Chamber leak and new turbo

[David Kunkle]

The now fired weld shop attempted to fix the original shoddy weld around a nipple and chamber seam. Thought they would grind it out and redo it. No. They put a lot of big ugly welds over the old and on the outside. I pressurized it, and the leak at the nipple is gone but not at the seam. I can tell it is simply pulling air in at the edge of the outside weld, travelling thru the inner and outer weld and into the interior. Lot of grinding and filling with JB Weld, and the leak is gone. Not something I want to leave permanently. My good weld shop is going to cut out the crapped-up section of chamber wall and replace it along with a new nipple I'm waiting on from LDS. Hate to tear it down AGAIN, but no doubt there are trapped air pockets in there- not to mention the JB smeared on the inside. That should at least put this small nightmare to rest, and I can move on. The good shop also welded on the new ISO100 flange for the new turbo. Their welds look like what I'd expect from someone that knows what they're doing.

Since the leak is fixed temporarily, I fired up the new to me Edwards turbo. Runs great and a lot quieter. After an hour, it bottomed out at 6.4 E-8 torr. Seems awfully low for not having done even a plasma cleaning, but my gauges have always seemed to be in good agreement. Wondering if that 220 l/s turbo at full throttle is just sucking everything out of there as fast as it can outgas?

[Jerry Biehler]

That is kind of low for that short of a period even with that turbo. What are you using for a gauge?

[David Kunkle]

I have a new Duniway Iridium broad range ion gauge- Operating Pressure: 4 x 10-10 Torr to 1 x 10-1 Torr. And a used Varian 845 from ebay. Calibration is unknown.

[Jerry Biehler]

Your emission settings may be wrong for the tube.

[David Kunkle]

Not sure what you mean by emission settings. The 6 pin jumper to connect either 564, 563 or UHV setting is on the 564 setting for a wide range gauge- confirmed twice by Duniway.

[Bruce Meagher]

Here are a couple things you might consider when investigating the accuracy of your pressure readings.

1) Create a pump down curve. Log the pressure vs time and plot it on a log-log scale. Are the results you get physically possible?
2) If you have a gate valve on your system close it off and log the pressure rise vs time. This will measure the leak rate (virtual + real). Do you have real leaks or are they all virtual (outgassing + diffusion)? How big are they?
3) Measure the ion gauge's filament bias voltage, filament voltage, filament current, and the grid voltage. What do you get?
4) On the back of my varian 845 there is a filament 1 2 select. What is yours set at?
5) I don't have a manual for the 845, but have you verified the correct pinout / wiring from the octal connector to your ion gauge?
6) Are you baking your system? If so, how are you doing that?

A picture of your current setup would be helpful.

[David Kunkle]

Chamber is still at the weld shop, but I pick it up tomorrow at least.

2) No gate valve currently on the system, but the backing pump does have an anti-suck back mechanism that seals the intake tube when powered off. When the leak was sealed with epoxy, I measured a leak down rate of about 2.5 mtorr/min. on this 3.6 cubic ft. chamber.
3)Don't know how to measure those currents and voltages on the gauge.
4)Mine has always been set on 1. Since I have the cover off, there appears to be no connections to this switch??? It has 3 pairs of connectors with a bridge between each pair, but no wires and I cannot find any floating/extra wires that may have gone on the switch.
5)Also bought the cable new from Duniway specifically for my gauge and controller. The cable connection to the 845 has always been pretty "loose". Tonight I crimped down on all the female ends on the 845 with a finishing nail. The connection is quite tight now on the pins. I'll check first if that alone makes any change to the vacuum readings when I get everything hooked up again.
6)No baking planned. Intend to do plasma cleaning.

Thanks.

[Bruce Meagher]

I personally would vent the foreline and not leave it under vacuum (or at least valve it off from the turbo).

If your leak rate was really 2.5 mTorr / min with a 3.6 ft^3 chamber (a ~4.25x10^-3 Torr l s^-1 leak) you couldn’t get below 1.9x10^-5 Torr with a perfect 220 l/s pump (assuming I did the math correctly). I suspect that since you don’t have a valve on the high vac side the leak rate you measured is really coming from the foreline/mechanical pump side. It’s my understanding most mechanical pumps are not meant to hold pressure when powered off. The leak you measure probably has no real value in diagnosing the accuracy of your ion gauge.

The voltage measurements to the ion gauge tube can be made with a simple DMM (the highest voltage shouldn't be over 200V). I believe the filament current can be a little high 3A-10A so make sure you investigate the values for your tube and your DMM. On most of the ion gauge tubes I own there are 4 pins on the bottom and one on top. If you look up the pinout for your gauge the measurements should become clear. Since your controller is second hand, and your pressure measurements are suspect, measuring these values might help pinpoint a problem.

[Jerry Biehler]

Not sure what you mean by emission settings. The 6 pin jumper to connect either 564, 563 or UHV setting is on the 564 setting for a wide range gauge- confirmed twice by Duniway.

Somewhere on the unit is a way to adjust the emission current for the tube, not all tubes are the same, especially if you went to a different kind of filament. I cant find a manual for the 845 so I am not sure. Usually the gauge controller is adjusted using a sealed off ion gauge tube like this: http://vacuumshopper.stores.yahoo.net/seirunbgatu.html

It is at a known vacuum and you can adjust the controller to read correctly.

[David Kunkle]

Not sure what you mean by emission settings. The 6 pin jumper to connect either 564, 563 or UHV setting is on the 564 setting for a wide range gauge- confirmed twice by Duniway.

Somewhere on the unit is a way to adjust the emission current for the tube, not all tubes are the same, especially if you went to a different kind of filament. I cant find a manual for the 845 so I am not sure. Usually the gauge controller is adjusted using a sealed off ion gauge tube like this: http://vacuumshopper.stores.yahoo.net/seirunbgatu.html

It is at a known vacuum and you can adjust the controller to read correctly.

Called Duniway today about emissions settings and the sealed gauge, and they were kind enough to send me a complete manual for the Varian 844/845 Ion Gauge Controller in PDF form. I haven't been able to find this anywhere on the net. So, if anyone is interested, I should be able to email a copy to you. They said to call back if I have any questions after going thru the manual.

The manual does give 3 methods of calibrating the ion gauge. One requires a second gauge tube mounted to the chamber which I will not be doing. The other methods involve a voltmeter and a resistor which I will probably try first. It makes no mention, but it does sound easy to use a sealed tube (as long as the tube's pressure has been recorded- Duniway's site only states the tube was sealed below 5E-6 Torr), then just adjust the gauge's pot so the controller's digital readout is the same as the sealed tube. Only downside is the $200 for the tube.

[Jerry Biehler]

I have seen sealed ion gauges on ebay now and then. Keep an eye out, they are bound to show up.