REMCO ENGINEERING ENVIRONMENTAL WATER SYSTEMS AND CONTROLS
REMCO ENGINEERING
ENVIRONMENTAL WATER SYSTEMS AND CONTROLS
RINSEWATER CONTROL REVIEW
FOR PRINTED CIRCUIT MANUFACTURERS AND METAL FINISHERS
Remco Engineering has been in the forefront in water control and
reuse since our beginning in 1987. Our first installation was a
shop rebuild with rinse water control timers on both spray rinses
and counterflow rinses. Since then we have sold hundreds of
timers to plating shops all over the USA. We have complied all
our rinse literature and applications notes in this one package
so that you can see for yourself why, when we say, "We can
reduce your water use by up to 80%," we mean it.
GENERAL INTRODUCTION
We have several methods to reduce the cost of running a wet
process area in a printed circuit board shop and treating its
effluent. These approaches will reduce your operating costs. The
main operating costs for a wet process area are, at present very
dependent upon water usage. We pay for water itself, we pay to
remove our waste from the water we use, and we pay for the sewer
hookup to get rid of our treated wastewater.
Most facilities use a lot of water hoping to reduce both panel
rejects due to contamination, operating costs by preventing
carryover of solution to other process baths, and to insure
compliance by diluting metals in the wastewater stream. No one
ever measures the water quality in a shop and usually the plater
is responsible for maintaining adequate rinsing. The one major
problem with water use is that no one knows what is the right
amount of water to use to rinse a particular panel after a
particular bath.
In this review, we will look at the main water use areas where a
comprehensive attack on excess water use will benefit you the
most. The four best things you can do, to improve panel rinsing
and reduce your operating costs are to:
1. Install general water softening (if you have hard water) and a
DI branch for critical baths,
2. Install counterflow rinsing,
3. Control water flow based on Work In Progress (WIP),
4. Segregate contaminants for removal and recovery.
We feel that each facility should clean up the water coming into
the shop using filtration, Iron oxidation (if necessary), and a
softener (if necessary) and Reverse Osmosis to remove all
hardness followed by a side stream DI branch for critical rinses.
After this the next thing should be to add a centralized or point
source water control system that removes water control from
employees who have other things on their mind besides water
reduction. Water use reduction and multiple use rinsing can save
a tremendous amount of water and every gallon saved is money on
the bottom line.
Multiple use rinsing, also called cascade or counterflow rinsing
saves a tremendous amount of water when used with flow controls.
If you refer to the Remco Engineering APPLICATIONS NOTE #2, you
will see that counterflow rinsing will have a huge impact on the
water usage of your shop. The ratio of water reduction from a
flow through rinse to a triple counterflow with dragout is about
6,000 to 1. If you were going to build a new shop and set it up
properly, you could process 1000 sq. ft of panels per day using
about 3 gpm. Increasing production to 2000 sq. ft. per day would
only increase the flow slightly.
This proposal has an appendix on waste water treatment as we feel
that the waste water treatment is actually incidental to water
control. If you control the water as you should, the waste water
system required is simplified. You cannot attack the effluent
treating effectively until you have your water flow controlled.
With controlled flows, the cost of the waste water system is
greatly reduced and the likelihood of reclaiming rather than
sludging the waste is greatly increased.
PROCESS TANK RINSEWATER CONTROL
The prime area for rinsewater control is the time process rinses
are empty but the water continues to flow. This is most obvious
during breaks or lunch when a plating shop is empty but usually
all the water is kept flowing. Some shops turn off the water
during breaks and lunch but do not turn it off when the work is
slow or there are natural breaks in the flow in the work in
progress (WIP).
Adjusting water use to reflect WIP is the prime method for
reducing flows in shops. Most shops have wide variations in WIP
volume but never adjust water flow rates to compensate. The
result is that even with a 30-50% variation in production, the
water flow usually will vary only 5-10%. Logically, if you get
good rinsing at the high production rates, why do you need the
same volume flow at low production rates? Or conversely, if the
flow rates are not adequate at the high WIP levels, why haven't
you modified your rinsing to compensate?
The three basic method for adjusting water flows are outlined in
the REMCO ENGINEERING APPLICATION NOTE #1, which is attached. We
feel that the best method is to use a flow control device (rate
limiter) with a timer to allow only a fixed volume of water for
each unit of production. This allows absolute control over the
volume of water used and is not maintenance dependent.
Expected results:
Without and changes in tankage or processing parameters the
overall rinse reduction should be 20-80% with a wide variation in
day to day usage due to changes in WIP. The result will be better
rinsing with less water. We can calculate the flow volume
required to guarantee a particular residual solids content in the
final rinse so that we know the quality of the rinse water.
Should your rinsing now be inadequate or you select to improve
the quality of the final rinses over what you are doing now, you
may actually increase water usage slightly unless you change the
tank configuration.
CONVEYORIZED SYSTEMS
SCRUBBERS/DEBURRERS
Conveyorized systems such as scrubbers, deburrers, etchers,
developers and resist strippers, require detailed attention as
the possibility of excess water use is obvious. Scrubber and
deburrers are usually designed to run about 8-10 gpm and
conveyorized etchers, developers and strippers are usually
designed around a 5 gpm per rinse chamber. The most obvious
solution is to just cut down the water flow rate but this can
have undesirable results. We would suggest the following
approaches.
For scrubbers and deburrers, standard recirculation units are
available that can recirculate over 90% of the water, using a 10%
feed and bleed. The addition of a pH adjust on systems if there
is an acid predip before the system will insure very low
dissolved metals in the 10% bleed overflow. Most standard systems
are not a feed and bleed recirculation unit and this results in
poor rinsing of the panels as the dissolved metals build up and
plug the spray nozzles and/or precipitate on the panels.
Another potential problem with the usual recirculation systems is
the change in pH over time, either increasing or decreasing,
resulting in either calcification in the system (as pH increases)
or dissolution of the metal as the pH decreases. The feed and
bleed system solves the first problem as the pH usually never
rises very high with a 10% feed and bleed and pH drop only occurs
if you are feeding after an acidic rinse. The declining pH
problem is usually solved by an on line pH adjust. A third
problem is the disposal of the scrubber/deburrer residue. The
usual Copper fines are not readily taken by most recyclers and
the usual systems of filtration are not efficient for Copper
recovery.
Conveyor systems usually run water down the drain as long as they
are on. If you watch your operator, he or she probably doesn't
shut off the system while fetching another load or while on
break. If there is a break in the workload, the water keeps
running. We suggest that you evaluate your present system,
including the possibility of recirculating your scrubber at a
feed/bleed rate of 10%. Should the preprocess rinse be acidic we
would supply a pH adjust system to maintain optimum pH. A
recovery method for the Copper fines using a modified
electrolytic plating unit to recover the fines as metallic sheet
will is an additional attractive possibility. This system will
allow the Copper now recovered as a fine sludge to be converted
to solid Copper at a nominal cost.
EXPECTED RESULTS:
1. The board contamination due to the buildup of solids in the
system will be reduced due to the feed and bleed system.
2. The problem of controlling pH in the recirculation rinse will
be solved.
3. The sludging of Copper fines will end and the metal will be
recovered as recycled Copper sheet for reclaim.
4. Rinse water usage of the scrubbers will be reduced by 90%.
ETCHERS, DEVELOPERS AND STRIPPERS
The conveyorized developers, etchers, and strippers present a
different type of problem as recirculation is usually not
considered for these systems. Most manufactures of these type of
systems will provide multiple counterflowed rinse conversions for
their earlier models, but the cost is usually high. Older systems
also usually have other problems such as leaking seals and
oversized spray heads. We feel that most of the problems with
conveyorized systems can be easily resolved with feed and bleed
type systems and/or an optical panel detector coupled to a rinse
water control timer and solenoid. In almost all instances, the
next process for a panel processed on a conveyorized system is a
wet process requiring rinsing and there is usually a cleaner step
first. Therefore, the requirement for absolute cleanliness
leaving these systems is usually overkill. The main reason for
the cleanliness requirement is to stop the process once it hits
the rinse (i.e., etching & developing). If the process is
stopped by a chemical reaction (or lack of one) the cleanliness
of the rinse is incidental in most instances. We could then use a
single rinse with feed and bleed, chemical (usually pH) adjust to
stop the reaction, and use the next process to remove any
residual chemistry from the panel. We do not feel that a 5% feed
and bleed recirculation system would adversely impact any of the
systems we presently know of, in fact in some instances, they
could help by adding point source waste water treatment to a
difficult to treat process (i.e., alkaline etchants).
We suggest that you evaluate and cost (for those systems not
flowing less than 1 gpm now) a feed and bleed recirculation
system to be placed on the last rinse. If multiple single flow
rinses are present now, we suggest that you counterflow them. You
should also look at the cost of supplying pH adjust/chemical
additions systems where it makes sense to do so. In addition, you
should evaluate rebuilding the spray system to remove excess
nozzles and downsize the nozzles presently in the system.
Installing water controls using a panel detection system
interfaced to a solenoid would allow much better water control.
We feel that 1-2 gpm is the upper limit that should be allowed
for all such systems.
EXPECTED RESULTS:
1. Water use will be reduced dramatically on all systems due to
recirculation and counterflowing the rinses, while spray head
pressure and volume are maintained.
2. Multichambered systems will seen a much greater reduction than
single chambered systems. (See the APPLICATION NOTE #2,
COUNTERFLOW RINSING).
3. Rebuilding the spray nozzles and reconfiguring the rollers
will increase the effectiveness of the rinses.
4. Adding timing control to rinses will reduce water use when
panels are not in the systems resulting in greater flow
reductions.
PATTERN PLATING PRECLEAN/RINSING
The pattern plating preclean line and post plate rinsing is
usually not an extremely large water user. Most shops use spray
rinses unless the aspect ratio is too large and then an immersion
rinse is used, sometimes followed by a spray rinse. Water control
on spray rinse is usually a knee or foot valve. We like the
concept of an immersion/spray rinse where the rinse tank is a
full immersion tank and a top spray is used to reduce dragout.
The single spray bars reduce total water use. Each panel gets a
thorough final rinse from the overlapping sprays. The immersion
tank flushes high aspect ratio panels and dilutes the process
solution before hitting the spray rinse. This configuration
conserves floor space and is usually a much more efficient rinse.
You should evaluate the preclean electrolytic process line for
areas of water reduction. The most common would be:
1. To reduce tank sizes in the process tanks to save space,
2. Change to immersion-spray rinsing to conserve water and
provide a cleaner panel.
3. Interface water control to central water control unit or point
source timers.
EXPECTED RESULTS:
1. Less floor space required for process.
2. Lower water usage.
3. Easier use of rinses, the operator doesn't have to hold down a
button to activate the rinse.
ELECTROLESS-DESMEAR/BLACK OXIDE
The electroless and black oxide lines are usually the highest
water users in any shop. Most shops have not considered multiple
counterflows and dragouts to reduce contaminate levels when
setting up the original lines. The result is that the water is
allowed to flow freely to hopefully reduce the contaminate
buildup in the rinses and to prevent carryover of cleaners. The
free running rinses usually keep the contamination within
acceptable limits but with increasingly high sewer, water, and
waste water treatments costs.
We feel that this area is the most likely candidate for change in
most shops to reduce the water flow and improve panel yield. The
greatest improvement is usually found by counterflowing all
rinses and by adding dragouts to reduce contaminate carryover.
Dragouts are generally unnecessary if your are planning to
recover metals with an ion exchange system.
The design of proper carriers for panels to lessen dragout and
using a lightweight hoist to increase "hang time" over
the process station will both lessen operating costs and reduce
contamination levels. Interfacing to a central or point source
water control system will take rinsewater control out of the
hands of the operator and back into process control where it
belongs.
Referring to our APPLICATION NOTE #2, COUNTERFLOW RINSING, you
can see that increasing the counterflow ratio greatly reduces the
amount of water necessary to maintain a clean panel. The actual
reduction in water uses is quite variable as many
process/procedural changes can be implemented at the same time
shop hardware changes are made that also augment the water
reduction/process improvement cycle.
We like to look at several approaches to water flow reduction in
these lines including remote water control devices (centralized
control), increasing the number of counterflow rinses, and adding
dragouts where metals can be reclaimed.
The evaluation of the wet process area for excess water usage is
always a fruitful one. Most of the excess water use occurs here.
We suggest that you do the following:
1. Evaluate water flows to see where increased counterflows could
significantly reduce water usage.
2. Evaluate the use of dragouts to reduce the contaminate
carryover levels.
3. Evaluate a timer controlled rinsing system to reduce water
use.
4. Evaluate carrier design to improve draining.
EXPECTED RESULTS:
If you can carry out a comprehensive plan to improve the
"dip and dunk" wet process area, we feel that a very
significant reduction in water use is possible along with a huge
reduction in waste water treatment operating expense. We would
expect the following:
1. A reduction in overall water usage by at least 75%.
2. A reduction in the contaminate levels on panel surfaces.
3. The recovery of metal from carryover into dragouts.
4. A significant reduction in waste water treatment costs.
Appendix A:
WASTEWATER TREATMENT
The treatment of our now reduce water flow has been made much
simpler. Previously the large volume of effluent required a
larger end of the pipe treatment system. If we divided the flows
into individual lines we would still required a major capital
expense for each separate line treated. Now that the flows have
been minimized, lets take a look at the possibilities.
SCRUBBER/DEBURRER SYSTEMS:
When the pH is maintained over 7.0, minimal Copper should be in
solution. The overflow of 0.5 gpm or less when the system is in
operation is passed through a bag filter rated at less than 5
microns to remove any residual Copper fines. The recirculated
rinse feeding the scrubber is pumped through a bag filter which
reclaims the Copper material as powder.
The Copper sludge can be recovered as sheet by using the full
filter bags as anodes and a Copper or stainless steel plate as a
cathode. The bags will self clean as the metal is plated out and
the bags can be reused instead of being thrown away.
CONVEYORIZED ETCHER, DEVELOPERS, AND STRIPPERS
With only 1-2 gpm or less from each of these systems, they are
easily segregated for treatment. For the etcher, using ion
exchange after pH adjust for Copper removal only requires a small
unit.
For the developer and stripper, pH adjustment followed by
filtering and Ion exchange removes any residual Copper.
Photoresist fouls most treatment systems and must be removed
before treatment. Developers generally have little or no metals
in solution and if regulations permit can bypass the treatment
system.
PATTERN PLATE
Using dragouts after the metal bearing process tanks (i.e. after
the micro etch and Copper plate) for metal recovery makes sense
if you are using a precipitation system (if you are using ion
exchange, do not use a dragout). When the metal is in a
concentrated waste stream, it is easier to remove a large portion
of it at a reasonable expense. Plating directly out of the
dragout tank allows the recovery of Copper or other metals as
reclaimable sheet. Any other process requires the Copper to first
be concentrated from a dilute solution and the recovered either
as sludge or be plated out. It is the old "stitch in time,
saves nine" analogy, but in this case it's dollars and not
stitches.
The rinses in the pattern plate line do not contain chelates
(usually, unless they are in the cleaner solution) and can be
combined in a central chelate free Copper containing effluent
stream. Using an ion exchange system with a chelate resin will
allow you to combine all Copper into one stream for easy
treatment and recovery.
ELECTROLESS-DESMEAR/BLACK OXIDE
These lines are the most chemically complicated in the typical
plant. Containing oxidants, reducing agents, chelates,
surfactants, strong acids and bases, they have a bit of
everything (even carcinogens).
Dividing the lines into chelated and non-chelated streams is the
simplest approach until Manganese becomes regulated as well as a
few other of the "interesting" compounds located
therein.
The most comprehensive approach is to recover concentrated metals
as above, using dragouts after the micro etches and electroless
Copper (bailout would go into the dragout to be plated down).
Manganese, if a problem can be reduce to a sludge in place and
filtered out as MnO2.
The separation of chelate vs. non-chelate lines allows us to
point source treat the chelated material with ion exchange and
add the non-chelate stream to the main Copper containing waste
stream.
LEAD PLATING AND STRIPPING
Solder lends its self well to removal by ion exchange. The only
real problem is with the Iron in some solder strippers. Strippers
are hauled or precipitated and the Lead recovered as a solid
compound. Rinses are closed looped through a ion exchange system
so no Lead leaves the plant. Lead is recovered as a salt an sent
to a reclaimer. Plating of Lead solutions is possible although
the anode cost is generally prohibitive unless a large volume of
Lead is recovered.
WASTE MINIMIZATION AND COST REDUCTION
If you presently use Ferrous Sulfate, DTC or Sodium Borohydride
in your treatment process you may benefit by investigating ion
exchange recovery systems as soon as possible. The high cost or
excessive sludge generation of these chemistries generally give a
very rapid pay back on a ion exchange metal recovery system. Even
a small system on just the chelate and sequestered streams would
solve many headaches.
RECYCLING
In some cases, recycling of a large portion of the shops total
water use may be possible. The cost and feasibility is usually
evaluated on a case by case basis. The recycling can be simple or
complex and usually requires much more segregation that point
source water reductions.
WASTE TREATMENT SUMMARY
In summary we have really the following waste streams to treat
with flow estimations for a shop plating 1000-3000 sq. ft. per
day:
1. Non-chelated Copper comprising all of the pure Copper rinses.
Total flow after a good water flow reduction program should be
about 10 gpm maximum.
2. Chelated Copper streams including some cleaners, electroless
Copper rinses and the alkaline etcher. The total flow should be
under 3 gpm which will lend itself to a small point source ion
exchange system or it could be combined with a chelate resin.
3. Lead bearing rinses including the Solder stripper and Solder
plate rinse streams. These are easily point source treated along
as iron is not present as a contaminant. The maximum flows off
this stream should be under 2 gpm. Recirculation of the rinses
with good filtration should be investigated as the EPA Lead
limits approach "0" .
All of these systems are augmented by point source metal recovery
from the dragouts which greatly reduce the operation cost of the
system and are usually cost effective for concentrations of 500
ppm metal or better. Putting a plate-out unit on a 10 ppm Copper
stream to get into compliance is a waste of money, it is much
more cost effective to recover the Copper in a passive Ion
Exchange resin and concentrate it into a plate type
electrowinning unit.
