Weekend passdown for Wk33
Long weekend ahead , so prepare the passdown for line people to follow up:
In all Auto-switch event onwards, to arrange the measurement for the IGBT RS
measurement that the run is processed 3hrs AFTER A/S and RS measurement
that the run is processed 6hrs AFTER A/S.> E.g. 8am A/S happened, measure the
IGBT run processed at ~~11am and run processed at~~ 2pm. If Not sure, pls verify with Engineer
PROCESS Alignment:
1. For all loading issue, add the loading op ID here to track. Ken will compile
and share with PROD team.
OOC due to Loading Issue:
R924A AVI failed due to wafer 21/22 loading issue. (Defects count 41/52)
R924B AVI failed due to wafer 01 loading issue. (Defects count 99)
2. For incoming issue, need to alert to previous module / relevant stakeholders
officially by email communication.
3. Do not use Profiler Coil for THK tuning and Do not use H2M for RES tuning.
4. For the qual results, need to be shared thru passdown OR email as official communication, not saved in common drive only.
5. For Back Transfer process, pls be reminded that the BT THK / DW qual is only
refer to F/L result. Test run result is not being considered <as per DPQ
system, we still need to key TR result> and team is working on APF system to
replace DPQ system.
NEW
6. For the intermittent loading issue, pls investigate the blackbox for the
lift down timing, life up timing and standby timing. Also, to check if the
operator did the wafer flat alignment at loading station OR at the workstation,
any PM activity for tool beside and etc. We will work out the package process
checklist for intermittent AVI OOC <based on fishbone analysis input>.
7. With immediate effect to stop any MFC Zeroing due to recent Res OOS case in
R919.
PROCESS ISSUES:
So far TC1 RS issue rate 50%.
- As per Whong we increased RS sampling from 1 sample/ day to 1 sample/shift for HU susceptor.
- Instructed above information to R908B/R931A/R928A EGA GUI and informed production.
- Gen 585 card has 52 fault code: 5=Serial watchdog No communication with Video
Keyboard; 2= Phase Sequence
- Video keyboard (VKB): Serial Watchdog Converter Error and Master and Slave
parameters (some) are not the same
- Reset the video keyboard and 585 card, still the same
- Aligned the VKB Master and Slave parameters , still the same
- Performed troubleshooting as per Gen Manual
- Wire continuity of Reactor to Gen RS232 cable - no OC, ok
- Wire continuity of HA142, HA141 & Gnd from VKB card 452.C2 to 436 to 585.C2 -
no OC, ok
- Check fuses for 452.C2 to 436 to 585.C2 - all ok
- Isolate VKB 452.C IC#13 JUN (timekeeper) using IC from spare card,
still the same, put back the orig
- Isolate 585 IC#24 (EPROM MEMORY) using IC from spare card, still the same,
put back the orig
- Measure VKB 450.3 pin 2-5 and 3-5:
> RS232 connector not inserted in the card: -12.24Vdc & -0.27Vdc
> RS232 connector inserted in the card: -10.69Vdc & -10.14Vdc
- Replaced the VKB 450.3 and 452.C card using old spare card- still same (old
card maybe faulty also)
- Put back original card for both and install back all original parameters
- Notice both R916 & R918 VKB 452.C card LED S0 & S1 is ON during idle
- Please continue to check and troubleshoot the 585 card
-Rotation shaft extend ipcrb R907 both side
- To do High Temp H2 leak check on BP/rotation and fill up the monitoring
checklist. Next monitoring date is 16/08/2017. Continue putting this and update
date follow up on passdown.
-Please help to fill up the R907 iCRB checklist located at: Checklist Rotation Shaft change from Half yearly to Yearly
B]R913 set up is priority over project. Once free, please follow up with start-
up checklist posted behind, on the machine gas panel cover.
C] R920 - Generator main breaker's relay temporarily bypassed
D] R926 - Generator main breaker's relay temporarily bypassed
E]HCL Fast command activation: to prepare 9pin-to-15pin cable; to plug-in to
Distributor I/O#3, J8 - (for R907, R908, R916, R919, R920, R921, R923, R926)
Strongly recommend friends to watch this youtube channel , as a form of travel in Turkey, Istanbul ! All the street views and shops are as good as you do the walking on the street !
Also a peep on Turkey this country , their daily life
https://www.youtube.com/c/FeelTheWorld/videos
Chronic issue resolved!
There is this chronic issue that haunted us for many years, with the team efforts in finding rootcauses, today we finally able to conclude something out of it:
Issue is High RS VS02.
Test that we plotted to run our hypothesis:
Run 6 t/w with different setup
1. I=0.63mA, IPFG=6A, RS=196.3 (1.42%) rem: clear scan map with 5 lines
2. I=0.63mA, IPFG=0.1A, RS=173.9 (0.42%)
3. I=0.63mA, PFG OFF, RS=174.4 (1.09%) rem: nonuniform oxide, RS should be lower
4. I=6.5mA, IPFG=6A, RS=175.7 (0.59%)
5. I=6.5mA, IPFG=0.1A, RS=172.8 (0.55%)
6. I=6.5mA, PFG OFF, RS=172.9 (0.41%)
Summary:
1. PFG isolation good
2. Small RS difference between high and low beam current (may be expressed as
0.5%, high current - OVERDOPE). Probably, high beam forces higher beamline pressure. It seems, by itself, faraday HAS NO STATIC LEAKAGE. Also, no scale error detected - no problem with dose processor.
3. PFG leakage can be expressed as 15-20 uA/1A PFG current.
4. Because day shift test beam was 2 mA and I PFG = 4A, expected leakage between 3-4%, what gives RS 95-96 Ohm/sq.
5. As mentioned above, in case low beam with high PFG arc current scan map observed. This map had lowest RS (and lowest leakage) when disk was on top and highest RS (and leakage) when disk in bottom position.
When disk travel up and down, because of 7 degrees tilt, space between disk and exit of faraday cage changing by about 0.7 inches (Space highest when disk on the bottom). Hence, leakage INCREASING with increments distance between faraday and disk. Leakage on the bottom is about 95 uA while 63 uA at top disk position. It may happen either exit fence magnet field weak or excessive distance between faraday and disk (or both together).
With assumption model:
1)correct magnet field,
2)linear dependance of leakage from space, and
3)travelling distance 6 inches, might expect extra gap between faraday and disk above 1.5 inches.
If this distance can't be found, weak magnet field should be the reason. (Rem: to eliminate effect of leakage, perhaps, smaller distance required, since magnet field drops not linearly, but as 1/x^3 (in some geometries as 1/x^5, because no magnet charges in nature)
6. Medium current t/w showed acceptable result, because beams difference (with high current test) is about 2-3 times, while PFG arc current - 8 times. Hence for medium current tests situation with leakage better about 3 times. But need to keep in mind, medium current test might be done with beam above 1 mA and with beam 0.5 mA, hence, result may drift signicantly.
7. Leakage getting not important in cases high beam current and low PFG arc current. Let say, Axcellis recommend DO NOT APPLY PFG for all ranges of boron implantations)
8. Leakage getting devastating in case of low current with PFG ON (let say, for 30 uA implant beam)
9. BLIND OR UNREASONABLE INCREASING PFG ARC CURRENT HAS STRONGLY NEGATIVE EFFECT.
Note: All leakage current estimations done on analysis of doping levels
Mid day issue report 21 Oct
****Tube06: EXTRACTCTON VOLTAGE ISSUE****
VESDA alarm activated due to burning smell and smoke from EXT resistor.We activated cleaned the holder and replaced the 3.9K Resistor.We also found Ext voltage stable only up to 55KV.
Decided to Vent the source and found Bushing heavily coated. Cleaned away and replaced the source as the Arc chamber slit too wide and suspect required more gas than normal. At last, conditioned with Argon and then released back to prod.
****Batch03: SF count A side 114 cavity 14 column 7****
Wait unload B side. Then do leak check A side. Was using both Black and Red H2 detector, found no leak. Check Belljar base condition- not much flakes.
Check Susceptor for spikes – none for finding.
Quartz items condition with flakes deposition. Baseplate checked- condition ok.
Engineer requested to wash quartz. Same time clean minor flakes on bell and baseplate. Then check Laminar flow-pass 0.45.
After quartz wash, put back. Redo susceptor centering. Since no dust count device, check shield door dust via wafer surfscan.
A side result -12-16. PM clean above sheild door. Then redo scan- result - 0-5. Then handover to Engineering.
**Etch09: Wafer back have acid stain**.
Both medium1 and 2 suction were too fast. He tries calibrate the suck back valves with regulate the top knot but med1 still. Try on regulate SV for activate both suck back valve till the suction is smooth and no have dripping of acid.
Decidd test run dummy wafer for 1 lot, ok. Encounter swivel pick chuck mark for both side. Clean proper and continue seasoning with dummy wafer till no mark see.
At last, calibrate the flow rate of medium 1 and 2 after found the medium1 adjust to max on the regulator of flow meter. Test few pcs of dummy and check the result, ok.
Mid day Issue - 14 October 2021
USYUL01 having PM1 Communication Error
This machine encountered error PM1 not ready. Upon check found PM1 lost comm to the main frame. Reset machine and initialized but still PM1 unable to communicate.
We remove power on PM1 controller and reset again, negative. Try change hdd from slave to master, -negative
So we put back master hdd and tried swap line card on the controller. Upon removing IOB-80 card, found some burn mark on component
Continue change the card, and reset but still PM1 unable to communicate. Change the whole PM1 controller, still negative
Other modules are ok and seems working fine. Checked PM1 LAN connection and LAN router unit. PM1 no link to router. Swapped LAN cable and tested cable is ok.
Proceed to change CPU card(mother board), PM1 able to link to router but still no comm. Again changed another HDD -ve. put back original HDD.
Renamed Persistent Materials and Persistent DataMC folder then reboot PC -ve. Renamed ":CWmanuserPM1datalogger.dat" and reboot PC -ve.
Find automation failed due to chambers offline error even though P2 to P6 are
online except P1 chamber. Recipes are checking ALL chambers to be online.
---
9B04 Layer time error.
During trial run, unable to get evap rate and Ni block deep hole. We aborted and changed filament. redo premelt Ni block.
Again encountered High voltage source error, no High voltage.We checked filament current ok. checked interlocks ok.
Finally found triode tube not growing. fine adjusted the tube base ok. So then continued premelt Ni block and test run.
Mid day issue
Today is relatively good day, issues are controlleabe:
Lot prev hold for stain after EWITO strip resist,only saw the slight discoloration at the test pattern. Did not notice the suspected TIW lack at the edge dice. As checked by LSZ, got history of remask at mask ITO due to wrong dev pg used.
Took photos of mask pattern for ITO/TIW lvls,& confirmed the affected pattern is TIW.chk tools but no downtime seen, refer to excel. No buddy lot at 5EWB03,run @bathlife 16 lots/14 hrs.no buddy lot at S4 as well,lots before/after are jetmos via level.
Check maestria cannot find any abnormality when compared with other run. Nearest lot of same pdt still at EPI.to discuss with N3.
Another lot W3 on-hold at IQC due to suspect ITO lack, affected wafer edges/mostly on the flat area. Saw in WSO have history of inspection for bubbles aft dep pese 2. Lot not yet 100% assessed. Single lot ran in 7A.
Lot checked 100% .12 wafer affected. Laser mark 05/24/23/25/13/14/16/11 affected few dice at 1st row of flat. Laser mark 18/04/03/01 affected 2 row from flat and right and left side of wafer flat. Affected dice metal peel same for same pattern of dice.
Informed masking to check. W534KP3 relog to ic as the affected dice affected with tiw lifting. relog to ic to relog to correct owner.
Report for the day 27 Sep
Today considered relatively good day for a start of the week:
R919
Inform maintenance to ensure external data logger is working before carry out test run . Every run to measue RS as instructed.
B side 2W TR Res Ave=68.7 (4.3ohm higher) compared to previous run with same dope setting of 29.8cc Res Ave=64.4ohm (OK) while A side Res Ave=64.9ohm (OK)
External data logger for both TR shows Dope flow extension on 1st Layer. This means that issue is intermittent and common. Since external data logger already in place to check dope, inj,dilute flow as per agreed with N3 to TR 2xprod for both side was arranged to check the flow again.
FILE UNABLE TO EXTRACT OUT. TOGETHER WITH MAINT, TRY TO EXTRACT TROUGHT WINDOWS INSTEAD OF USING SOFTWARE, STILL UNABLE TO EXTRACT OUT THE DATA. PHYSICALLY FILE IS EXISTING, BUT UNABLE TO EXTRACT.
Check bothside FL result Res is on target. Partial rework aranged on R99B . Profile OK Lot Moved out
As per morning discussion to change inject valve to vent and bell spv for line breaking will be done as one with change drum spv
External data logger was also fixed by Yuni . To continue new drum qual and TR for 6w8L (Together with Maintenance to check external data logger after both side TR completed.)
Please ensure production stop the tool during data extraction . Every run need to measure Res profile .Every end of shift to pull out ext data logger data and publish to team until further notice.
RS issue & Changed drum
Per advised, changed V57 and V74 (did pre and post line breaking before and after replacement. Did H2 leak check - passed. Pressurized test for 1hr - okay/passed.
Cycle purged Dope line. Replaced corresponding EV for the replaced valves - resistances: EV53 before=429ohms, after=430ohms / EV54 before=431ohms, after=427ohms.
Encountered problem in datalogger, other valves having reading when they supposed to get zero; found that terminals are actually cannot be left hanging. Put 10KOhm resistor across +signal and GND so that +signal won't be left hanging if pressure switch is OFF. Tested all gases (and response in external datalogger).
All are okay. Please extract the data after first TR to ensure that datalog is recording. 2]Gas Tech changed the silicon drum (during troubleshooting). Empty the saturator and topped-up done okay.
Weekend mid-day report:
Today have to work from home during weekend for mid-day report:
**R765 –**
B side RS DW OOC due to B-point low and T-point high on cavity #07. However previous run RS DW on cav#07 is within control. Suspect due to tapping issue or measurement error
So have to arrange to measure same wafer and other cavity for further check. Next run arrange to measure RS profile. RS DW-11.47% VS SPEC=15.2%. To continue production, however next run still RS DW OOC-16.78%
Next to perform treatment with bake and arranged next run wait for result. B side Res DW OOC after susceptor treatment. Recovered after change profile from 2,6-16 to 2,6-0 while A side profile changed from 1,3-8 to 1,4-8.
A side RS DR improved but DW OOC after change profiler coil . Current convert to 6B
**PROCESS ISSUES:**
**R932**
Conversion from H42 to F49. Both side GR high : A side 1.14 & B side 1.17
Temp check A side 1096 vs 1100 degree C ; Temp check B side 1098 vs 1100 degree C
Suggest down for maintenance to check TC/BUBBLE/H2 carrier and so on . Current maintenance attending R8B S/F issue
Due to urgency on F49 Automotive, to change Sx from 17 to 14 and continue production. GR ~ 1.02um/min. Every run to measure thickness 1pcs, compare GR based on Center point and publish result per shift.
**R929**
As per N3 team request to adjust H2 carrier pressure to original 1.5bar. Test run bypass HCL
After Replaced V22(SX pneumatic valve to bell. Arranged 1x1074 test run GR still low-1.29 um/min . Escalated result to N3 team
Mid day report 20 Sep
Today mid-day report are as below:
Lifting
Level: Etch Wet Active
Summary:
1 lot seen with lifting after etch metal and suspect it is due to over etch at Etch Wet Active. We raised NCL and moved the lot
Lot on hold to do 100% T-84 testing due to NCL lot. 100% .T4 testing done and passed, moved to monitor yield. Unfortunatly, LM23 RES_MET_THIN value has difference compared to other LM.
Batching TS2 Hold Proc
Summary:
Operator Ying Ing fdback STS2 running metal1 rcp but time not counting. And Techinician Yunis arrived & saw the rotor turning & time stuck on Step 18 3:13 drying step.
He stopped machine from running & unloaded wfrs. Put in white 50wafrr box.
We simulate together with technician how to recreate the time freezing & rotor still turning is by pressing HOLD PROC button. Upon pressing this, alarm will be heard,& it had to be cleared to be erased from list of alarms/errors, hence suspect operator cleared the alarm.
Operator who ran wafrrs is Evelyin she's OT from G2. From estar, process held 26mins into the recipe & ran 39mins more on the same cycle before stopping.
Lots rewashed & no issue encountered. Inspected 5wfrs each lot can only see slight pitting worst on edges.
Requested human interview form from trainer & advance hold lot with msg ID "WOE NCL" before move out to confirm if ncl needed. 2lots ECD is in spec at slope etch level. NCL chopped
Evaluation for backside clean
A little update on the project we follow this far:
MEMS BACKSIDE CLEAN EVALUATE
BACKGROUND:
ENGINEER evaluate for backside clean, this wafer after dry etch sent to coat recipe front side, then do 1% HF 5min to clean backside. After HF clean will strip resist use SVC14, but now SVC14 still pending vendor, so as Aigen comment we sent to strip resist use batch540. now I done 1pc wafer(slot01) R-ST-RIN at batch540, but under scope can see some red colour defect at wafer edge, refer attached pic.
Please follow up to take another wafer use EKC to strip resist then inspect. Take LM#02 together with LM#01 to do EKC , LM#02 after strip inspect under scope can see the same defect as LM#01. Please help sent 2pcs wafer(LM#01,02) to do 100% CAMTECK scan. 2pcs wafer I put in the slot01~02.
2pcs send to CAMTEK 100% scan, handcarry to QC operator. Please follow up to data. The 2pcs gave done scan and review, lot parked at EWS etch bay rack.
As today excel sheet update is: Pending Huni getting approval of using SVC14 from wet oxide team. To pend until Monday for Huni.
Night dinnot handle the lot. Please follow hld comment. Check with wet ox engineer, the EKC bench already changed to use chemical SVC14, now send 23psc of Ww12 to Gold fab coat resist but encount . DNS machine transport problem (all have coat), Please follow up with coat front side, just follow template.
HF 1% step done at EEWB1. Lot parked at UBM Engineer table awaiting for the WOE SVC14 availability due to they have high WIP. Shift Engineer already informed to handover to G3 shift to allocate.
EKC strip done, merge LM#1/2,send QC to 100% CAMTEK, Please follow Result , Please follow to collect data.
LM#03~~24,LOW yield<80%,is 32%~~65%,lot in QC, As consulted with Chrisi, already take LM#03 to use chamber 3 to strip resist, now wafer is at tegal bay pc desk.Please take LM#04 to wash S4(EKC),then sent this 2pcs to do CAMTEK.
LM#04 washed and scan result is 47%,less than 80% low yield, defect all at center.LM#03 97.7% pass,which restrip using TG1C. Need to compare the LM#03/04 pre and post re-strip CAMTEK data. But at the end of the shift I check the CAMTEK data QC operator give me, only 16pcs,Please help to get the others data and send Chrisi.
No time to follow up due to run 6 lots RND with 100% AEI. Please send lM#03 ~~to strip 5E1C then scan CAMTEK. Machine still down under MS, still on going troubleshooting. Please follow up to send lM#03~~25 to strip 5ETE1C then scan CAMTEK.
This morning N3 talk to device, and afternoon decide to send lM#03~25 to strip 51C then scan CAMTEK again, now machine running product, when finish, Please follow up to strip and CAMTEK.
LM#3~25 now running "R-ST-RIN" in 5ETE01. Please follow up to send to QC and request to scan CAMTEK 100%.Feedback from MS some pcs have rinser fault timeout, please help check in the lot history and check wafer.
3pcs encounter "rinser fault timeout" alarm, MS resume and the 23pcs already finish rinse-strip-rinse, check 3pcs alarm wafer history, done process already. Send to QC to 100% CAMTEK, Please follow up result.
Problem appear everywhere
When there is problem, we started to argue who is right who is wrong.
Debate went on and on.
**INCOMING BROKEN**
Summary: operator Josh seen 1pc broken when opened box, LM19/slot22 when about to run bomba. broken 12-6 o’clock and seen chip on 12 o’clock with no acknowledge on r/s. Lot hold to production. dry etch also no clamp point and key out with no issue.
M/l sent to bomba already haven’t inspected. Awaiting supervisor to agree to scrap. Inspected slot 21(LM18) after bomba no s/c, supervisor Xiong say to wait for their boss decision.Lot hold to 250HM.
We checked gasonic JS11, clamp point of tool is at the flat area(small portion of the center flat) when the arm takes the wafer. unable to find the broken wafer at the lot.
**BROKEN WAFER**
BACKGROUND:-1pc from slot 6 broken into many pcs at CD01 aft operator Yun measure ECD.
She placed the lot on the stage and some of the wafers are potruding out. Measured wafer is from slot 25. Aft the arm placed backnowledge the measured wafer from slot 25, it went down and this affected wafer was not sitted properly in the carrier (potruding out too much) that the arm hit the wafer.
She realized that the wafer was not sitted properly but as there is an alarm from the etching bench, she immediately attended to the alarm and did not properly place the wafer to be slotted in. Debris seen for slot5. sent for re-qdr and srd. Pls help to inspect for scratches for slot 5
Mid day report 16 September
Not an easy day with problems in 3 areas, so we have to one by one settle.
**STS1 MAESTRIA ALARM**
MF:Min(Tank 2 Active(Conc Scaled)) alarm encountered for via run confirmed fluctuation in the maestria graph; already changed controller but still encountered alarms,changed valves and key to prod monitor no more alarms, to monitor all maestria data w/ in shift to check for spikes
To extract the maestria data for all runs within shift at end of shift to check for any spikes and saved in Y:Wet EtchWet Oxide. JS1 Monitoring if no alarm encountered throughout shift. To extract that particular run if got alarm.
**DI water heater temp out of tolerance**
Mc autodown. As per estar, enountered alarm after 20mins of processing time. MS able to resume the process.MS reset the temperature controller and error/alarm removed.MS performed empty run 2X no issue. key up tool for monitoring. Asked ms to check as per OCAP,no more alarm within shift.
Lots affected rewashed and inspected 5pc each no issue. moved out.
Same alarm aft 20mins into the rcp max temp seen was 109C. chk estar,process finished aft alarm. rewashed wfrs then insp 5pcs each no issue.ms reset temp controller as per ocap, then run empty run same alarm happened,max temp seen on controller panel was 95C. Reset system then do empty run no alarm.run another 2x empty run no alarms encountered, key up to prod.ms feedback no spares for this machine, need to consult expert from diff; currently high wip hence let run prod
**transfer jig BOMBA**
new transfer jig T3ET06 at G4 and saw the wafer can't transfer from teflon carrier "BOMBA 08/12 307" to pp carrier, but it is ok for other Teflon carrier,and this affected carrier can easily transfer at other transfer jig in same bay.informed Uyne to check the transfer jig and the carrier, informed opr not to use them. another carrier "BOMBA 06/10 002 "also can't transfer on the jig. Both carrier put beside scope.
Small tool owner,sun adjusted the height of pusher for transfer jig, 06 so that the H-bar of the 2 carriers wont hit the pusher causing difficulties during wafer transfer. Sup tested out the carriers with su yan.Pls help to perform test runs using the 2 carriers with T06 tonite.
Tried both carriers in T06 & T05 no issue w/ 5x load/unload, also tried other prod carriers no issue; can see enough clearance for wafer and carrier when transferring
Metal Thin issue happened
Brand new issue happened in machine , we term it metal thin.
history of the flow is so:
Lot load into MS02 by Sami, he went to washroom and informed piao ting to unload the lot when done etch. Due to miscommunication, she thought she was asked to take care of another bench, WB09. The lot was overetch for extra 7 minutes (Total 17munites) when she realized the MAS alarm.
Photo taken and compared with good wafer. Awaiting N3 to consult Device regarding the acceptable overetch factor. Informed etch sup, Eder to HIF the opr.
Awaitng monday to attend MRB before send lot for rework. After attended MRB, approved. Lot sent for freckle etch 520s, done & signed on runsheet. Lot adjusted backward to op 5931 for clean pre metal dep, Batch engr already checked steps on rework form. A/w MAS end bath to strip metal, freckle, to follow up
**Avago 6" TiW Etch**
Refer to email,find back TiW A & TiW B test wafer lots (5inch) in F2.Recycle 2lots to strip TiW (if haven't strip), Posistrip Clean (WB05), Sput TiW & Mask TiW (Follow Avago process step)
Not find at FD,TIW bay and engr racks already check- Checked all test wafers lot at F2 etch bays, N2 box & tables, only found 5pcs scribed with TiW B (#19,20,21,23,24), which sputtered with AlSi, measured all wfrs with 2um except #23 & #24 with 0.67um(used), wfrs placed at eng table with on hold slip.
2 lots find at T8,take back to strip TIW 2hours each at end of BL,at end of shift will strip done,to follow do inspection and posistrip clean,after posistrip clean to change chimical.If no residue seen,to sent to sput TIW/mask TIW(normal Avago process step).2 lot park at bay 1 table.
After posistrip clean, WB16 chg sol. Instruction form attached to sputter/ Mask TiW according to Avago prod process flow. Lot TiW A=24pcs, lot TiW B=18pcs.
Both lots done sputter TiW end of shift and passed to G2 bayleader Ewe, awaiting to proceed gasonic descum due to high WIP, to follow up status Collect data Lot set auto hold at 5068 etch ALSI move in and move out,this is the last lot which product from 5" change into 6",need to collect data before/after etch. To take some photos measure metal line and con hole at centre/edge 1 pc.
Midday issue 13 Sep
Start of week is monday blue. All issues are not easy issue.
**HARDENED RESIST**
4 lots at bomba dirty. Hardened resist like previous lots. All with 4.4um blue resist. All from bathcing59, as per masking, only tool for blue resist. All lots wash in 2EWB23, tanks slightly dirty. Wafers seen with hardened resist. Lots rewashed in bench and changed chemical. Aft rewash, wafers clean, moved out.
**AG PITTING**
Two lots hold for residue. Inspected the lots but defect look like Ag pitting most on wafer edges. Defect not able to remove aft rewash. Both lots keyed rework route and send rework. Runsheet taken out and stamped NCL already.
Inspected #25 suspect not residue since have defect on open & metal area but majority on metal defect seen around flat area. Defect at the 1st 4 rows from flat area and edges, unable to ink off as >4%. NCL stamped on r/s but in wso havent raise . Need to confirmed w/ qc op lot done 100% scan DF alr,mum lot vacuum sealed 2x
**RND lots:**
sample etch with both lowest EWMA and current ER if available. Don't move out till excel file completed in RND folder . Advance hold this device before etch boron base due to previously encountered lots having residue. When lots reach, to do 1.5min IPC and calculate etch time based on incoming thickness and current e/r (1.2 oef). Adv-hold msg: "KK913_WOE"
Weekend problems
Sunday the issues still ok, not too many pop up. But still a few from them:
**R&D UNDER ETCH**
**WET SINKER**
3 rnd lots underetch (whole lot - whole wfr affected). Measured remaining thickness ~1200A highest. Diff flow from std productions. All 3 lots from BRC4 (all 3 lots batched) . temperature check on monitor at 32.2C, measured at 32.5C.
Incoming thickness for 1st ox at 14833-14843A. No splits on template. ER for 8A: 1450A/min. Time on runsheet is 9m20s, based on calculation etch time should be ~11mins
Took #25 from batch3 to re-etch 1min, inspected ue cleared. Rest of wfrs havent re-etch. Lots hold for device to verify etch time
RESIDUE STRIP WET RESIST
BACKGROUND:
Whole lot seen with rsd on s/l. Buddy run in B04,BL: 29 PL1 inspected no similar rsd seen. Rsd seen randomly across the wfr but edges affected more. Severity <3%.
Initially lot max delay aft oven. , Buddy run/lot bfr/aft in no issue. Buddy run/lot bfr/aft no issue.
PWB1 CONVERSION TO AZ300T
----------------------------------
Refer to YS's email. Pls help to cascade to operator about change. Already paste the conversion checklist on the bench. Chemical outside the grey area lift.
Conversion checklist pasted on tool. PM tech Sel feedback they will perform pm and conversion at the same time. Surfscan passed at 9p. Temp set to 82+/-3C
Release all metal etch/tre process. 1st run, inspected 10pcs under DF, seen with slight particles in die and photodiode. Photos saved. Expedite lots to supr and adv hold at every step to inspect. Tool up for prod.
Problem never rest even on weekend
One bad thing about production line over 24/7 is problem never rest even on weekend. We have these problems as example:
Line issue over weekend:
**N+ LIFTING**
2 lots seen with lifting at alignment pattern only. Inspected die seen with slightly higher undercut but no lifting. WB8B ER: 1494A/min within control. All wafers affected. Discussed with mask engineer Juce, already advance hold at incoming mask USG for them to try to align as he cannot gauge now if can align or not. Common tool from Cell8, no downtime yet.
Lots already reached and informed G3 mask engineer. To follow up with her to check alignment. Informed Pudin to help check if any misalignment issue.
Done mask, no alignment issue feedback by masking engr. Lot was proceeded to etch, and hold after due to random oxide residue across wafer. Did not trigger adv hold at incoming.
Already asked to bake+ipc lot pls follow to try re-etch .Already advance hold this product at incoming to add IPC step. After IPC, etched both lots 30s but after etch inspect only mildly improved, still have random residue
**N+ OX RSD/LIFTING**
Aft N+, whole lot seen with ox rsd localised at lower LHS/RHS edge. Also seen with similar lifting on alignment as previous 2 lots.
Extracted commonality. Common tool is CELL8. Lot etched in day, ER :1552A/min. etch time is 9min20s.
Refer to excel summary. To hold the lot first pending discussion with masking. Do not re-etch 1st.
Advance hold this pdt at N+ move in to inspect incoming and sample 1pc. If similar issue seen, to send another 1pc for remask+std etch
**AU OverEtch**
BACKGROUND:
Operator hold for LM25 due to suspect Au **OverEtch**. Refer to photos. Seen whole lot affected, affecting LHS/RHS edge. Crown/flat ok.
Buddy lot in 1B no similar issue. Lot done in GS1. No mc down during running of this lot. Inspected 100%, not all wafers affected. Affected wfrs: #25/22/21/20/17/16/14/12/11/10/9/7/6/5/3 affecting the RHS edges only, not severe as prev lot.
The die saw pattern **OverEtch** for this lot is milder than prev lot. Commonality, GC1, for affected lot, no downtime for MRC. At Etch wet gold, 3rd run aft change sol
Buddy runs in MRC all terminated w/o hold. Lot was baked 5hrs after oven. see line is practising flat 90deg at AuTa etch
Started to map but if based on the die saw, only LM25 is have reject, the rest of the affected wafers, the die saw pattern still "curved". Pls verify further
Production undercut issue
Just when you think going to be a peaceful weekend, QC detected with suspect undercut. Inspected, whole lot affected with undercut on capacitor. Defect looks same for all 25pcs and is affecting the whole wafer.
Etch capacitor done in WB11 @er:878 A/min. No machine down. No abnornal delay in WSO. Lot ran before: W2543 already terminated without hold. after: W15V24 also no hold.
Mask capacitor have down time when running this lot in WW54 due to machine hang. Checked with masking engr but say the machine problem is vacuum, and will not cause overexpose on wafer.
Checked with G2 bayleader and G3 bayleader, they used the fix time for this prod for this capacitor lvl, which is 11m. As per G2 bayleader, if ER at high side, and used 11m will see high undercutut, thus they use 10.5min.
As per G3 bayleader, she always used 11m, she do see slightly high under cut for some device(she forget de device) but not as bad as the current affected lot.
Asked G4 Operator Wanita both uses etch timing based on incoming thickness calculation. Usually etch time is around 10.5min.
She was on OT tonight, she said all the while she is following the r/s timing (11min) for this product/level. When told her about the lot history on bench, she said "maybe" wrong etch timing used.
Checked the r/s for of the affected lot + other lots that she etched, nothing written on r/s. The 15.5min lots ran before and after the affected lot, already terminated.
Tagged WW1YK8 (same pdt), measured incoming thickness ~7200-7300 and took 1pc LM19 to etch 15.5min, the rest did std 11min. Seen higher undercut for LM19. Refer to photos. Adv hold this lot after metal etch to check.
Pls help to follow up HIF w/ JJ since delay time from etch to semitool very long ~51mins (13:47-14:38)
Interviewed JJ with Leng, accor to JJ, she followed procedure and sometimes 11mins can see slightly bigger undercut thus reduced time to 10.5mins. She was also the one who inspect the lot aft etch, to her the undercut is still acceptable thus move out the lot. She recalled that she changed the timing to 11mins for WB11 when she wants to etch the lot. The 51mins delay frm etch to srd is due to her break time, no one to transfer the lot to srd
Help tag same prdt after metal etch strip resist/freckle & take photo for similar structures to compare undercut . Previous tag lot WYK8 now at clmode after etch contact,to take photo after ewmetal strip resist & to compare undercut of lm19 & lm13 . N3 plan to discuss w/ device on this issue
Mid day issues
**H2O2 spike error**
We found cannot spike H2O2. Check H2O2 vessel, full level.
Check and found main H2O2 solenoid spoilt. Replace solenoid with spare from 8th floor office, ok. Test spike H2O2 for all tanks, ok.
Found qdr reclaim valve leak, replaced new valve. Also found solenoid cda tubing water inside. Check found h202 vessel N2 pressure valve leak. H202 leak to solenoid. Replaced the N2 pressure valve.
Purge all tubing. Took out solenoid for clean. Fixed back, test found h202 main valve also leak. (h2o2 go to cda tubing). Await for spare.
**AVAGO STAIN ON PHOTODIODE**
Lot detected by QC with stain in photodiode, inspected random few pcs only found random stain in photodiode DF/BF, other area/oxide no stain, ref to pic
Last lot etch in WB14, lot bfr W37 at 5406 FINAL ANNEAL, no detect of stain inline, to inspect. Lot prev encountered max delay in WB05 & released by bacthing eng to rewash during sputter pure Al, no other related history found. To inspect wfrs & verify if stain found in other area
Inspected not all wfrs affected. Stain/rsd seen only in photodiode. Affected wfrs #14/19/07/25/05/18/21/24. Defect easily detect under scope, at random location.
Inspected lot bfr, W33 aft anneal, no similar defect. Buddy run in DFRF4, W015 - terminated, W16 at qc no similar defect. #24 sent for gasonic, inspected stain remained in photodiode at same location, ref to pic taken bfr/afr gasonic. Prod gross die 7517
Pls help to continue mapping,lot at F2 table. Lot is inspected by Ew155 SAw. Inspected Ww18 with F2 TiW etch bay scope udr 20x, stain on photodiode can be detected.
#10 done scruber w/o brush (pg1), no improvement. #16 to send for x-section. FA alr submitted
Machine Arm hang half way
Machine Arm hang half way
Feedback by operator Alifu, when he proc wafer initially, there was no abnormality, timing and recipe was correct, he returned when the alarm sound was heard, signalling the qdr process end but instead he saw the arm on top of chemical tank.
Then lot on RHS (W3) of arm is inside the chem tank, Left hand side lot (W58) just hanging on the arm. he pressed stop process but the machine no response. He called maintenance & move arm manually to qdr side, but alignment/position not good.
We take lot & put in qdr, started manually -use carrier handler & took W53 in chemical tank & put inside qdr as well.
After qdr sent for std srd, on hold pending disposition. check machine/estar do not have record of any alarm. once maintenance reset the tool, the arm immediately recover. home position & chemical/qdr positions no problem. Maintenance running auto-qt to further test & chk if will recur.
W53 advance hold at etch met to perform sample. W58 hanged re-prime 5mins and advance hold at etch met incoming to perform sample.
Maintenance checked arm alignment at acid tank slightly out & adjusted encoder position .Previous data, QDR -480,0 ACID-7676,0 adjusted, -481,85 -7660,85.
Monitor so far so good.
Vacuum pump pressure not reach setpoint within time
Vacuum pump pressure not reach setpoint within time
Tool encounter vac pump pressure not reach setpoint within time. Checked all testwafers inline all have oxide. Measured oxide is ~10K.
Good results come from the one sputerred at Tool 2 with 3.2um (ER=.89) and the one from prod ER tw BOX2 (ER=.88). The other 2 wafers, was oe on edges.
Currently still have 1 wafer good ER tw left (Tool 2,3.2um). To use later for scheduled ER check. Or choose from prod BOX2.
Withdraw 1 box of fresh bare Si test wafer, now ongoing grow 10K oxide in RG2/sputter in Tool 2 and Tool 7 with 3.2um .
Also, took 5pcs of standard BHF Etch rate test wafer (10k) and sent to Tool 7 to sputter 3.2/3.3um to retry ER check later.
Afternoon took Etch rate from Tool 2 test wafer ,done er 0.86. checked 1st lot. Continue run prod wafer.
1 lot fresh test wafer, grow 10K oxide in FR02 in previous shift. Already send to sputter 10 wafer in 2, another 5 wafers in TOOL 7 with 3.2 alsi recipe. Remaining unsputter test wafer to be kept besides SP07.
BHF ER test wafer done 2 wafer sputter in Tool 1 and 3 wafer sputter Tool 2. All wafer kept besides sp07. And request operator to do er on fresh test wafer sputter from Tool 2 AND Tool 7 each.
ER chk using 2 wafers - from Tool 1/Tool 2. Pre Tool 2: ave 3.26, PST: 0.54, er:0.98. For 3sec51, pre: 3.24 pst: 0.237 er: 1.0, seen with oe at the edge. Now inform operator to use TW frm Tool 2
ER chk for TOOL 7/Tool 2 done. TOOL 7 seen with same uneven edges hence unable to measure post thk. Tool 2 er:0.92. no ue detected for production lots. Informed operator to do 1 lot er tw for Tool 2 but not able to find any dummy/tw.
Sent 1 lot of TW to strip metal and to dep at Tool 2. MAS prod rcp #03 changed agitation from 10s to 15sec however no lots to run. 1st run to inspect for any oe . Check 1st run W332 timing ok,no oe/ue.
No head no tail
We came to some issue that unconclusive.
Taping relog due to same defect as prev case, photo taken - Pattern distorted at dice edges. Peding investigation and pulled commonality.
No common tool for wets side. Only common tool is Washing1.
Verified the 2 lots, only 1 wfr each affected. Batching feedback another 2 lots similar defect, bth detected at Mask. All seen with straight line "stain" defect under macro view localised from 1oadlock to wafer center, 1 wfr per lot only.
Previous also another lot 1 wfr similar defect inked off by diff. Our prev lot inked off also similar defect and location thus tats y they hold for us.
Also checked AW5P4 similar defect (now at backlap), prev at Mask. CTT on hold for masking but was waived by diffusion
When inspected under scope, defect affecting at base opening like chem seep thru, base pattern not distorted. Suspect could be resist not adhere properly at base opening area, like sc b4 mask base
Commonality points to IP01 but no correlation to defect type. Mask/Etch/tube all diff tools. Checked with diff, 1st ox measure 1 wfr post thickness at slot 1 for both center/source.
Estar shows no abnormality/interference. Wfrs running flat down, if susp stain dripping, it shld not be coincidentally at the same particular area, most prob like prev stain issue, patchy
Also the current lots no AVI scan at base level. No agreement from diff/mask. Pls help to send email for the 4 lots and relog to 1st owner. Keep device,and MQA in
copy
All the affected wafer 69 wafers but our tool ran many product.but so far no other product affected. Discuss with masking engr,and pls relog to 1st owner.
Mid day report for new week
As monday morning, we need to update the status weekly report:
Metal 2 ALCU strip wet resist
Lots detected with bubble like defect at wafer edges during inspection (slot25,LM#04)-> at wafer right edges. #04 was scanned previously during Batching like AVI not capable of detecting this small bubble defect.
Slot25,LM#23)->at wafer edges. #23 was scanned previously during and passing with low count. Defect look like underlayer resist bubble/water stain happen during poly/contact layer. Shadowish contact rainbowish colour observed.
Concurrently, there is one lot failing AVI OOC at CONTACI (#11/12=~39/38 count) with similar bubble stain localised at wafer right or left affecting 1 dies. Please help to perform AVI 100% and send lot for rewash to see if stain can be removed.
Both lots released and advanced hold after Etch wet tantalum to perform 100% AVI scan to see if can detect.
Please help to tag current 6 lots from poly level to scan 100%.
100% scan LM08-19 seen with same defect at the LHS/RHS edge. Lots sent for rewash but no improvement seen. Lot still on-hold . <5 counts,no similar map all moved out.
Advance hold at Pad to check if similar defect as WMK lots -> done scan 100% but cannot see the bubble defect.maj ul particles and false cnts.move to monitor yld as inst
MAX DELAY TRIGERRING FRM BOE TO TMAH
- If max delay>15min, error msg Peed001 Max delay exceeded will prompt out.
- to perform sample rework. If sample ok, to KEY REWORK ROUTE for other wafers
- NO MANUAL REWORK ALLOWED. To re-read the specific task and acknowledge.
Additional eSTAR alarm activation for BOE process . Estar will be activated if:-
i) Operator did not attend tool within 1min aft lots finish process, lots will
auto hold
ii) Operator did not unload lots to tse within 1min, lots will auto hold
Abnormal pattern on test wafers
We need to urgent activated the line support for test wafers mixture.
George prepared 7 wafers bare Si wafers placed beside PSE, which to send for standard SiN. Mask Via(any product - to take note of what pdt)
Need to Etch Via (To etch using fix time; normally use endpoint. To ask dry spin what is the average etch time used for and add additional 20-30s. Ensure that it is
etch to clear)
Sputter std Ta/Au as per the pdt flow. Need to etch in 3B-> 10mins -> inspect FS + BS(see roughness) with photos. Etch 65mins and inspect both FS+BS to check for any Au seeding
Mask Via done, sent for etch via fix time 60s, inspected wfr via cleared, sent for gasonic and sts. Wfrs passed to A4 operator to sputter.
Etched 1pc 65mins in GSEB4B (PG2) observed did not have any bubbling while processing. Opened the door at the 1st 5mins, 30mins, then last 5mins of the process but did not see any bubble. Check wafer did not have any black particles (but got lot of scratches & debris like defect). Etched another 65mins & inspect again still no black particles seen.
Sent the same wfr to 5min BOE in 4A+65min TSE in 4B. No bubbles observed during the 1st 2mins of the process. Seen with Si attacked at the backside but FS no black parts
2 wafers sent to Mask Pad (manual align). 5 wafers placed back to back (including prev wafer etched) for 5mins BOE+65 mins TSE. Inspected the 5 wafers backside Si attacked but FS no black parts.
2pcs etch pad done.to confirm simulation to be done. 2 wafer which done pad etch sent to strip resist.
Pls verify location of wafers or of already done sts as no handover from G3 according to G4 operator. 2 sample done strip resist and vac seal. keep on rack.
Draw another 2 testwafer lot to follow CSPPUM6E-6F product flow. 2 lots passed to batching to dep passivation with insruction form. Please hand carry every steps.
Line chronic issues that need followup
We have few chronic issues that need constant followup, and today I updated the status as below:
**Under etch issue**
Dry etch feedback whole lot seen with ITO ue. Inspected only 7 wfrs and as above seen with the circular signature around the edges of wfr with slot #14 most severe.
Remaining wafers only seen with the whitish ring at the wfr RHS. Checked under scope, brown pad seen everywhere.
Buddy run inspected good. Lot run before, tagged back to inspect also good.
Inspected under DF, photodiode no rsd. Etch Rate check before run at 344 and 355 (at edges). Talked to Hyini but do not want to accept back lot. Pls help to send email to them
**Black Particle issue**
Lot 44/45 hit with severe/mild Black Particle. Rework done and moved.
Both tools released. As per Engineer, to check severity. If severe, to ground tool and inform him. If BLACK PARTICLE still mild, to continue to run.
Took the Au Etch Rate wafer prepared to check ER in 3B(2mins), no loss observed. Pre ave: 9279, Post Ave: 9278. Wfr placed back in the blue box slot 25. To etch using 65mins and measure pre/post RS reading. To check backside (w photos).
Au Etch rate wafer run 65mins no loss.pre avg:9266,post avg:9262 A . Another 7 lots hold for black particles, To chk lot list in common drive. 2 lots severe lots unable to clear particles even after 2x rwk. maybe need to dip
Another 5 lots done rework & inspection no more.
Need to ask TA to change chemical for 3B and run. 4B to remain down. 50/65mins process to run only in 5B at BL 10-50.
Mid day report Sep 2
**Mid day report to boss**
Batching feedback incoming seen 3 wfrs with haze #7/16/19, wfrs at 1F Tube. He feedback he split out the 3 wfrs from mum lot. Previous oper is Tube2- LMODE.
From scope, seen the 3pcs with stain like defect, under scope got crystal/oxide residue-like defect. Unable to confirm if can further re-process,ask engineer also unsure.
Lots Sent to rewash but no improvement. Confirmed other wafers w Epi, no stain seen. No recipe in AVi to scan for ink off. Raised req form to create rcp 2x but encounter some error when submit req. Pls help to bring the wfrs to AVI bay, lot at 1F Bay, or just manually ink off from elogbook if still cannot submit request,.
Inkoff in elogbook then passed lot to epi engr.fb he will sample grow to chk if stain will not get worse or cause other abnormality.
---
**Washing tool issue:**
MS removed N2 nozzle and adj angle to potrude .Ss passed using Via rcp, no stain
But encounter resist residue: Defect decrease across slot, meaning slot 5 most
severe whereby slot25 lesser.
Current flow rate: T2:12LPM, T4: 10LPM. Requested MS to reduce flow rate for T2 to 8.5LPM.
Lot reviewed aft scrubber/gasonic counts reduced
Hit 2 lots with the same water stain mark issue. After re-SRD and rescan cleared. Tool down for MS to check.
MS adjust door & increase n2 purge, via ss 30000p, mid 7000, start. Re-edit Via rcp for step 26 dry step frm 1-> 8mins, total dry time 15min. K6: 14:35mins. Observed at step 26, chamber alr seen with stain but actually slightly better than before.
Supervisor inform MS to drain IPA tanks and run again the edited recipe.
Some abnormality in tool
Today we found some abnormality , and after performed the check, result as this. Incoming lot found broken after opening box to align wafers by Operator. Incident witnessed by BL. Suspect both wafers already broken even before opening the box as OP slowly open the box and already learned from previous similar incidents. It is impossible to be broken during LM operation as wafers already been transferred to wacker box and silicon debris are all in the carrier and box.
Possible root cause of wafer breakage:
1.LM OP closed the wacker box forcefully.
2.Transporter see the box mildly open and close forcefully
3.Batching operator during wafer planning see the box mildly open and close forcefully
4.Batching operator open the box fast and 1 wafer stucked on wacker top cover groove and when put back broken. Aside from LM#15 and 16, LM#14 has high defect count due to scratches induced by LM15 onto its wafer surface.
23 wafers already completed EPI growth including LM14 w/scratches. Merge lot and hold under 240WH [SHOM] to investigate. Since G4 SHOM didn't HOLD wafers in WSO under G2 SHOM right away after find broken wafers, so G2 SHOM, re logged to G4 SHOM as owner.
ENG PROJECT/EXPERIMENTS:
Deployment effectiveness:
1. Pls audit the black box usage and compliance, there is observation that some
opers are still using etch scope for post bacthing inspection.
2. Pls feedback the FL conversion bay leaders give results after our deployment.
Next week we will continue and start:
1. Fixed dope excel deployment to BL
2. N+ / 2w / FL TR alignment
Projects status up to date
Today we had a long meeting for progress of the whole year projects:
• **87 Qualifications remaining** :
• On time execution + analysis & anticipation/ back-up plan -> missed out items lessons learnt (qual task, scopes, semitool link missed out etc ..)
• **G4 closure**:
• PR event link / flexibility follow up
• Rack ID change on time
• Staffing model & training anticipation
• **6 Tools shutdown on time**
• Line audit
• Shutdown or decontamination checklist
• On time execution
• Resources allocation to the HYT consolidation Project after organization merge.
• Daily PCRB
• SWR turn at 10.5 : Module Project owner & Production : To focus & escalate
• T4 closure delay (min 1 week) -> (Alloy -> LTO)– Midas Detector Quote/PRPO/Installation late. SWR done.– (*Lack of Focus).*
• GDSTRE01 (Paddle SD-> Nzone)/ (Lydop) / (Pzone) – Gain higher than std. Device requested to continue with EWS results. – *(additional iPCRB request)*
• 88B2 (Boron ->Pzone): JU7 re-trial (2 wafers) pending WIT1, with retrial of U87 sample and full run, *(Technical issues)*
• CHC2 (Oxidation -> Pzone) : To follow the qualification similar to 765B2. 2 wafers to send as trial & 9 trial wafer to send for SRP testing > *(additional iPCRB request)*
• 5URA1 (SD-> Nzone): WIT1 OBSERVED ISSUE . PENDING WIT2. Device request for EWS instead of wit2 (as gain seen higher for swr lots) – *(additional iPCRB request)*
• 4YS3 : To start qualification in W52/W01 for Pgm 15 thin oxide measurement – *(Lack of Focus / Change priorities)*
• JDUY4 (Poly-> NiTi) : STACK MEMORY FULL (NEED TO REMOVE SOME OBSELETE RECIPES FIRST) - NI SINTERING, NITI SINTERING SINCE ONLY W03 INPUT NITI *(Lack of Focus / Change priorities)*
• UYP07 (PiUYha) : Delay in initial hookup of auto-fill due to PRPO > concurrently delay in release for 4IOP07 conversion. *(Lack of Focus)*
• 1YTE1 (Nanospec backup): Focus on the above tube qualifications > *(Change priorities)*
Smooth problem resolved
Today is quite productive in a way - many issues able to resolve one by one:
SPC HOLD
================
Tool spc hold. MGGGW625 step back to remeasure. and 2EWB22 done ER. Uv tool down while measuring er wafer.wait for uv tool to up to measure er.
Lot step back and remeasure thickness within ctrl. No issue with uv tool when observed, compared with nanospec thkness comparable.
ER high, temp verified actual 22.5 bth edge and center. Display 22.5C. Flow rate 13Lpm vs display 15Lpm (within spec). Checked ER using wfrs frm 3 boxes, all high but from one of the box, pre thkness 900+, ER slightly on the high side, 214 but those pre thkness 1000+ or <800A, ER alr out. Alr informed optr to use thkness arnd 900+A. MS found no issue with tool, ER rcp used correct(rcp 1:45s). Also towards end bath, req to change chem, and chk ER within ctrl on high side at 211. Rls tool to monitor. 1st run thickness verified using nanospec within ctrl
ECD LOW OOC
-----------
RTK604508 have LN oos for LM03/LM20 pt7. ~44.1 (spec: 49.5-52.5). Checked, no focus issue when measuring pt7. Lot ran in JEWB3B, partner lot is RE04423 no ecd issue.
Ave LN ooc ~~50.0, but still within spec. MCD~~54.5 on target, delta ~0.33 no drift as well. Measured the LN using QC scope for LM03/LM20 at R01C26, LN is 51.56/51.43 respectively. Stepped back lot,Also done 100% already haven't seen data yet.
Lot done 100% cd measurement. checked no focus issue. still seen 9.07 for slot 3/4/5/7/11/22/23 .this 7 wafer pt7 with -9.07.again check lm 03 and lm -4 under scope, pt 7 reading with 51.03 -51.49. Photo taken.no scratch seen on wafer surface.
From 100% data, pts 7 getting -ve value, under scope no abnorma;it with slope pattern at pt 7. Step back to remeasure, for LM3/12/20, no -ve value, also no issue with ivs focus. All data within ctrl. Photo taken of ivs measurement data. Lot moved
BS RING Probelm
Lot on hold by prev shift due to bs ring. Checked that lot was on hold for masking but disagree. Checked most severe is slot 25(#02) and slot 5(#01) but still within the 3mm width criteria. Moved. Buddy run in 3B, no on hold for bs ring. BL:126
Lots not adv hold and is under prod. At etch wet n-buried, event ASRHwB33AM
keyed and no timing shown. Prev rnd lots launched using 19mins to etch at this lvl. Need to highlight to device engr abt this issue.
Smooth start of the week
Today a smooth start of a week, line issues are manageable:
Chamber Pins moved UP TOO Fast - Right Chamber wafer broken when clearing wafer
calibrated lifter pin speed replaced the lifter pin cda regulator while clearing the prod wafer, had an alarm of robot extension error, robot pan was extended on right chamber hitting the slanted wafer on chuck suspect wafer move/slide while calibrating the speed of lifter pin unable to clear wafer due to lifter pin alarm
after clearing the broken wafer and fixing lifter pin issue, cycled and checked
surfscan.
Both chamber failed so cleaned the loadlock and shuttle area. We checked again the surfscan, left side still failed @ 11/19. Right chamber pass.
We checked Left chamber Surfscan passed at 14/1. Also checked b/p and l/r passed.
run seasoing wafer. no alarm
Back chamber lifter pin move too fast alarm First, recover the wafer, checked lifter pin up/down speed is 0.8/1.6.
Next tune lifter up/down speed and run seasoing wafer no alarm
Observation Alarm 4626 THROTTLE VALVE SYSTEM ERROR.
Secured throttle valve wiring cable & Tighten throttle valve and baratron wiring connector. Check bp lr passed.
Test run dummies wafer monitoring no issue.
Throttle valve system error so we clear production wafers and reboot the tool and secure throttle valve interface card.
Verify bp & lr ok . Ran few lots of dummy wafes, no error occur
Engineering problem during weekend
Engineering issues that we faced last 2 days , Too many until the width of page may not enough to write:
1.) 8" BRE6: Residue Issue at EM layer (LOT # START WITH "C")
Do not rework any BCD lot for resist residue until further notice. If AIT
send for rework, pls inform PO or hold until morning. Refer to email sent by Kelvin regarding rework. AIT will hold the lot if residue counts>30 and etching will manual rework.
If resist residue <30 counts, lot can move as per normal. Device will automate rework route soon and future we will rework by WSO. Feedback if information is not clear.
2. MG40 OOS lot disposition need full lot measurement:
Many of you may know MG40 OOS lot disposition need 100% measurement. If any
lot encounter OOS for CD and thickness, pls sort wafers and do all 25 wafers
measurement before move lot to next level. We may have more OOS due to contact ROX bring down. I need team help to do 100% measurement if ROX after metal etch showed OOS.
As for OOC, no need to do 100% measurement. AFTER Remeasure , Please send email to Ehah. Pls adjust step to remeasure good wafer before move out the lot.
3.Miscellaneous:
18 inch wafer random scrap, please put details in WSO and SMY:
- which tool ID
- which wafer ID
- what happened
- process actions if have any
4. For R4 lots incoming for this prd HHYS(hold for PO), pls execute:-
- Sort to LM# before PROC
- Proc ONLY 1 lot in HYE4 no buddy lots allowed.
- Note: Loading Sequence: FR-SEQ
5. ARAM801 & ARAM802 POST PM RELEASE PROCEDURE
+++++++++++++++++++++++++++++++++++++++++++
a. Post PM only can run RSW lot only. Metal2 recipe to link >15 RF hours.
b. If there was no wip for RSW, pls do resist seasoning to reach RF counter
up to 15. (You can withdraw wafers from wfer maangement).
c. When tool is down for PM or maint team feed back for "open chamber" .
Pls delink metal 2 recipe.
If chamber B (PM/open chamber) delink PTR
If chamber C(PM/open chamber) delink ATR
