2020-01 Tank opening report

Date

2020/01/14 00:00to 2020/01/31 00:00to 2020/02/12 00:00 

Attendees

• @ptikkane@helsinki_fi

• @httyrvai@helsinki_fi

• @piekie@helsinki_fi
• @mimanner@helsinki_fi
• @atacan@helsinki_fi
• @kmizohat@helsinki_fi

Goals

• Perform a thorough scheduled service of the 5-MV tandem accelerator
• Replace the corona control rod pressure seals

Action items

Note on the installation of new resistors

The distance between the tube resistor end-caps is in the original construction about 7 mm (= 25 mm - 18 mm). According to the original documentation, the spark gaps at the accelerator tube between the electrodes is adjusted to 2.8+- 0.2 mm:

This means that the tube spark gaps fire before the gap between the resistor end-caps.

The new resistors from Nicrom have protecting end caps with an outer diameter of ø21.4 mm, meaning that the gap reduces to 25.0 - 21.4 mm = 3.6 mm, which is still larger than the accelerator-tube spark gap. This means that it's safe to install the new resistors without modifications. Care must be taken, however, to check that the resistors are fixed tightly and with tight tolerances to keep the end-cup to end-cup distance at 3.6+-0.2 mm or so.

The new resistors seem to fit well, leaving a gap large enough.

Note on the corona rod seals

The original axial seals are inadequate, as it turned out. Axial seals are specified typically for a pressure difference of 1 bar only and 10 bar max. They will be replaced with X-ring seals:

The larger diameter axis needs four x-ring seals, two for the ø25 mm axis,

For the outer diameter of ø42 mm, two X-rings are needed:

The smaller has dimensions ø14.4 mm x ø28 mm, that fit two inner X-rings

and two outer rings

Because the seals need some lubrication, the muffles that accommodate the seals must include lubrication holes to insert some grease in the space between the sealing X-rings. Must install lubricating nipples.

It turned out that the main rod was severely worn out and had to be replaced. The inner rod is replaced, although it was still quite OK. Unfortunately, the old and new rods were not directly interchangeable, but had to be taken to the machine shop for slight modifications.

The base of the corona rod flange was modified to make a O-ring groove for a ø42.0x2.0 mm ring.

The end of the corona control HV-cable is very difficult to make a pressure-tight joint. It must be replaced with a joint that is secure by design. Use a 1/16-inch connector by Swagelok, part Item # SS-100-1-2ST, Male Connector (SAE/MS Straight Thread) - Fractional (for details, see here.) The end of the HV-cable is brazed on a 1.58-mm (1/16") male pin which is then tightened with a standard SS-ferrule on the body. The part has an O-ring seal and the old part must be modified to include the SAE/MS 5/16-24 female thread. The part is in EGP84.3dm rhino3d-file.

The first two pressure tests failed at 6-7 bar(g). We first thought that the fluxed joint was the reason and Pietari made it again with no luck. Finally on Monday February 3rd we realized that the leaking part may be the welded seam of the spare Russian inner tube. This was confirmed with a classic bicycle tyre leak test using pneumatic air and immersing in the tested part in water bath:

The final pressure test on February 5th was successful, the tank (puupäähattu) was filled with He gas up to 18 bar (g) as shown in the pressure regulator "PAINEKOEVENTTIILI". However, the digital gauge shows only up to 15 bar, so that the pressure was lowered accordingly. The pressure dropped from 15.11 bar (at 13:17) to 14.65 bar (at 14:11) which is quite reasonable (0.46 bar) regarding the complexity of the whole system. It's decided that the corona rod mechanism is safe to install back in use.

Note on securing the corona rod end

There are a small risk that the fixing of the corona rod outside of the tank fails (this has happened once) and the pressure pushes the rod completely out leading to a gas leak. To prevent that to happen, the original design had securing ring at the end of the outer tube. However, there wasn't such a ring installed nor there was any groove for the ring!

A suitable ring for ø25 mm axis, DIN471, is

Note on the charging belt

It was decided to install a new charging belt, this time a white Forbo-Siegling one. The main properties of the belt are

• Total length L = 10900 mm
• Width 560 mm (old was 540 mm)
• type 900277, E 8/2 U0/U2 MT-NA white FDA
• k1% is 6.5 N/mm
• Elongation at break longitudinal 26%
• minimum fitting elongation 0.3%
• maximum fitting elongation 1.0%

A suitable elongation would thus be 0.5 to 0.8%, corresponding to 3.25 N/mm to 5.2 N/mm. For a 560 mm wide belt, this corresponds to 1,820 - 2,912 kN. In practice, about 1,8 to 2,9 kN is OK.

But be aware of the fact that we have two branches! This means that 1% elongation in the total length is only 0.5% on each branch. A properly installed and tensioned belt is about 70 mm longer than without tension. Neglecting the effect of the upper and lower rollers, this implies that the distance between the rollers increases for about 30 mm during tensioning.

A note dated 23.05.2006 says (P.O.T.): Uusi hihna on Siegling E8/2 U0/U2 MT-NA white, leveys 540 mm. Hihnan kireys on siis 8 N/mm eli 8 x 540 N = 4.32 kN. Bahnerin mukaan noin 80% tästä riittää, eli 3,46 kN (per puoli, kun on kerran kaksi kireysmittaria!) What to think about this? If we follow this recommendation, the new belt would be tensioned to 0.8x560x8 N = 3.58 kN. Perhaps this is right! The fitting elongation is just for the fitting purposes, and after that the belt is tensioned for the proper tension for running! It makes more sense. At the tension recommended by Klaus Bahner, the belt is elongated by 0.8x8/6.5*1% = 0.98%. This is about the same as the k1% value.

For comparison, the key properties of the previous green/blue belt were

• Total length 10900 mm, width 540 mm
• Type E 10/M U1/U3 NA green, 900068
• Total thickness 3.1 mm ± 0.35
  
• Weight 3.3 kg/m² ± 0.35
  
• k1% value is 5 N/mm
  
• Breaking force 105 N/mm
  
• Elongation at break longitudinal 15 %
• Elongation at fitting min. 0.3 %
• Elongation at fitting max. 1 %
• Friction coefficient of driving face against steel panel according ISO21182 0.28
• Permissible operating temperature-10/80 °C, for a short time 100 °C14/176 °F, for a short time 212 °F
• Hardness of top face coating as perDIN 53505 (Shore A) 8
• The recommended tension for this belt was 0.65*5*540 N = 1.755 kN, corresponding to a 0.65% elongation

Note on drive belts of the alternators

The drive belt of the Russian 400 Hz generator must be installed before the belt can. The previous belt is Contitech FO-Z XPA 1450 Lw L=L EL (w=12.7 mm)

The drive belt of the turbopump generator can be installed later, The previous belt is Contitech AVX10 x 635 La

The generator panel has to be modified to be more informative. See the corresponding InTouch panel.