>>> 'rodi' problem from evan -- Sat Jan 14 10:12:02 1989 <<<

Refilled the sodium metabisulfate tank.  It is emptying at a much faster
rate due to increase of lab use.
Added 1 full jar to a full tank and lowered stroke to 20.  I think this
will get us through the long weekend holiday.

>>> 'rodi' problem from evan -- Tue Jan 31 09:58:01 1989 <<<

The UV sterilizer lamps were replaced.  The hour meter reads 24647.
The quartz tubes were not cleaned at this time.

>>> 'rodi' problem from evan -- Mon Mar 13 17:19:24 1989 <<<

The di loop has been broken at the sink in the tylan chase.  It is
bad for the rodi system to have stub lines with no flow.  This is the
ideal breeding ground for bacterial colonies as it warms to ambient
temperature.  Please reconnect  and open any di lines when finished
servicing or modifying equipment hook-ups.  The quality of our H20 affects
the operations throughout the lab.

>>> 'rodi' problem from rnorman -- Tue Mar 14 08:44:48 1989 <<<

Problem: rodi deck hose at the sink in the tylan chase was not
        connected to the loop.

Solution: The deck hose was reconnected and each leg, supply and return, was
          ran awhile before reconnecting to the main loop. I apologize for
          my forgetfulness in reconnecting this line after repairing the
          lam chiller problem. The deck hose is now operational.

>>> 'rodi' problem from hoagland -- Tue Mar 14 08:40:05 1989 <<<

I was in the process of cleaning some stainless fittings for the
Tylans yesterday when I discovered the DI deck gun missing from
the sink in the tylan service chase.  No message regarding it's
removal or fate was forwarded to me.  Removal of important utilities
like this without consideration of staff who use them is both
thoughtless and unprofessional- who did this?????  Respect for the
needs of others and established procedure is critical for effective
control of contamination in a cleanroom... Someone on our staff 
obviously doesn't care.

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
Return-Path: galewski
Received: by argon (5.57/1.16)
	id AA16552; Wed, 5 Apr 89 13:26:38 PDT
Date: Wed, 5 Apr 89 13:26:38 PDT
From: galewski (Carl Galewski)
Message-Id: <8904052026.AA16552@argon>
To: bob
Subject: sink6 comments 

Tank 6:			17.3 Mohm
1st rinse:		15.9     , dipped low but came up fast
2nd rinse:		16.9     , dipped only to 12 and then up easy

Rinsing definetly looks better.  Lets see if it translates to less
oxide nucleation.  Will compare results today to Friday's run both
at 900 degC bake.
>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

>>> 'rodi' problem from bob -- Wed Apr  5 13:42:53 1989 <<<

The resin beds on the recirc loop and on the makeup were exchanged
4/4/89 as part of the normal 90 day cycle. Reports by users infere
this may not be often enough. The interval for exchange of all
resin tanks will be moved up to 60 days. 

>>> 'rodi' problem from bob -- Mon Apr 17 09:55:48 1989 <<<

Make up rate on RO membranes exceeds normal rate. Conductivity
is 60+ microsiemans, and should be around 5. There is indications that
we have erroded through on the membranes. A new set will be put on
line ASAP.

>>> 'rodi' problem from evan -- Mon Apr 17 17:00:14 1989 <<<

problem: leak-through in membrane in tube #1.
solution: took tube #1 offline. installed tube #2, which had a new
membrane installed some time ago.  Tested system with only tube #2
permeate flowing.  results were 7 microsiemans and 1.5 gpm.  Restored
permeate line from tube #3 and got total (two tubes online) results
of 10 microsiemans and 2 gpm.
   Tube #1 will be rebuilt using new membranes.  When completed, it will
be installed to replace tube #3 which apears to be getting worn out.

rodi is up and flowing

>>> 'rodi' problem from evan -- Tue Apr 18 16:56:10 1989 <<<

Low make-up rate and high use caused the tank level to drop to the
point where the circulation pump shut off.  Tank has since refilled to
the level needed for pump to restart.  tank should completely refill overnight.
Lab users should not use excess DI until new RO membrane is installed
tomorrow. Thank-you.

>>> 'rodi' problem from evan -- Wed Apr 19 15:58:34 1989 <<<

The new membranes were installed into tube #1.  The seal on the defective
membrane has been rebonded and taped.  At the time scheduled for switching
over to the new tube configuration, there was too little DI in the tank to
allow a shutdown and still have a safety factor.  The switch will be done first
thing Thursday morning when the tank is full.

>>> 'rodi' problem from evan -- Thu Apr 20 09:59:37 1989 <<<

The new membranes, as well as the 'newest' used set, are now online.
The make-up rate is 3.25 gpm @ 6 microsiemans conductivity.  The reject
rate is set at 6 gpm.  The new set of membranes are working as expected.
There should be no trouble with low tank conditions.
rodi is up and flowing

>>> 'rodi' problem from evan -- Fri Apr 21 08:49:08 1989 <<<

The RO membrane cylinders have been labeled #1, #2 & #3.
The changes made during this weeks installation are as follows:
1) 2 new 40" membranes were installed in cylinder #1
   This was placed in the leading position.
   Test Data: Make-up = 3 1/4 gpm at 5 microsiemens; reject at 4 gpm.

2) Cylinder #2 contents are unknown.  This was the offline spare.
   It was placed in the trailing position.
   Test Data: Make-up = 1 5/8 gpm at 7 microsiemens; reject at 4 gpm.

3) Cylinder #3 was taken offline. It contained a single 40" membrane which
   was removed.
   The 2 used 40" membranes (removed from cylinder #1; one repaired) will
   be installed and 'pickeled' in cylinder #3.  This will become the spare.
   New adapters will have to be made to convert this cylinder into a double
   40" membrane configuration.


The current RO configuration of #1 leading and #2 following gives make-up
rates of:
          3 3/4 gpm at 6 microsiemens; reject at 4 gpm.
          3 1/2 gpm at 6     "       ; reject at 5 gpm.
          3 1/4 gpm at 6     "       ; reject at 6 gpm. --> CURRENT SETTING



   

>>> 'rodi' problem from evan -- Mon Apr 24 09:10:28 1989 <<<

Pricing for complete H20 quality testing from Balazs:

 -Basic tests: TOC, bacteria, SiO2, particles & residue         $260
 -Anions & Cations by Ion Chromatography: F-, Cl-, NO2-,
          HPO4-2, Br-, NO3-,  SO4-2, Na+, K+, NH4+ & Li+        $150
 -H20 sampling kit: various collection vials                    $ 30
							    _________
						Total           $440

>>> 'rodi' problem from phillip -- Mon Apr 24 14:14:58 1989 <<<

P: Resistivity of water at sink6 is the subject of debate.
   No definitive method of measuring same to end controversy.
S: Connected Thornton probe and meter to d.i. input of sink6.
   The correct probe position shows the water to be 18+ Mohms
   at delivery to this sink.
   (Also, substituted existing Balsbaugh gauge with a known
    good, factory calibrated gauge. Both had the same high
    reading.)

>>> 'rodi' problem from evan -- Thu May  4 15:59:55 1989 <<<

I received and installed the PVC parts for RO tube #3.  This tube now contains
two 40" membranes.  These were formerly located in tube #2 and were removed
when suspected of rupturing (very high make-up flow and high conductivity >60).
The membrane PVC middle connector had broken.  A brine seal was also loose and
was retaped.  The membranes themselves seemed ok and had been in the leading RO
position.

After these repairs, tube #3 was ready for service.  I took tube #2 offline
and placed tube #3 in it's place in the trailing position.  It was flushed
for 1/2 hour and the system conductivity lowered to 4 microsiemens.  I 
restored the connection to the polishing beds and found the make-up flow rate
at 3 1/2 gpm.  This compares to the tube #1&#2 set-up results of 3 1/4 gpm @
5 microsiemens.  

I have left the RO system with tube #1 leading and tube #3 trailing.  I will
consult with bob as to which tube we want to have in the trailing position
and which for the spare (to be pickeled and sealed).

We are in very good shape with the RO system.

>>> 'rodi' problem from phillip -- Tue May  9 13:40:48 1989 <<<

Replaced leaking spigot valve on the black chemical room sink.

>>> 'rodi' problem from phillip -- Wed May 10 14:55:40 1989 <<<

The Balsbaugh 900 resistivity meter in the ro shack has
been replaced. We now have a Thorton 832 digital readout
installed. There is a 4-20ma output card available for
this unit. I will check the price and availability of
it tomorrow.

 The water reads 17.9Mohms resistive as it leaves for the
lab. Nothing has been calibrated yet. .

>>> 'rodi' problem from evan -- Thu May 11 10:57:15 1989 <<<

I will be away Friday and Monday.  Someone will have to check the sodium
meta-bisulfite tank on these days.  It will be filled late Thursday.

>>> 'rodi' problem from bob -- Mon May 15 09:18:59 1989 <<<

Refilled Sodium mete-bisulfite solution 48gallons/250gms. Removed
ROguard filter and lft the housing MT as there isn't a spare. Replaced
2 20" upstream filters with 5u prefilters. The RO output rate is excellent
with greater than 92% rejection. 
RODI is up and healthy.

>>> 'rodi' problem from evan -- Thu May 25 13:05:16 1989 <<<

The final status of the RO membranes.  Tube #1 in leading position. Tube #2
in the trailing position.  Tube #3 OFFLINE, membranes sealed in Sodium Meta-
Bisulfite solution, ready to use.

>>> 'rodi' problem from evan -- Fri Jun 30 11:40:51 1989 <<<

I will not be in on Monday.  Someone will have to check/refill the sodium
metebisulfite tank in the di room on monday.

>>> 'rodi' problem from bob -- Tue Jul 25 11:38:44 1989 <<<

The rodi system was sterilized on Saturday, July 22nd.
Sanilclean from King-Lee Chemicals was used @ 1% for a 
period of 1 hr. All lines to the equipment were sanitized. 
Sterilization was done with the old filters in place and 
with the system in the recirc mode, a departure from previous 
sterilizations.

New resin beds and filters are now in place and the
resisitivity is excellent. Calibration of the monitors 
will be performed shortly to assure accuracy in readings.

The next sterilization will be scheduled during the
Christmas Holiday. The pmstat will now include 30 day
advance reminders for sterilization on Memorial Day and 
the Christmas Holiday.
s

>>> 'rodi' problem from voros -- Tue Jul 25 16:16:46 1989 <<<

Ken Lutz has a glitch problem on one of the power lines in 524.
His and the PFM electrician's investigation lead them to a power panel
in closet 554, which is just on the other side of our rodi system.
Ken does not know if the rodi gets its power from PNL"5LF" and
we could not figure it from the drawing that he has (on my table).
They claim that the power distribution on this panel is very
uneven (6 amps on one phase and 40 amps on another), which may
be the source of the problem. They also note, that there is an
on/off type of noise every 30 sec.
Anyway, they are planning a shut down of this panel for further
investigation after we give them the go-ahead. Pls follow up.
Thank you.
Katalin
From voros Wed Jul 26 08:05:24 1989
To: lutz@noyce
Subject: PNL"5LF"
Cc: bob phil evan $ec/rodi-p voros

Bob and Phil checked if we have anything hoked up to that panel.
It looks like none of our equipment is tied to it; however, to be
on the safe side, pls let me know when the shut down of "5LF"
will occur, so we can keep an eye on the rodi system in 546.
Thank you for your cooperatin.
Katalin


>>> 'rodi' problem from evan -- Fri Sep  8 13:31:01 1989 <<<

The Metronics pump was serviced by installing a new diaphram and check
valve balls and seats.  These came from the spare parts kit #SP-U1.
The stroke adjustment was becomming hard to turn.  It was removed and
the gears and bearing surfaces were lubricated.  The pump is back in
service.
.

>>> 'rodi' problem from evan -- Fri Sep 22 16:14:35 1989 <<<

problem: low flow on sink6 di h2o
cause: the dominique-hunter filter seems to have a higher flow restriction
then the Brunswick BSTC .1 micron.
solution: replaced the dom-hunter with a Brunswick  and the flow rate returned
to a higher volume. This factor is being considered in the decision of what
filters will be ordered in the future.

>>> 'rodi' problem from evan -- Fri Oct  6 16:00:52 1989 <<<

A leak has developed on the inlet fitting of the RO pump.  I attempted to
tighten but the flat of the pvc reducing bushing is hitting the pump case.
This whole fitting will have to be replaced. 

>>> 'rodi' problem from bob -- Mon Oct  9 13:52:43 1989 <<<

The filter DI filter element in CV-2 was replaced due to low flow.

>>> 'rodi' problem from evan -- Mon Oct 16 16:32:25 1989 <<<

problem: leak in RO pump fitting.
cause: cracked PVC reducing bushing
solution: replaced bad bushing with a brass fitting. used pipe sealant compound
instead of teflon tape since this was a 1 1/2 in. fitting.
leak has stopped.

>>> 'rodi' problem from evan -- Mon Oct 23 09:18:38 1989 <<<

Both sets of resin beds have been changed.  This is in keeping with the 
new policy of quarterly exchanges.

>>> 'rodi' problem from evan -- Wed Nov  1 16:14:40 1989 <<<

The DI tank dropped to the pump shutoff level around 15:00.  The system was checked and found in proper operational mode.  A check of the lab revealed no
excess h2o use.   The cause will be looked into further tomorrow. The tank
level sensors will be examined.
From jimmoon Wed Nov  1 15:04:03 1989
Received: by argon (5.57/1.16)
	id AA06149; Wed, 1 Nov 89 15:03:59 PST
Date: Wed, 1 Nov 89 15:03:59 PST
From: jimmoon (James Moon)
Message-Id: <8911012303.AA06149@argon>
To: evan, lab_equip_probs, phillip
Subject: utilities problem
Status: R

We're running very low on DI water.  Just a trickle in sinks6-8.

From bob Thu Nov  2 08:08:01 1989
Received: by argon (5.57/1.16)
	id AA24398; Thu, 2 Nov 89 08:07:52 PST
From: bob (Robert M. Hamilton)
Message-Id: <8911021607.AA24398@argon>
To: rodi
Cc: procestaff
Subject: Water usage
Date: Thu, 02 Nov 89 08:07:50 PST
Status: RO

The lab ran out of DI late yesterday afternoon. Usage of
the lab was light; however, we went through a lot of water!

At 7:00 AM this morning, the storage tank level is 8" below 
the full line and the makeup is on and running, a normal
mode.  Currently the makeup system is set to actuate when 
the tank drops 8 inches.

The system appears to being operating correctly leading to
the conclusion that yesterdays loss of water was through
usage, a conclusion that is difficult to believe with light
lab usage.

The usage by the tylan wet-ox will be checked. There is
continuous flow @ 1/2 gpm through the wet ox on bank1 and
this rate is controlled through an orifice placed at the end
of the system in the lab drain. A check will be made to make
sure this orifice is in place.

Please note any excessive water usage. A prime suspect is
the goosenecks in the old lab as they have no timers on them
and can be left on and unattended.


>>> 'rodi' problem from bob -- Mon Nov  6 08:13:08 1989 <<<

Follwing up on the unexplained higher than noraml use of
water in the Microlab the following check was made:

After the power shutdown, the DI storage tank level was
checked. It was full. This, in spite of the makeup being 
down for 6 hrs. The DI feed to the tylan wet-ox was valved 
off at the start of the power shutdown.  Indications point 
to no loss of water in the DI system during the shutdown.

It appears that our shortage of DI can only be explained by
very high use within the lab.

>>> 'rodi' problem from evan -- Sat Dec  2 14:17:24 1989 <<<

I happened to log-in from home and found alarm messages from the resistivity
meter.  It is normal to find 3 messages after Saturday a.m. due to the back
flush of the sand filter.  A check of DI tank level sensors alerted me that
the flush mode had stuck.  I came down to the lab at 13:30 and found the
RO pump off due to low pressure.  The roof tank was empty, recirc system was
off.  I restarted the RO makeup and will stay around to verify that the system
is operating.  I checked around the lab and found no h20 running.
Because the back flush, tank, recirc and RO makeup systems are very interrelait will take some testing during the week to find the cause of this malfunction.
The DI loop should be operating normally by 16:00 after the tank has time to
refill.  Since no one is using the lab, everything should be fine.

>>> 'rodi' problem from evan -- Sat Dec  2 20:25:37 1989 <<<

problem* rodi system off
cause: ro pump off due to low pressure
solution: reset pump and started refilling tank. 
cause of low pressure might be due to the fact that when system is in back
flush of sand filter mode, the pressure to ro system is limited to the
height of tank minus pressure drop across sand filter minus pressure drop
across pre-menbrane filters. I changed the pre filters as they were showing
a drop of 7-8 psi (indicating time for a change).  I will check out at what
pressure the ro system shuts down as well as what pressure is normal during
back flush.  An interesting set of circumstances well worth looking into.

Once again, the accurex monitor saved a situation that would have been
a big disaster on Monday a.m.  The tank would have been empty and it
would have taken a couple of hours to restore.

>>> 'rodi' problem from evan -- Thu Dec 14 16:42:58 1989 <<<

The sand filter back flush seems to be using too much di h2o from the tank.
I will be contacting continental water since we do not have a manual for
the controller. 

>>> 'rodi' problem from phillip -- Fri Dec 15 06:25:56 1989 <<<

Several reports of insufficient d.i. late last night.
Checking the system today notes the tank 'full' and the
system in normal operating mode. All systems related to
the water are 'up'.

>>> 'rodi' problem from voros -- Fri Dec 15 08:21:46 1989 <<<

Date: Thu, 14 Dec 89 22:06:08 PST
From: denlinge@palladium.berkeley.edu (Darrell Denlinger)
To: lab_equip_probs@palladium, phillip@palladium, saunders@palladium
Subject: misc problem

DI water is grinding to a halt.  There is none in sink8.  There is a 
trickle in Y1. I noticed the pressure was lower than normal at about 
7:00pm. It has gotten much worse since then.

>>> 'rodi' problem from evan -- Fri Dec 15 14:56:05 1989 <<<

the h20 shortage was due to the sand filter back flush cycling on at the
wrong time.  this controller is being repaired.  it has been turned off
for now.
di supply is back to normal

>>> 'rodi' problem from bob -- Wed Dec 27 14:28:57 1989 <<<

The resistivity of the DI has begun to dip. It now 16.74 megohms, down from
the 17megs it is normally at. A pair of 3.61 resin beds have been ordered for 
delivery on Dec 27th and will be installed when they arrive.

>>> 'rodi' problem from bob -- Thu Dec 28 14:41:37 1989 <<<

New resin beds have been installed. 

The automatic backflush on the sand filter was serviced. The
timer and its functions were normal however the piston that
selects normal-backflush-resettle-normal, during the backflush cycle,
was allowing excessive water to flow in the resettle mode. Instead
of .5gpm being used, a full 15gpm continued to flow in this mode.
This depleted our stored volume.

The piston has been replaced, realigned and tested. The rodi
system is back up and running normally. The resistivity has
risen on sensor one substantially after the resin bed change
however sensorstat still shows 16.74 megs. The resistivity
monitor needs to be repaired or serviced to reflect real numbers.

>>> 'rodi' problem from evan -- Wed Jan  3 14:47:15 1990 <<<

Problem: chem pump not stroking fully at the speed of h2o meter pulses.
Cause: wear of piston; the pump runs at near top speed and the piston does
not recover to 'home' position after stroke.
Solution: removed every other magnet (5 of the 10) from the h2o meter pulse
generator.  This causes the pump to stroke at 1/2 the former rate.  The piston
can now easily stroke the full amount of the setting.  I have increased the
setting and should be delivering the proper amount of solution to the feed
h2o.  A new pump will be ordered since a loss of this pump would be very
bad for the RO membranes.

>>> 'rodi' problem from evan -- Wed Jan 17 15:34:06 1990 <<<

The Metronics chemical pump has been replaced.  The new model has a knob
for power setting which lets the user adjust the power applied to the driving
coil.  It is set to a point just past the stalling point.  Since the coil is
using only the power needed rather than full power at all times, the life of
the drive system should be longer. I will be fine tuning the stroke setting 
by close monitoring usage over the next few days.
Stroke is initially set at 40%.  I have also replaced the magnets in the flowmeter so that all ten are in place.

>>> 'rodi' problem from evan -- Thu Mar  1 11:31:55 1990 <<<

The di h2o loop is not running due to power transformer repair.

>>> 'rodi' problem from evan -- Wed Apr  4 13:29:16 1990 <<<

problem: alarms on low DI tank and low resistivity
cause: power interupted during YAG laser installation.
solution: reset breakers and checked rodi system
rodi is up

>>> 'rodi' fix from bob -- Sat Apr  7 21:35:36 1990 <<<

The reverse osmosis de-ionized water system has received it's
semi-annual preventative maintenance.

The system has been sterilized, all lines have been purged
and all the filters in the loop replaced. New resin beds are 
on line.

All of the equipment in the lab dependent on di is now up and
functioning. The current resisitivity is at 14 meg ohms. The
di should be at theoretical maximum, 18.2 megohms by Sunday
morning.

The di and the lab are back up and ready for use.

>>> 'rodi' problem from bob -- Mon Apr  9 14:22:34 1990 <<<

After 36 hours of operation the flow in all the di loops has
been adjusted to 1.75 gpm. The loop is balanced. A flow restrcitor
of 1.5 gpm needs to be added to the loop the custodians water comes
off of. This is not urgent and will be added to Evans techjobs.

>>> 'rodi' fix from phillip -- Fri Apr 20 06:31:43 1990 <<<

P: Many blimp messages stating low water in system.
S: The r.o.d.i. water is at 96 inches (full tank).
   I believe the water was simply used up at that time
   when user mail coincided with blimp mail.
As of this writing, the system is up and running.

>>> 'rodi' problem from phillip -- Fri Apr 20 09:00:35 1990 <<<

The top of the tank has come off. It might have
happened during the recent earthquake. A tank level
of ~72 inches makes it impossible to put it back on
at this time.
(It is resting on the top of the tank and not IN the tank.)

>>> 'rodi' problem from evan -- Mon Apr 23 09:23:00 1990 <<<

The loss of di h2o on thursday evening was probibly due to an unexpected
cycling of the sand filter back flush.  In responce to a problem of the
flush occuring too late in the morning two weeks ago, I checked the timer
and found it was running three hours slow.  On Thursday afternoon, I reset
the clock ahead three hours to the correct day/hour.  The cycle start must
have been tripped during this operation and a back flush cycle started a few
hours later (this is a week long mechanical timer so it's actions are very
slow).  The back flush uses a nearly 500 gal of h2o directly from the tank.
Combined with normal lab usage, it takes several hours to make-up enough
to restart the recirc pump.
Sorry for the inconvienience caused.  This is very unlikly to happen again.

>>> 'rodi' fix from bob -- Mon Apr 23 12:05:09 1990 <<<

The recovery of the rodi system has bee improved. We are now at
46% and will monitor quality and membrane life. This will cut
Microlab consumption of icw.

>>> 'rodi' fix from evan -- Mon Apr 23 15:21:23 1990 <<<

A 1.8 gpm flow restrictor has been installed in the di loop feeding the
custodial sink by the techroom.
