First up is a check scan of 18O4+. There is a definite vertical offset compared to earlier today. Otherwise, it correlates well enough. The same beam taken with a greater vertical offset follows, together with the modified emittance panel.

A mass scan from A/q of zero to 25 is in progress. Data file (large): masscorr_mcis_090708.txt. The source is optimized for 18O4+. Fishing for 18On+ states and their emittance scans to follow. Attached is one screenshot of the Gabe's Calibrate interface. I ran it several times with different peak selections from the same data. I entered into the EPICS mass correlation program a best average of the calibrations for the two constants.

* 18On+ emittance scans at present ion source settings: complete today.

* 18On+ frequency measurements: Thursday.

* 17O source gas: open end of Thursday.

* 17On+ frequency survey: Friday.

* 17On+ emittance scans at idealized source settings: Monday.

* Development completion: Tuesday.

* 18O4+ setup for beam schedule: Wednesday.

Tune saved as 090708_1200.snapmcisfc6. I settled on the 15.486GHz optimal frequency, and tweaked that for effect. Then I ran a few emittance scans. During that process, I noticed that the beam intensity had stepped down and up a few times. So I tweaked the frequency again to return 30enA on FC6. Note that in this process the beam shifted slightly in the horizontal plane. The emittance shape is different also. Attached are emittance scans in sequence, the last one being the "real" one for this test. Finally, before and after RF screens.

15.480GHz => 30.0nA

P.S. Appended is a stripchart of the 18O4+ beam on FC6 following the emittance scans (magenta trace, between T-5 and T-4h). Not bad, eh?

Starting frequency: 15.7864GHz => 870pA on FC6  Step size: 1MHz. Selected peaks shown. There were many smaller resonances in-between also.

Strong peaks at:

15.8594GHz => 7.5nA

15.9834GHz => 1.3nA

16.0354GHz => 1.3nA

16.107GHz => 1.5nA

16.183GHz => 0.9nA

16.247GHz => 1.4nA

16.284GHz => plasma extinguished

16.248GHz => plasma re-ignited

15.860GHz => 7.5nA

15.685GHz => 3.5nA

15.518GHz => 2.5nA

15.486GHz =>  27.5nA

15.479GHz => 4.3nA

15.277GHz => 2.8nA

15.130GHz => 3.7nA

15.035GHz => 6.2nA

14.903GHz => 3.8nA

14.799GHz => 2.6nA

14.763GHz => 11.8nA

14.554GHz => 8.8nA

14.455GHz => 4.2nA

13.588GHz => 1.1nA

13.398GHz => 3.0nA

13.281GHz => 2.7nA

13.223GHz => 7mA (unstable)

13.153GHz => 7.5nA (unstable)

13.061GHz => 10.0nA (unstable)

13.019GHz => 40.0nA

13.002GHz => 50.0nA (unstable)

12.986GHz => plasma extinguished

13.089GHz => plasma re-ignited

15.4798GHz => 30.0nA 

After accessing the OLIS cage and opening the 18O source gas valve, I restarted the Supernanogan source last evening around 18:00 ( just to the left of -15h on the attached stripchart), with the line set up for 18O4+. The beam on FC6 grew nicely overnight. MCIS pressure moved from ~6.5e-8 to ~7.1e-8 Torr since the 18O source gas bottle was opened last evening.

As noted earlier, we have 17O and 18O source gas attached to the Supernanogan source chamber. Both are valved off presently. The source has been running for a while. I set it to 18O4+ on background, which gives about 125epA on FC6. Now let's stop the source, open the bottle and try the same again.

I opened the 18O valve just around 18:00. Beam is on now, around 150pA. That's a good sign. It's growing too. Even better. The MCIS tank pressure moved from ~6.5e-8 Torr to ~6.6 as I opened the valve, which is located upstream of a calibrated leak. Note that both source gas bottles have 10-5 calibrated leaks in series, so it will take a while to reach equilibrium. I attempted a couple of emittance scans, but it's too early yet.

Keerthi and I were front and centre doing ISIS emittance rig testing this morning. He ran it, and I tried to find ways to run the data in Matlab afterwards. Main limitation up there at the ISIS Console is lack of any ssh-enabled terminals. Keerthi's wireless laptop is okay, but this stuff pushes the poor little thing to the edge. Plus, there were some X display issues on it. He has Matlab installed locally, but it takes a bit of a rocket scientist to run it, and once you do get it going, Matlab complains that the data file format is incorrect. We believe it's a matter of line endings: Windows versus the rest of the unix/Linux/Mac/VMS world. So it goes. All in all, a successful morning. There is enough data to think about how it all works, and we have a couple of hours of R&D again tomorrow morning before production beam. I'll put in a few emittance panels here in a while. Keep in mind that these are with beams on the order of mA, not nA!

DataFile    Pulser     Arc       90%     86%     63%    cut%    #71    #72    111    112    113    30    nor. Emit.
 95152        5%        6A      4.01    3.51    1.91     1%                    0      4k     4k             0.101
102032       10%        6A      1.59    4.02    2.1      1%                    0      4k     4k             0.116
102904       20%        6A      4.89    4.29    2.21     1%                    0      4k     4k             0.122
103602       30%        6A      5.18    5.11    2.7      1%                    0      4k     4k             0.131
105650       40%        6A                               1%                    0      4k     4k        
             60%        6A                               1%                    0      4k     4k        
115852       70%        6A      6.21    5.56    3.07     1%                    0      4k     4k             0.157
120728       70%        6A      6.07    5.41    3        1%    2159            0      4k     4k             0.154
122216       70%        6A      5.96    5.32    2.94     1%           2247     0      4k     4k             0.151
123149       70%        7A       (no tuning)             1%           2247     0      4k     4k             0.147
124609       60%        7A       (bit of tuning)         1%     all tuned           all tuned                    

Meanwhile, OLIS and the Supernanogan were off overnight in preparation for transformer tap selection for the sake of the CSB. See WP#2009-07-07-7 for more details.Turbos were turned off in preparation for attaching 17O and 18O source gases. Both are attached now, and valved off presently. I restarted the Supernanogan with about a nA of 18O2+ at FC6. Let it cook for a bit.

Well... my browser blew away just before I submitted version one of this entry. Let's try again.

Keerthi changed the tune this morning to 10kV on the MCIS bias in order to allow the full throw of the Enzel lens power supply. (Remember, for the moment the panel displays twice what the physical power supply is actually delivering, because it is a 10kV supply, whereas the control system expects a 20kV supply. Such will be the reality in the fullness of time.) He looked at 18O in background gas. It's there all right. We have an up-coming TIGRESS run that calls for 18O delivery. The schedule shows 21Ne for that run, which goes from July 15 - 20.But in fact 21Ne was just a placeholder and 18O is really required. That put us in a spot, because we had not ordered a backup bottle of 18O source gas, and our present bottle would only give us a very few days of operation from the MWS. However, the same bottle would run for months in the MCIS.Pick an option. We floated this situation at the Beam Delivery meeting today. The user really wants 18O, not just an A/q=18 or 9 pilot beam. Good. We can delivery plenty of 18O4+ from the MCIS. Gordon confirmed that 18O from the MCIS is a done deal, and the general attendance concurred.

Run plan for the short term:

1. Complete 18O background studies today.

2. Shut down OLIS for transformer work in the morning.

3. Spin down turbos for source gas change.

4. Install 18O source gas tomorrow morning.

5. Restart vacuum system and source mid-morning.

6  Condition as required.

7. Check intensity and emittance of 18O4+.

Note that 18O4+ is the only charge state that fits within the acceptance range of the DTL. However, it is on the lower end of the Pre-Buncher. Historically a low-level pre-amp was required on the Pre-Buncher in order for it to operate properly. That's between Marco and Ken Fong et al.

Well... Firefox crashed again. What's up with that?

I am having an argument with the OLIS emittance rig. It is returning zero data on 18O scans. Try with 12C, where there is oodles of beam. Aha! A spot! A couple more passes, and we have a "perfect" 12C1+ emittance scan. Here it is. Note: Slits are at 5mm setting.

Now I am trying a scan of 18O2+ with some 1.5nA on FC6. Here we go... Faint, but it's there. Slightly different vertical offset. Done.

Prepare for orderly shutdown of source.

We had 16O3+ on FC6 on Friday. An IOSIOC reboot occurred mid-afternoon Friday unfortunately. Operators restored beam at about 06:00 on Saturday. Beam came back fine. I requested FC3 in at 18:00 Saturday to get some temperature correlation with total beam. Maybe there is a correlation, but a longer stretch is needed. Something stopped at 18:00 on Sunday. Keerthi told me this morning that he found steerers off, and switching them on restored beam instantly. Good. Plot attached. Red is FC6 beam. Brown is FC3 beam.

This morning Keerthi began studies on 18O background. Optics are now set for 18O2+ delivery. We will pursue that today. Tentatively the run plan for this week will include:

1. 18O background studies.

2. Repeat emittance scans of the charge states of 16O with slits open.

3. Prepare emittance versus charge and frequency versus charge for 16O, 17O, 18O, 20Ne, 21Ne and 22Ne.

16O2+ beam on FC6 = 10nA as before. Move to 16O3+ A/q = 5.33...

A frequency scan between 13 & 14GHz. I panned closer around two peaks at 13.11 and 13.12GHz. Best frequency so far is 13.1321GHz, which provides a reasonably stable 150pA of 16O3+.

Later... more scans, etc. One of several possible solutions for 16O3+: ~450pA at 13.6348GHz. Let it cook for a while.

Here are some pics and data from total MCIS beam on FC3 overnight. A bit inconclusive, but there may be a temperature correlation.

Let's take a look at varying frequency versus beam current for 16O2+. Starting RF values shown below. FC6 = 1.09enA.

What I did: Pan RF frequency between about 12.5GHz and 14GHz. Plasma seemed to extinguish at a frequency under 13GHz.

What I found: Two peaks on the low side, right around 13GHz.

How I proceeded: I panned across those two peaks for stability. I found that the stronger peak in fact was noisier, and also at risk of losing plasma.

Screengrabs:

1. Starting RF settings

2. Lower window of "good" RF frequency.

3. Higher window of "good" RF frequency.

4. Original beam plot.

5. Beam plot at middle of "good" frequency range.

6. Beam after frequency restored to starting value.

7. Frequency survey correlation plot.

As it turns out and as you can see, the original frequency not only provides the same intensity of beam, the beam is also visibly more stable. (Keerthi found this one.)

Lesson learned: Discovery of the ideal frequency is an iterative process. At least three passes, each successively finer, are required.

Happy Canada Day!

I noticed a long-term trend in beam intensity. I have every reason to believe that the source itself is rock-solid. I think it is temperature variation affecting the Philtek power suply on the separator magnet. What bothers me is that the stability program appears to not be compensating for this. I am looking into this now. Here is a plot of beam current, mag field and temperature (at AHU-1, the best I could do). There appears to be a delayed response to temperature. I have turned off our Hall probe for an hour or so to watch the response on a stripchart.

Well, there were results, but not what I expected. The beam became nearly out of control while the stability program was disabled. What's more is that the correct field came back immediately upon turning on our Hall probe again. Here are a couple of stripcharts, one snapped right after turning on the probe, and again a while later The stability program was disabled during the middle couple of hours or so. Note that the beam is oddly more stable in the short term now also.

There seemed to be some small change to the beam over the week of TACTIC running. I would like to poke around a bit. The emittance plots suggest some vertical movement at least. The two plots are the before (emittance) and after (emittance). The two tune saves are 090624_0930.snapmcisfc6 and 090629_1730.snapmcisfc6 (present tune) respectively. Here is how they look juxtaposed. I restored the downstream elements. Let's see what that looks like in an emittance scan. They compare well! All changes in beam position can be completely attributed to tuning changes. End of story. I have restored the tune and centred up the emittance plots. Done.

First, with upstream slits opened, then with both open. Original slit setting was 1mm. Finally, restore slits to 1mm and see if it reproduces. It does.

The first emittance scan sort of got interrupted; it auto-restarted just before the previous scan had finished. I thought the resulting plot was somewhat interesting. The emittance isn't half-bad either. A better scan follows. TACTIC took beam again for a while this afternoon to shake down a timing discrepancy. We got it back around 4pm. One more emittance scan for good measure.

Setup teething pains... The Start button on the emittance panel worked this time. But ICR called, requesting a few minutes more for Marco. No problem. When I went to restart the scan, the button no longer worked. Restarting X11 made no difference. However, there is a significant update available for X11, so I am installing it now. I perceive that it works, yes, significantly faster than its predecessor. Good, all good.

Supernanogan development starts today. The TACTIC run, which ended around 08:30 today, appears to have been a success. TACTIC used our beam for setup and pilot, then took 8Li from downstairs. Marco is doing some EOB checks now. We will begin at 10:00. First part of our run plan will be to compare the present 16O2+ beam with how it looked when we started, and identify what might have changed.

Proposed Run Plan for Today &ff

1. Compare present 16O2+ beam with start of TACTIC run and identify change factors.
2. RF frequency scans versus each charge state of 16O.
3. Identify frequency with largest bandwidth for each state.
4. Emittance scan for best beam of each state.
5. Compare with earlier emittance scans.
6. Select one clean beam and test whether or not RF amplitude changes best frequency.
7. Comprehensive mass scan.
 

We handed over a 16O2+ beam on Wednesday morning to Sophia for TACTIC setup. On our side, everything went well. TACTIC switched to RIB on Saturday night. See ISAC elog entry. We remain on standby for pilot beam. Schedule switched to development tomorrow morning. Attached is a stripchart of our beam for the last several hours. As good as it gets! Save file: 090629_1730.snapmcisfc6.