Beautiful fitting.

Quote:

EDIT: I used an IFIT (inverse fast idiot transform) to change the x-axis of the plot from Hz to m. I think xlabel('Frequency [Hz]') is in my muscle memory now.. I have redone the beam fit, this time omitting the M2, which I believe was superfluous. I have made the requested changes to the plot, save for the error analysis, which I am still trying to work out (the function I used for the least squares fit does not work out standard error in fit parameters). I will figure out a way to do this and amend the plot to have error bars.

[Jenne, Steve]

We removed the old SRM and PRM from their cages, and are temporarily storing them in the rings which we use to hold the optics while baking.  Steve will work on a way to store them more permanently.

We then hung the new SRM (SRMU03) and new PRM (SRMU04) in the cages.  We were careful not to break the wires, so the heights will not have changed from the old heights.

The optics have not been balanced yet.  That will hopefully happen later this week.

Quote:

I have redone the beam fit, this time omitting the M2, which I believe was superfluous. I have made the requested changes to the plot, save for the error analysis, which I am still trying to work out (the function I used for the least squares fit does not work out standard error in fit parameters). I will figure out a way to do this and amend the plot to have error bars.

 Are you sure about your x-axis label? 

EDIT: I used an IFIT (inverse fast idiot transform) to change the x-axis of the plot from Hz to m. I think xlabel('Frequency [Hz]') is in my muscle memory now..

I have redone the beam fit, this time omitting the M², which I believe was superfluous. I have made the requested changes to the plot, save for the error analysis, which I am still trying to work out (the function I used for the least squares fit does not work out standard error in fit parameters). I will figure out a way to do this and amend the plot to have error bars.

 I restarted the elog with the restart script as it was down.

 Here is Crystal 724 polished side 2 with all photos along the length stitched together

The 40m's new undergrad Kevin Kuns was introduced to 40m safety hazards. He is new and needs guidance as specially with 2W laser work.

Peter King will train him on Friday to LIGO-laser standard.

ETMY sus damping was restored

We found ref-cavity HV was off yesterday afternoon. It was turned back on.

 The Guralp Box appears to be back in working order.  It's reinstalled with the 2 seismometers plugged in.

In order:

* Koji suggested retouching the through-board solder joints on the broken channel (EW2 = Gur2X) with a bit of solder to ensure the connections were good.  Check.

* "C7", one of the giant 1uF capacitors on each channel is totally bypassed, and since that was one of the original suspects, Rana removed the (possibly) offending capacitor from EW2.

* Rana and I isolated the craziness to the final differential output stage.  We tried each of the testpoints after the individual gain / filter stages, and found that the signals were all fine, until after the output stage.

* I started to remove the resistors in the output stage (with the plan to go through the resistors, capacitors, and even the amplifier chip if neccessary), and noticed that 2 of the 1k resistors came off too easily, as if they were just barely connected in the first place.  After replacing only the 4 1k resistors, the craziness seemed to be gone.  I poked and gently bent the board, but the output wouldn't go crazy.  I declared victory.

* I reinstalled the box in its normal spot, and put Gur2 (which had been out by the bench for use as a test signal) back next to the other seismometers.  We are in nominal condition, and should be able to do a huddle test this week.

I looked at the time traces of all the seismometer channels, and they all looked good.  I'll put a spectra in in the morning....I'm too impatient to wait around for the low frequency FFTs.

Attached are the before and after pictures of the output stage of EW2 / Gur2X.  The "before" is the one with the OUT+ and OUT- words upsidedown.  The "after" picture has them right side up.

Koji and Kevin

We unpacked the Innolight 2W laser, took an inventory, and scanned the operations manual.

[Edit by KA]

The scanned PDFs are placed on the following wiki page

http://lhocds.ligo-wa.caltech.edu:8000/40m/Upgrade_09/PSL

We will measure the P-I curve, the mode profile, frequency actuator responses, and so on.

Jenne, Koji and I opened up the package from Raicol and examined the crystals under the microscope. The results were mixed and are summarized below. There are quite a few scratches and there is residue on some of the polished sides. There is a large chip in one and there appear to be gaps or bands in the AR coatings on the sides.

There are two albums on Picassa

1. The package is opened ...

2. The crystals under the microscope.

I went to Ottavia, and tried running yum update.  It was having dependancy issues with mjpegtools, which was a rpmforge provided package.  In order to get it to update, I moved the rpmforge priority above (a lower number) that of epel ( epel -> 20 from 10, rpmforge -> 10 to 20).  This resolved the problem and the updates proceeded (all 434 of them). yum update on Ottavia now reports nothing needs to be done.

I went to Rosalba and found rpmfusion repositories enabled.  The only one of the 3 repositories in each file enabled was the first one.

I then added priority listing to all the repositories on Rosalba.  I set CentOS-Base and related to priority=1.  I set CentOS-Media.repo priority to 1 (although it is disabled - just to head off future problems). I set all epel related to priorities to 20. I set all rpmforge related priorities to 10.  I set all rpmfusion related priorities to 30, and left the first repo in rpmfusion-free-updates and rpmfusion-nonfree-updates were enabled.  All other rpmfusion testing repositories were disabled by me.

I then had to by hand downgrade expat to expat-1.95.8-8.3.el5_4.2.x86_64 (the rpmforge version).  I also removed and reinstalled x264.x86_64.  Lastly I removed and reinstalled lyx. yum update was then run and completed successfully.

I installed yum-priorities on Allegra and made all CentOS-Base repositories priority 1. I similarly made the still disabled CentOS-Media priority 1.  I made all epel related repos priority 20.  I made all lscsoft repos priority=50 (not sure why its on Allegra and none of the other ones).  I made all rpmforge priorities 10.  I then ran "yum update" which updated 416 packages.

So basically all the Centos control room machines are now using the following order for repositories:

CentOS-Base > rpmforge > epel > (rpmfusion - rosalba only) > lscsoft (allegra only)

I'm not sure if rpmfusion and lscsoft are necessary, but I've left them for now.  This should mean "yum update" will have far fewer problems in the future.

There is a clearly a difference in the divergence angle of the x and y beams - maybe 10-20%. Since the measurements are outside the Rayleigh range and approximately in the linear regime, the slope of the divergence in this plot should be inversely proportional to the waists - meaning the x- and y- waist sizes should differ by about 10-20%. You should check your fitting program for the waist.

Frank noticed that this particular SR560 had an offset on the output which was unzeroable by the usual method of tuning the trim pot accessible through the front panel.

I tried to zero the offset using the trimpots inside, but it became clear that the offset was due to a damaged FET, so Steve ordered ~20 of the (now obsolete*) NPD5564.

I replaced this part and adjusted the offsets and balanced the CMRR of the differential inputs mostly according to the manual (p. 17). There are a few notes that should be added to the procedure:

1. It can sometimes be that the gain proscribed by the manual is too high and saturates the output for large offsets. If that's the case, simply lower the gain, trim the offset, then return the gain to the specified value and trim again.
2. The limit in trimming the offset is the stick slip resolution in the trim pot. This can potentially leave the whole preamp in an acoustically sensitive state. I tapped the pots with a screwdriver after tuning to make sure it was in more of a 'sticky' rather than 'slippy' region of the knob. A better design would allow for more filtering of the pot.
3. In the CMRR tuning procedure it says to 'null sine wave output' but it should really say 'null the sine wave component at the drive frequency'. The best CMRR tuning uses a 1 kHz drive and leaves a residual 2 kHz signal due to the distortion imbalance (of the FETs I think).
4. The CMRR tuning upsets the DC offset trim and vice versa. The best tuning is gotten by iterating slightly (go back and forth once or twice between the offset and CMRR tuning procedures).

It looks like its working fine now. Steve's ordering some IF3602 (low-noise, balanced FET pair from Interfet) to see if we can drop the SR560's input noise to the sub-nV level.

Noise measured with the input terminated with a BNC short (not 50 Ohms) G=100, DC coupled, low-noise mode:

In order to measure the transfer function of the RC cavity's foam, I've turned off the servo so that the room temperature noise can excite it.

The attached plot shows a step response test from 2 weeks ago. Servo is nominally still working fine.

NOTE: This measurement is wrong and only remains for documentation purposes.

Koji asked me to take a profile of the output of the 1W NPRO that will be used for green locking. I used the razor-scan method, plotting the voltage output of a PD vs the position of the razor across the beam, both vertically and horizontally. This was done at 6 points along the beam path out of the laser box.

I determined the beam spot size at each point by doing a least-squares fit on the plots above in Matlab (using w as one of the fitting parameters) to the cumulative distribution functions (error functions) they should approximate.

I then did another least-squares fit, fitting the above "measured" beam profiles to the gaussian form for w vs z. Below is a summary.

It seems reasonable, though I know that M² < 1 is fishy, as it implies less divergence than ideal for that waist size. Also, like Koji feared, the waist is inside the box and thus the scan is almost entirely in the linear regime.

Steve and Koji

We aligned the peeping mirrors to look at the surface of the ITMs.
They had been misligned as we move the positions of the ITMs, but now they are fine.

I pulled the Guralp breakout box from the rack, and it's sitting on the EE bench here.  The game plan is to check out the Gur2X channel.  

Rana and Steve have been investigating, and found that the X channel has been funky (which has been known for ~a month or two) when the seismometer has been plugged in, and also when the seismometers have been unplugged, but the box is left on.  The funkyness goes away when the box is turned off.  Since it's not there when the box is off, it seems that it's not a problem with the cable from the box to the ADC, or in the ADC channel.  Since it is there when the box is on, but the seismometer is unplugged, it's clear that it's probably in the box itself.  

Preliminarily, I've connected a set of BNC clipdoodles to the input testpoints, and another set to the output.  They're both connected to a 'scope (which is on it's battery so it's not connected to any Ground), and when I tap on the circuit board the input trace is totally unchanged, but the output trace goes kind of crazy, and gets more fuzzy, and picks up a DC offset.  Koji is concerned that some of the big capacitors may have an iffy connection to the board. 

Investigations will continue Monday morning. 

The 1979 vintage RF spectrum analyzer HP4195A  sn2904J01587 shipped out  for repair today to http://www.avalontest.com

It has a 25 MHZ oscillation when you go  below 150 MHZ in your sweep....atm1 with the larger amplitude shows this 25 MHZ

Atm2 is displaying  full sweep-sign scans from 1 to 500 MHZ.....here one can clearly see the three segment of the scan:

1, large amplitude 25 MHZ oscillation dominating the spectrum up to 150 MHZ

2, the mid section from 150 MHZ  to 300 MHZ with medium size amplitude is normal

3, from 300 MHZ to 500 MHZ the amplitude is decreasing.......showing the disadvantage of using a 300 MHZ oscilloscope

Bob, Steve, and Koji

We opened North heavy door of the BS chamber in the afternoon.

In the evening, Koji worked on the PRM/SRM removal.

- Cleaned up the OPLEV mirrors to create some spaces near the door.

- Clamped PRM/SRM.

- Removed OSEMs. Made a record of the OSEMs. The record is on the wiki (http://lhocds.ligo-wa.caltech.edu:8000/40m/Upgrade_09/Suspensions)

- Found the SOSs are quite easy to remove from the table as they are shorter than the MOSs.

- Put a new Al sheet on a wagon. Put the SOSs on it. Wrapped them by the Al foils.

- Carried it to the clean room. They are on the right flow bench. Confirmed the wires are still fine.

- Closed up the chamber putting a light door.

 I redid the PZT Phase Modulation measurement out to 5 MHz for both the Innolight and the Lightwave.  The previous measurement stopped at 2MHz, and we wanted to see if there were any sweet spots above 2MHz.  I also reduced the sweep bandwidth and increased the source amplitude at high frequency to reduce the noise (the Lighwave measurement, especially, was noise dominated above 1MHz).  I also plotted the ratio of PM/AM in rad/RIN, since this is the ultimate criterion on which we want to make a determination.

It looks like there is nothing extremely useful above 2MHz for either laser.  There are several candidates for the lightwave at about 140 kHz and again at about 1.4 MHz.  The most compelling peak, however, is in the innolight at 216 kHz, where the peak is about 2.3e5 rad/RIN.

Below about 30kHz, the loop suppresses the measurement, so one should focus on the region above there.

 To prevent this confusion in the future, at Koji's suggestion I've created a Makefile.no_parse_mdl in /home/controls/cds/advLIGO on megatron.  The normal makefile is the original one (with correct parsing now).  So the correct procedure is:

1) "make yep"

2) Modify yep.c code

3) "make -f Makefile.no_parse_mdl yep"

Please put those numbers onto wiki somewhere at the green page or laser characterization page.

Koji

ITMX was aligned with regard to the 40m green oplev.
Now both cavities are aligned.

Next thing we are going to do is to remove PRM and SRM towers.

As well as the oplev construction for ITMs.

We anticipate the drift of the stack. So we need to revisit the alignment again.

Some tools and the level gauge were removed from the table.

Picture of the ETMX - reflection from the ITMX is hitting the mirror and Jamie's windmill.

0. The suspension tower had been placed on the table close to the door.

1. Brought the OSEMs from the clean room. Connected the satellite box to the ITMX suspension.

2. Went into the chamber. Leveled the table.

3. Released the mirror from the clamp. Put and adjust the OSEMs.

- Note that the side OSEM is located to the south side of the tower
  so that we can still touch it after the placement of the TT suspension at the north side of the SOS tower.

4. Clamped the mirror. Moved the SOS tower according to the CAD layout.

5. Leveled the table again.

6. Released the mirror again and adjusted the OSEMs.

7. Turned on the end green laser pointer.

- The spot was slightly upside and left of the mirror. Adjusted it so that the spot is at the center.

8. Align ITMX in Pitch

- The spot was hitting the tube. Moved the pitch bias such that the beam get horizontal.

9. Align ITMX in Yaw

- Moved the SOS tower such that the approximate spot is on the ETMX. If I hit the right spot I could see the tube get grown green because of the huge scatter.

10. Adjusted the OSEMs again and check the alignment again. Repeated this process 2~3 times.

- Bias values at  the end of the work: Pitch 0.7800 / Yaw 0.270

11. Close up the chamber

- Remove the level gauge. Some of the screws are still in the Al ship in the chamber.

- Close the light door.

Data looks perfect ... but the fitting was wrong.

Vn = Vdn + Z * sqrt( 2 e Idc ) ==> WRONG!!!

Dark noise and shot noise are not correlated. You need to take a quadratic sum!!!

Vn^2 = Vdn^2 + Z^2 *(2 e Idc)

And I was confused whether you need 2 in the sqrt, or not. Can you explain it?
Note that you are looking at the raw RF output of the PD and not using the demodulated output... 

Also you should be able to fit Vdn. You should put your dark noise measurement at 10nA or 100nA and then make the fitting.

 Same thing for the Innolight Mephisto.

Not unexpected values with dn/dT around 11E-6 K^-1 and coefficient of thermal expansion = 8E-6 K^-1 and a laser resonator length of order 10cm.

The beams from the Innolight and Lightwave NPROs were both incident on a 1GHZ New Focus PD. Mott and I swept the temperature of the Lightwave and tracked the change in frequency of the beatnote between the two. The Innolight temperature was set to 39.61C although the actual temperature was reported to be 39.62C.

Freq. vs temperature is plotted below in the attached PDF. The slope is 2.8GHz/K.

The data is in the attached MATLAB file.

 The elog crashed when I was uploading a photo just now. I logged into nodus and restarted it.

Currently, the y end plant is yep.mdl.  In order to compile it properly (for the moment at least) requires running the normal makefile, then commenting out the line in the makefile which does the parsing of the mdl, and rerunning after modifying the /cds/advLigo/src/fe/yep/yep.c file.

The modifications to the yep.c file are to change the six lines that look like:

"plant_mux[0] = plant_gndx"  into lines that look like "plant_mux[0] = plant_delayx".  You also have to add initialization of the plant_delayx type variables to zero in the if(feInt) section, near where plant_gndx is set to zero.

This is necessary to get the position feedback within the plant model to work properly.

#NOTE by Koji

CAUTION:
This entry means that Makefile was modified not to parse the mdl file.
This affects making any of the models on megatron.

 Here's another measurement of the noise of the REFL11 PD.

This time I made the fit constraining the Dark Noise. I realized that it didn't make much sense leaving it as a free coefficient: the dark noise is what it is.

Result: the transimpedance of REFL11at 11 MHz is about 4000 Ohm.

All the details and data will be included in the wiki page (and so also the results for AS55). Here I just show the comparison of the transfer functions that I measured and that I modeled.

I applied an approximate calibration to the data so that all the measurements would refer to the transfer function of Vout / PD Photocurrent. Here's how they look like. (also the calibration will be explained in the wiki)

.

The ratio between the amplitude of the 55Mhz modulation over the 11MHz is ~ 90dB

The electronics TF doesn't provide a faithful reproduction of the optical response.

I measured a jitter modulation caused by injection of a signal into laser PZTs.

The measurement has been done by putting a razor blade in the middle way of the beam path to cut the half of the beam spot, so that a change of intensity at a photodetector represents the spatial jitter of the beam.

However the transfer function looked almost the same as that of amplitude modulation which had been taken by Mott (see the entry).

This means the data is dominated by the amplitude modulation instead of the jitter. So I gave up evaluating the data of the jitter measurement.

One dell is in the clean room for the suspension work.

I can't find the DELL laptop anywhere in the lab. Does anyone know where it is?

Also one of the two netbooks is missing.

 1) True. My bad. In my elog entry (but not in my fit code) I forgot the impedance Z= 750Ohm (as in the fit) of the resonant circuit in front of the square root: Vn = Vdn + Z * sqrt( 2 e Idc )

2) That is exactly the point I was raising! The measured dark noise at resonance is 2x what I expect.

I also admitted that the data points were few, especially around the 3dB point.

Today I'm going to repeat the measurement with a new setup that lets me tune the light intensity more finely.

Something must be wrong. 

1. Physical Unit is wrong for the second term of "Vn = Vdn + Sqrt(2 e Idc)"

2. Why does the fit go below the dark noise?

3. "Dark noise 4 +/- NaN nV/rtHz"   I can not accept this fitting.

Also apparently the data points are not enough.

Putting my hands ahead, I know I could have taken more measurements around the 3dB point, but the 40m needs the PDs soon.

From the measurements of the 11 MHz RFPD at 11Mhz I estimated a transimpedance of about 750 Ohms. (See attached plot.)

The fit shown in the plot is: Vn = Vdn + sqrt(2*e*Idc) ; Vn=noise; Vdn=darknoise; e=electron charge; Idc=dc photocurrent

The estimate from the fit is 3-4 times off from my analsys of the circuit and from any LISO simulation. Likely at RF the contributions of the parassitic components of each element make a big difference. I'm going to improve the LISO model to account for that.

The problem of the factor of 2 in the data turned out to be not a real one. Assuming that the dark noise at resonance is just Johnson's noise from the resonant circuit transimpedance underestimates the dark noise by 100%.

What??? I don't see any gray trace of Rs in the plot. What are you talking about?

Anyway, if you are true, the circuit is bad as the noise should only be dominated by the thermal noise of the resonant circuit.

#
#                 gnd
#                 |
#                 Cw2
#                 |
#                 n23
#                 |
#                 Lw2
#                 |
#   gnd           n22
#   |             |
#   Rip           Rw2
#   |             |                   |\
#   nt- Rsi-n2- - - C2 - n3 -  - -  - |  \
#            |    |      |   |        |4106>-- n5 - Rs -- no                                                            
# iinput    Rd   L1     L2 R24    n6- |  /     |           |
#    |- nin- |    |      |   |    |   |/       |         Rload
#           Cd   n7     R22 gnd   |            |           |
#            |    |      |        | - - - R8 - -          gnd
#           gnd  R1     gnd      R7
#                 |               |
#                gnd             gnd
#
#
#

In order to minimize the diffusion of more dust particles into the vented IFO vacuum envelope

BEFORE opening chamber:

-Have a  known plan,

-Heavy 1" thick door requires 3 persons- of  one experienced and one certified crane operator and steel tow safety shoes

-Block IFO beams, be ware of experimental set up of other hazards: 1064,  visible or new-special installation

- Look at the particle counter, do not open above 6,000 particles of 0.5 micron. Construction activities are winding down. See  plot of 35 days since we  vented.

-Have clean door stand for heavy door, covered with merostate at the right location and dry-clean screws for light covers,

-Prepare lint free wipers for o-rings,(no solvent on o-ring!) Kimwipes for outside of chamber and metal covers, methanol and powder free gloves

-Wipe with wet Kimwipe-tissue of methanol around the door, chamber of interest and o-ring cover ring

-Cut door covering merostate and tape it into position,..if in place...check  folded-merostate position, if dusty... replace it

-Is your cleanroom garment clean?.......if in doubt ....replace it

-Keep surrounding area free and clean

-Make sure that HEPAs are running: PSL-enclosure, two mobile units and south end flow banch

-Check the tools: are they really clean? wipe it with wet Kimwipe, do you see anything on the Kimwipe?

-You are responsible to close chamber ASAP with light door or doors as you finished for the day.

Merostate cover down is appropriate during daily brakes.

Kiwamu and Koji

ITMY (south) was aligned with regard to the 40m-long oplev with the green laser pointer. Now the cavity is waiting for the green light injected from the end table

Some tools and the level gauge were removed from the table.

BAD news: I could clearly see scatter of the green beam path because of the dusts in the arm tube. Also many dusts are seen on the ITM surface.

Picture of the ETM - reflection from the ITM is hitting the mirror and the suspension structures.

1. Shoot the ITM center with the green beam.

- Two persons with walkie-talkies required for this work.

- Turn on the end green pointer. We could see the long trace of the beam sliced by the beam tube wall.

- Look at the tube peeping mirror for the CCD.

- Adjust yaw such that the beam trace on the tube wall is parallel to the arm.

- Adjust pitch such that the beam trace on the tube gets longer. This means that spot gets closer to the ITM.

- Continue pitch adjustment until some scatter appears on the ITM tower.

- Once the spot appears on the tower, you can easily adjust it on the mirror

2. Adjust pitch/yaw bias such that the reflection hits the ETM.

- Initially the ITM alignment is totally bad. ==> You clealy see the spot on the wall somewhere close to the ITM.

- Adjust pitch/yaw bias such that the spot goes farther as far as possible.

- Once you hit the suspension tower, the scatter is obviously seen from the peeping mirror.

- You can match the incident beam and the scattering of the reflection. You also can see the reflection from the ETM towards the ITM as the spot size gets huge (1/2 tube diameter).

- We found that the bias is ~-2 for pitch and ~-6 for yaw.

3. Go into the chamber. Check the table leveling.

- Open the light door.

- I found that the table is not leveled. Probably it drifted after the move of the weight (i.e. MOS removal).

- Removed one of the round-shaped weight. Moved the other weights such that the table was leveled.

4. Remove the bias for yaw and rotate suspension tower such that the reflection hit the center of the ETM.

- Removed the yaw bias. This makes the reflected spot totally off from the ETM.

- Rotate suspension tower so that the beam can approximately hit the ETM.

- Look at the peeping mirror, the beam is aligned to the ETM.

5. Adjust OSEMs

- Push/pull the OSEMs such that we have the OSEM outputs at the half of the full scale.

6. Adjust alignment by the bias again.

- Moving OSEMs changes the alignment. The pitch/yaw biases were adjusted to have the beam hitting on the ETM.

- Bias values at  the end of the work: Pitch -0.8159 / Yaw -1.2600

7. Close up the chamber

- Remove the tools and the level gauge.

- Close the light door.

 I noticed that someone, that wasn't me, has edited the wiki page about the netgpibdata under my name saying:

 " [...]

* A4395 Spectrum Units
Independetly by which unites are displayed by the A4395 spectrum analyzer on the screen, the data is saved in Watts/rtHz"

Lately I've been trying to sort out the problem of the discrepancy that I noticed between the values read on the spectrum analyzer's display and what we get with the GPIB interface.

It turns out that the discrepancy originates from the two data vector that the display and the GPIB interface acquire. Whereas the display shows data in "RAW" format, the GPIB interface, for the way the netgpibdata script is written, acquires the so called "error-corrected data". That is the GPIB downloaded data is postprocessed and corrected for some internal calibration factors of the instrument.

 I restored damping to all SUSes except ITM-east. The ITMX OSEMs are being used in the clean assembly room.

This morning, at about 12 Koji found all the front-ends down.

At 1:45pm rebooted ISCEX, ISCEY, SOSVME, SUSVME1, SUSVME2, LSC, ASC, ISCAUX

Then I burtestored ISCEX, ISCEY, ISCAUX to April 2nd, 23:07.

The front-ends are now up and running again.

Albeto and Koji

We took the tip replacement from the blue tower.

I am looking at http://www.cooperhandtools.com/brands/weller/ for ordering the tips.

The burnt one seems to be "0054460699: RT6 Round Sloped Tip Cartridge for WMRP Pencil" We will buy one.

The replaced one is "0054460299: RT2 Fine Point Cartridge for WMRP Pencil" We will buy two.

I like to try this: "0054460999: RT9 Chisel Tip Cartridge for WMRP Pencil" We will buy one.