NOTE: To see a detailed view of each image, click on the image.
Tests of the effect of beam heating on the Moller target are underway at Kharkov. These tests are necessary, as the electron beam will deposit energy in the iron foil, thus heating it up and depolarizing it. The heating tests were performed as follows: Two 4 micron iron foils are deposited on 30 microns of aluminum backing. the aluminum has a hole in it where the beam would pass through. These two foils are then mounted parallel to one another. A resistor is then placed between them such that it is perpendicular to the foils, and touching each of them. See figures 1-3. One foil is insulated from the support with mica.
Figure 1.
Figure 2.
Figure 3.
The resistor consists of a sapphire rod of cylindrical cross section 6 mm in diameter. The sapphire has good thermal conductivity but is a good electrical insulator. Around this rod is wrapped insulated copper wire which is the resistive material. The ends of the sapphire rod have magnets attached to them. The tips of the magnets come to points of diameter 0.5 mm. These come in contact with the foils at either end. See Figure 4. The magnets provide mechanical contact between the resistor and the foil. The length of the resistor is a bit larger than the separation of the foils to provide pressure for electrical, mechanical, and thermal contact. There is no solder in the contact.
Figure 4.
The resistor is kept in place by four thin stainless steel wires which are each attached to an end of the sapphire rod (at the junction of the sapphire and the magnetic tips.) See Figure 5.
Figure 5.
This apparatus is placed in high vacuum (10-8) so that the only
source for cooling of the resistor is the two foils. (Radiation is said
to be small.) The resistance as a function of the temperature was calib-
rated with a thermocouple and R(T) was found to be very linear. (I saw
this plot.) Resistance is about 6 Ohms at 300 K and about 1 Ohm at 77K
The idea is to change the current, I, and measure the resulting
voltage.
Change I--> measure V--> get R--> gives T.
This is done for different cooling. Namely, no cooling and liquid nitrogen
cooling. In practice we intend to cool the target with 15-20 K gaseous
He, but these measurements serve to check the calculations of the target
heating and cooling.
As previously mentioned with LN2 cooling, with no heat input(77 K)
the resistance is about 1 Ohm. After heating, resistance goes up to 12
Ohms which corresponds to about 500 K. Higher temperatures cannot be
obtained because the aluminum and insulation on the copper wire will melt.
The resistance of the foil is not negligible, but it is small and
they measure it. They measure with and without foil to make a subtraction.
I did not see the final results as they want to make some more
measurements, but the conclusion is that the measurements and the calcul-
ations are in agreement at the 20% level. This means that without Be
backing, we should be able to run at 30 microamps as far as hearing is
concerned.
One issue discussed is that this test applies heat at the surface
of the target, whereas the beam heats in a volume defined by the inter-
section of the beam and the target. This is probably a small difference.
The basic idea of measuring the target magnetization is to measure the induced emf in a coil wrapped around the iron foil which is generated when the magnetizing field is reversed (Faraday's Law). The induced emf has a contribution from the Fe foil magnetization, and other contributions which must be subtracted. Tests were preformed with a Fe foil, about 8 cm long, 1 cm wide, and 6 microns thick. A pickup coil consisting of 200 turns of 80 micron copper wire is wrapped around it near one end. A similar coil is placed nearby for subtraction. Helmholtz coils provide field of 100 Gauss. See figure 6.
Figure 6.
A DAC provides a sawtooth wave of +-5 Volts which is fed into a X2 amplifier which drives the Helmholtz coils. It seems to me the Helm- holtz coils are in series. They said they were in parallel, but perhaps I misunderstood. Anyway, a plot of voltage on the Helmholtz coils versus time is shown in the plot below (Figure 7). X-axis is time in milli- seconds, Y-axis is Helmholtz voltage.
Figure 7.
A hysteresis curve is shown below (Figure 8) with Helmholtz voltage on the X-axis (essentially I) and B on the Y-axis.
Figure 8.
The curve does not flatten out on the left and rightmost sides as one would expect at saturation due to the "air" contribution which is to be subtracted. See sketch below (Figure 9):
Figure 9.
In figure 7, the time dependence of the pickup coil voltage is plotted along with the Helmholtz voltage. The sharp spikes are due to regions of large slope in the hysteresis curves. The low DC level is due to the "air" contribution. When this is subtracted using the signal from the coil which is not wrapped around the foil, one obtains the following plot (Figure 10) in which this DC level is absent:
Figure 10.
The corresponding subtracted hysteresis curve is shown below (Figure 11), where, I believe, it has been averaged over many cycles.
Figure 11.
Noise in one cycle gives error of 10% in the polarization. They average over many cycles. Softer iron foils are desire (or more field) to make a more narrow hysteresis curve. They want to anneal foil to make it softer, but can't use aluminum backing because melting is a problem. Copper is better. It's O.K. for annealing, but difficult to remove for beam hole. Copper is O.K. for target cooling purposes, but if the beam has halo, it could cause a problem with backgrounds. They need DSP Programmable Amplifier 1402 ($1875) but want to confirm when they go to CEBAF. They will need an ADC. Total needed as of three years ago: $8-10K. They do not care if it's VME or CAMAC.
The Kharkov group provided detailed drawings of the target design. The plan is to have it constructed, and to bring it to CEBAF May 10. The following aspects of the target design were discussed as they were of interest to the Kentucky group. Entrance and exit of target have flanges which are rotatable and removable. (Flanges of this type are desirable for the pipe which goes through the bore in the shielding insertion for easy assembly and dis- assembly.) The geometry for the downstream side of the arget is shown in Figure 12 below.
Figure 12.
In words, a stainless pipe protrudes downstream from the Helm-
holtz coils. A short piece of pipe is welded to it. This piece is
threaded (rezba) on the outside. The most downstream edge of this piece
has a knife edge which makes contact with a gasket to make the vacuum
seal. A flange with corresponding threads on the inner edge screws onto
this and provides compression with a flange on the beam pipe. The beam
pipe flange has the groove for the gasket. The holes in the flange on
the downstream side are threaded, as there is no room for a nut. This
is all screwed together with "cnulbka" which I think is a threaded rod.
*flange-toflange distance of target: 250 mm.
*downstream flange 2 3/4" O.D., 1.375 " I.D. MDC part number
110014
*All their chamber and pipes are stainless.
*They don't care if we use Al or stainless
*We should use stainless flanges and stainless coated Al pipe.
The followin geometry for the upstream connection of the pipe
which goes through Q1 was proposed by the Kentucky group and o.k.ed by
the Kharkov group:
Figure 13.
This allows for removal of the pipe without torching anything. The Kharkov group requested that the following information be obtained for them from MDC: For the 6" O.D. flanges, part number 110025, what are the dimensions of the Cu gasket, inner, outer, and thickness. The target ladder has 5 positions-two Fe foils, a blank, a screen for viewing (BeO or ZnS) (there was some discussion as to the need/usefulness of this), nonpolarized target (provided by the sup- port fram for a Fe target). The Kharkov group needs a 30 V power supply, 5 Amps for their ladder stepper motor. They have one, but it is too heavy. Need to check into this before their visit in May. The cooling of the foil is provided via copper wires attached to a heat exchanger coolded by He gas and to the target ladder. This allows for a flexible connection enabling continuous rotation of the target. The issue came up that some idiot might accidentally rotate the target several revolutions, breaking something. A mechanical stop or a limit switch is desirable.
They have a draft desing of electronics, but want to check some things when they go to CEBAF. Will ship polarized target chamber and stepper motor controllers after May 10.
They have only 8 or so FEU 110's, but many FEU 30's. They showed me a mounting system which can accommodate either tube. Looked at FEU 30 with 5 nsec LED pulse. With Sr-90 source had 1.5 Volt signals with about 8nsec rise time. To vary the rate, the tube voltage was changed with a Sr-90 source. Further tests are needed. A pulser mimicking a lead-glass pulse is needed. (I believe we have one at Kentucky. Need to check.) Need to look at gain shifts versus rate, and tube recovery parameters. I passed on the results from Hall B. They will look into this in the next week. Each PMT has different custom made base. I brought back to Kentucky one lead-glass block, a FEU-30 PMT, and the support structure for the base. They take -1.8kV, -2.5kV max.
They want to build scintillators. They want a design in which the scintillators are staggered to enable attaching PMTs directly to the scintillators (figure 14):
Figure 14.
We had envisioned one on top of the other as shown (figure 15).
Figure 15.
The Kharkov design requires tubes of diameter of about the width of scintillators (currently 2 cm). One possibility under consideration for 12.5 mm Hammamatsu tube 4124 coupled directly to the scintillators. The tubes are small enough that scintillators may not need to be staggered.
Agreed that the detector shielding must be easily movable. Access from the top is probably best, with roof sliding off to beam left, but need to check interference with beam pipe. They have draf to design for box that contains detector. Final design depends on hodoscope design (see above.)
We need a collimator to reduce the rates on the detector. Since the theta acceptance should be large to decrease errors du to intra- atomic motion, we wish to collimate in phi. One possibility is col- limating up and down. We have easy access to motion along the beam- line, and Victor suggests this geometry for converting lateral to up- down motion (figure 16).
Figure 16.
Another option is to collimate just upstream of the dipole (or possibly just upstream of the last quad). This collimator could consist of two rotatable disks, one behind the other as shown in the sketch (Figure 17). This should be investigated in the simulations.
Figure 17.
This would be best placed after the last quad if there is room, as the trajectories should be pretty much the same there ind- ependent of energy.
Kharkov group says they need one month for visa preparation. They want to come May 10.
Sergei (forgot last name) believes he has learned of some limitations to Raytrace which have important implications in HRS quadrupole issue. He would like this to be brought to the attention of people at CEBAF. Fax him Offerman's latest report on the issue. Respectfully submitted, 3/21/96, DSD.Back to the Moller Homepage.
This page maintained by Bruce Deboard and Kirsten DeNoyelles.