The LIGO Executive Committee Agenda for Monday August 08, 2005 will be:
(Meeting time: 10:30 am Pacific Time)
Change Request CR-050008, LIGO Document Management System (WBS 1.2)
Special Items:
The Annual
Report for FY 2005 LIGO Operations and the Request for a Supplement and an
Extension for FY 2007 and FY 2008 LIGO Operations have been submitted to the
NSF. I thank everyone for your support.
-pel
No report
>From: Ed Chargois <chargois_e@ligo.caltech.edu>
>From: Linda Turner - turner@ligo.caltech.edu>
>From: Cleveland Mak <mak_c@ligo.caltech.edu>
|
Week Ending 08/04/05 |
In |
Out |
|
Packages |
27 |
8 |
|
Faxes |
23 |
22 |
>From: "Cronin, Holly" <Holly.Cronin@caltech.edu>
>From: "Brambila, Ruth" <Ruth.Brambila@caltech.edu>
>From: Florence Kaufman <fkaufman>
>From: Gina Salone <gsalone@ligo.caltech.edu>
>From: Ed Jasnow <jasnow@ligo.caltech.edu>
>Irene Baldon
[last week]
[this week]
>Julie
Hiroto jhiroto@ligo.caltech.edu
>Dorothy Lloyd
>From: Jim Covington <jamesc@ligo.caltech.edu>
A written summary of the document management committee’s findings and recommendation has been prepared, reviewed by the members of the committee, and distributed to the executive committee. A change request has been distributed to the Executive Committee for consideration during the meeting scheduled for Monday, August 8, 2005.
>From: Cindy Akutagawa <cindy@ligo.caltech.edu>
>From: Bill Tyler tyler@ligo.caltech.edu
Nothing significant to report this week.
With the 4k back to full locking after swapping out ITMX, drag wiping ITMY, and dodging the impact of the Montana quake, efforts to understand and nail down parameters with regards to thermal compensation (TCS) are underway. Previously, annulus heating was required on both optics; now, central heating is required on Y. The ITMX needs central heating too but that awaits a swap of the TCS laser. So it seems clear that things are better than before (we've posted near-S4 inspiral ranges, no annulus TCS heating is required, and less TCS heat is required overall). However, not all experimental results hang together in a coherent way, such as the current cooling curves/spot size measurements intended to assess core optic absorption: look for more elogs to be posted this weekend.
Some highlights from the week of elogs are linked below:
Stable multi-hour locks with 5 W into the Mode Cleaner were achieved during the last week. The best inspiral range was 7 Mpc, not up to our best but respectable. The angular control loops transfer functions were re-measured and re-tuned to give more stable power up.
The Faraday isolator heating issue was revisited as a part of the interferometer stability investigation. The transmitted beam was found to be well thermally compensated but the rejected beam moves in yaw as power increases. The thermal drift of the ifo reflected beam is comparable to that measured in March 2005 using the WFS quad photo-detectors on the reflected port.
Other commissioning activities: - possible clipping of the beam on the reflected beam was investigated but no improvement in the interferometer noise was found after the re-centering of the beam on the top periscope mirror. - the bulls eye detector was installed on the ITMY pick-off beam to be used in the TCS servo. The controller for the TCS servo was also installed and control software was set up. - the X arm photon calibrator was tested. The ETMX displacement from the radiation pressure is consistent with a rough power calibration by the scope and the design. - a first attempt was made to measure the backscattering from the in vacuum support structure by exciting the 1-3 micron motion at the micro-seismic frequency using HEPI as an actuator.
There have been eight different visitor groups at LIGO in the last eight days. These visitors included a group of REU students from LA Tech, a group from a math/science magnet high in St. James Parish, LA, and 21 middle school science and math teachers receiving Professional Development training provided by LIGO.
The Local Educator's Network will participate in "An Evening at LIGO", Thursday evening, August 4th, providing them with an opportunity to meet the staff at LIGO and enjoy an evening out with their families.
The two Research Experience for Teachers participants, Rutherford McNair and Steven Griffin gave their final presentations today, capping off a productive summer of work at LIGO. Their efforts to tie the new exhibits into classroom learning are greatly appreciated.
Shannon attended the USENIX Computer Security Conference in
1) HPLF news: Two weeks ago IPG claimed that they could not find anything wrong with the 100 W laser. The reason why it was sent back to IPG was that I noticed 10 % power loss and fluctuations both in power and current. However, during the final stage of IPG testing on week ago, a failure occurred burning up the pump fiber. IPG proposed that in addition to repair the current system they will prepare a back-up system in order to improve our situation. We will receive the repaired laser in about 2 weeks and the backup in about 3-4 weeks. After discussion with the IPG engineers we decided that the back-up laser will also run at 1064 nm since a change to +1070 nm would not improve reliability.
2)
Finished the installation and test of the MC WFS
relief servo. Have been participating in some of the heating experiments of the
Faraday Isolator.
3)
Have been discussing some REFL port beam
stabilization design issues with Luke Williams at UF. We worried about a
conflict between the mirror mounts actuators and the roof of the ISCT1
enclosure. After measurements we found it to be OK. Most of the mounting
hardware are ready and on route to LLO. I also tested the absorption of the
Edmund Optics black glass, which will be used to prevent back scattering of the
light transmitted through the periscope 2 inch mirrors. Seems good. Discussed
electronics issues with Rich Abbott and Jay Heefner
who are visiting LLO this week. Jay installed the software which will be part
of the servo and explained how to set it up. I have been testing it and it
looks good.
Oddvar is on travel.
1) After the implementation of the Bull's eye detecor code, including the WFS5 interface with the pico motor controller, there has been an intermittent drop out of communication between the controller l1iool0 and the MC Autolocker script. The problem was tracked down to an overflow of open socket connections on the ethernet port. A workaround of the problem was found that temporary fixed the problem. It invloved increasing the number of open files in the mv162 Kernel. Still looking for a permannent solution.
2) Helped Doug Lormand in implementing the weather station setup for the X-End station.
[current report]
Rusyl and Allen are preparing the scope of work for the move of LDAS installation to the temporary location and for the power and HVAC work in the current LDAS room. This was discussed during Ed Jasnow's meeting yesterday. According to the current schedule, that we discussed today with Mike, Rusyl and Allen, the actual work to move LDAS downstairs should start during the second week of September.
1) Processed SG20 and SG21 MDC frames:
http://www.ligo-la.caltech.edu/~igor/S4/p3/ OUTPUT_SG20_S4_DARM_ERR.1.triple/plots_1.9_with_ampl_cut
http://www.ligo-la.caltech.edu/~igor/S4/p3/ OUTPUT_SG21_S4_DARM_ERR.1.triple/plots_1.9_with_ampl_cut
2) Processed LIGO-GEO quadruple coincidence triggers:
http://www.ligo-la.caltech.edu/~igor/S4/p3/OUTPUT_LIGO-GEO.quadruple/ plots
3) Working on SG22.
[report from week ending 7/28]
1) Discussing with Robert Schofield details of HVAC seismic isolation; discussing with Allen, Rusyl, Shannon, Mike and Joe the details of LDAS upgrade and LDAS move.
2)
node39 died; most
likely power supply failure.
3)
Called StorageTek to
find out how to move the tape robot and whether they provide any support for
it; they can do it for $1200 but probably we can do it ourselves as well:
tapes, tape extension need to be removed, drives should be switched into the
locked mode before moving.
4)
running badblocks on the nodes revealed several bad drives. However
badblocks and dd do not
necessarily agree on which drives are bad.
5)
LDRed from PSU SG20_S4
and SG21_S4 to LLO, LHO and CIT.
6)
Interviewed several candidates for LDAS/GC
position.
1) Siong and I finally figured out why waveburst results on LIGO-GEO data did not make sense: it turned out that S4 GEO frames start from non-integer number of seconds (they have 15000 ns offset). Waveburst assumed that a start time of a frame is always an integer. As a result there was some discontinuity in the time series, processed by waveburst, on the boundary between GEO frames that resulted in excess of triggers each 60 seconds and high noise variability. I have fixed waveburst to handle non-integer start time of frames and so far at first glance the new LIGO-GEO waveburst triggers look reasonable (but there is still some work left to do to make sure).
2)
Generated LIGO only production triggers with the
version of waveburst in which a minor bug in the
calibration (that was discovered during the code review) was fixed. Two waveburst modes were used: 64-2048 Hz and 64-4096 Hz. The
study is under way to check whether these sets of triggers overlap in the
common frequency subband.
3)
Running waveburst on
SG20_S4 and SG21_S4 MDC frames.
4)
Continue testing waveburst
on LIGO-VIRGO simulated data.
Most of the activity this week has involved
aligning the arm cavities. NAC was fairly easy to align and is done. SAC is
proving to be more difficult, but we are making progress.
Rick read a paper on silicate bonding and
presented it to the group. Chinyere will leave after this week, and she will give her
final SURF presentation on Friday, August 5.
No report.
Mark Barton gave a talk on his recent work on "Modeling Wedged Optics"
[Viewgraphs at http://www.ligo.caltech.edu/~e2e/Minutes05/050804/ ]
(from Hiro) Sany Yoshida was visiting Caltech to study the suspension models of Virginio's and Mark Barton's.
There was a discussion about the modeling of mechanical parts of the SimAdvLIGO package which has been setup by Matt Evans to simulate advanced LIGO. Right now, a preliminary version of Mark Barton's multi state pendulum model is used, but the seismic isolation part is the one of LIGO I. Sany will lead the effort to collect necessary informations about mechanics part of advanced LIGO, and place appropriate suspensions on top of the advanced LIGO SEI (whatever is available now).
During Sany's visit, Mark modified his suspension model written in Mathematica to support more anticipated physics effects, like wedge angle.
(Biplab) Studied limitations and accuracy considerations for the FFT code written in Matlab for Advanced LIGO Arm Cavity. Started calculating diffraction losses. Using MIT FFT code for LIGO-I to study effects of transverse displacements of beams and optics.
(Monica) The Length Sensing and Control (LSC) code written by Matt last June (just the lock acquisition procedure) is now implemented in my e2e package for the 40m. Modeler runs well with no errors. Tests to repeat the lock acquisition procedure are in progress.
Hiro helped David, Keita's SURF student, to resolve problems simulating Output Modecleaner (OMC). One bug was found in calculating the mode miss match in the triangular cavity module, and Biplab fixed that. After fixing the bug, various test runs showed consistent mode spacing changes when the geometry of the OMC is changed.
Hiro's major work is still programming, modeler and new FFT.
(Bruce)
(Melody) Continuing with fixing the PRs. Fixed PR 415, 472, and 396
Chatterji;
Continued work on coherent burst analysis pipeline. Investigating alternative statistics to take into account sky-dependent sensitivity of the network and the effect of calibration errors. Modified pipeline code to allow allow "compilation" of a stand-alone executable from the Matlab source code. Successfully ran a simplified version of the pipeline on the Caltech cluster under Condor. Initiated statistical tests and validation studies using ideal white noise.
Gave summary talk http://emvogil-3.mit.edu/~shourov/q/documentation/presentations/G050332-00.pdf
"Searching for gravitational-wave bursts with the Q Pipeline"</a> for the LIGO Science Seminar. The talk focused on the application of the Q Pipeline to the S2 H1H2 data set, as well as the current status of the algorithm and future plans.
Continued working with S. Hormoz on implementation of an directed sky search extension to the Q Pipeline. We are also investigating statistical tests for H1-H2 consistency tests that are robust against calibration errors.
Working on an updated version of the Q Pipeline as well as tools for control room use and detector characterization during S5.
Mandic:
I performed some checks of the S4 H1L1 result with time-shift, including a comparison with Philip's recent calculation.
I also worked on the presentation of the preliminary S4 results for the LSC meeting.
Mendell:
The focus for the past week has been on finding the 1 Hz harmonics in the S4 data. These appear in DARM_CTRL and SFTs made from h(t) for H1 and L1, but apparently not in H2. I am also making final changes to speed up the StackSlide Monte Carlo Simulations so that S4 results can be produced.
Yakushin:
1) Processed SG20 and SG21 MDC frames:
http://www.ligo-la.caltech.edu/~igor/S4/p3/OUTPUT_SG20_S4_DARM_ERR. 1.triple/plots_1.9_with_ampl_cut
http://www.ligo-la.caltech.edu/~igor/S4/p3/OUTPUT_SG21_S4_DARM_ERR. 1.triple/plots_1.9_with_ampl_cut
2) Processed LIGO-GEO quadruple coincidence triggers:
http://www.ligo-la.caltech.edu/~igor/S4/p3/OUTPUT_LIGO-GEO.quadruple/plots
3) Working on SG22.
Searle(ANU visitor to Caltech):
Had idea to use incoherent estimate of null power to determine level of suppression, which can be formulated as F-statistic-like test of significance. Still working on the details, but tests seem promising and results better than null or null+residual formulations considered so far.
Implemented production of incoherent power estimate, but not statistical tests yet as we still need to work out either the expected distribution of the two semi-independent quantities, or some pessimistic estimate of it.
Also preparing talk for LSC meeting.
Sutton:
I've spent most of the week studying the LIGO cheese / bursts space and chirplets as maximum entropy waveforms in that space. I've also been helping out with the coherent network analysis development, and helping Maria Principe with her network simulator project.
LDAS:
After successfully pushing out the release last week, we decided to focus on the issue of the disappearing threads in the diskCacheAPI that was new to the 1.7.0 release. The 1.7.0 release had a workaround for this problem but the cause was unknown at the time of the release. We were able to add enough debugging statements to the code over the weekend to identify the cause of the disappearing thread and trace it down to an extremely rapid scan of a mount point in the file system (less than 1/100 of a second of scan time). The fix was simply to add a mutex lock to an update function call for the thread's state variable. This has been tested and verified to have no ill effects on performance. Now having the fix in hand, we are planning to patch up to other minor issues and have a 1.7.1 release of the software early next week. We will not require any of the site to up- grade, it will be their choice based on their satisfaction with the 1.7.0 workaround. Once this is completed, we will begin our efforts to port to Solaris 10 and Fedora Core 4 for the 1.8.0 release.
Fixed up a documentation issue with the LAM (message passing interface) installation guide to properly reflect our migration to FC3 RPMs.
Change the color of the system error giffs associated with the missing threads in the diskCacheAPI from red to yellow as this is not a fatal job issue but a system issue. This has now been fixed and will be in the 1.7.1 release.
Identified the source of all users submitting tests jobs via the control- MonitorAPI's client from the submitter's id to Mary Lei. This has now been fixed and will be in 1.7.1.
Discovered several failure modes for SAM-QFS using LDAS on the DEV and the TEST systems. These issues have been reported to the hardware group. These also continue to demonstrate how successful LDAS is a system challenging application since as is so often the case it finds these issues when other data management and data analysis applications do not.
TCLGLOBUS:
Current status of the Globus GASS Transfer:
A total of 64 functions from globus_gass_transfer.h have been SWIG- wrapped. 40 functions (client/server, send/receive bytes, and set/get attributes) have been tested.
Fixed the following two functions: globus_gass_transfer_get_type and globus_gass_transfer_get_status. The functions had incorrect typemaps caus- ing huge memory leaks.
Finished writing GASS Transfer server-initiated test case. Still testing the complete GASS Transfer client/server testsuite for any memory leaks or other failures.
Corrected several incorrect
contact links on the webpages based on inputs from
OSG/GRIPHYN/IVDGL:
The iVDGL Annual Report went out to the NSF this week. It covered LIGO's involvement as a Tier II center and recent activities on developing the inspiral application for use on the OSG Grid. Kent Blackburn was aske to take the lead along with Craig Prescott on the Tier II section of the document and to be the lead on the LIGO application section. The report has been submitted to the DCC with LIGO document number T050127-00-Z.
A number of refinements to the OSG-ITB test bed were made including NFS mounting of /home VOMS accounts on Worker Nodes(WNs).
A backup mechanism using a cronjob has been designed to backup the VDT configuration files of the Condor Central Manager on a daily basis.
A cronjob has been developed to daily cleanup stale gram_condor_logs produced by GridCat.
Log rotation and truncation scripts have been developed to manage a number of log files produced by Condor and Globus services.
A new Fedor FC3 kernel (2.6.12-1.1372_FC3smp) has been tested on the Condor Central Manager and plans exist to migrate this kernel to the WNs in the near future.
To automate maintenance, a replica of the Yums repository on the production cluster has been established on the Condor Central Manager. This repository will serve the WNs.
According to the latest OSG-ITB condor statistics, 150 consecutive GridCat jobs have run to completion with zero failures among this set.
A problem with Grid Exerciser jobs being submitted with two-phase commit requirements enabled, was reporting to the GridEx team and no email response has occurred within the last 48 hours.
Two Terabytes of hard drives have arrived for our OSG testbed cluster these will be installed once the configuration stabalizes and down time can be established. We wil use this extra disk storage to explore SRM/dcache as a storage resource manager and virtual filesystem for clusters.
Caltech
(Dan Kozak)
(Stuart Anderson)
MIT
(Keith Bayer)
(Igor Yakushin)
(Greg Mendell)
(Ben Johnson)
(Keith)
(
On travel
(Christine)
(Christian)
(Mike)
(Veronica)
(Larry)
Mail statistics July 28-August 03, 2005
|
Mail Statistics |
August 04, 2005 |
|
Rejected Messages |
26,619 |
|
Virus Messages |
950 |
|
False Positives |
229 |
|
Accepted Messages |
14,399 |
|
Total Messages |
41,018 |
No report this week.
From: Ken Mason kmason@ligo.mit.edu
Purchase orders have been issued to ASI for the blade resizing design and to Southern Enterprises for the fabrication of the test stand. Both suppliers have promised a completion date of 9/23/2005.
From: "Joseph A.
Giaime" jgiaime@ligo.phys.lsu.edu
From: Janeen Romie romie_j@ligo.caltech.edu
From: ctorrie@ligo.caltech.edu
1) 2nd Mode Cleaner
I have prepared the second mode-cleaner for LASTI. Rich (LASTI) is coming next week to take the suspension apart and prepare it for cleaning and baking. The replacement parts have been completed. The control osems and fixtures have also been prepared and arranged ahead of the visit.
2) 2nd Structure
Mike Gerfen, Tim (RAL) and I met to discuss the 2nd structure for Stanford. Mike is keen to apply lessons learned from the 1st structure to the 2nd. RAL are keen to be involved in this and we will continue discussions on Monday.
3) Quad build
Today I built a single chain triple pendulum from a fixed top mass. The plan
is to now suspend a single chain quad. Mike and Russell were helping me
remotely from
4) Installation fixtures
Ken Mailand and I met with Mike Gerfen today and have a follow up meeting next week. Each of us have some actions from our opening discussions. Mike believes it is a job his shop could do. One of the first jobs would be to get
5) Drum ended wires
The order is in for more drum ended wires. These will be used for a combination of testing, installation in LASTI and interaction and subsequent testing with RAL.
I also attended the Techmart training and Mike Perreur-Lloyd and I met with SolidWorks representatives to discuss some issues with the new release.
From: Ken mailand kmailand@ligo.caltech.edu
Making simplified S/W drawings of the Adv. LIGO optical component assemblies for import to Zmax, for Mike Smith, and converting Mechanical Desktop LHO full site layout to S/W
The exit air HEPA filter for the large bake oven has arrived.
Will meet today with Calum and Mike Gerfin @ces re. the Quad installation fixture part manufacture.
From: Rolf Bork <rolf@ligo.caltech.edu>
Alex and I have been working on porting CDS software to a dual AMD64 computer for use as the quad suspension controller for LASTI. We are using a hard real-time Linux system to allow us to run the real-time front end code at 2048Hz on one CPU while running the EPICS interface software on the second CPU. As an initial test, we are porting the current LSC front end and Epics code. The code is now partially functional, but still has a few bugs. We've had to make code modifications to run on a 64-bit architecture and the latest Linux does not support the same PCI bus accessing calls that Linux 2.4 had. I think we just have a couple of more 64bit issues to work and should have it fully functional by next week.
From: Helena Armandula ahelena@ligo.caltech.edu
See Suspensions
From: Bill Kells kells@ligo.caltech.edu
In anticipation of the LSC meeting (where I will not be, but David Blair, et
al will give a presentation) we have completed a full "round" of
discussion and clarification of the issues on better estimating the parametric
instability problem as it would apply to AdL. (This
is in parallel with completion of reviews of several of the pending
From: Peter King pking@ligo.caltech.edu
A 300 mW OEM version of the Innolight NPRO was ordered.
The first revised cut at estimating the labour costs for the PSL was completed. There are a few other labour tasks that I failed to include such as costs related to procurement, and shipping and handling. Estimates for shipping are being worked on.
From: Michael Smith <smith@ligo.caltech.edu>
Aabeg has implemented the 5/3 beam expanding input telescope and measured the spot size at the output of the optlev receiver to be less than 4 mm diameter. He adjusted the focus of the optlev receiver by displacing the input beam laterally and minimizing the signal observed on the QPD; this minimizes the coupling between displacement and angle. He is in the process of measuring the QPD output voltage as a function of input beam angle.
Shasta is trying to calibrate the output of the lock-in amplifier by inserting known neutral density filters in the laser beam path. Subsequently, she is preparing to measure a Lambertian scatter surface and a sample of the oxidized, polished stainless steel baffle material that is being considered for the advanced LIGO baffles and beam dumps.
From: Riccardo DeSalvo
desalvo@ligo.caltech.edu
Riccardo
Working on the bid requests for the production of the LASTI OMC HAM SAS system.
Juri
SPIE conference presentation
Francesca
I began the measurement of GAS feedback behavior.
To begin, I leveled my system. Then I carried out several types of measurements. I connected the output of the LVDT to the digital oscilloscope (without the connection from amplifier to coil), in order to study the answer of my system when I change the load and find the minimum of frequency. I found that we have this minimum of frequency (0,288 Hz) near to 0 volts. After I made the same measurement connecting the amplifier to coil (placing proportionality, integration, differentiation equal to zero), in order to see if the amplifier introduced noise in the system. For last, I measured frequency and voltage of LVDT whit proportionality equal to 1, and equal to -1. I’m studying the results of these measurements.
Chiara
Out to SPIE
Marco
Writing thesis
John, Marco
We have now mounted the mirrors using two reinforcing rings rather than one spacer. This week we will test whether any improvement has been made. The mirrors will also be scanned again this week. Improved input optics have been implemented including a new mode-matching lens. I am optimistic that all of these efforts will improve our coupling to and ability to lock to the fundamental mode.
Justin, Sean:
This week we mounted the top silicon flex joint, and acquired a Q factor that was ~ 2.9 times higher than the maraging steel case. Then we tried to progress on to mounting both silicon flex joints, and in the process, we broke ALL of our flex joints. THis is an issue. In the meanwhile, we have been working with ANSYS to try and get our simulation to converge. Due to difficulties with this, we have decided to take the simulation down to a 2D level. We will spend next week putting together the nuts and bolts of the 2D simulation and trying to acquire more flex joints to break.
Anamaria and David:
We moved the whole setup in the enclosed room and coupled all the LVDT's to the blades. The connections are all good. We need to stabilize the floor in the enclosed room (it's soft) and teach LabVIEW to acquire all 4 signals at the same time.
Glenmary,
The wire stressing machine is working to an extent. The piston moves both backwards and forwards with (very) manual control. The wiring connecting the solonoids to the pump and piston is dodgy, and we're going to redo it. Also, the pressure release valve seems to have vanished.
Machine should be working with (proper) manual control soon.
Alberto
My parts have just left the manufacturers and I think I'm getting them next few days. Now we are going to set up the GAS prototype in the lab to be ready to mount the arms.. I'm trying to improve the theoretical model of the system I'm projecting, looking for the best set-up to move the center of percussion of the whole system (spring + arm) as closer as possible to the flex joint. Unfortunately the simplification I thought I had found revealed to be incorrect so the moments of inertia of the single parts can't be neglected. Meanwhile I’m working on my SURF-LIGO report and on my Italian Senior thesis.
Ilaria
I have continued my work with Ansys. I sketched some shapes and I simulated their movement with this software. Since Ansys is a general purpose finite element modeling, the body can be sub-divided up into small discrete regions known as finite elements. So, the first function I used was the mesh. I began with a mesh discretization (896 elements) and then observed and recorded the solution interesting, the resonance frequency, for the first 6 modes. I repeated the problem with a finer mesh (until 18924 elements). Than I could conduct the convergence tests: for each frequency on the x-axis I set the number of elements, on the y-axis I set the frequencies corresponding to the number of elements derived from the mesh. I plotted it and I saw that the finite element model converged. Through this simple example I understood that it is necessary to conduct convergence tests on the finite element model to confirm that a fine enough element discretization has been used.
Maria Paola
I received the laser and the silicon detectors I ordered, and I started to build a raw version of the whole circuit, connecting the laser to a 5 V power supply and to a signal generator which generates a square wave of frequency 47 KHz and amplitude 0.5 V. Then, I fixed the photodiode in front of it, and I connected the photodiode to the circuit that amplifies and rectifies the input signal, consisting of a transformer, a diode bridge and a capacitor. The diodes used are germanium diodes, having a voltage drop of about 0.3 V, which is quite lower then the one other kinds of diodes have. I got a final output voltage of about 4 V. Now I'm going to figure out how to optimize the whole circuit, by properly choosing each single component, so as to produce a continuous output voltage of about 12 V, that is the supply voltage a common stepper motor requires.
Nicky
Restarting analyzing creep data.
For additional information about this report, contact Stan Whitcomb or Phil Lindquist