November 2014 Survey
Survey introduction
The November 2014 survey was the 11th survey carried out using a format where individual cleanroom users were asked about their view on various parts of using DTU Danchip's facilities and in particular the cleanroom facilities.
The questionnaire was send to 334 registered users (based on the logging of cleanroom use) unfortunately 13 of the e-mail adresses was obsolete and thus the mails bounced, which brought the total number of potential answers to 321 . In total 74 persons answered the questionnaire which means an answer fraction of 22% regardless of whether or not the bounces should be included in the total.
This page deals with the comments given as part of the questionnaire. Danchip values this feedback highly.
Comments
The comments appear here as they were written in the feedback field of the questionnaire. In a few cases they have been anonymized or the worst spelling mistakes corrected.
Comment 1: Working a lot with the e-beam I have experienced several times that the load team did not arrive as expected. Everyone have, however, been very nice and helpul when I found someone to ask for help, so it seems like a small coordination problem - and I think it might already have been taken care of
Action: We will investigate this issue - "no show" of the loader team should not happen
Comment 2: I would prefer to have more readily available TiO2 deposition using the Ionfab. The machine is prone to breaking down, and spends most of its time in etch mode.
Action: Working on loading issues with Oxford
The deposition performance is less than satisfactory and is in our perspective, due to the fact that the machine is primarily an etching tool with a deposition add-on.
Comment 3: It is becoming more and more cumbersome to use the computer terminals in the cleanroom. Everytime I want to use my email I have to set up my account, and wait for outlook to update my inbox. It takes up to two minutes to check email in this way. As a result I now use the web based outlook instead. But this is not as convenient. The citrix solution is also very slow. transfering files to the cleanroom drive (U) was very fast and convenient. With citrix it takes 2-3 minutes before I can start transfering my files.
Action: Several projects between Danchip and AIT
We are aware that the computers are running a bit slow, we are working together with AIT to find ways of improving this.
Comment 4: Lately almost all delays in my fabrication have been due to equipment availability, never due to poor performance or anything else. This is mainly a problem on the D-RIE Pegasus, Sputter system lesker and Electroplating Ni.
Action: Evaluate need for extra capacity
Use of Danchip has increased over the past many years which means that in general tools are utilized more than previously. Danchip keeps track of bottlenecks on individual tools to gauge when investments needs to be made in more capacity. At the moment the Cryofox can supply extra capacity for the Lesker for plating bases. The ASE and Metal ICP can provide extra capacity for the D-RIE but we do not have a tool which can give extra capacity for the plater yet.
Comment 5: Mine topkarakterer kan I jo ikke bruge til andet end "feel good" men det må I så nøjes med.
Action: Thanks!
Comment 6: I have only experience with XPS, it was indeed really good in the beginning of this year but these days the results are, in most of the times unexpected. I think we need to make sure that the instrument is as good as it should be as running a standard samples once in a while (I know, this reduces the life time of the X-ray gun, but can't help).
Action: We will look into the matter
Comment 7: When you get training in Danchip you get training in a process and not in a machine. I would like to get more info about the machines when I am trained, how and why I should alter some parameters, instead of just using a "recipe". I know that using the recipe is faster and more efficient for a process, but if you need to change your process later it's best to know more about why you are pushing the buttons.
Action: No immediate action
We understand and appreciate your interest. Feel free to contact the relevant process specialist for deeper explanation and reference to literature.
Comment 8: The equipment I have to use has been broken since 2 months and will not be fixed before Christmas, I don´t think this is something that should happen.
Action: No action
It is an unfortunate fact that sometimes our tools break in ways which take expert assistance and/or long time to fix.
Comment 9: Generally pretty good.
Action: Thanks.
Comment 10: In general I am very satisfied with the Danchip (staff & organization), however, the performance of the machine we are using (XPS) is unsatisfactory, see example below: XPS machine is often giving very unexpected and uncomprehensible results. e.g. Carbon is often present in huge amounts although the sample should be absolutely free of it. Maybe exp. chamber suffers from unstable samples.
Action: We will look into the matter
Comment 11: The Danchip personal and the facilties at DTU Cleanroom are very professional. I do think through, that there should be more follow up on how the users get dressed. I think that some shoes are falling down, and sometimes I see hair sticking out from the hoods.
Action: Encourage everyone to help keep the cleanroom clean
Keeping the cleanroom clean is a joint effort between all users of the cleanroom. So if you see someone not dressed or acting properly, you can help by telling them in a friendly way.
Comment 12: Some users use SEM ZEiss to look at structures that are in the range of microns. They should be directed to SEM LEO, to avoid overbooking on the 2 good SEM.
Action:
It would indeed be nice if people would choose the appropriate tool for the task, but we do not control people's work to the extent that we can choose the tool for each user process. When it comes to SEM's the users should also consider using the Cen facility instead of Danchip.
Comment 13: It would be nice to choose from which equipment you get email notifications when a status change is made in labmanager. It is a little bit annoying to recieve 20 emails a day with information that is irrelevant to you. The machines you have had training in could be required and others optional?
Action: This exact feature is implemented in the next version of LabManager as announced at the user meeting
Comment 14: The list of equipment in service or limited use seems to be ever growing. The down-time of the equipment is by far the main source of dissatisfaction.
Action: No action
We keep a close watch on tools being down and for the past years we have had an average uptime of more than 85%. This is an area which is very important for us and which we will follow closely so we do not agree with the list growing.
Comment 15: The general lack of XL cleanroom suits is very annoying, and it would be nice if the shoes with buttons were exchanged with those with a strap.
Action: We will investigate this further
Cleanroom suits and boots are rather expensive so they are only replaced when worn out.
Comment 16: Very professional, and friendly staff with a high level of knowledge and compentence as well as customer care.
Action: Thanks
Comment 17: Henrik Nielsen at thinfilm shall have a big thank you for making my work at the Physimeca more easy.... :-)
Action: Thanks!
Comment 18: In some time period the Pegasus is too heavely booked to make processing "on the fly". I mean, when starting the week, the machine is already booked entirly. To jump in quickly in some empty slots is not possible. Sometimes the same counts for the SEMs too.
Action: No action
We do keep an eye on bottlenecks to see if we need more capacity. The cleanroom is generally open 24 hours a day and for very popular tools we recommend using the tools outside normal working hours.
Comment 19: IT is much more difficult to use now. Many would like to install Google Chrome but it is being deleted frequently. SEM availability is very limited for companies since we are the only to pay for booking. Access to SEM images is very problematic. Hard to get it to work in practice. Missing an extra level between "Satisfied" and "Unsatisfied".
Action: IT issues are being worked on
SEM's are very popular tools, but please realize that DTU management has already paid for all DTU users.
Comment 20: It would be nice to get training to new machines from someone that has actually used the machine a bit and knows a bit about the process so that basic questions can be answered - beside introduction regarding simple machine operation (as which button to press when).
Action: No action
We put an emphasis on releasing tools fast and allow users to develop their processes.
Comment 21: I would appreciate very much if the Ionfab300 was in deposition mode more often. Would it be possible to get black wafer boxes for 4 inch wafers? I often have samples that need to be protected from visible light.
Action: No action
The deposition performance is less than satisfactory and is in our perspective, due to the fact that the machine is primarily an etching tool with a deposition add-on.
The blue waferboxes should be completely opaque for visible light.
Comment 22: The performance of equipment is too low in the particle contamination. Making particle measurement in the particle counter part of the QC would in my mind improve the performance. This should how ever also be done for equipment like the RCA and maybe even for the resist spinners.
Action: Input to QC evaluation
The particle counter was acquired to help us quantizise particle contamination. We are using it for some of our QC processes already and evaluate whether more should include it.
Comment 23: It should be possible to subscribe to status changes for the few select machines that one actually uses, instead of 95% of all LabManager emails being irrelevant. This degradation of quality of information also means that messages of actual importance are easily overlooked. Also, there is a need for better logging, such that a single wafer can be backtracked through the different processing steps. The current fragmented logs do not provide such an overview. Keep up the good work!
Action: This exact feature is implemented in the next version of LabManager as announced at the user meeting
Comment 24: Very frindly, skillfull, and professionel staff. Access to state-of-the art, equipment incl. an experienced operator to assist you.
Action: Thank you.
Comment 25: I am only using 2PP and SEM LEO and because i book one week in advance it is always available.Well there where one time the SEM LEO couldnt be used. There were one time where the clean room with SEM was closed, because of some gas, but that isn't your fault. When I had a question the staff always give good answers.
Action: Thank you.
Comment 26: We had an issue with a technical error on the Alcatel (shutter didnt open all the way) which essentially destroyed a batch of wafers. We are also worried that there might be issues with the quality of the silver in the Alcatel. But we are not certain. Otherwise I am very satisfied with Danchip.
Action: No action
If you want to look further into the Alcatel issue, do contact the Thinfilm group
Comment 27: I especially like the renaming of the different flows on the SSE spinner. It was extremely clear to me which processes were which and implied a certain Danchip standard in them.
Action: Thank you.
Comment 28: Maybe make a public calender where it is possible to see planned cleanroom maintenance and other major maintaince?
Action: No immediate action
Good idea, we are not certain how to implement this for good effect but we will keep your suggestion in mind.
Comment 29: Thanks for supporting our collaboration!
Action: You are welcome!
Comment 30: My process has been very difficult to complete since some of the tools I use (pegasus and technotrans electroplater) is either booked a lot or undergoing maintenance. Neither of them is directly your fault, and I appreciate well maintained equipment, but you could maybe consider creating some sort of priority system such that everybody gets a chance to use the equipment within reasonable time. Its a problem that users can book a machine for weeks at a time if others also need to use it. Take a look at the bookings of the electroplater! You could restrict users and "only" allow them to book the "popular" machines for 50% of the time inside normal working hours (2.5 working days). This will result in a more fair division of the available time. If they then want more time they're free to use it in the evenings or weekends (or if it isn't booked a day in advance). This is a rule that's easy to implement and will stop unfair booking of machinery. 2.5 working days a week should be sufficient for most users but you could of course allow exceptions for some users upon agreement. Christmas wish list: another Ni electroplating machine as it is a HUGE bottleneck at the moment.
Action: Evaluate whether more tools should get booking rules
Thank you for your suggestion, we do work with booking restrictions on some of our tools and we will consider the ones you mentioned further.
Comment 31: My student have experienced problems during her work with our Silicon Drift Detector. It is a demanding project because we need to use the entire wafer (both sides) for one circuit. We have noted holes (scratches) in the resist we use to cover the backside of the wafer when we process the frontside. The SSE-spinner is our prime suspect. The scratches are in roughly the same position on several wafers.
Action:
Comment 32: It would be very helpful, if emails regarding a certain tool are only sent to those who have the license to use the tool. I move all messages from the cleanroom in a special folder where I never read them. There are simply too many messages. That way, I may not know vital information until it is too late.
Action:
Comment 33: 1. Most of engineers are helpful and keep their promise in mind while some of them not. It seems necessary to evaluate the training to know what the trainee thinks of the trainer. 2. Generally speaking, most of equipments are well maintained. However, in my opinion, some of heavily-used equipments need much more care. Take SEMs for example. Two SEMs are out of use or limited use recently which results in a terrible time for booking SEM Zeiss. Besides, the repairement seems not moving fast. 3. It is a little difficult for me to understand why there is no FIB system inside cleanroom though there are indeed some in DTU CEN. As a tool for milling and restoration, FIB system is a standard cleanroom inclusion in many other places like where I worked before. 4. Another thing is maybe a issue about the balance between efficiency and equality. Sometimes, E-beam writer is booked heavily by several users in one week. Of course, I agree with 'first come, first server'. However, it is somehow not fair that other user cannot find a time slot. In most cases, we need the previous results before we can decide what to do next. A worse practice seems some users book the time anyhow and delete it just one night before in spite of notice in email. This unscheduled behaviour not only does harm to other users' right but also decreases the utilization rate of the equipment. Actually, I am in doubt that this behavior is abuse of booking rules. I suggest every user who wish to user E-beam writer in the following week can get at lease one chance and booking rules can be revised to minimize deleting. 5. It is rather weird for me that cleanroom is partly closed in some areas before 16:00 (actually from 14:00) last Friday without any beforehand notice. It resulted some problems for my work flow. If I know the area I need to use will be closed in the afternoon, I will try to finish what should be done in the morning. I need scheduled work plan to avoid unexpected project delay. So I suggest cleanroom can provide more detailed schedule in the similar situation in the future. I am definitely satisfied to work inside cleanroom and with cleanroom engineers but there is still some aspects which can be improved in my point of view.
Action: