5600 XPS system MCD channel plate replacement procedure.

Posted on April 28, 2021 by Randy

This procedure will explain how to replace the channel plates (technically, they are Chevron plates) in the MCD detector used on Physical Electronics (PHI) 5600, 5700 and 5800 X-ray Photoelectron spectrometers. You may also want to check out our blog post on checking the MCD pins continuity.

Set up a work area with some aluminum foil and lint free cloths. Vent the system.

Mark the MCD feedthrough flange to confirm the orientation of the flange when you replace it later.

Remove most of the bolts and nuts that secure the MCD feedthrough flange to the analyzer. The MCD feedthrough flange has some springs on the inside so there will be some tension when you remove the flange. There are 3 studs that are used to guide the MCD feedthrough flange when it is removed or installed.

Remove the last few nuts or bolts and carefully lower the MCD feedthrough flange and place it on the aluminum foil. Incidentally, most food grade aluminum foil has oil on it. If possible, use UHV compatible foil when working with electron optics.

The MCD assembly is located inside the 6″ CF flange where you just removed the MCD feedthrough flange from. You can remove one of the analyzer braces in order to gain better access to the MCD assembly.

At this point put on a pair of gloves as you will be working inside the analyzer and want to prevent oil contaminator from your hands. Use tools that have been cleaned with Isopropanol or Methanol.

Remove the screw that holds the capacitor in place.

Remove the capacitor by pulling on the pin that is connected to the MCD assembly. Do not pull on the capacitor. The pin wire is silver soldered to the capacitor and it can come off easily, so be very careful with the capacitor.

Remove capacitor by pulling on the pin connector

Remove the POS and NEG pins from the MCD assembly. Just pull up on the pins, keep the wires attached to the SHV feedthrough pins.

The MCD assembly is held in place with 2 slotted screws. Remove one of the screws completely. Be careful as it is easy to drop.

Loosen the second screw and then reach up and hold the MCD assembly as you remove the screw. Ease the MCD assembly down and out of the 6″ CF flange.

Set the MCD assembly on a clean working surface.

Pull up on the pins and insert a small Allen wrench or rod to hold the pins in place. The pins are spring loaded and by inserting a small Allen wrench or rod into the holes in the pins, that will prevent the pins from contacting the plates or 16 channel ceramic when replacing the plates.

Remove the two inner slotted screws.

Hold the MCD assembly as you rotate it 180 degrees so that the grid screen is facing up.

Hold together and rotate so that grid is facing up

Remove the grid screen. Set it to the side.

Remove the ceramic. Set it next to the grid screen. As you remove parts, set them down next to each other in sequence. That will make it easier to reassembly the parts in the correct order.

Remove the gold spacer. Sometimes that part will stick to the bottom of the ceramic.

Remove the Teflon tubes

Remove the top plate.

Remove the second plate. Note that the plates have a little dot on them. Those dots need to face each other. So, the bottom plate has the dot facing up, and the top plate has the dot facing down.

Use some clean air or nitrogen to blow off the 16 channel ceramic. Feel free to dust off the MCD assembly and plates frequently. Dust is to be avoided as much as possible.

Install the new bottom plate. It sits on the gold spacer which is on top of the 16-pin ceramic. The hole in the plate needs to line up with the pin. The pin is centered in the hole in the plate. The little dot on the plate needs to face up.

Install the new top plate. The dot needs to face down and the pin should be centered in the hole in the plate.

Install the gold spacer. The tab on the spacer should be above the pin.

Install the Teflon tubes.

Install the ceramic over the Teflon tubes. The top of the tubes should be flush with the ceramic.

Set the grid screen on top of the ceramic.

While holding the grid screen and ceramic in place, rotate the MCD assembly 180 degrees so that the pins face up. Since the screw are not get installed, the only thing holding the grin screen and ceramic in place is you.

Rotate MCD assembly so that pins face up

Install the two screws to secure the grid screen. You may need to rotate things just a little bit. Tighten the screws firmly but not too tightly or you may crack the ceramic.

Remove the Allen wrench or rods from the pins. Rotate the pins a little bit to make sure that they are seated.

Measure the resistance between the POS and NEG pins. You should have 15 to 24 meg ohms. From the other pin to POS or NEG should be about 1 Meg ohm.

The MCD assembly is now ready to install back into the analyzer.

Before you insert the MCD assembly, prepare one of the screws for mounting. A pronged grabber works well as does a starter screw driver. In a pinch you can use a regular screw driver with a little bit of tape to hold the screw on. The tape needs to stick to the screwdriver so that it all comes out when you remove the screwdriver.

Insert the MCD assembly (it can only go in one way or else the screw will not line up) and rotate it slightly to make sure that it is seated properly. Tighten the one screw to where it is just snug.

Install the second screw and tighten it to where it is just snug.

Slightly rotate the MCD assembly to make sure that it is seated properly and then tighten both screws firmly.

Reconnect the POS and NEG pins. Refer to the drawing to make sure that you do not insert the pins backwards. Also make sure that the POS and NEG wires are centered as the MCD feedthrough flange will need to clear the POS and NEG pins.

Insert the capacitor pin. Be careful not to stress the capacitor.

Reattach the capacitor wire screw and tighten the screw. The capacitor should be perpendicular (not tilted) as the MCD feedthrough flange will need to clear it.

Remove the used 6″ copper gasket.

Install a new 6″ copper gasket on the MCD feedthrough flange.

Carefully line up the MCD feedthrough flange and slide it onto the 3 guide studs. Install the 3 nuts to hold the MCD feedthrough flange in place and then slightly rotate the flange to make sure it is seated. The springs on the MCD feedthrough flange fit into the 16 holes in the MCD ceramic and press against the metal discs which in turn make contact with small capacitors inside the MCD assembly.

Install the bolts and lightly tighten the MCD feedthrough flange.

Measure the capacitance of all of the pins on the MCD feedthrough with respect to the vacuum chamber. The four center pins should read approximately 100 pf (open). All of the other pins should read 250 to 300 pf. If not, you need to drop the MCD feedthrough flange down and inspect to see where it is hanging up. Sometimes the spring pins do make contact with the MCD assembly and need to be adjusted slightly.

Once you have the correct capacitance on all of the pins, tighten all of the nuts and bolts.

The vacuum chamber is ready to pump down and baked.

Installation complete.

Note: Set the MCD multiplier voltage to 1800V in the software so that after the bake out you start outgassing the new plates at a lower operating voltage.

Here is a side view of the MCD assembly.

72-100 Electron Multiplier Supply Notes

Posted on April 2, 2021 by Randy

The 72-100 electron multiplier supply is used on older Physical Electronics (PHI) Auger electron spectroscopy systems to provide voltage to the cylindrical mirror analyzer (CMA) and secondary electron detector (SED) electron multipliers.

The 72-100 comprises a digitally controlled 0 to +4000V variable switching supply, a 0 to 300V fixed supply, and a voltage to frequency monitoring circuit. The digital control from the PC software is converted to an analog output which drives a switching supply.

The AES and SED 72-100 board addresses are set by changing the segments on SW2 as follows;

Switches set up for AES SW2 segment 3 closed

Changing the switch settings allows you to swap the 72-100 boards to confirm whether or not there is a problem. For example, if you think that your SED 72-100 board has a problem, you can swap it out with the CMA 72-100 after first changing the CMA 72-100 address to SED (SW2 segment 4 closed). If the SED image works after swapping the boards, then you know that your SED 72-100 board does have a problem.

Once you confirm that your 72-100 is not functioning properly there are a some simple things that you can try. ***CAUTION – high voltages (up to 6kV) are present on this board, refer servicing to qualified personnel who have been trained to work safely with high voltage.***

Note that the CMA 72-100 is controlled by AugerScan software and the SED 72-100 is controlled by AugerMap software. You can also use the AugerScan diagnostic menu to send commands to the board per the 72-100 manual calibration procedure.

Make sure that the card rack power is OFF and then extend the 72-100 board. Use the high voltage extender (the one with red wires on the bottom section).

For testing the AES 72-100 board, you can first check the analog output from the DAC to see if you are getting the proper control voltage. Note that the card rack power needs to be ON when measuring voltages. For the high voltage supply, measure from the left side of R 32 with respect to the left side of CR 3.

With the electron gun supply off, set up and acquire a survey in AugerScan. The DAC output voltage should ramp up from 0 to about 5 volts as the multiplier voltage is being set by AugerScan. Since there is no electron beam current, the AES electron multiplier voltage will default to the PC (pulse count) voltage which is typically 1800 to 2200V. 5 volts on the DAC output corresponds to 2000V on the multiplier supply POS output.

If the DAC voltage output is correct, then the problem is probably related to the switching supply transistors. With the 72-100 board on the bench (card rack power OFF when removing or inserting the 72-100), test the diodes and transistors for shorts. Note that two legs on all of the power transistors will show as shorted because those points are connected to the switching transformers.

One common problem that causes the HV output to be zero is that Q7 and Q8 are defective, even though they may check out OK with the diode test. If the DAC voltage is correct but you do not have any high voltage output, then try replacing Q7 and Q8 with the correct 2N5337 transistor. Some 72-100 boards have 2N3725 transistors instead of 2N5337 transistors installed. The 2N5337 can dissipate 6 watts of power but the 2N3725 can only dissipate 3 watts. The lower power rating of the 2N3725 can result in poor switching characteristics.

Q7 and Q8 are located under the black high voltage cover. It is held in place with 3 screws. Remove the cover and you will see Q7 and Q8.

Assuming that replacing the transistors solved the issue, you can adjust the high voltage by adjusting R7 to match the software. For example, if the AES PC voltage is 2000 V then adjust R7 for 2000V between the POS and NEG cables. Note that you need to use a high voltage probe when measuring the POS output voltage as most DVMs are limited to 1000 V or less.

R2 is the 300V adjustment and you measure that voltage from the NEG cable center pin to the outer cable or vacuum chamber.

The easiest place to connect your high voltage probe to measure the AES high voltage is between the center pin of the J6 HV POS IN cable to the 96A/B / VF4 preamp and the center pin of the NEG cable on the CMA electron multiplier.

When testing the SED 72-100 then you need to control the SED voltage in the Auger Map TV dialog box. As you change the SED voltage in AugerMap you should see the DAC output voltage change. 2000V would correspond to about 5V on the DAC output. For the SED high voltage, measure between the center pins on the 97 SED preamp POS and NEG cables. Set the SED voltage in Auger Map to 2000V and then adjust R7 on the SED 72-100 board for 2000V. Remember that you need to use a high voltage probe when measuring the SED and CMA POS output voltages.

Regarding the V/F circuitry on the 72-100 board, those components are not used and so can be removed if your board has them installed in sockets. Some 72-100 boards do not have those ICs installed at all, but others do and they may be soldered in. If they are soldered in just leave them as is; it is not worth the effort to pull them off the board if they are soldered in place. The ICs which can be removed are:

U1, U2, U3, U4, U5, U6 and U9.

72-100 with V/F ICs removed from sockets

Finally, the 1K ohm filter resistors on the 72-100 mother board can fail (burn and become open). In that case the 72-100 board may be working properly, but the voltage will not get out to the cables. You can ohm out the resistors on the motherboard to see if that is an issue.

72-100 R1 R2

And here is the schematic that shows both resistors:

If you can’t repair the 72-100 board yourself, RBD Instruments provides a 72-100 repair service and loaner 72-100 boards.

Cylindrical mirror analyzer fringe field termination ceramics

Posted on March 14, 2021 by Randy

Physical electronics (PHI) cylindrical mirror analyzers ( CMA) use fringe field termination ceramics to reduce the fringe fields from the end of the cylinders.

Abstract an early patent:

Field termination plates for cylindrical electron analyzers are provided wherein the plates are constructed of an insulative material coated on the interior surface with a high resistance, electrically conducting coating. Spaced concentric rings of relatively high conductivity material in electrical contact with said coating are provided; the rings providing equi-potential regions on the plates, thereby minimizing field fringing near the ends of the cylindrical tube electron analyzer.

The PHI CMA (cylindrical mirror analyzer) utilizes conical and flat termination ceramics to eliminate electrostatic edge effects between the inner and outer cylinders. These ceramics are essentially gold rings with resistors in between that divide the outer cylinder sweep voltage down in even steps. The result is a very high throughput and even energy distribution of the Auger electrons. If a CMA has a poor contact on a termination ceramic, the results are noisy data and poor energy linearity.

Single pass AES CMAs have just 2 terminating ceramics, a conical at the front of the CMA and a flat at the base. Double Pass AES/XPS CMAs have 3 terminating ceramics, a conical at the front, a center and a base.

The conical and flat ceramics are essentially resistors and so the total resistance between the inner and outer cylinders add up like this:

The table below lists the values on the combined resistances of the older PHI CMAs.

If the VM or IC to OC (inner to outer cylinder) resistances are off significantly in your CMA then you probably have a contact issue between a conical or flat (base) ceramic between the outer cylinder or the inner cylinder. Sometimes the resistances of the conical ceramic can be off due coating caused by years of sputter depth profiling.

A poor electrical contact can result in high background counts or extremely high noise levels in the data due to arcing. If you suspect that your CMA has a contact issue with a terminating ceramic then you will need to tear down your CMA to where you can add improve the electrical contact by adding some thin copper or gold shim foil between the suspect ceramic and cylinder. If you need some guidance on how to do that, please contact RBD Instruments.

The pictures below show the conical and flat ceramics from a 25-120A CMA. You can see the gold rings and also the thin film resistors.

RBD TechSpot

A blog on the repair, operation and calibration of surface analysis systems and components including electron spectrometers, sputter ion guns and vacuum related hardware. Click on the Index tab below to see a list of all posts. Visit our website at http://www.rbdinstruments.com

Category Archives: General Optics and Vacuum

5600 XPS system MCD channel plate replacement procedure.

72-100 Electron Multiplier Supply Notes

Cylindrical mirror analyzer fringe field termination ceramics