CCDTL 6th technical meeting for production of 7 CCDTL modules for Linac4 (ISTC contract #3888 and #3889), 1-3 March 2011 at BINP


BINP Novosibirsk: Alexey Tribendis (ISTC Projects 3888/89 manager, deputy RF group leader BINP), Yaroslaw Kryuchkov (design group leader)

VNIITF Snezhinsk: Mikhail Naumenko (ISTC Projects 3888/89 sub-manager, design group leader VNIITF), Gennady Ostashkov (Technologist), Piotr Taushkanov (Technologist), Dmitry Vavasov (Design Engineer)

CERN: Alessandro Dallocchio (responsible for the mechanical design of Linac4 accelerating structures), Frank Gerigk (ISTC Projects 3888/89 coordinator at CERN, responsible for accelerating structures in Linac4), Jean-Michel Giguet (responsible for installation of Linac4 accelerating structures)

List of talks

Indico meeting CCDTL-Tech-6

Supporting documents in EDMS

  • EDMS 1135929 Q measurements on first module
  • EDMS 1135930 Tuning range for different angles of the tuners on the coupling cell
  • EDMS 1135932 Financial status of project (restricted access)
  • EDMS1136307 Tolerances for laster tracker target holder shafts and holes
  • EDMS 1136210 Polishing "wheel" used for Helicoflex surface areas
  • EDMS 1142472 Memorandum on revised construction and payment schedule

Technical summary

1. Q Measurements
2. Coupling Cells
3. Metrology
4. Test results at VNIITF of the first module
5. Discussion on disassembled module
6. Support
7. Project Schedule
8. ISCT contract situation
9. Repair of pumping ports
10. Next Meeting
11. Action list

1. Measurement of Q values

Compared to 2D simulations (that means neglecting the influence of joints, coupling holes, surface roughness, stems) 82% of the theoretical Q value was achieved for tank3. Measured: 43500, 2D calculation: 53100. In the first tank a value of 77% was achieved: measured: 40200, 2D calculation: 52000. Apparently a different machining technique was used for tank1 and 3 resulting in a better surface for tank 3. It will be clarified at VNIITF what the difference was, with the goal of using the better technique for the remaining tanks. Both measured tanks have better results than the prototype tanks of project #2875, where 76% and 78% of the 2D Q value was achieved.

2. Coupling cells

  • The frequency of coupling cell now at 354.7 MHz (vacuum) with having 0.5 mm surplus length on the coupling cell. The surplus length, which was added because there was an uncertainty in predicting the coupling cell frequency, corresponds exactly to the frequency delta of 2.5 MHz.
  • The "0" position for the tuners is chosen at +20 mm penetration. This results in a tuning range of approximately: -250 kHz +750 kHz per tuner,
  • During the tests the tuners were mounted at 60 deg angles (because of space constraints of the support) instead of vertically. This resulted in a smaller tuning range than expected. Simulations carried out during the meeting showed that once the tuners are vertical the tuning range is as expected. See figure below (for 2 tuners).


3. Metrology Tests at BINP

  • 4 half tanks have been measured at BINP, one external flange diameter is non-round by 0.56 mm, the other 3 are 0.16,0.12,0.23 mm. This is too much to use them together with the drift tube alignment tools, (the position of the drift tube is given by the centre of the beam hole and the centre of the flange diameter). To avoid that this happens for the remaining tanks a change is made in the drawings for manufacturing, and the existing half tanks will be corrected either at BINP or VNIITF.
  • On the CERN drawings the minimum diameter for target holders should be 10.000, is this practically usable? AT BINP they have some test pieces with 9.9xx mm, which one should fit?

  • The holes of some of the laser target holders were "too precise" 10 mm diameter +2/-12 um, with the effect that the target holders could not be inserted in all holes.
  • The position of the holes that were usable were measured, and also the position of the alignment "bridges" with respect to the beam pipe openings was measured, - the data will be put together soon,
  • A comparison between VNIITF and BINP CMM measurements and laser tracker measurements at BINP is in good agreement (distance of holes on alignment bridge),

4. Test results at VNIITF of the first module

presentation slides

Alignment test

  • A first alignment test was made at VNIITF to see if everything can be assembled. Two old machine supports were used for this test onto which 3 levelled frames (with a smooth surface) were mounted.
  • A test of the lateral precision was done: no force was applied during assembly and then the lateral position of the cavities was checked with respect to a precision bar. The bar has a precision of 80 um over 3 m: max. lateral variation that was measured: 0.19 mm. The measurements were taken with respect to the feet, which are used for lateral alignment.
  • Vertically blocks were put on the alignment bridges and then the distance between the half tank axis and bar was measured. The maximum distance variation was 0.189 mm (total).
  • For this measurement the beam axis was determined as the centre of the beam pipe and the centre of the helicoflex groove (50 um on diameter).

Machining procedures

  • Sequence: welding of feet, all other welds, machining of beam hole and beam pipe connection, machining of inner surfaces with respect to the beam pipe surface, machining of outer surfaces including the reference surfaces on the feet,

Acceptance tests, etc

  • vacuum leak checks were done,
  • vacuum baking, 1 hour 350 deg to check Cu plating, no degradation found, remaining modules will not be baked,
  • water channels checked at 16 bar, and also that they are not blocked,
  • transport under nitrogen from VNIITF to BINP was successful,

5. Discussion on disassembled module

  • One damaged joint area for a stem was found, and several of the big half tank joint areas (Cu plated) have small defects. On the RF port there are machining marks perpendicular to the joint direction, and also there are some pits on the big joint areas and one or two on the RF port. On one coupling cell flange there was a Cu bubble.
  • It was agreed that these areas will be corrected by polishing before assembly with Helicoflex joints. It was also agreed that more care will be taken at VNIITF of all areas which take Helicoflex joints. CERN will provide their method of polishing these areas.

  • On one nose there was a circular "scratch" were one can see the stainless steel below, (this is one of half tanks, which will be repaired due to the wrong grid direction, that means it will be replated)

  • On the two half tanks with the wrong grid direction in the vacuum port: these ports will be remove and new grids re-welded. After this procedure the half tanks will be measured and if there are unacceptable deformations the half tanks will be re-made. If the deformations are acceptable we will use the repaired units. VNIITF will then remove the Cu plating and re-plate them,

  • On one large joint area one could see some "liquid shaped traces", which most likely come from chemical treatment, when the cavity was moved from one bath to the next due to a rubber sealing, which was not perfectly tight.

  • On the Cu plated areas which hold the joints, the Cu plating often went over the edges, it was recommended to remove these parts, to avoid that when the half tanks are assembled, pieces of Cu are detached and fall into the area of the joints.

6. Support and installation procedure

Installation procedure

  • first the centre half tanks are put onto the support and fixed together,
  • then the half-tank with the wave-guide port is fixed onto the support,
  • mounting of coupling cells on both sides,
  • mounting of half tanks that connect to the coupling cells,
  • outer half cells,

This is different from the procedure that was used to mount the first module at BINP in that the half tanks are first fixed together and then put onto the support. To be able to assemble the half tanks before connecting them to the coupling cells we need a seperate support table to assemble the half tanks. For this we should foresee a minimum distance of 10 cm between the half cells. For mounting the drift tubes one can face the back sides of the half tanks. To do this the separate support has to be long enough for longest half tanks back to back (taking into account the length of the couplings cells sticking out in the back).

When mounting the 2nd cavity we must be sure that the support does not tilt to one side, for this we may foresee an additional support to be put under ends of the frame during installation, which keeps it from tilting.

As a back-up, one can even do a better alignment, if we use measured position of the cavities instead of just pushing them against the reference surfaces, this could also be used if we don't find a supplier who can do the required tolerances of 0.1 in transverse/longitudinal, on the support pads for the cavity. On the non-reference side there is a small L-shape, because it cannot be machined completely flat, when the machining is done on the assembled support.

Support features

  • VNIITF put the first module onto the design of the CERN support frame and the integration of the complete module seems ok.
  • The excentric screws are not yet optimised.
  • The pads should be made of hard steel (or surface hardened) and the surface should be smooth.
  • Concerning the grooves in the pads, which support the cavity: these should have tolerances on surface roughness, and the grooves should have a parallelism of 0.1 (same as vertical alignment surface). Only the grooves on the side with the alignment surfaces must have this tolerance, on the other side they can be free. The same tolerances should be applied to the separate assembly table.
  • The support pad of the inter-tank elements must be dismountable, so that we can put the complete tanks, then move them so that we have enough space to connect the coupling cells (we need ~80 mm of movement along the beam axis). The support pads for the cavity are already long enough to allow this movement .
  • The original "ears" for transport were not accepted by CERN transport. This was replaced by putting vertical cylinders with holes through the frame. The bulk pieces onto which the ears were fixed can probably be taken out, but i) the transverse intertank support bars should not rise, ii) the cavity support pads should still be connected vertically to the transverse intertank supports.
  • The parallel flatness on all support pads on the frame is defined as 0.1 mm (horizontal and vertical), this value is confirmed and it is the one that was used for machining the test frame at BINP. After adjusting the feet of the test frame an accuracy of 50 um was achieved. Roughness on pads of Ra 1.6 was not relaxed so that we can easily slide the cavities on the support.

7. Project Schedule

  • VNIITF announded that module no 2 will be completed by the end of March (now under final machining, soon to be copper plated). Then every 2 months one module will be completed. The last module will take 3 months due to Christmas and be finished in February 2012.
  • Transport to BINP takes around 10 days.
  • The 1st module will be returned with the truck that brings the 2nd module, so that it is back at VNIITF in April 2011. It will then come back to BINP together with module 3.

  • For the assembly at BINP 3-4 months are necessary. For the last modules this time is expected to go down to 2-3 months.

  • It was agreed that the payment schedule be linked to the following milestones:
    • 260 k$ to 3889 will be paid after completion of 4th module at VNIITF.
    • 148 k$ to 3888 will be paid after arrival of the 3d module at CERN.

8. ISTC Contract Situation

  • We need to make sure that after the expiry of ISTC, that the CCDTL belongs to CERN and can be shipped from BINP to CERN , this question should be cleared in September and we should foresee a back-up solution via BINP

  • The project needs to be extended, 6 months would bring us to May 2012, which is needed for #3889 and to cover the assembly of the last module, we need 9 months for #3888 so that the last 3 months are covered, which are for assembly and commissioning at CERN.

  • The present financial situation is listed in EDMS 1135932.

  • A memorandum must be written for the project extension, construction and payment schedule.

9. Repair of Pumping Ports

Principle: cut the tube where it becomes conical, machine out the grill, and weld a new cone (weld from inside) with a grill. This way the weld is farther away from the cavity and there is not so much risk for deformation. That means that the grill will be lower than in the original design, which will slightly increase the RF losses and the cavity needs to be tuned differently (can be done with drift tubes). The inside will be a continuous weld but thin, on the outside they will make some welding points to give the needed stability so that the pump can be carried. The detailed procedure can be seen here:

10. Next Meeting

The next meeting will take place when the 2nd module is ready for vacuum tests at BINP. This should be in June

11. Action list:

action institute/person status/result completed
verify the exact diameter for the laser tracker target holders CERN pending  
send photo or drawings of the "polishing wheel", which was used at CERN to prepare the Helicoflex surfaces CERN, J.M. Giguet done EDMS 1136210 2011-03-17
clarify the ownership of the CCDTL tanks after expiry of the ISTC, this should be clarified in September with the ISTC CERN/BINP/ISTC pending  
clarify tolerances of laser tracker target holder shaft and its hole CERN done shaft: +0.000 - 0.0005 inches, hole: -0.000/+0.015 up to +0.018 mm, see EDMS1136307 2011-03-18
memorandum for project extension, fabrication and payment schedule CERN, F. Gerigk/BINP, A. Tribendis done written and signed: EDMS 1142472  
remove material defect on the inside of the tank, on the surface where the stem Helicoflex joints sits BINP/VNIITF in progress see EDMS 1145292  

-- FrankGerigk - 16-Mar-2011

Topic attachments
I Attachment History Action Size Date Who Comment
PNGpng tuning_range.png r1 manage 75.2 K 2011-03-16 - 18:13 FrankGerigk  
Edit | Attach | Watch | Print version | History: r6 < r5 < r4 < r3 < r2 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r6 - 2011-05-17 - FrankGerigk
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    SPL All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright