking . In this case follow the diagram in section withmk3 except that connect the cables from both WRM modules to head stack #2, the read/write stack. That is:
(Mark III (Mark IV (WRITE
Formatter) I/O BOARD) DRIVER#1) (STACK2, R/W)
J11 ------->J7--J8------->J5(EVEN)J7 -------> EVEN VLBA HEADS
ODDS J3<------- MON I/O
WRITE#1 POWER------>J1
(Formatter) (REC I/O) (DRIVER#2) (STACK2, R/W)
J12 ------->J9--J10------->J6(ODD)J8 -------> ODD VLBA HEADS
EVENS J4<------- MON I/O
WRITE#2 POWER------>J1
Finally, connect a T-BNC connector to RDM Monitor A output and add a
cable from it to the Mark IIIA Inchworm Controller ``Odd'' reproduce
signal BNC connector. This provides feedback information of the
playback power level for FS ``locate'' and ``peak''
commands. You may have to use an attenuator here---the
exact amount of attenuation is not known yet (10--35dB?) and it may
need fine-tuning according to the procedure found in FS ``Narrow Track
Calibration'' manual, describing the prerequisites of automatic
calibration.
With this setup you can use the WRM bypass check described in section first-tests .
Handling the head block requires some experience in its assembly. Please refer to Mark~IIIA narrow-track upgrade documentation for head block mechanical drawings.
First you need to remove the electronics boxes that are fitted to the inside of the tape transport door. Power cables to both upper and lower electronics boxes are removed, too. (The power wires of the lower box can be spliced away from the ribbon-type cable which leads to the Honeywell/Metrum Power supply. Its AMP Faston lugs will be replaced by the lugs of the new IOM power cable.)
Please see the Mark~IIIA upgrade head block drawings for information on which screws to loose to make the complete head block slide out of transport door. The old septum divider follows the block and it comes out of the hole in the door.
Remove old head interface cards: unscrew the card retainer and lift the cards gently out of head stack connectors. Retain the cards! It might eventually be possible to upgrade these cards up to the current HIB standards with all 36 tracks---thus they might be reused in the ``full'' four-head upgrade.
The septum divider is fitted with three screws (two from front, one from under (when the assembly is held on table headstack facing upwards)). Unscrew these and remove the old divider. For EVN: New UNC 2-56 screws have been provided to replace existing metric Mark~IIIA septum divider mounting screws. In some cases the existing metric screws fit better than the new UNC 2-56 screws. For JPL: The existing UNC 2-56 screws can be reused. Mount the new septum divider (with the premounted HPB cards).
Insert the new HIB cards into the head stack connectors and observe the order, shown in figure ham-back.eps . Mount the card retainer and affix the HIB grounding lugs to the screws.
Carefully ``thread'' the new septum divider with its cards into the tape transport door hole. Mount the head block with the screws you dismounted in step dismount .
Mount the plate using the spacer pillars and screws provided. The pillars can be screwed directly into the transport door threadings.
Plug the HAM connectors to the receptables on the plate. The cables from the plate are connected to the RDM and WRM rear panel connectors according to the labels attached near IDC connectors.
It is very important that WRM write signal cables (WRM*SIG*) are not grounded via CPM! Early versions of CPM assemblies had their WRM signal cables grounded. This can cause uncontrollable noise and oscillation in HAM read preamplifiers. The write signals are so strong that they do not require a grounded shield in cable. If grounding is absolutely required, the cables may be grounded at WRM module rear panels.
Next Chapter, Previous Chapter
Table of contents of this chapter, General table of contents
Top of the document, Beginning of this Chapter