From mcvittie at snf.stanford.edu Wed Sep 1 10:34:55 2004 From: mcvittie at snf.stanford.edu (Jim McVittie) Date: Wed, 01 Sep 2004 10:34:55 -0700 Subject: Comment pquest SNF 2004-09-01 07:48:02: Chlorine pressure Message-ID: <4136083F.9C123678@snf.stanford.edu> Elmer, I suspect that my request to Mike to get the CL2 line back down to 10 psi, where has been for a long time, resulted in the regulator knob being turned too far. Since it is a long line and all the etchers use relatively low flows.it took a while for the line to drain. I just checked the line pressure, it now at 12 psi. Lets stay at this pressure and see if the GaAs users can ran their processes. If the minimum flow is still too high, we will have to go back down to 10 psi. By the way, I have contacted MKS, who made all the MFCs for the Pquest. I have asked them about getting a procedure for adjusting the control valve on the CL2 MFC so we can get the valve to close more and get a lower minimum flow. Yesterday, I did talk to an etch engineer with lots of Cl2 MFC experience. He thought that running at 10 psi was fine as long as we can make the needed flows on all our etchrers. The fact, that we have been running with a line pressure of 10psi well over a year with no CL2 problems says that we should be ok. By the way, I checked the CL2 flows in the main recipes on the other etch tools. They as follows: Lam poly etch -- 50 sccm CL2 MFC with max process flow of 40 sccm P5000 - Poly -- 100 sccm Cl2 MFC with max process flow of 20 sccm P5000 - Al -- 100 sccm Cl2 MFC with max process flow of 30 sccm PQuest -- 43 sccm Cl2 MFC with max process flow of 20 sccm Note that only one of the P5000 processs chambers is run at a time. This means that our max CL2 flow from the source line for all our tool is 90 sccm. I will check that we can deliver the max process flows to all the etchers at the same time with no problems. Jim -------------- next part -------------- A non-text attachment was scrubbed... Name: mcvittie.vcf Type: text/x-vcard Size: 422 bytes Desc: Card for Jim McVittie URL: From mcvittie at snf.stanford.edu Wed Sep 8 13:03:29 2004 From: mcvittie at snf.stanford.edu (Jim McVittie) Date: Wed, 08 Sep 2004 13:03:29 -0700 Subject: PQuest Flowmeter Problem Update Message-ID: <413F6591.B7A97B6A@snf.stanford.edu> CL2 flow meter/controller on the PQuest has three problems: 1. The control valve does not fully close at low flow settings. This means that for a given line pressure, there is a flow for which we can not go below. Normally, this is not much of a problem since we usually do not operation flow meters at the bottom of their range. However, the Pquest is the exception. Most of the GaAs etch processes use Cl2 flows below 3 sccm. The flow is kept this low so that the processes can be run at 2 mT of pressure. Note that the Cl2 flow meter has a maximum flow of 43 sccm of Cl2. To get Cl2 flows down to 1.0 sccm from our flow meter, we have to operate at line pressures at or below 10 psi. At these low pressures, we are sensitive to the Cl2 flows to the other etcher on the Cl2 line. In particular, both the Lam and P5000 etchers share this line. When either the Lam or P5000 start drawing Cl2, the Cl2 line pressure does drop a bit but it does appear to be able to deliver the required Cl2 to all the tools. As expected when the Cl2 pressure dropped, the minimum Cl2 flow from the Pquest Cl2 flow controller also drops. This is not a major problem, if the flow setting is above the minimum flow. However, if the flow setting is below the minimum flow, the flow will also drop until the flow setting is achieved. This last point is presently an issue with the Pquest because of the offset problem described below. 2. On the Pquest to adjust for errors in the D to A converter, there is an adjustable offset for all the flow meters. Unfortunately, all the offsets on the Pquest were set incorrectly some time in the past. I think a former technician got confused with the setup procedure. For the CL2, the correct offset error is 1.8 sccm low. This means that a setting of 1.0 sccm actually called for a flow of -.0.8 sccm. Currently to get a flow of 1.0 sccm of Cl2, you need to set the flow to 2.8 sccm. We plan to correct all the offset errors later this week. For now for all the flow meters, you need to adjust your flow settings to give desired readout flows that you want. Except for a small error for the Cl2 flow meter, all the read outs for the flow meters on the Pquest are accurate. 3. Finally, there is a gain adjustment error the Cl2 flow readout. Using the rate of pressure rise procedure for measuring flow rate. I have determined that the chamber volume is close to 75 liters and as stated above the flow read outs are accurate except for a 13% high error for the Cl2 flow. The correct Cl2 flow is 87% of the readout number. I expect this error has been there for sometime. For now, I am not going correct this error. You will be informed when we make is correction. Thanks, Jim McVittie -------------- next part -------------- A non-text attachment was scrubbed... Name: mcvittie.vcf Type: text/x-vcard Size: 422 bytes Desc: Card for Jim McVittie URL: From scaccag at stanford.edu Tue Sep 28 15:43:07 2004 From: scaccag at stanford.edu (Luigi Scaccabarozzi) Date: Tue, 28 Sep 2004 15:43:07 -0700 (PDT) Subject: Cl2 flow calibrated Message-ID: Hi all, last friday, the Cl2 (and BCl3) flow was calibrated. We (with Elmer) reset the offset of the Cl2 flow, so now you get what you set. The minimum flow should be around 0.6-0.7 sccm depending on the pressure on the line, so, to have repeatable results you should set a number > 0.7 sccm, or better > 1sccm. We reset also BCl3 and Ar11(the lower one), then we ran out of time. Unfortunately I realized later that everybody uses Ar1 (the upper one) and not Ar11... Anyway, Jim McVittie said that he would check again the flows and calibrate them. Please Jim, can you send an email what the job is done? Thanks Gigi From mcvittie at snf.stanford.edu Tue Sep 28 18:00:14 2004 From: mcvittie at snf.stanford.edu (Jim McVittie) Date: Tue, 28 Sep 2004 18:00:14 -0700 Subject: PQuest Flowmeter Test Results Message-ID: <415A091D.8CC57488@snf.stanford.edu> All, See attached Word file for flowmeter calibration results for PQuest. I used rate of pressure rise method to measure the flows for the Cl2, BLCl3 and Ar MFCs. All the MFCs were zeroed by Elmer on Monday. Jim -------------- next part -------------- A non-text attachment was scrubbed... Name: PQuest-MFC-Cal-9-28-04.doc Type: application/msword Size: 163328 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: mcvittie.vcf Type: text/x-vcard Size: 422 bytes Desc: Card for Jim McVittie URL: