CA2125678A1 - Ablative coating removal method and system using pulsed light and optical feedback - Google Patents

Abstract

Pulsed light sources (14), such as a flashlamp or laser, remove coatings (24, 26) from substrates (28) via the ablation method. A
photodetector circuit, sensing reflected light (27) from the surface being ablated, provides a feedback signal (194) that indicates the color of the surface being ablated. The boundary between the coatings or substrate surfaces is distinguished by a change in color between an upper coating (24) and an undercoating (26), e.g., between a topcoat of paint and a primer coat of paint, or between a coating (26) and the substrate surface itself. The color determination thus provides a measure relative to when one coating has been removed and another coating remains. The photodetector circuit also provides feedback information relative to the quality of a stripped work surface for quality control of other purposes.

Description

WO g3/12905 PCI~/U~i92/10878 ..~

.. . .

- ~BI~TIVE CO;P.q~ING REMOVAI. MET~OD AND 8'Y~:TE~
U ING PUL8ED LIG~Iq? AND OPTIC~L FE:15DBAClR ~ :

The present invention relates to a material ~ -removal process and system, and more particularly, to a material removal process and system that uses pulsed light from a flashlamp, or equivalent pulsed high energy light source (such as a laser), to ablate the materia~ to be removed; and also uses optical feedback from the surface being ablated to determine when the proper amount - ;
of material has been removed. ;~ ~
. .... ~ ., . ,. , - :
Back~round of the Invention Material coatings play an important role in - ~;
our manufactured products based society. Coatings pro~ide im~unity to corrosion, the Dal insulation, shielding, as well as appearance enhancement, and an aid `;
in identification.
During the li~e o~ many manufactured products, ;
such as bridges, aircra~t, automobiles, and ships, painted coatings require removal and replacement for a variety of reasons. For example~ refurbishment of the paint on aircraft is a regular maintenance item.
Commercial airlines repaint their aircraft about every 4~
5~years of service. The~United states military typically repaints its aircraft after three years of service, or lëss. ~Coatings on the ex*erior surfaces of,largelships or~bridge reguire periodic re~urbishment in order to ~ ;~
prevent or inhibit corrosion.
The removal of~paint~from the surfaces of aircraft presents special problems. Such ~urf~ces are large, irregularly shaped, and relatively delicate.
Because the surfaces of aircraft ~re typically lightweight aluminum or organically based composite ~ materials, such surfaces and the underlying substrates . ~ ,: ::.

. ~ - .

WOg3/12~S PCT/US92/10878 2~ 78 are particularly susceptible to damage while undergoing paint removal that could degrade their structural integrity.
Many different methods have been used to remove painted coatings. One type, the "particle medium blast"
(PMB) method involves impinging the surface to be stripped with particles such as BB's, plastic media, steel shot, wheat starch, and/or sand. However, PMB
methods that are energetic enough by themselves to remove hardened coatings such as paint may damage delicate surfaces such as are found on aircraft and automobiles if they are not carefully managed. For example, if the impinging particles dwell too long at one location, the impinged surface may become pitted or stress hardened.
This is especially important with regard to the surfaces of aircrafl: since pitting or stress hardening may change the loading on that por~ion of ~he aircraft. PMB may also damage putty joints often found on aircraft between surface plates.
It is also known in the art to apply chemical compounds to painted surfaces in order to chemically breakdown the layers of paint, thereby stripping the paint away from the surface to be exposed. However, such - ;
compounds may pose a risk to human health, are usually toxic, and often not biodegradable. Overall, these types of compounds are difficult and costly to dispose of because they present serious environmental problems. ;
` Mechanical paint removal techniques are also known in the art. For example, U~S. Patent No.
4,836,858, entitled "Ultrasonic Assisted Paint Removal Method" discloses a hand held tool which uses an ultrasonic reciprocating edge placed in contact with the surface to be stripped. Unfortunately, employment of this tool is labor intensive and relies upon the skiIl of a human operator to use it effectively. Further, control .

WO93/12905 PCTtUS92/10878 ; ~
2 ~ 2 ~ $ r~ ~ ~

of this ~ool is a problem when applied to aircraft ~ ~
because the aircraft surface may be damaged if there is ~ ;
excessive tool dwell at one location.
Radiant energy paint removal techniques are 5 likewise known in the art. One such system uses a laser `~ ~ ' and video frame grabber in a video controlled paint `~ ;
removal system in which paint is stripped from a surface ;;
using the output of the laser to ablate the paint while a ~ ;
video camera converts images of the surface being ~j -stripped into electronic data signals. The data signals are used to control the laser output. A processor compares the data signals with parameters stored in a memory to det¢rmine whether sufficient paint has been removed from the surface being stripped. If an insu~ficient amount of paint has been removed, then the surface continues being irradiated by the laser. If the irradiated area has been adequately stripped, the processor directs the laser to ablate another area.
While the basic approach of ablating and "looking" to see i~ the proper amount of paint has been removed is sound, doi~g the looXing using a video camera is extremely data ~
intensive, requiring an enormous amount of data to be ~-generated, gathered and analyzed in real time. Thus, real time control of video controlled paint removal ~y~tems is extremely difficult. What is needed, therefore, is a system and method wherein an ablation -~
removal process can ~e easily controlIed in! real time ~ -without extensive data handling and processing requirements.
~ ~he difficulty associated with removing paint or other coati~gs is compounded when the basic substrate material over which the coating has been placed is non~
metallic. For example, the use of composite structures is becoming increasingly more common. Such structures ~-35- are typically manufactured, for example, of fiber ~, ~ .,, ~ ,.. _ .. ~ . , ~ ~ J.t ~ r t J ~S~J ' S~J`)~

4- - .
:. , rein~orced ~poxy or otheir thermoset or ther~oplastic compo~ite~. Many aircra~t and automobile~ exten6ively ~mploy pla6tlc compo~lte6 ~or sur~ac~ structure6. such structure6 ars painted for a variety o~ rea~ons including ae~th~tics, ident~ficatiGn, and camou~lage.
However, ~uch painted ~ur~ac~ doteriorate und~r the aatlon o~ woather and the mechanlcal foroei6 to which they are 6u~ectedl thu6 r~quirlng remoYal and replacemen~. Di advantageoualy, oth~r than h~nd sandin~, there are no ~uitable m~hods ~or removing pa~nt ~rom the ~iur~ace~ of ~ch compo~it~s. PMB and mechanica~ grinding methods ~hat are 6u~ficiently ~ .
energetic by the~selve~ to remove the p~int ~ra AlSO
~u~.fic~ently energetla to damage th~ composlte mat~rials. Sl~ilarly, the u~e o~ chemic~l co~pounds to remove the palnt i~ not a sati~ctory ~olution because such chemical~ tend to attac~ the compo~ite~, a~ WQll as ~he pAi~t. ~ence, there ~ a sritical need in the art ~or a ~ans ~or safely and e~flclently removin~ palnt or 20 other aoa~ing mat~rials ~rom compo~i~e ~u~trate materlal~ wit~out aompro~i6ing tho in~gri~y o~ th~ .
un~erlylng ~ubstratQ co~po~ aterial.
In U~S. Pat~nt No. 4,58a,885, there i~ ~
dl~clo~ed ~ methcd o~ ~nd apparatu~ f~r the re~oval of palnt and the like Prom a ~ubstrat~ that u~ili2es two source~ o~ r3diant energy. A p~l~ed la~r ~ u~ed ~ a fir~t ~ource o~ r~diant energy a~d generates a hi~h inten~ity~ea~ of radiant energy th~t vaporize~ a ~ ~urface portion, e.g.., paint, to be re~oved from th~
subs~rate. A quartz halcgen la~p i~ uo~d a~ a s~cond sourc~ of radiant energy. ~he quartz hal~gen la~p ha~ a ~iven spectra~l range, and ~ s used to examinc the ~ubs~rato i~ betwe3n la~er pul~Rs. Th~ r~ ct~d light f rom the halogen lamp is conv~rted to an elec~ronic signal. Such elect~onlc ignal i6 compared ~o a previou~y stored electronic ~gnal. When the~e is a s~ $9~ T

' . . ,:
' ' .
, ~...~,,"':
~ ~ 2 ~ ~ ~ 8 : .,:.-:' '., :' substantial mism~tch betwe~ he pre~ent ~lectronic 61~nal and the storsd electronic Bignal ~ th~n imping<~mQnt from the hi~h ~ntenslty laser cQase6~
For ;~ va~iety of rea80n~ , ~nown palnt remo~al ~: :
S te~hnlque~ for rQmoving paint or other coatings from large ~ur~ac~s, have not proven wholly 6atis~actory. It can thus be appreciated that ~oatlng romoval, and . -: .
par~icularly, the ~emov~l o~ palnt ~ro~a large ~nd often , . .
dellcate surfaceJ, ~uch a~ are ~o~md on a1rcr~t and :
10 automobiles, ~ a problem 'cha~ has not yet been ;.
~atis ~a¢torily 601~red .

~Y c~the Invention -The present inv~ntion advant~ge<; usly provide~
. :: :
a syst:em and meth~d for remov~ng paint, or slmilar -15 caatings, that address the ab~ve and other n~ds.
. . .. ~
.: ~ .::
,. . ~.':''' ~.

'~ '~'''~ ' ' : ' . ~., .
.
.~

:, '. : .

~ :'''` ' ~;

`:;~ '~ .: ' ~7~ , S3~EEl ~ ~;

~ W093/1290~ PCT~US92/10878 ~2~7$
," ~, In accordance with one aspect of the present invention, pulsed light sources are used to remove ',! coatings from su~strates via the ablation method. For purposes of the present application, ablation is defined as the rapid decompositlon and vaporization of a material resulting from the absorption of energy by the material ,~ and is associated with the generation of pressure waves radiating from the surface of the material. The amount of material removed by the ablative process of the invention is controlled using a photodetector system that measures the;color intensity of the light reflected from the substrate at the particular location where the material is being removed. Advantageously, because most ~-paint and other coatings can be differentiated by the lS color between, e.g., a topcoat(s) and a primer undercoat, or between an uppercoat and the substrate on which the coating is placed, the photodetector system is able to readily ascertain in real time when the topcoat(s) has been removed and only the primer undercoat remains, or when all the upper coats have been removed and only the substrate surface remains. Immediately upon making such a determination, i.e., immediately upon determining that only a primer undercoat remains, or that the substrate ~ ;
surface has just been exposed, appropriate control ~signals are generated which result in irradiation of ¦ ~ another area of the substrate by the pulsed light source.
.~ .. ...
In accordance with another as~ect of the invention,~a~pulsed light source in combination with a photodetector system is scanned across a work surface in a controlled manner so as to systematically remove all coatings on the work surfa~e down to a prescribed color.
Such prescribed color may ad~antageously be that of a ~-prescribed undercoat, e.g., the primer coat, or that of ~-~
the substrate. In some applications, the feedback 35 provided by the photodetector system may also be used to -~

:
~' ~

W093/12905 2 ~ 2 5 6 ~ ~ PCT/US92/10~78 ~- ~

- 6 - :

indicate the character of the stripped work surface ~or quality control purposes. In addition, the photodetector system may sense position information on the work -`
surface, which positio~ information may be used by a robotic controller to control the scanning operation.
The coating removal system of the invention includes an ablation removal device, uch as a flashlamp or a laser, and a photodetecting circuit, housed within a single scanning head~ A flashlamp, or flashtube, is a :
gas filled device which converts electrical energy to optical energy by passing current through a plasma ;~
typically contained in a transparent tube through which ::
the optical energy is transmitted. Appropriate control :
circuits and positioning devices, coupled to the scanning 15 head, pos;ition the scanning head at a desired location `~ .
ab~ve a work surface having coatings thereon that are to -~
be removlad, and generate the requisite control signals needed to operate the ablation remo~al device and the photodetQcting circuit. Advantageou~ly, the output ao signal(s) from the photodetecting circuit is ~are) used as a feedback signal(s) to provide optical feedback to the circuits that control and position the ablation ` `
removal device, thereby enabling the scanning head to be `~
., - . -.
effectively ~canned across the work surface at a rate and ~ ~h 25 with an inc~dent intensity of the ~urface that - -efficiently and safely removes a desired coating from the work surface without damaging the work sur~ace.
The photodetecting circuit of the invention lncludes at least one array of photodiodes. The ~ 30 photodiodes of each array are optically filtered so a to : d~tect a prescribed wavelength, or band of wavelengths, representing a desired color, e.g., one o~ the primary colors, or other selected color. The photodiodes of each -~
array are positioned so as to monitor re~lected light ..
35 energy from the work surface. Typically, the photodiode ;

.' :'' ,'~ .':

,. ~.

WO93/l2905 PCT/US92/10878 2~2~

arrays are gated ON only at a time that allows them to sense or collect light energy associated with a trailing edge portion of the main optical pulse generated by thé
ablative device. Alternatively~ a seconda~y or auxiliary light source within the scanning head may be pulsed ON at an appropriate time within the ablative removal cycle to adequately illuminate the work surface with sufficient viewing light to provide a source of reflectad light suitable for detection by the photodiodes in the array.
The electrical signals generated by the photodiodes in response to sensing the reflected light of each array are converted to digital sensor data and processed by digital circuitry in a sensor controller.
Parallel monitoring of the optical output from the 15 ablative removal device provides a basis for normalizing -th~ optical information from the photodiode arxays. The normaliz~sd sensor data is then temporarily stored and comp~red with permanently stored reference data representative of the desired color of the work surface once the prescribed coating has been removed therefrom.
The results of this comparison form a basis for a feedback signal directed to a remote computerized controller. That is, since different coating layers as well as su~strates are ~haracterized by different colors 25 and reflected opticaI intensities, the feedback signal ;
af~ords a method for real time control of the coating removal process and allows selective removal of successive coating layers. Hence, the feedback signal may be acted upon by the remote computerized controller in order to control the ablation removal process in a desired manner. Further, as a function of the particular pattern of the photodiodes used in the photodetector arrays, tbe output signals from the one or more of the photodetector arrays allow optical information from the entire width of the irradiated surface on the structure ~ .

WO93/12905 PCT/~S92/~0878 - ~-~ :.
2 ~ 2 ~ ~ ~$

s to be processed. ThP area irradiated by the light source when the ~can speed is zero is referred to as the "footprint." Optical information reflected from the "footprint" advantageously provides increased spatial ---~
; 5 resolution and sensitivity of the system to coating ;~ I¦ anomalies, e.g., repair patches on an aircraft skin that ;~
may have been hidden by subsequent over coats of paint.
Noreover, such spatial capability provides a means for asslsting a robotic controller, or equivalent positioning ~! 10 device controlled by the remote computerized controller, ' in maintaining the scanning head in a prescribed orientation, e.g., a level position, above the work surface.
In accordance with yet another aspect of the invention, appropriate cooling means for limiting the temperature of the ~tructure in the vicinity of the ~irradiated area is also included within the scanning ~; head. ~Such cooling means typically includes a nozzle for directing a particle stream, e.g., a jet of CO2 pellets, at the area from which material has been ablated, as well as an appropriate vacuum system for remo~ing all expended ~
particles, gases, and vapors associated with the ablative ;
removal process. The particle stream also advantageously ~ -cleans the ablated surface.
~ 25 ~ The present invention may be characterized, in 3 accordance with one embodiment thereof, as a method for l~ removing material from a structur~. Such method includes - -l ~ the steps~of: (1) irradiating a target area of a structure~havinq~at least one layer of material formed on ~ -1~ 30 a ~ubstrate with radiant energy having an intensity !~ sufficient to ablate`the layer of material; '~
~ (2)~monitoring reflected radiant energy from tha target ~ :
¦ area to sense the presence of a prescribed color ~-different from the color of the layer of material being ablated; and (3) controlling the irradiation of the ` '.

.
'' ~

WO 93/12905 PCI'/US92/10878 2 ~ fj r~ ~

_ 9 _ target area with the radiant energy in step (1) as a function of the color sensed in step (2).
In accordance with another embodiment, the present invention may be characterized as a system for removing a layer of material from a structure. Such system includes: (a~ irradiating means for irradiating-a target area of a structure having at least one layer of material formed on a substrate with radiant energy having an intensity su~ficient to ablate the layer of material: ; i 10 (b) monitoring means for monitoring reflected radiant ~;
energy from the target area for the presence of a ~;
prescribed color different from the color of the layer of material being ablated by the irradiating means; and (c) feedback means for controlling the irradia~ion of the target area with the radiant energy from the irradiating means as a function of the color sensed by the monitoring ~-means. 'rhe prescribed color, in turn, is detected -througA the use of photodetection means. Such photodetection means detects reflected radiant energy ;
having a prescribed wavelength, where the prescribed wavelength is oharacteristic of the prescribed color. In ~ ~
a preferred embodiment, such photodetection means ;
lncludes: (a) means for dividing the reflected radiant energy into a plurality of optical channels; (b) fir~t detection means for detecting if the reflected radiant energy in each of the plurality of channels contains a respective wavelength; and (c) processor means for analyzing the respective wavelengths detected in each ~ ~ ;
optlcal channel to ascertain whether the prescribed color ~ ;-30 is present in the reflected radiant energy. -~
Still a further embodiment of the invention may be characterized as photodetector apparatus useful for ~ -examining the surface of a structure. Such photodetector apparatus includes: (a) illuminating means for illuminating a target area of the surface of the :::
~ : .
, 7 8 ``
- 10 - .;,, structure with pulsed radiant energy; and (b) monitoring means for monitoring reflected ra~iant energy from the target area for the presence of a prescribed color.
It is thus a feature of the invention to 5 provide a coating removal system and method wherein ;~
coatings may be selectively removed using a photodetector ? feedback system in conjunction with an ablation removal process.
It is another feature of the invention to ~ 10 provide such a coating removal system and method wherein ~
; a photodetector system ascertains the color of the work ` ~;
~, surface from which the coatings are being removed, and uses ~uch color determination as an indicator of whether ~ -~
the coating has been sufficiently removed.
It is an additional feature of the invention to provide~such a coating remoYal system and method that reduces~the risk of damage to frangible substrates such as composites.
It is a further feature of the invention to 20 provide such a coating removal system and method that ; ~ ~
includes in a single scanning head: ~1) radiant energy - ` ~`
` ablative removal means, such as a flashlamp, for removing material coatings off of a work surface of a structure;
(2)~photodetector means for optically detecting when a ~ -25 desired~coating has been stripped from the work surface;
and~3)~cooIing and cleaning means for limiting the temperature of the structure and for removing any residue of~the ab~lated material from the stripped work surface~
~ ~ ~Advantageously, such scanning head may be scanned across 3 30 the work surface as a function of feedback signals sensed ~y the photodetector~means.
It~is another feature~of the invention to provide`a photodetector system that generates an output signal(s) that indicates the presence or status of 35 substrate surfaces or coating layers of a structure.

. ..
~ .

WO93~12905 PCT/US92/10878 ~2~ 6 ~ ~

Such photDdetector output signal(s) provide an indication of the color of ~he work surface, which indication may `
advantageously be used for varied purposes. When the -photodetector system is used as part of a coating remo~al system, for example, such output signal(s) may be used as a feedback signal(s) for one or more of the following purposes: (1) to control the coating removal process, i.e., to limit the exposure of the stripped surfaces or layers, thereby preventing damage to the work surface or coating layers; (2) to position and orient, e.g., level, the ablative removal system above a desired location on the work surface relative to topological landmarks on the ;
work surface; (3) to enable the safe and efficient operation of the ablative removal device, as by, e.g., turning on the ablative removal device only when certain conditions are satisfied, and/or by controlling the output power of the radiated energy generated by the ablative removal device (4) to provide real time feedback to a remote controller that controls the scan rate of the ablative removal device; or (5) to monitor the formation o~ frost and/or condensation on the work surface from the application of a particle stream, which particle stream may be used to cool and clean the structure. When the photodetector system is not used directly as part a coating removal system, or in addition to being used as part of a coating removal system, the output signal(s) from the~photodetector syst~m may al$o be~used, for example, to monitor a previously stripped section of the work surface for the~purpose o~ quality control or surface anomaly detection.

Brief Description of the Drawinas ~ The above and other aspects, features, and ~
advantages of the invention will be more apparent from~ -35 the following more particular description thereof, ~ ~

, , ' .

WO93/12~ PCT/USg2/10878 2 ~ 2 ~ ~ 7 ~
^ - 12 ~
: , presented in conjunction with the following drawings, wherein: -Fig. 1 is a block diagram that diagrammatically illustrates the main components of a coa~ing removal system made in accordance with the present invention;
Fig. 2 is a schematic diagram of the scanning head used with the coating removal system shown in Fig.

Fig. 3 is a block diagram of the photodetector circuit;
Fig. 4 is a waveform timing diagram that illustrates one timing arrangement that may be used for operating the photodetec~or circuit in accordance with the present invention; ;~
Fig. 5 is a waveform timing diagram that illustrates an alternative timing arrangement for operating the photodetector circuit; ;`~
Fig.~6 is a diagrammatic representation of one ~-~
embodiment of the photodetector arrays;
Fig. 7 shows a diagrammatic representation of ;
an alternative embodiment of a photodetector array of the ~-present invention; and ~-~ Fig. 8 is a flow chart that depicts the basic method used by the system shown of Flg. 1 to romove coatings from a substrate.
Like reference numerals are used to represent ~like elements in the various figures~and~the accgmpanying ~-description that follows.

Deta~iled DescriDti~n of the Invention ~ The following description is of the best mode presently contemplated for carrying out the invention.
This description is not to be taken in a limiting sense, but is made merely for the purpose of describing the general principles of the invention. The scope of the ,:' .

W093/12905 PCT~US92/10878 2 ~ 2 ~ 6 l ~
.~ - . . ..
. 13 invention should be determined with reference to the '! claims. ~
-~ It is noted that the present invention combines :~:
;, an ablative removal technique with a particular optical 1 5 detection technique in order to remove one or more ;, coatings ~rom a substrate surface. Ablative removal ~ techniques using radiant energy, e.g., a flashlamp, ;
:1 combined with a different type of detection system, as well as optical detection techniques using a plurality of ! 10 photodiode arrays, combined with a non-ablative coating ,1 . ' il removal mechanism, are the subject of other patent ~ :
~ applications filed by applicants.
Referring first to Fig. 1, there is shown a block diagram that diagrammatically illustrates the main 15 components of a coating removal system 11 made in accordance with the present invention. Advantageously, the system 11 removes coatings 24 and/or 26 from a ! substrate 28 without damaging the substrate. (Note, the coated substrate 28 may hereafter be referred to as the ,! 20 ~work surface~ or ~structure~ 22.) Further, the system 11 includes a digital control processor 200 that coordlnates and controls the scan rate of optical energy ~ 18 and particle stream 30 across the surface of substrate 1l 22. Control is effected using feedback pro~ided by an 25 optical detecting circuit lOO tha~ detects the optical ~, character of the surface of the work surface 22.
. Referring to FIG. 1, data processor 200 (which ~ ~may be an IBM AT or AT compatible personal computer, or g ~eguivalent) generates output signal 5 to enable particle 3 30 stream source 6, output signal 7 to enable vacuum system ~¦ 37, output control siqnal 12 to control light control circuit 13 (which may be of a type well known by those :~
~killed in the art), and output signal 202 to pro~ide path and speed instructions to robotic controller 204.
! 35 Particle stream source 6, in turn, is coupled to nozzle `` ' ~",:.',.

WO93/12~5 PC~/US92/10878 2 L ~ 5 ~ r~

. .
32, which nozzle is adapted to direct a stream 30 of particle~, explained more fully below, across the surface ~
of the workpiece 22. Similarly, vacuum system 37 is ~ :
coupled to exhaust nozzle 36, which exhaust nozzle is positioned to receive the residue 45 of any materials that are ablated by radiant energy 18 generated by ablative light. source 14 and/or the spent particle .
stream. Light control circuit 13 generates a control signal 15 which establishes the repetition rate and pulse -~
10 width of the output of ablative light source 14. In some .:~
embodiments of the invention, light control circuit 13 ~ . .
also generates another control signal 17 which ~urns on auxiliary light 29 for a desired time period during the :::
coating :removal cycle, as explained more fully below in connection with Fig. 5.
Nozzle 32, ablative light source 14, auxiliary ` ~ ::
light source 29 (when usedj, and exhaust nozzle 36 are all housed within a scannin~ head assembly lO that is .. ..
adapted to move above the work surface 22 as controlled 20 by robotic po~itioner l9, as indicated by the arrow 21. `~
Advantageously, electrical, optical, and other coupling to the elements within the scanning head assembly 10 is achieved through appropriate flexible cabling 31, thereby .
facilitating movement of the scanning head assembly, including the elements housed therein, while allowing the ~: control cir uits for such elements, such as the particle stream source 6, the light control circuit 13, and the vacuum system 37, to be stationary at a position remote from the ~canning head assembly 10.~ :
In order~to provide a feedback signal to the :-system 11 that:allows it to control the coating removal process, a photodetecting circuit 100 detects thè optical condltion at the work surface 22 by monitoring radiant energy 27 reflected from the work surface 22. The photodetecting circuit 100 receives the optical signals WOg3/129~ PCT/US92/1~X78 - ~
~2~ ~J~.'3 ::
- 15 - - ~
~ . -27 and generates electrical feedback signal(s) 194 ~ ~ :
therefrom that are conveyed to the control processor 200.
The control processor 200 processes ~he feedback signals 194 and converts them into a composite output signal 202.
Robotic controller 204 transforms signal 202 into control or instructional signals 206 that direct the path and speed of robotic posi~ioner 19. Such instruction signal 206 directs robotic positioner l9 to move the scanning :.
head assembly across the work surface 22 so as to effectively scan ablative energy source 14 and particle stream 30 across the surface of the structure 22 in -accordanc:e with a prescribed pattern. The path of robotic controller 204 is de~ermined in accordance with a suitable path generating processing routine implemented by data processor 200 in accordance with techniques well known by those skilled in the art.
As seen in FIG. l, the photodetector circuit 10~ is preferrably located within or attached to the scanning assembly lO, with the output signal 194 of the 20 photodetector circuit 100 being coupled to the remotely .:
positioned control processor 200 through appropriate :~
~lexible electrical cable.
Referring next to Fig. 2, th~re i~ shown a schematic diagram of the scanning head assembly 10 used with the coating removal system 11. As seen in Fig. 9, the scanning head assembly lO, comprising optical energy source 14 and reflector 16, i^q supported by robotic positioner 13 at a predetermined standoff distance "d"
from the surface of structure 22. The optimum standoff distance "d" for ablative removal of coatings is a : function of the amount of output power contained in the radiant energy 18 output by the ablative energy source -14. In general, the closer the source 14 is positioned :~
to the work surface 22, the more power there is to ablate .~
35 the upper coatings 2~ and/or 26 coverin~ the substrate ~ ~-: :--' .

WO93/12gO5 PCT~US92/10878 ~
i , :
2 ~ 7 ~ - ~

~8. However, care must be exercised to prevent too much ; ablative power from being delivered, else more than the desired coating(s) may be ablated. While the ablative power may be controlled by adjusting the repetition frequency and pulse width of the light 18 generated by the light source 14, the intensity of optical energy 18 incident on the surface of structure 22 is preferably controlled by simply controlling the standoff distance "d". Initially, an approximate distance "d" for nominal output power levels of the light source 14 is determined experimentally. For example, where the ablative energy source is a flashlamp, as described in applicants' copending applications referenced above, and where such flashlam~p provides an incident intensity at the surface o~ the structure of about 1-10 joules/cm2 and has a pulse width th,at may range from about 1000-2400 microseconds (~sec) and a repetition rate of 4-5 Hz, and further where a coating of paint having a nominal thickness of 4-8 mils overlays an aluminum substrate, the initial standoff distance "d" is on the order of 1 to 3 cm.
Robotic positioner 19 is controlled to move the assembly 10 along a predetermined path at a controlled scan fipeed over the surfa~e of structure 22 so that ablative energy source 14 and particle stream 30 may be directed to scan and ~mpinge, rsspectiv~ly, the coating or coatings formed on the surface of substrate 28. The radiant energy (light) 18 from the source 14 ablates the ~ ~
coating to be removed in the immediate area of exposure ~ -to the radiant energy 18. The particle stream 30 limits the temperature rise of structure 22 as a result of absorbing optical energy in the form of heat provided by light 18. Robotic positioner 19 may be implemented as a CIMROC 4000 Robot Controller manufactured by CIMCORP
Precision Systems, Inc., Shoreview, MN. The scan speed is functionally related to the output signal 194 by a WO93~12905 PCT/US92/10878 17 - 2~

. . ~ .
function bounded by upper and lower limits~ as described more fully in the referenced patent application. Such function may be increasing or decreasing, depending on the particular application. Material remsved from the surface of substrate 28 and the expended particle stream 30 after it impinges stxucture 22 are collected by vacuum system 37 through nozzle 25 mounted to housing 12.
Particle stream 30 i5 provided by particle stream source 6 which may provide gas, liquid, or solid particles, or any combination of particles. For example, particle stream source 6 may be a gas tank if particle stream 30 is a gas, or a carbon dioxide pellet source of the type commercially available from Cold Jet, Inc., of Loveland, OH. The particles which comprise particle stream 30 are delivered to nozzle 32 via duct 34.
Note that as depicted in Figs. 1 and 2, the ~;
system 11 is configured to remove an upper layer 24 from the subst:rate 2~ while leaving a lower layer 26, e.g., a primer paint coat O Such removal is only exemplary, as the ~ystem 11 could just as easily be configured to remove both layers 24 and 26, leaving the surface of the substrate 28 exposed. The mechanism by which the system 11 determines when the proper layer has been ablatively ~ ;
removed i8 to monitor light 27 reflected from the surface 22 for~a~specific color. such monitoring assumes, of course,~that a distinguishing color difference exists ~etween the layer 26 to be removed and the layer 24 to remain,~or between the layers 24 and 26 to ~e removed and the substrate surface. This assumption wilI almost 30 always be true. The reflected light 27 is directed to -the photodetector dircuit 100. The reflected light that is ~onitored may be either the trailing edge of the ~ ~;
ablative light pulse 18, as explained more fully below in ~ ;~
connection with Fig. 4, or light obtained from an ;
:, ~

', .- ;- ''', -' ~.

2 1 ~ $

auxiliary light source 29, as explained more fully below in connection with Fig. 5.
Regardless of the source of the reflected light 27, the reflected li~ht is monitored by the photodetector S circuit 100 for the presence of a specific wavelength, or -~
~ a band of wavelengths, characteristic of the color of the i layer or surface that is to remain. Immediately upon detection of such characteristic wavelengths, the control processor 200 is notified via the signal 194 so that it knows that sufficient material has been removed at the present location o the incident radiant energy 18.
~ Thus, the control processor 200 immediately generates the ! requisit~e control signals so as to move the scanning head l~ 10 to a new location, adjust the scanning speed, and/or adjust the output power of the light source 14, in order to assure that no further material is removed at the location where the characteristic wavelength was ~ -detected~
In the preferred embodiment, as seen in Fig. 2, the light source 14 is placed within a water cooled housing 12. Water, or other suitable coolant, enters and exits the housing 12 through parts 44 and 46.
~i Advantageously, the particle stream 30, directed at the ~ i -1 surface 22 at an approximate angle ~, helps to keep the l 25 lens~cover 20 of the housing 12 clean from debris and other foreign matter that might otherwise accumulate ~ thereon. The angle ~ will typically range from 5 to 60 !: degrees, but is not felt to be critical for ablation as j ~descrlbed herein The reflected light 27 detected by the ~ photoconductor circuit 100 should be received from ¦ immadiately behind the same area impinged by the incident ablative light source 18. In some embodiments, in order to monitor the status or condition of the surface 22 over a wide "footprint", it is desirable that the reflected ,i i ~i :

- 19 ~

footprint area, i.e., tha~ area from which the re~lected light 27 is received, actually be somewhat larger and behind the irradiated footprint. The irradiated footprint may be referred to as the "target area" because it is the area at which incident light 18 is directed.
For such wide area monitoring to provide useful information, it is necessary that the photodetector circuit 100 have spatial distribution resolution capabilities so that it can detect not only the presence of a characteristic wavelength, but also a particular narrow area or region within the monitored area whereat the characteristic wavelength originated. Such spatial distributiQn resolution is advantageously provided by using a plurality of photodetectors arranged in a suitable pattern within a photodetector array, as described more fully below in conjunction with Figs. 6 and 7.
A preferred ablative light source 14 is a water-cooled flashlamp that is housed within an appropriate housing. A suitable flashlamp for use within such a housing is available from Maxwell Laboratories, Inc., of San Diego, CA.
The photodetector circuit 100 will next be -described. It is the function of the photodetecting oircuit 100 to de~ect the optical character of the surface of structure 22. In its simplest form, the photodetector circuit lOO simply includes a single ~photodiode selected to detect a particular characteristic -wavelength. Wavelength selection is made by choosing a -- 30 particular photodiode/len6/filter co~ ination (which are -commercially available components), or by selecting a broadband photodiode and manually placing a removable or replaceable filter in the optical path leading to the photodiode. In this manner, only optical signals of the characteristic wavelength successfully pass through the "`, . ~....

. . .~

WO93/1290~ PCT/US92~10878 ~ ~
.:
~ ~ 2 ~

. .
filters and are detected by the photodiode. All other optical signals are blocked by the filter. Thus, if it is known that the primer coat is blue, for example, and if it is desired that the primer coat remain, then a blue filter may be placed in front of the photodiode so that the photodiode only detects blue light. If a subsequent coating removal operation requires that all layers be removed down to the substrate, and if the substrate is, e.g., yellow, then the blue filter may be removed and replaced with a yellow filter., i.e., a filter, or combination of filters, that only allows yellow light to pass therethrough.
Using a single photodiode as the photodetecting ~`
circuit 100 only provides limited resolution of the reflected light to be analyzed, and does not provide additional infor~a~ion, such as spatial distribution data, that may be detected. ~ence, it is preferred that more than one photodiode be used, and that an appropriately processed optical, digital output signal ~`
194 be generated ~rom all of such photodiodes. For example, a digital weighted sum average ("WSAV") signal may be generated from all of the output signais from the individual photodiodes in the array. A block diagram of one type of photodetector circuit lO0 that achieve this ;~
function is shown in Fig. 3. ~s seen in Fig. 3, the reflected light 27 from the surface 22 is received over optical fiber bundle 25. ~The advantage of receiving the input signal as an optical signal is that it prevents electromagnetic interference (noise) from a~fecting the quality of the signal received. ~As seen in Fig. 3, at the heart of photodetecting circuit 100 is a processor 148. Such processor 148 may be realized using any suitable microprocessor circuit capable of operating at a modest clock speed, e.g., 5-lQ MHz. By way of example, processor 148 may be implemented using an Intel ~X51FB

W093/12905 PCT/US92/1087~
~2J~7~

) imbedded proc2ssor. Coupled to the microprocessor 148 ~ ;
is a conventional random access memory (~AM) 151, a ~; - conventional read only memory (ROM) 150, an analog-to-digital (A/D) converter 152, and an analog multiplex circuit (MUX) 144. The incoming light signals are split into three data channels~ Each channel is designed to select a particular characteristic wavelength, or band of characteristic wavelengths. For example, the channels may be respectively designed to receive and process wavelengths characteristic of the colors, red, blue or yellow. In this manner, photodetecting system 100 is :~
i able to :receive and analyze optical energy from selected :
portions, or from all, of the entire optical portion of the electromagnetic spectxum.
The optical data received in each data channel `~
is filtered and continuously monitored by photodiodes ~ ;
~ containe~ in the photodiode arrays 106, 118.or 130, and '~ is temporarily stored in response to receiving an : ::
appropriate clock or shift signal obtained ~rom the . ~
ao processor 148. Each photodiode in the array, as ~ :.
I explained more fully below, represents the light received ¦ from a defined area or "pixel" of the reflection .:
! footprint, i.e., the monitored area from which the t reflected light 27 is received. The data temporarily ~ 25 held ln the photodiode arrays is then serially ! transferred, under control of the processor 148, through .::~
appropriate channels, including the MUX.14~ and the A/D ~ ~:
152, into the processor 148. The processor 148 processes .
the data in a prescribed manner. For example, the~
processor ~ay divide the signals received in each data channel by a corresponding normalization signal obtained :: -from a sample optical energy 18' of the light 18. Sa~ple optical signal 18' is provided to photodetecting circuit 100 throuyh lens 23a and fiber optic bundle 25a. Fiber optic bundle 25a may penetrate housing 12 as shown in .

~ .. . :.

W093/12905 P~T/US92/~0878 ~ ~ 2 ~

FIG. 2 Cptical energy 18~ is filtered and provided to photodiode circuits 156, 158 and 180, and is used to normalize the amplitude of received signals so that each is independ~nt of variations in the incident light intensity.
As seen in Fig. 3, each optical data channel includes an optical filter 102i that attenuates all light except light of the characteristic wavelength that is received from the reflection footprint. Preferably, at least a portion of the reflection footprint is located somewhat behind the area on structure 22 which is .~ : :
impinged by particle stream 30. Filters 102; are available commercially from numerous vendors for any ~ :~
desired ~wavelengths. The ligh~ that passes through the - :-filter 102~ is received and temporarily held in a photodiode array 106, 118j or 130. By way of example, ~ ~:
the photodiode array may be a 1 X n photodiode array, ; ~ :
where n is a positive integer, as for example 1024. The :
photodiode array receives and transforms any received light 104 transmitted thro~gh filter 102~ into a series of electrical pulses 108 having amplitudes corresponding to the intensity of the received light, as controlled by an appropriate clock signal 143 generated by the processor 148. The rate of the clock signal 143, by way of example,:may range from 2-25 MHz. The electrical pulses 108~ are amplified in amplifiers 110, 122 or 134~ Track-and-hold circuits 114, 126 or 138, receive.signals 112, ~
124 or 136 and generate a DC analog signal 116, 128 or ~:.
l~O that corresponds to the average peak pulse amplitude of~el~ectrical pulse train 112, 124 or 136 in response to ; receiving a hold signal 142a from parallel interrupt timer (PIT) 142.
Analog signals 116, 128, and 140 are coupled through MUX 144 to flash A/D converter 152 over signal line 145. Control of MUX 144 is effected by signals 147 .
::

WO93~1290s PCT/US92/10878 ~ 2~&7~
-~
- ~3 -generated by processor 148. The A/D converter 152 thus generates a digi~al data stream 154 corresponding to the signals 116, 128, or 140 that is directed as an input signal to processor 148. Processor 148, operably coupled 5 to ~AM 151, stores the digitized optical data thus :.
received in RAM 151. ROM 150 has stored therein a :~
suitable operating program that controls the operation of . . .:
the processor 148.
Photodetecting circuit 100 also includes a plurality of ablative light source reference channels.
Each such sample channel includes a photodiode circuit, 156, 168 and 180, with each receiving as an input a : :
sample 1~3' of optical energy 18 directed to the surface .
220 Each sample channel fur~her includes an appropriate optical filter 1021, l022, or 1023 that filters out all . :
but a de~;ired wavelength or band of wavelengths. The photodiode circuits 156, 168 and 180 function similar to the photo~iode arrays 106, 118, and 130, transforming any light transm~tted ~hrough the filter 1021, 1022, or 1023 into a series of electrical pulses having amplitudes corresponding to the intensity of the transmitted light. ~:
Electrical pulses 158 are provided to amplifiers 160, 172 ~i ~
or 184. The resulting amplified pulse train is directed ~.: .:
to track-and-hold circuits 164, }76 or 180 which generate DC analog output signals 166, 178, and 190 representing the peak~pulse amplitude of the amplified pulse trains in response to receiving hold signal 142b from PIT 142. The - :~
signal thus generated for each sample channel is provided to NUX }44.
~ The photodiodes 156, 168, and 180, and their associated filters 1021, 1022, and 1023, respectively, receive sample optical signal 18'. In this wayO the signals directed to the MUX 144 through the respective sampled light data channels correspond to a sample of the light source used to provide the reflective light 27 to ' W093/1~905 PCT/US92/10878 2 ~ 2 ~

the photodetector circuit 14. Such sample of optical t signal 18' is used to normalize the light detected through photodiode arrays 106, 118, and ~30 so that variations in the intensity of the incident light source 5 do not adversely affect the processing of signals 116, .~:~
128, and 140 into an appropriate output control signal 194.
; As also seen in Fig. 3, a summing amplifier 181 sums the output of the respective sample channel amplifiers 160, 172 and 184. The resulting summed output signal is directed over signal line 183 to one input o~ a threshold detector 185. The other input of the threshold detector 185 is a reference voltage that is generated by digital-to-analog (D/A) converter circuit 187 as a function of a digital reference signal 189 determined by the processor 148 and conveyed to D/A circuit 187 via signal l~ine 186. The signal 189 is provided only during a sample window. Hence, the threshold circuit 185 receives the reference voltage that enables it to respond to the summed output signal 183 only during fiuch ~ample window. If the summed output signal 183 exceeds the threshold reference voltage during the sample window, :
: which only happéns if there is incident light present during the sample window,~then the output of the thre~hold~detector 185 goes high and functions as::an interrupt signal to the processor 148 causing it to enter a data sa~ple mode ~: In the data sample mode, the processor 148 serially receives optioal data from the photodiode arrays 106, 118 and 132 through the optical input channels and stores such data upon receipt of a reset signal 198a gene~ated by processor 148. Also-during the data sample ;~
mode, 6ample optical data may be received from the :
photodiodes 156, 168 and 180 through the ~ample channels. :
Parallel interrupt timer (PIT) 142 controls the timing of : .

W093/12~5 PCT/US92/1~878 2 ~
- 25 - :
,' '~ . ~ ' . ~
the particular data streams which are read by processor ~:
148 and stored in RAM 151 by hold signals 142a so that, ~
for example, data originating from a first input channel ~ .
including photodiode array 106 and photodiode 156, are . .
read together. PIT 142 similarly controls when processor 148 reads data from the second input channel that includes photodiode array 118 and photodiode 180, and ~ : -from the third input channel, which includes photodiode :.
array 118 and photodiode 168.
The processing routine stored in ROM 150 and .. ~:
implemented in processor 148 causes processor 148 to -:~ .
determinle the quotients of: signal 1~0 divided by signal 190, signal 128 divided by signal 178, and signal 116 :~
divided by signal 166, in order to normalize the outputs .-of the photodiode arrays for variations in the intensity of the output o~ light 14. Signals 166, 178, and 190 need~be æampled only once every data sample cycle, e.g., .~ ~-once every 100 clock signals 143 if photodiode arrays : ~:
106, 118, and 130.each have, for example, 100 diodes. . ~`:
Such normalization allows photodetecting circuit 100 to evaluate the optical character of the surface o~ -.
structure 22 as the output of light source 14 degrades ~:
over time.
The processor 148 generates the output signal .
25 194 and tran~its such signal to the control processor ~ ;
200. If needed, such signal can be converted to an optical signal using an appropriate conversion circuit in ;~
order to allow the transmission of the signal to be done optically over a fiber optic transmission cable, thereby -~
- 30 rendering the signal much more immune to electromagnetic noise. If so converted, an appropriate optical receiver ~ .;
circuit is used at the other end of the transmission line in order to convert the signal back to an electrical signal ~uitable for use by the control processor 200. ~:
Fiber optic transmitters and receivers suitable for such ~ .

''`

W093/12905 PCT/US92/1087B : ~
2 ~ 2 ~
- ~

purpose may be implemented using, e.g., a Litton Fiber optics Transceiver, Model E03675-2.
By way of example, the value, S, of signal 194 may be determined by processor 14~ in accordance with the 5 following equation: ~;
~ m ~ Signal 116~ Signal 128i~ m ~ Signal 140 S ¦~ Sign~l 190 ) ~1~ Si~nal 178 ) ~1~ S gnal 166)~

where i represents a particular photodiode in the photodiolde arrays and m represents the number of photodioldes in photodiode arrays 106, 118, and 130.
~he processor 200 uses information contained in the signal re~eived from the photodetector circuit 100 as a feedback signal to generate an address for a look-up table stored in the processor 200. The look-up table contains~scan speeds corresponding to the part1cular address used. Thus, when addressed, the contents of the addressed cell of the look-up table are retrieved and transformed into suitable scan speed control signals that :
comprise, in part, signal 202, directed to the robotic controller 204.
~ :The control signaI 202 comprises a composite ~:
: ~control s;ignal that~:also includes "path" control ::
~instruction~. ~Thus, composite signal 202 provides both ;~
.; ~ipath and speed:control~instructions to robotic cqntroller ~ 204. Robotic controller 204 then generates com~and : 25 ~signals 206 that direct the operation of robot~c positioner~l9, w~ich may be implemented using a~CIMROC
4000:Robot Controller manufactured by CIMCORP Precision Systems, Inc., Shoreview, MN. A suitable robstic controller i~ typically included as part of any robotic - ;
30 system:sold by vendors of commercial robotic positioners. -~ :
;
' '~

W093/12905 PCT/US92/10878 ; ~ ~
2 ~ r~

, Thus, in summary, the purpose of robotic positioner 13 is to position the scanning head 10 so that the surface of structure 22 is scanned with optical energy 18 provided by ablative energy source 14 and : -5 particle stream 30 in a predetermined path at a scan -speed dependent on the optical character of the surface of the structure 22 as determined by photodetecting ;
circuit 100. The scan speed is controlled so that substrate 28 of structure 22 is not damaged as a result of structure 22 absorbing excessive optical energy which is transformed into heat. --The temperature gradient through structure 22 is controlled to prevent damaging substrate 28 while layers 24 and/or 26 are being removed to expose layer 26 or substrate 28. Two approaches may be used to achieve this pu~)oce. In a first approach, the speed at which -~
the scanning head is moved across the surface 22 i controlled by determining an appropriate scan speed, ;
standoff distance "d", mass flow rate and temperature of particle stream 30. This approach i8 described in applicants' aforecited patent ~pplications. In a second approach, the scanning head may be incrementally moved across the surface 22 in small discrete distances. Also~
the duty cycle of the ablative light pulses may be controlled to prevent excessive temperatures in the substrate. This incrementa} approach is described further below in conjunction with Fig. 8.
Turning next to Fig. 4, there is shown a '~
waveform timing diagram that illustrates one timing arrangemenk that may be used for operating the photodetector circuit in accordance with the present invention. As seen in Fig. 4, an ablation light pulse 220 is generated beginning at a time Tl. Such light pulse is emitted from the ablative light source 14. ~-During the trailing edge of the ablation light pulse 220, , I
2 ~ 7 ~
.

i.e., at a time T2 seconds after the start of the pulse 220, interrupt signal 18sa is generated by comparator 185, as represented by sample pulse 222. If, for example, the light pulse 220 has an approximate duration of 1000 microseconds, then the time T2 may lie in the range of 800-900 microseconds. Such interrupt signal 185a defines the sample window referred to above that places the photodetector circuit 100 in its data sample mode. Further, because the sample window occurs while the ablative light pulse is still present, the light from the light pulse 220 may be used to provide the source of the reflective light 27 used to examine the color of the surface ;~2. The ablative removal cycle comprises the time, T4, between ablative pulses 220. Such time T4 may 15 be selected to b~ any suitable value to provide the -necessary power output, but typically will range from about lO to 5000 microseconds, corresponding to an ablative pulse rate of between 0 to 1000 Hz, where a pulse rate of 0 Hz corresponds to a ~ingle pulse.
In applications where light source 14 ~s a gas filled flashlamp for generating pulsed light, the data sample mode may correspond to a period when the optical energy generated by the flashlamp is at or near a -miminimum, as for example, at a level corresponding to amplitude 220a,~ shown in~FIG. 4. As is well known, the output of a flashlamp is at a mimimum when the flashlamp ~ ;-is energized by a "simmer"~current. The "simmer"i current ---is that level of current sufficient to maintain the gas contained in the flashlamp tube in an ionized state.~
Even when energized~with a simmer current, a typical flashlamp~would still generate sufficient optical energy to illuminate the surface of the structure being proce~sed. In some applications of the present invention, it may be desirable for the data sample mode 35 to correspond to an interval in the pulse period of the -~

~ W093/12905 PCT/US92/10878 ,. --: ~ -flashlamp when the flashlamp is energized by the si~mer current. Such interval would be established by selecting an appropriate sample window, as previously discussed.
~eferring to Fig. 5, a waveform timing diagram is shown that illustrates an alternative timing arrangement for operating the photodetector circuit 100 ~- -when an auxiliary light source 2g is used. As seen in Fig. 5, ablatiYe pulses 220 are generated at an appropriate rate defined by the ablative period T4. At sometime after the control pulse 226 has gone low, the ; auxiliary light 29 is pulsed ON by control pulse 17, provided by light control circuit 13 in response to receiving signal 12 from control processor 200. Control processor 200 generates such signal 12 based on the value of signa;l 194 provided by photodetecting circuit 100.
While the auxiliary light is ON, a photodetector sample pulse 230 is generated, corresponding to interrupt signal 185a, which effectively place~ the photodetection circuit in the data ~ample mode. In such mode, the 1 20 photodetection circuit examines the reflected light 27 to j determine the character of the surface 22. Thus, as seen for the approach shown in Fig. 5, the ablative process comprises ablating the surface material and looking to -~
see if sufficient material has been removed.
~Fig. 6 shows a diagrammatic representation o~
one embodiment of the photodetector arrays 106, 118 and 130 that may be used with the photodetector circuit 100 of th- present invention. As ~een in Fig. 6, the ;~
reflcctive light 27 is preferrably split by, e.g., a 1 x 3 optical splitter of a type commonly known and available in the art, to divide the opt~cal sig~al into three optical paths. Each optical path directs the light `27 traveling therein through a respective filter, 102~
1022, and 1023, also referred to in Fig. 6 as filters El, F2 and F3. Each filter is selected to pass only a .'- ';~ ~ :~
-.

WO93/12~5 PCT/US92/10878 2 ~

wavelength or band of wavelengths characteristic o~ a prescribed color to be detected a~ the surface 22 being ablated. The light passing through each filter is then focused to fall upon an m x n photodiode array 106, 118 S or 130, where m and n are integers. Using an m x n photodiode array in this manner offers the advantage of being able to detect the relative spatial position of the reflected light from the surface 22 of the material being ablated, as well as its color characteristics. For example, if the re~lection footprint is optically focused to cover the entire surface area of the detector 106, and if such reflection footprint is larger than the irradiated ~ootprint, then an area 221 may appear on the surface of the detector 106 that represents the ablated area, as measured by the wavelength that passes through the filter Fl, while the area around the perimeter of the area 221 on the surface of the detector 106 would represent th~ non-ablated area. In other words, some of the individual photodiode elements that make up the surface of the diode array 106 would receive light of the passed wavelength, and others would not~ In this manner ;~
the array 106 i~ able to provide a rough pixel-by-pixel resolution of the surface 22 of the material being ~ ;
ablated as seen through the particular wavelength that ;~
the array 106 is adapted to receive. When such ~
infor~ation is combined with the other arrays 118 and 130, a great deal of information can be learned about the ~`
character of the surface 22 being examined by the , incident light 18. ~ ~ ~
. , As will be appreciated by those of skill in the -art, when a photodiodé array is used as described in Fig.
6, a somewhat different data processing scheme may be ;~
employed than is described above in connection with Fig.
3 in ordex to process and analyze the array data.
However, such processing schemes are well known in the `'' W093/12gO5 PCT/US92/10878 ~ ~ -- 31 ~

art, and are commonly used to pxocess the data obtained from large diode arrays, such as CCD arrays, in imaging applications.
With an array as shown in Fig. 6, the present -invention provides more than just an ablative coating removal system. This is because the photodetector circuit 100 may be used indep~ndent of a coating removal system to examine the character and quality of surfaces, ~.g., for quality control or damage control purposes. : :~
When used as a photodetector system in this manner, all that is required is to pulse ON the auxiliary lamp 29, or other non ablative light ~ource, and direct such light to the surface to be examined so that it is reflected ~ :
therefrom to the photodetector circuit 100. The 15 photodetector circuit 100 then processes the received ~
reflected light in the manner described aboYe to -~ :
determin~ the character (color) of the:surface being ~ ::
examined. Further, when used as part of an ablative coating removal system, the additional spatial distribution information provided by the individual diodes o~ each array provides a much more complete ~picture" of the effectiveness of the coating removal .~
process, and further helps the control processor 200 to .
better define an appropriate scan path. Moreover, such additional spatial diætribution data allows the control . .
system-200 to level, or otherwise orient, the saanning : head 10 relative to the scanned surface of.the structure 22.
~ ~ ~Fig. 7 shows a diagrammatic representation o~
an alternative and simplified embodiment of a photodetector array of the present invention~ ~s seen in ~:
Fig. 7, a~ ingle m x n detector io6 ~ is utilized to receiv- reflected light 27. An appropriate lens assembly -~
focus-s the light 27 through a replaceable filter assembly 102' and onto the surface of the array 106'.

';'':, ''.' .

' .:.' ~:

W093/l2905 PCT/US92/108~8 ~ 21?JS~ 32 -The replaceable filter assembly 102' is selected to pass only those wavelengths characteristic of a particular known surface that is to remain after one or more over coats are removed ~sing the ablation process of the S present in~ention. The simplified embodiment of the detector array shown in Fig. 7 may be used, for example, when the ablati~e removal process is being used to remove paint down to the primer coat from a large number of airplanes or automobiles, all of which have the same color of primer coat. Should the need arise to ablate away coatings down to a different color undercoat or substrate surface than can be detected by f ilter assembly 102', then the filter assembly 102' is simply replaced ~ ;
with an alternative filter assembly that can detect such ~ - ;
different colox.
Referring next to Fig. 8, a flow chart is shown ;~
that depicts one method that may be used by the present -;
¦ invention to ablatively remove coatings from a substrate.
As seen in Fig. 8, a first step of the method, shown at block 302, involves the setting of initial parameters used to get the process started. Such initial parameters lnclude, for example, the coordinates of a starting ;~
location for positioning the scan head, the scan path, an initial standoff distance "d",~an initial ablative pulse 25 energy (amplitude and pulse width), and an initial - ~;
¦ ablative pu}se duty cycle (fre~uency). The initial -~
I -para~eteræ also include~setting an index control ~-¦ variable, "i", to a starting value, such as 0. ! ~, Once the initial parameters are set, the scanning head i8 moved to the starting location of the I prescribed scan path, L; (block 304). Once moved to tbis I position, the timing circuits within the processor 148 `;
I (Fig. 4) determine whether it is time to generate an ablation pulse (block 306j. An ablation pulse may be --generated, for example, at a frequency of 4-5 Hz. When .
.

WO93/12905 PCT/US92/10~7~ ~
.
7 12 ~

':- :, ' it is time to generate the ablation pulse, such a pulse is generated (block 308) ha~ing a pulse width and amplitude as controlled by the parameters previously set. -~
After the ablation pulse has been generated, the incident ;~
S light used to illuminate the area being ablated is summed for each light channel that is used (block 310~. As explained above, in one embodiment, such incident light may be derived from the trailing edge of the ablation pulse (Fig. 4). In another embodiment, such incident 10 light may be derived from an auxiliary light that is ;
pulsed on at an appropriate time (Fig. 5). In either event, if it is time to sample the reflected light, determined at block 312, then a determination is made as to whether the sum of the incident light, performed at 15 block 310, is greater than a pre~cribed threshold (block ~;-314). If not, then that means that there wiIl ~e no `
refle ted light of sufficient amplitude to provide any useful information. Hence, the reflected light is not monitored, and control of the process returns to block ~ `~
ao 306, waiting for the generation of the next ablation pulse. If the sum of the incident light is greater than the prescribed threshold (block 314), then the data collection mode of the photodetector circuit 100 is begun (block 316).
Once the data collection mode has been initiated, the reflected light from each channel is received and stored (block 318) as digital data. As this `-is being done, a determination is made as to whether such -~
data should be normalized (block 320). If so, a normalization process is carried out (blocks 322, 324). -~`
The data from each channel is then analyzed to determine if it is characteristic of a prescribed wavelength, A, representative of a prescribed color (block 328). If not, then the ablation parameters are adjusted (block -330), as required, and the next ablation pulse is ': ~
~' ~
, 2~2~7~

generated (blocks 306, 30~). If 50, then a determination is made as to whether the scan path has been completed (block 332). ~f not, the index is incremented (block 334), the scanning head is moved to the next scan path location, Lj (block 304), and the process repeats. If the scan path has been completed, i.e., if all locations along the designated scan path have been ablated, then the process is stopped.
As thus described in Fig. 8, the scanning head is incrementally moved along a desired scan path, with the scanning head b~ing positioned at specified locations ~-along the scan path only for so long as is required to ablate the desired layer(s) at that location. The desired alblation may require a single abIation pulse, or multiple ablation pulses, with the determination as to whether t:he layer has been removed being made by analyzing the reflected light from the ablated location ~-for the pre~ence of a prescribed color.
Not included in Fig. 8 i5 the control for the 1 particle stream 30. It is contemplated that the particle ~tream 30 may be enabled at all times during the ablative removal process. If so, and if the stream is made up oP
CO2 pellets or cold gases, frost or condensation may form around the ablative site. Advantageously, the photodetector circuit ~no, while performing its monitoring function, ascertains if such frost or condensation has formed, and if so, an appropriate ~;
control signal is generated to make appropriate adjustments, e.g., turn OFF the particle straam for an 30 appropriate time. ;~
While the invention herein disclosed has been described by means of specific embodiments and applications thereof, numerous modifications and variations could be made thereto by those skilled in the art without departing from the scope of the claims.

Claims (10)

What is claimed is:

1. A system (11) for removing a layer of material (24, 26) from a substrate (28), comprising:
a flashlamp (14) that generates pulses of radiant energy (18) having an intensity sufficient to ablate said layer (24, 26) of material;
a control circuit (13) that controls said flashlamp (14) so that it irradiates a target area on said structure with a series of said pulses of radiant energy (18) for the purpose of ablating said layer (24, 26) of material;
a photodetection circuit (100) that monitors radiant energy (27) from said flashlamp (14) that is reflected from said target area for the presence of a prescribed color intensity that is different from the known color intensity of the layer of material being ablated by said pulses of radiant energy; and feedback means (200, 204, 19) that controls the control circuit (13) so that the target area is irradiated with said pulses of radiant energy (18) from said flashlamp )14) as a function of the color intensity sensed by said photodetection circuit (100).

2. The system as set forth in Claim 1 wherein said photodetection circuit (100) includes a first set of photodiode arrays (106, 118, 130) that detect a prescribed wavelength within the radiant energy (27) reflected from said target area from said flashlamp (14), said prescribed wavelength being characteristic of said prescribed color intensity.

3. The system as set forth in Claim 2 wherein said photodetection circuit includes:

a plurality of optical channels, each being adapted to receive a portion of the radiant energy (27) of said flashlamp (14) that is reflected from said target area;
said first set of photodiode arrays (106, 118, 130) being located within said plurality of optical channels to detect if the reflected radiant energy (27) in each of said plurality of optical channels contains a respective wavelength; and a processor (148) that analyzes the respective wavelengths detected in each optical channel to ascertain whether said prescribed color intensity is present in the reflected radiant energy.

4. The system as set forth in Claim 3 wherein said photodetection circuit (100) further includes a second set of photodiode arrays (156, 168, 186) that detect the intensity of incident radiant energy (18') falling upon said target area from said flashlamp (14), and wherein said processor (148) further normalizes the detected reflected radiant energy (27) detected by said first set of photodiode arrays as a function of the detected incident radiant energy (18') detected by said second set of photodiode arrays in order to remove variations in the intensity of the detected reflected radiant energy caused by variations in the intensity of the incident radiant energy.

5. The system as set forth in Claim 1 further including a particle stream source (6), controlled by said feedback means (200), to deliver a particle stream (30) through a nozzle (32) pointed at the target area in order to impinge the ablated material with said particle stream (30) and clean said substrate (38).

6. A method for the selective removal of a material from a structure (22), comprising the steps of:
(a) irradiating a target area of a structure having at least one layer of material (24 or 26) formed on a substrate (28) by generating a series of radiant energy pulses (18) using a flashlamp (14), said pulses of radiant energy having an intensity sufficient to ablate said layer of material;
(b) monitoring radiant energy (27) from said flashlamp (14) that is reflected from said target area in order to sense the presence of a prescribed color intensity different from a known color intensity of the layer of material being ablated; and (c) controlling the irradiation of the target area with said radiant energy in step (a) a a function of the color intensity sensed in step (b).

7. The method as set forth in Claim 6 further including detecting the reflected radiant energy (27) from said flashlamp (14) in a plurality of optical channels, determining if the detected reflected radiant energy (27) in each of said plurality of optical channels contains a respective wavelength, and analyzing the respective wavelengths detected in each optical channel to ascertain whether said prescribed color intensity is present in the reflected radiant energy.

8. The method as set forth in Claim 7 further including detecting the intensity of incident radiant energy (18') from said flashlamp (14) that falls upon said target area, and normalizing the detected reflected radiant energy (27) from said flashlamp (14) each of said plurality of channels as a function of said detected incident radiant energy so as to remove variations in the intensity of the detected reflected radiant energy caused by variations in the intensity of the incident radiant energy.

9. The method as set forth in Claim 8 wherein the step of monitoring the reflected radiant energy (27) from said flashlamp (14) comprises monitoring a trailing edge of the pulse of incident radiant energy (18) generated by said flashlamp (14).

10. The method as set forth in Claim 9 further including impinging the target area with a particle stream (30).