>>> 'tylan10' problem from galewski -- Sat Jan  7 17:07:48 1989 <<<

Vacuum stuff:
-------------
Pump output filter pressure is up a notch.  Clogged?

Also will run the tube alot (1-2 times a day) for the next 2 months or
so.  Maybe a good idea to keep an eye on the oil?

The turbo on masspec is very sick, will not run.  Finally seized?

Electron multiplier on masspec reading low by ~1.5 orders of magnitude.
Can it be regenerated or must it be replaced?

Electrical stuff:
-----------------

Will (as said above) run alot of characterization runs the next 2 
months.  Spending alot of time calibrating and profiling the system.
Would, therefore, appreciate to be in the loop if the system (slave
cpus) have to be worked on for bank 3.

Noticed that display on tube 10 ROP much brighter today than last 
night.  Can some of the problems we are seeing be caused by voltage
fluctuations in the system?

Finally, I had trouble loading a recipe today:

-> First when loading the H-flag on the S-zone was stuck on so zone 
would not fire eventhough temp was below set-point.  Went away when 
restarting the recipe.

-> At that point the AMP board started giving low temp readings (saw 
this from the open cal-inputs reading less than 1373.7) and zones went
into full on (bad failure since if not caught only the overtemp shutoff
would stop it).  Back to normal after wiggling the connector with the
12V on it at the back of the AMP board.  I have seen this problem
before so I can live with it.  Since I have begun the 
characteriziation I do not want to fix it now (will have to reboot
the slave cpu to do it).  Just letting you know how the system 
can be made more reliable.  Can cause the problem by slamming the door
by the AMP boards.  Good idea to check the temperature readings after 
working back there.  

				Thank you,
				  Carl

>>> 'tylan10' problem from galewski -- Sat Jan  7 21:13:09 1989 <<<

Re: Vacuum Stuff
----------------

The pressure on the vacuum exhaust filter only goes up during high
flow.  Did not use to do that so eventhough looks ok with low flow
probably should change the filter.

			 	Thank you,
				  Carl

>>> 'tylan10' problem from galewski -- Sat Jan 21 19:24:05 1989 <<<

Probably time to check the oil.  
From galewski Mon Feb  6 12:57:43 1989
Received: by argon (5.57/1.16)
	id AA27845; Mon, 6 Feb 89 12:57:40 PST
Date: Mon, 6 Feb 89 12:57:40 PST
From: galewski (Carl Galewski)
Message-Id: <8902062057.AA27845@argon>
To: bob, evan, hoagland, parrish
Subject: Working in pump cabinet of bank 3
Cc: voros
Status: RO

-------
How many times do I have to say this?

There is a bad connection on the back connector to the AMP board
for tube 10.  If doing work back in the cabinet there is a good
chance that the board is upset. 

It is easy to check for.  The symptom is an artificially high 
temperature reading and the tube stops firing.  The fix is to
press and wiggle the connector slightly.

This is what happened today: 1) The tube was down to 200 degC when
I got into the lab.  2) Somebody had been working in the cabinet
and left the door open.

I cannot go on working with this tube unless either it is fixed
(NOW!) or you take care when working in there and check for the
symptom.  My patience is getting worn very thin by having to deal
with problems like these.

			Thank you,
			  Carl


>>> 'tylan10' problem from galewski -- Fri Feb 10 20:52:54 1989 <<<


Cracked the T/C well.  By inserting profile T/C too fast.  Got
it plugged up and cooling under vacuum untill tomorrow.  Will
replace T/C well with on of the cantilever rods (same size).
Will use the cracked one as a cantilever rod if not damaged too
much.  Do we have any spares?  Can be a cracked one.

Probably will have to clean the system but will try first not to.
Will let you know.

				-Carl

>>> 'tylan10' problem from galewski -- Sat Feb 11 14:06:36 1989 <<<

T/C well was cracked in half.  Took a spare from the chem room area.
Cleaned it with Don Lyons help and installed it.  Found it to be
3.5 shorter than the ones I use.

Hope it is not a problem taking it.  Tried to call Bob for authorization
but not home.  Please order one on my account to replace the one I used.

				-Carl

>>> 'tylan10' problem from evan -- Wed Feb 22 15:37:47 1989 <<<

EPISTNBY aborted itself again.  I started it again but it appears that
something besides the molex connector.

>>> 'tylan10' problem from evan -- Mon Mar 27 09:36:40 1989 <<<

Carl,
EPI.TRAP found at step 25 at 9:30 Monday morning.
I ack'ed and re-ran.
It looks like tyhere is still a glitch.  Not sure when status was last
checked.

>>> 'tylan10' problem from jlou -- Thu Jun 29 15:50:21 1989 <<<

Hydrogen purifier doesn't work. Need to check the H2 sensor.
It seems that the reactor shows a small leak. One time the base pressure was 65 mtorr. Now the base pressure is between 70 to 75 mtorr.
Could Bob help me checking it.
                                          Thanks
                                                               jlou

>>> 'tylan10' problem from bob -- Fri Jun 30 15:14:11 1989 <<<

Problem: H22 purifier produces no flow.
Cause: OT switch tripped
Solution: opened chassis and reset switch. New gaskets installed and the
          unit is back in service.

>>> 'tylan10' problem from phillip -- Wed Jul 19 07:17:04 1989 <<<

P: Chiller motor is burnt out. It needs replacing.

>>> 'tylan10' problem from phillip -- Wed Jul 19 09:53:38 1989 <<<

P: Chiller will not run.
S: This problem seemingly cleared when work was started
   on it. The chiller has been running for over two hours 
   now. It will be monitored today for performance. As it
   stands, it is up and working.

>>> 'tylan10' problem from phillip -- Wed Jul 19 13:36:52 1989 <<<

P: Motor finally & conclusively failed. Will work on replacing
   it ASAP.

>>> 'tylan10' problem from phillip -- Wed Jul 19 14:37:58 1989 <<<

The motor for the chiller ($93.00) is on will-call @ Grainger's.
The truck is reserved for tomorrow ,7/20, between 8:30 & 10:00am.
Mario will pick up the motor and I should have the chiller back
together by noon.

>>> 'tylan10' problem from galewski -- Wed Jul 19 15:44:59 1989 <<<

The profile thermocouple (the thin one) has been broken in several places
during the disassembly of the rear tubing.  I usually have it pulled out
so that the temperature at the front of the boat can be read quickly.
Good idea to push it in (slowly if tube hot) before doing any work that
can damage it (basically only plumbing work on tube 10).

Can it please be fixed?  Single junction thermocouple with a long
slender 2 hole alumina rod.

			-Carl

>>> 'tylan10' problem from galewski -- Wed Jul 19 15:57:34 1989 <<<

While I am at it.

I would really like to have a logout on the wand for tube 10 that
requires that you enter the deposition time and the flow rate of 
dichlorosilane.  These numbers would then be multiplied together
by a program (or script) and when a set value is reached then
James would be notified e-mail.  This way the oil would be checked 
when a certain volume of dichlorosilane has gone thru the pump.
Safer to have it automatic like this.  It would not be a very 
big task to implement this in the wand according to Mudie.

Another two cents worth.

			-Carl

>>> 'tylan10' problem from phillip -- Thu Jul 20 11:52:44 1989 <<<

P: Chiller will not operate.
S: Replaced chiller pump motor ($100.00).
   The cooling system is again on-line.

>>> 'tylan10' problem from galewski -- Thu Jul 20 12:27:41 1989 <<<

We do not seem to have any 50 micron 10" hydrex filters.  I need one
for the city water backup.  Can we have some in stock?  No other equipment
use them?

				-Carl

>>> 'tylan10' problem from galewski -- Fri Jul 21 12:31:43 1989 <<<

Not a problem but want to mention it here.

The N2 flow meter for the N2 blanket on the chiller does NOT use
40 lpm of N2.  I just checked it and it uses a little less than
1 lpm when adjusted for 1 scfh.  Used the total flow readout
for the lab to measure it.

			-Carl

>>> 'tylan10' problem from parrish -- Thu Aug  3 14:50:38 1989 <<<

oil filters has been changed

>>> 'tylan10' problem from galewski -- Mon Aug  7 13:13:39 1989 <<<

Amp board connection so bad now that system is completely unreliable.
When can the edge connector be changed?  Is Friday 8/11/89 a firm
date?
				-Carl

>>> 'tylan10' problem from galewski -- Mon Aug  7 13:17:41 1989 <<<

Forgot to mention characteristics of problem.

Usually indicate too low a temperature and tube gets too hot.  However
today found temperatures indicated too high and tube cold.

			-Carl

>>> 'tylan10' problem from evan -- Mon Aug 14 17:05:18 1989 <<<

The connector to the amp board was replaced.  A tylan MFC connector was
used with it's better strain relief.  The unused spaced were filled with
empty contacts in order to increase the grip on the circuit board.  The tube
fired up in good working order.

>>> 'tylan10' problem from bob -- Tue Aug 15 16:20:11 1989 <<<

It appears there are still problem with the amp board on tylan10.
The connector has been ruled as it was replaced. A spare amp board
will be returned ASAP to Tystar for repair/calibration and then
tested in tylan10.

>>> 'tylan10' problem from galewski -- Tue Aug 15 16:34:26 1989 <<<

Now amp board does not work properly.  The spike output varies up and
down by 30 deg. destroying any possibility of stabalizing the furnace
at the correct temperature.  Do we have a working replacement board?
If not can we send one out to Tylan to get fixed ASAP?  Have several
important runs that I need to do this week.

>>> 'tylan10' problem from hoagland -- Fri Aug 18 13:08:15 1989 <<<

problem:  User reports AMP board not working properly- efforts
          to secure or repair boards connectors have little effect
          on board sensitivity to vibration.

cause:  To be determined.  Several components on board were found
        unsoldered.

solution:  Replace board with working unit from tube3.  Tube10 is
           up and working.  The new AMP boards tolerance to vibration
           has yet to be determined.

>>> 'tylan10' problem from hoagland -- Fri Aug 18 17:47:34 1989 <<<

Pounding on the pump chassis doors and manipulating connector
cables has verified that the swap AMP board for tube10 is "vibration
proof".  This will hopefully end conclusively the many previous
temperature control problems of this tube.

>>> 'tylan10' problem from galewski -- Fri Aug 18 20:45:42 1989 <<<

Is there a way to adjust the amp board sensitivity somehow with
a pot or something.

Reason:  After trying calibration and only getting to within
3 degrees or so, I used calibrator to check accuracy of amp
inputs for calibration T/C.

Result:	load	reading ok, within +0.2 -0.5
	center	2.5 - 3.5 deg low
	source	1 - 2 deg low

>>> 'tylan10' problem from hoagland -- Mon Aug 21 14:24:54 1989 <<<

problem:  User reports calibration TCs reading 3.0C low
          in center zone and 1.5C low in source zone with respect
          to actual temperatures(using reference calilbrator 
          attached directly to TCs).

cause:  Replacement AMP board recently installed may read
        TCs slightly lower than previous one.

solution:  Correction offsets were programmed into DTC card.
           Displayed readout for both primary and secondary TCs
           was reduced to increase actual temperture when tube
           hits setpoint.  Center zone displayed temperature was
           dropped 3.0C(S3 switches 5,6,7,8 set to ON-OFF-OFF-OFF)
           and source zone displayed temperature was dropped 1.5C
           (S4 switches 1,2,3,4 set to OFF-OFF-ON-OFF).

>>> 'tylan10' problem from evan -- Tue Aug 22 16:49:26 1989 <<<

The new DCS cylinder has been installed.

>>> 'tylan10' problem from galewski -- Tue Aug 22 18:29:54 1989 <<<

I think the offsets maybe should be adjusted in the other direction.
Checked it again since clibrated poorly again.  Measured the following
through the relay:

zone	calibrator	front display	calib input	error
----	----------	-------------	-----------	------
L	835.0		835		835.0-835.2	0

C	850.0		850		844.5-845.5	-5

S	835.0		835		831.5-832.0	-3.5


My question is if the dip switches adjust only the diplayed temp
but maybe also the read temp into the DTC?  The error seems to 
have increased rather than decrease with the offset change.

				-Carl

>>> 'tylan10' problem from hoagland -- Wed Aug 23 11:29:34 1989 <<<

Mike Arkon of Tystar, their resident electronics technician, indicated
all documentation there supports the notion that a reduction in the DTC
board offset should raise the actual tube temperature(with DTC software
version 4.0).  This has been a standing procedure on all temperature
control boards; the hardware which replaced our AMP and DTC cards.  This
offset should apply equally to our setup, however Tystar has no DTC cards
to verify this hypothesis.  To resolve this ASAP, Carl's observation was
noted and a decision made to reverse the earlier offsets and increase
rather than decrease the displayed temperature.  This has been accomplished
and results will be posted after Carl runs another calibration.

>>> 'tylan10' problem from galewski -- Thu Aug 24 07:45:22 1989 <<<

I do not know what to make out of it know.  Tried again to calibrate
at my deposition temp of 835/850/835.  According to Steves prediction
the C and S zones are now low.  The C and S zones calibrating in at
5 and 4 degress low respectively.  The story does come more complicated
however.

Found now the following:

			Thru wiring		Direct to amp board
Zone	Calibrator	CALIB 	delta		Calib	delta
----	----------	-----	-----		-----	-----
L	834.0		843.0	0		834.2-	+0.4
						834.5	

C	846.0		847.5-	+2		849.7-	+4
			848.5			849.5

	Same wire > 	848.2-	+4
 	as L zone >	850.0

S	833.0		832.2	-0.8		835.7-	+3
						836.5


Tentative conclusions from this:

1)	The L zone is not as senstive to its input conditions.
	Reads the same thru wiring, from actual T/C, and direct
	into amp board.  C and S seem to be sensitive to that.

2)	The old wiring from the calibration input terminals by
	exhaust still used by C and S (L's had to be replaced)
	affect offset.

3)	The C and S inputs noisy, i.e. larger variation in readings
	than those seen on the L input.

4)	The offset as measured by the direct input (less ambigious)
	seems to have swung too far reading high.


Best from this to replace the wiring (maybe shielded?) to the inputs,
set the offsets to zero, and try a reworked board from Tylan when
they come in.  As long as the readings are noisy I do not think I
can hope to get better control than within 2 - 3 degrees.  Have to
get going on the experiments anyway.

Note:
-----
THE CHANGE IN TEMPERATURES EXPECTED MAKES FIXING OF THE PROFILE
T/C A HIGH PRIORITY SO THAT i CAN MEASURE THE NEW PROFILE.

			Thanks,
			 Carl

>>> 'tylan10' problem from hoagland -- Fri Aug 25 13:03:43 1989 <<<

TC wiring between the fire wall connector panel and AMP board, which
was previously water damaged, has been replaced with shielded teflon
coated wire per Carl's request.  The DTC card offsets were also reset to
0C.  Tube10 is up and available for calibration.

>>> 'tylan10' problem from galewski -- Tue Aug 29 19:01:30 1989 <<<

If it is not one thing it is another.

The DCS MFC has gone bad.  Have seen this problem before.  It cannot
hold set point with DCS but can with N2.  The flow with DCS never is
steady but varies for example between 25 and 40 when you ask for 32.

Noticed that we have a DCS MFC sitting in the Tylan parts cabinet.  Is
it ok and can I use it?  Would like to swap and send out the broken one
for repair.

				-Carl

>>> 'tylan10' problem from evan -- Wed Aug 30 14:29:38 1989 <<<

Carl is installing the UNIT DCS MFC.  We will be sending out two tylan 280
DCS MFC's for repair. 

>>> 'tylan10' problem from galewski -- Sat Sep  9 17:38:01 1989 <<<

If it is not one thing it is another, but it is always impossible to
depend on anything.

After pump down during recipe the tube aborted and would restart
but not light up the yellow LED's.  It looks like DTC board is dead.  
Its LED's fire on and off in unison, but the heater will not fire.
Tried resetting boards but no effect.  Remeber Hoagland saying 
he knew this symptom meant something on DTC board bad.

Looked in the spare boards and we do not seem to have a spare DTC.  
If we had a complete working set I could have been up again with
only a calibration to do.  As it is now the tube is getting cold 
and I cannot do anything about it.  Happens after a week of work 
getting the tube running right again.  Time wasted.

Why can we not have a spare set of boards!!!!!!!!???????????
Promises, promises, but never any action on this.

				-Carl

>>> 'tylan10' problem from evan -- Mon Sep 11 12:02:04 1989 <<<

problem: tube not firing
cause: DTC board malfunction
solution: replaced DTC board with board from tylan3.  Reset boards and
ran DTCSET AND KMSET.  Loaded MANCAL, which is holding at step 10. (500,
525,500 oC.).  
tylan10 is up and ready for use.

>>> 'tylan10' problem from evan -- Thu Sep 21 16:51:55 1989 <<<

Found tylan10 shutdown.  Attempted to restart but unsure of cooling panel
restart procedure. Chiller is off.

>>> 'tylan10' problem from galewski -- Fri Sep 22 00:18:14 1989 <<<

problem:
chiller is not working.

analysis:
no flow when chiller started.  sounds like motor spinning but
pump not turning.  since system cool turned off city backup since it
makes filter clog leaving it running.

editorial:
cooling system backup worked as intended.  shut off tylan heater,
turned off chiller, and ran city water through flanges at 2gpm to
cool them while tube cooling down.

			-Carl

>>> 'tylan10' problem from bob -- Fri Sep 22 07:41:13 1989 <<<

Carl reports a likely failure of the coupling used between the 
motor and the pump on tylan10's chiller. Phillip will go
over this repair with Evan this AM. 

Carl has tied the flow monitoring of this chiller into his SCR 
firing board so that a failure of the chiller automatically turns 
the heaters off.

A sterilization of the chiller is not needed as the
city water used for bypass cooling is chlorinated.  
s

>>> 'tylan10' problem from evan -- Fri Sep 22 15:59:18 1989 <<<

problem: tube shut off due to compressor shutdown
solution: phill and I removed and inspected pump/motor coupling.  It appeared
ok.  The motor was run alone and it also ran fine.  The coupling was reconnectedand the pump came on normally.  The compressor was put back into automatic
mode. galewski is running the tube up to temp.
The recent fluctuations of the city h2o pressure may have been part of cause
of the shutdown.
tylan10 is up

>>> 'tylan10' problem from evan -- Mon Sep 25 15:57:18 1989 <<<

problem: broken ceramic tubing on profile T/C.
solution: replaced broken tubing with a new 5' thinwall double bore tube.
The rebuilt T/C has been installed into tylan10

>>> 'tylan10' problem from galewski -- Wed Oct  4 22:25:06 1989 <<<

Problem:
Hydrogen MFC 280 suddenly way off set point.  MFS 460 will not let it
run.

Analysis:
Looked by symptoms like flow controller getting bad with H2 but not
with N2.  But worked poorly on calibrator with N2 also.

Fix:
Tried to blow clean sensor, leakchecked to 10e-9, and calibrated.  
Put back into system and seems to control for now.  Will run over 
night and see if still ok.  Less steady in last digit than previously 
making me a little nervous about long term reliability.

Editorial:
We did not have any parts at all for 280's.  The area around the
calibrator was extremely dirty.  Had to wipe area and calibrator 
clean before I dared work on the MFC there.  Wipes completely black 
from the dirt that looks like it is deposited from outside air.  Must 
be a better way to get circulating air into the room than open window.

>>> 'tylan10' problem from bob -- Mon Oct  9 12:22:28 1989 <<<

A mass flow controller has been installed in tylan10. It is working 
however not to factory specs re: stability and linearity. Carl is
looking into sending his Mod.# 280 back to tylan for service.

>>> 'tylan10' problem from galewski -- Thu Oct 12 07:34:12 1989 <<<

Not really a problem but could you check the oil in the Tylans for
any adverse effect of really long runs.  Just finished a 6 hr deposition
and will do several in a series and do not want to find out too late
that the pump is getting damaged.  The next one will start around 3pm.

			Thanks James,
			    Carl

>>> 'tylan10' problem from bob -- Thu Oct 12 16:28:48 1989 <<<

The oil filters on tylan10 were changed and the pump topped off with
oil. Steve Hoagland replaced the hoses on the filtration unit as they
had become very hard. tylan10 is up and ready for use.

>>> 'tylan10' problem from galewski -- Mon Oct 30 20:48:56 1989 <<<

Would like to clean the quartz ware and service the cooling system
since power will be off over the weekend and system will be cool.
Tue. next week to etch the quartz ware is good for me.  Can we make 
that a firm date?

			-Carl

>>> 'tylan10' problem from bob -- Mon Nov  6 07:52:36 1989 <<<

tylan10 is down for maintenance. The quartzware will be cleaned shortly.

>>> 'tylan10' problem from bob -- Wed Nov  8 10:50:51 1989 <<<

Qty. 6, p/n PA 5-10P, 5micron X 10" filters and Qty. 1, p/n
DIMN1002-1 10" mixed bed cartridge have been ordered as
spares for tylan10. They will be needed in 6mos.-1yr for the
recirc. cooling system.

tyaln10 remains down for quartzware cleaning.
 
s

>>> 'tylan10' problem from evan -- Mon Nov 13 16:33:21 1989 <<<

problem reported: faulty N2 MFC
symptom: high tube pressure at normal flow
report: removed n2 MFC and checked it on the calibrator.  It was working
within spec. and was replaced.  The problem must be elsewhere.

>>> 'tylan10' problem from evan -- Wed Nov 15 15:54:19 1989 <<<

I checked the exhaust system and found the manual butterfly valve almost
closed.  With valve wide open and 300 sccm n2 flowing, P=570 mTorr.

Carl, is this the proper way you would check for the throughput? 
Also, I would like to know the normal base pressure test.  Do you ever
turn off the n2 bleed to get the gauges zero'd?  What pressure do you
set the bleed for, etc. etc...

>>> 'tylan10' problem from evan -- Mon Nov 27 16:32:42 1989 <<<

Tylan 460 controller is not auto-zeroing for the MFCs.  This is causing
flow fault alarms.  An attempt to rezero has caused a loss of RAM offsets.
tylan10 is down

>>> 'tylan10' problem from evan -- Tue Nov 28 16:44:45 1989 <<<

problem: DCS flow fault on tylan 460
cause: MFC zero drift
solution: The UNIT MFC was rezeroed with the DCS line shut off.
According to the spec sheet, there should be an access hole in the case
for the zero pot.  Because there was no hole, the case cover had to be removed.
Unfortunatly, the screws are located in the side.  A right angle socket drive
phillips bit was shortened to enable me to remove screws.  I did not replace
these screws at this time.  Either the screws should be removed before install-
ation or an access hole drilled in all future UNIT MFCs if they are mounted in
close quarters.

tylan10 is up (running EPI.TRAP, at step 35)

>>> 'tylan10' problem from jlou -- Thu Nov 30 20:29:13 1989 <<<

I found a serious temperature drift. When I set to 900 degree, It 
goes up to 1000 degree. If I set to 850 degree, it goes up to 910
degree. Also, When I use thermocouple to measure front zone 
temperature, I broke the thermocoule glass tube and created a leak.
I immediately turn off the pump and power. Now I only feed 500 cc
N2 to the tube.
Could Even replace a glass tube for thermocouple?
                                 Thanks
                                                       Jesse
From galewski@janus.Berkeley.EDU Fri Dec  1 01:14:06 1989
Received: by argon (5.57/1.16)
	id AA04973; Fri, 1 Dec 89 01:14:03 PST
Received: by janus.Berkeley.EDU (5.57/1.34)
	id AA11723; Fri, 1 Dec 89 01:14:00 PST
Message-Id: <8912010914.AA11723@janus.Berkeley.EDU>
To: jlou@argon (Jen Chung Lou)
Cc: evan@argon, galewski@argon, hoagland@argon, lab_equip_probs@argon,
        parrish@argon, robin@argon, tom@argon, galewski@janus.Berkeley.EDU
Subject: Re: tylan10 problem 
In-Reply-To: Your message of Thu, 30 Nov 89 20:37:03 -0800.
             <8912010437.AA03479@argon> 
Date: Fri, 01 Dec 89 01:13:54 PST
From: galewski@janus.Berkeley.EDU
Status: R

Forgot to address the serious temperature drift problem.  The
temperatures where not correct because the tube has not been
calibrated at any other set points beside 500/525/500.  It
should be calibrated also at the following points and in the
given order:
		 L / C / S
		-----------
		500/525/500
		1040/1075/1040
		980/1000/970
	  	880/900/880
		835/850/835

Operating at radically different set points than these will
require further calibration.  I have not had time to calibrate 
the system myself as it requires about a week to do. 

			-Carl


>>> 'tylan10' problem from evan -- Fri Dec  1 13:15:11 1989 <<<

Tylan10 thermocouple sheath will be replaced on Monday by evan and carl.
The tube will then be calibrated at all needed temperatures.  The heaters
are off and the tube purging at 5SLPM.

>>> 'tylan10' problem from evan -- Tue Dec  5 11:30:23 1989 <<<

The replacement of the thermocouple sheath is proceeding but has been
slowed by the repair work on the lindberg sink. The boats and baffles
will also be cleaned at this time.  Calibration should start sometime tomorrow.

>>> 'tylan10' problem from evan -- Thu Dec 14 16:37:48 1989 <<<

tylan10 is back together.  the liner, cantelever rods, sheaths, boats and
baffles were cleaned. A new front flange o-ring was installed. A new 
thermocouple well was installed.  Two 25 slot boats were borrowed from
the dummy holders of tylan9 and tylan5 until new 50 slot boats arrive.
Carl will start the calibration.

>>> 'tylan10' problem from evan -- Thu Dec 21 12:20:28 1989 <<<

tylan10 is down during repair of air line

>>> 'tylan10' problem from evan -- Thu Dec 21 14:26:52 1989 <<<

The air line has been repaired.
tylan10 is up

>>> 'tylan10' problem from galewski -- Wed Dec 27 15:52:20 1989 <<<

Found that the N2PUMP connection was broken.  Not supervital but
controls the dilution of H2 during normal flows.  Looked like the
connection had been broken during the pump change for 9 since I
found the connector wedged behind that pump.  Fixed it myself.
Reporting it so that it will hopefully not happen again.

			Thanks,
			 Carl

>>> 'tylan10' problem from galewski -- Wed Dec 27 19:22:04 1989 <<<

Problem:  Tube aborts in the middle of runs because of .GNTLK (gate
	  valve interlock).  Impossible to complete a run.

Cause:  Sensor for .GNTLK is pressure switch on N2 ballast and 
	dillution.  Tried to set it at as small a sensitivity as 
	possible, just barely detecting if the N2 turned off.  Still 
	get error.  Trying now to use a bypass (shorting plug) to see 
	if error related to sensor or elecrtronics.  Do we have any 
	spare pressure switches?

				-Carl

>>> 'tylan10' problem from galewski -- Wed Dec 27 20:16:38 1989 <<<

Problem:  Still get .GNTLK error even with a short.

Cause:  Either connectors or electronics bad.  Will try to change
	connector and see.

>>> 'tylan10' problem from galewski -- Wed Dec 27 22:03:28 1989 <<<

Keeps giving me a GNTLK fault.  Gives fault almost continously
so maybe can find it now.

Changed connector, checked continuity (OK back to MFS 460) and 
still have not found problem.  Will have to check relay next.

				-Carl

>>> 'tylan10' problem from galewski -- Wed Dec 27 22:38:55 1989 <<<

I give up for tonight.  Found that the PS17 supply is putting out
37.8 V instead of 24.  Have a vague memory of having had this
happen a couple of years ago.  Then Steve Hoagland fixed it by
changing the regulator.  Also seem to remember that he said that
the regulators for the PS17 are not possible to get hold off.
Anyway I turned the PS17 off so as not fry anything unecessarily.
On top off this problem the GNTLK fault is not happening con-
tinously anymore so I cannot track it.

			-Carl

>>> 'tylan10' problem from galewski -- Thu Dec 28 17:18:01 1989 <<<

Moved GNTLK to from 24ret to digital GND.  Put a 240K resistor on
PS 17A giving output voltage of 26 V instead of unloaded 38 V.  We
have the output transistors for the PS 17A in stock but as Henry
predicted changing them did not affect the high voltage.  The 24 V
in not regulated and therefore goes high when unloaded.  Put resistor
on it to elliminate the spike that the voltage swing could cause.  
Reconnected the pressure switch.

Will try it tonight and see if tube will run reliably.

			-Carl

>>> 'tylan10' problem from evan -- Thu Jan  4 11:19:38 1990 <<<

tylan10 is down for the day for a change of DCS cylinder.

>>> 'tylan10' problem from evan -- Thu Jan  4 14:25:46 1990 <<<

A new cylinder of DCS has been installed.
tylan10 is up.

>>> 'tylan10' problem from galewski -- Mon Jan 15 16:53:29 1990 <<<

Had a lot of problems with the o-ring of the gate valve rolling out
of the groove and not sealing.

Pulled gate valve twice.  Second time I tried a black viton o-ring
and put a little grease on inside circumference (hoping it will make
o-ring stick to groove better).  Hard to put on without twisting the
o-ring.  Is there a tool?  Special o-ring?  Trick?  Would be nice to
find out.  Maybe from the manufacturer.  

Cycled the gate valve 100 times before installation the second time
without the o-ring rolling out.  Has worked once in the system now.
Maybe ok for a while but the above information would be nice for next
time.

			Thanks,
			 Carl

>>> 'tylan10' problem from bob -- Tue Jan 16 10:43:06 1990 <<<

The new gate valve on tylan10 has been problematic since
its install. Carl has serviced it. If problems continue
a retrun to the vendor will be made and a replacement will
be sought. Similar gate valves in tylan's 9-11 work well.

>>> 'tylan10' problem from evan -- Tue Jan 30 14:51:43 1990 <<<

Oil pressure sensors have been added to the pump and connected to the
bank3 warning light system. If pressure drops to 6 psi, the yellow lamp
will light. If pressure continues to drop to below 4.5 psi, the red lamp
will light, the pump breaker will trip and the vlsi area buzzer will sound.

>>> 'tylan10' problem from galewski -- Thu Feb  1 20:08:13 1990 <<<

Tube is not firing, reading erratic temperatures on TYCOM, L,G,H
labels not making sense.  Have seen this behaviour in the  past
and due to DTC board.  Do we have a fixed one in stock?  If so
can we please put it in?  Will check in on this tomorrow.  But,
if changing the board fixes it the procedure to get it started
is as follows:

1) Load and run DTCSET  
2) Load and run KMSET (use recipe from galewski #2)
3) Load and run MANCAL at 500/525/500

				-Carl
 

>>> 'tylan10' problem from evan -- Fri Feb  2 12:03:27 1990 <<<

Swapped DTC card, then firing board, and then CPU and still have no firing.
Will get together with galewski this afternoon.

>>> 'tylan10' problem from evan -- Fri Feb  2 17:47:18 1990 <<<

spare board replacement did nothing
tube started heating up on its own, no firing indicators.
turned off breaker and pulled firing board out
very big mystery

>>> 'tylan10' problem from evan -- Mon Feb  5 16:29:32 1990 <<<

Replaced firing board and turned on breaker.  Ran EPISTNBY (which had been
loaded from Friday)and tube started firing.  I then tried loading DTCSET and
when I ran it, no firing.  I could not get anything else to fire. (20 tries).
I have returned everything to where it was Friday (firing board out, EPISTNBY
in ready) to see if it fires up tomorrow.  
I also noticed that there was a considerable delay for the T/C's to start
climbing and they continued to climb for 30 min.  I think the phantom rise
on Friday was due to the board firing but me not noticing because I was in
a panic as soon as I noticed that tylan11 had stopped firing after a circuit
card switch. 
It looks like the problem is related to the loading of programs but still
no idea where or how.

>>> 'tylan10' problem from evan -- Wed Feb  7 14:02:01 1990 <<<

Discovered that the amp board power supply was loading.  replaced the amp
board and reset everything.  recipies loaded ok and tube is firing.
ran DTCSET, KMSET, EPISTNBY, and now going to load MANCAL.
tylan10 is up

>>> 'tylan10' problem from galewski -- Tue Feb 13 17:30:42 1990 <<<

The o-ring has jumped out of the groove again on the gatevalve
since it will not shut.  Not critical during the calibration
which I am doing now, but will have to fix in the next couple
of days so it does not delay starting the expts.  Maybe this
time James can help me?  He seems to have a better touch than
me since he has not encountered the same problem with the other
tubes.  Or maybe I am using the wrong o-ring size?

			-Carl

>>> 'tylan10' problem from galewski -- Thu Feb 15 20:14:49 1990 <<<

O-ring that we installed this afternoon jumped out of the groove
after only three cycles when installed in to the system.

Tried again tonight, this time using grease in the groove to 
hopefully stick the o-ring better.  Also used a black viton
o-ring, and cycled the valve many, many times before installing.
Works in the system for the moment.  Will see tomorrow.

When trying to find the popped out o-ring that had slipped down
towards the pump I noticed the inlet screen looking full of brown
powder.  Took the liberty of putting in a clean one I saw on James
cart.  The one I took out is in a glove on the cart.  Hope that
was ok?

			-Carl

>>> 'tylan10' problem from galewski -- Fri Feb 16 19:46:03 1990 <<<

esCooling system shut off heater and went into backup because of
fuse blowing for power to status panel.  Will go into longer
anaylsis when I get back to the office and some food.

			-Carl

>>> 'tylan10' problem from galewski -- Fri Feb 16 20:42:35 1990 <<<

Ok this is what I know.

Fuse blew because outlet Phil wired for me on the back of the
chiller was putting out 197 volts.  It is wired with half of
the 204 that the chiller is using and with the ground as return.
The return and ground tied together with a screw, no starwashers.
I thought that there maybe was some resistance so I put a couple 
of starwahers in.

Also thought that maybe the leakdector had done something, since
Bob and I smelled a strong smell of something burned around it.
BY THE WAY, I have asked before that the leak detector not be 
plugged in where it was since has popped breakers on that circuit
before.  PLEASE plug it in where it is now, that circuit is almost
unused.  (Bob I think that maybe the smell inside the gas cabinet was
probably drawn into it by the exhaust.)  Bob and I thought maybe the 
heatere tape on the DCS was doing something but after checking it,
it seem to work ok.  Just does not get very hot.

Plugged chiller in and got 118V.  I restarted chiller and
then found that valves would not close.  The oultet was now only 
putting out 68V.  Then I disconnected the valves but left the 
status panel plugged in and got 98V.

Got frustrated and worried so I connected the status panel and 
valves to the outlet by the chiller.  Can run the system like 
that but not as protected as if running of the same power as 
the chiller.

Then when I tried to reset the firing board the SCR overtemp
alarm would not reset when turning off the power to the firing
board.  Took board out and checked relay K3 but ok.  Put board
back in and now works again.

Conclusion: I might be in business for the weekend if the gate
	    valve continuous to work and the power to the chiller
	    does not act up.  Problem with circuit seems to be
	    something related to the common (=ground) for the chiller.
	    The chiller does not notice since it is running off
	    the two hots but the power to the status panel and
	    valves is affected.

			-Carl

Ps.  I am leaving the recipe loaded in a wait step and will run 
     it tomorrow if everything looks ok then.  I am not running 
     the tube unenabled.

>>> 'tylan10' problem from galewski -- Sat Feb 17 17:14:07 1990 <<<

Had problems with MFS 460 saying that I had leaks for DCS and H2.
Would not improve with going through a power off autozero.  Was 
able to improve zero on H2 with pot accesible through hole in
cover.  Nice feature.  For DCS, however, there is no hole as
Evan discovered before.  When I took off the cover there is what
looks from the instruction manual to be an autozero board which
should be present if it was a UFC-1500.  Thought I bought a UFC
1400 without the autozero because I did not want it to fight
with the MFS 460 autozero.  This would agree with there not
being a hole for the zero pot.  The pot under the cover says
"do not adjust".  Did you adjust when we had this problem
before Evan?  Well the output reads -84mV and bounces around
+-10 when idle.  The MFS 460 and it seem to agree after a
couple of rezeroings.  I will keep my fingers crossed.

BY THE WAY, I would have been dead in the water if I had not
been able to get at a good multimeter, but fortunatly the
yellow one was in the tech room.  Can that be its resting
place for a while?  Only until I do not have to do this any-
more.

			-Carl

PS. Maybe the Unit DCS MFC was misslabeled?  S/N is C15-7751 TSV.

>>> 'tylan10' problem from galewski -- Sat Feb 17 17:37:58 1990 <<<

Sorry, had leads reversed.  The "zero" output from the DCS MFC is
+84mV.  Not really the zero the autozero feature is to provide.
Or maybe instead of misslabeled, it is a 1400 but with a 1500 case?
All very confusing, but if it will only work for a few runs this
weekend I am happy.

			-Carl

>>> 'tylan10' problem from galewski -- Sat Feb 17 23:01:32 1990 <<<

Still have to problem with the DCS mfc.  Tweaked it down from 84mv
to 4mv while idle.  Noticed the race between the MFS 460 to think
it found its zero before the mfc did.  Maybe better to reset the
460 completely so it takes much longer to zero from scratch.  Well
anyway the mfc read about 8mv by the time the 460 zeroed itself.
Should give more margin since the DCS leak error occurs when the
DCS mfc get to above about 90mv or so (did not catch it in the act).
Will keep on trying.

			-Carl

>>> 'tylan10' problem from galewski -- Tue Feb 20 00:27:07 1990 <<<


After observing the output of the DCS mfc for several runs with the
multimeter I think I know why it kept giving DCS leak alarms during
the hydrogen cooldown after deposition.  After the depostion using
DCS the output stays elevated by about 20mV until the switch to 
nitrogen.  The sensor responds to the pressure surges in the system
since the flow controlling valve is on the other side.  Well anyway
with the "zero" output at 86mv the additional 20mv got over the
100mV limit which means that there should be flow, and the 460
alarmed.  I still do not understand why the mfc has what looks like
a autozero board.  Also do not understand the dcs indication on
the MFS 460 showing 0.0 +- 0.1, when the mfc output is rock steady
at 4mV.  Is the 460 noisy by itself, at least it does not seem to
alarm on its own noise.

				-Carl

>>> 'tylan10' problem from galewski -- Tue Feb 20 00:34:49 1990 <<<

BY THE WAY, I am only reporting all this for information purposes
right now.  Still getting succesfull expts. so want to leave every-
thing alone for the moment.  Do not want to disturb the good karma
which seems to have settled on the system.  I just hope it lasts
a few more days.

				-Carl
From phillip Fri Feb 23 09:40:06 1990
To: /micro/micro/labroot/cmnts/tylan10-p

P: Tylan tube10 ancillary equipment malfunctioning.
S: Tracing circuit revealed the 117Vac feed into a
   common junction box had a grounded neutral lead.
   Repaired ground and had Carl reset the cooling
   system.

(The parts necessary to provide a neutral conductor
 for the tylan10 chiller are now on order.)


>>> 'tylan10' problem from phillip -- Tue Feb 27 14:30:03 1990 <<<

The chiller for this system is DOWN. The tube is on city water
back-up for cooling. This will all be reassembled 2/28/90.

( The problem with the voltage on the neutral phase on the panel
 feeding this system is still there. This will also be worked on
 2/28/90 starting @ 6:15am.)

>>> 'tylan10' problem from evan -- Wed Feb 28 15:33:24 1990 <<<

tylan10 chiller is down during electrical repairs.  Loaded and running
EPIVAC per galewski's instructions.  The purpose is to make sure the tube
is cool during long durations of city h2o backup cooling.

>>> 'tylan10' problem from evan -- Fri Mar 16 11:56:12 1990 <<<

Replaced 50 micron city h2o filter on chiller system.

firing breaker reset after auto shutdown on low h2o pressure alarm

tylan10 is up

>>> 'tylan10' problem from evan -- Fri Apr 13 13:52:55 1990 <<<

Calibration was lost during work on bank3.  KMSET and DTCSET have been run.
MANCAL is running at 500,525,500.  Calibration will continue at the other
temps.
tylan10 is down until calibration is complete.

>>> 'tylan10' problem from evan -- Fri Apr 13 16:57:09 1990 <<<

Calibration complete at 500/525/500.
Calibration started at 1040/1075/1040 but stopped for the weekend.
Will be continued on Monday.

>>> 'tylan10' problem from evan -- Mon Apr 16 14:13:52 1990 <<<

Calibration complete for 500/525/500, 1040/1075/1040, and 980/1000/970.
880/900/880 is running now. Calibration should be complete tomorrow pm.

>>> 'tylan10' problem from evan -- Tue Apr 17 09:17:07 1990 <<<

tylan10 will be down all day Wednesday for bank3 repair.  Calibration
continues today.

>>> 'tylan10' fix from evan -- Wed Apr 18 16:57:00 1990 <<<

work on bank3 complete for now.
calibration will be completed Thursday

>>> 'tylan10' problem from galewski -- Tue Apr 24 08:42:16 1990 <<<

Just logged in a checked the status on Tylan 10.  (Cannot help it.)
Anyway, I noted that the pressure in standby seems a little high.
Is reading about 935 on the Tycom.  Should be about 776 on the Tycom
according to my note book (9/9/89).  Maybe good to check the that the 
pump is ok.  Could also be the butterfly valve not fully open. The
N2 MFC can also be out of wack.  Has been in there for a long time
(several years) without any attention.

		Just trying to help,
			Carl

>>> 'tylan10' problem from nam -- Thu Apr 26 12:12:14 1990 <<<

Alarm went off with red "OT" light on.
I pushed alm ack, but it didn't work.  Pushing alarm cancel made the
alarm silence.  
I don't know who was using the tube, but he(she) has to take care of 
wafers as soon as possible.

>>> 'tylan10' problem from jlou -- Thu Apr 26 16:22:52 1990 <<<

This morning I did a high T dummy run to bake the whole reactor.
I found the temperature drift was induced by a loose contact.
I already fixed it. Now the temperature readout is stable.
Tylan10 is up.
