Vacuum pump rebuild

Every fusor and fusion system seems to need a vacuum. This area is for detailed discussion of vacuum systems, materials, gauging, etc. related to fusor or fusion research.
Tyler Meagher
Posts: 30
Joined: Thu Mar 08, 2018 10:29 am
Real name: Tyler Meagher

Re: Vacuum pump rebuild

Post by Tyler Meagher »

I finished putting everything back together. I attached three vacuum gauges to the intake port and then started the pump. After a few seconds the top gauge was at -29.5 Hg. The thermocouple gauge was at 65 millitorr. The pirani gauge was 49.5 millitorr. The first picture shows the pump running with the pressure gauges. After a few minutes something weird started happening. The pressure started going up and bouncing around. It went as high as 150 millitorr, but it was changing really fast. I started the pump a few times and it did the same thing.

Then something weirder happened. I put my hand over the exhaust port for a second and all of a sudden the pressure really dropped. When I removed my hand the pressure went up and started bouncing around again. I placed my hand on the exhaust port longer and the pirani gauge went as low at 11 millitorr. My dad and I are trying to figure out what might be causing this.

Here is a link to video. https://youtu.be/o6wUOpPbQJs

Tyler
Attachments
IMG_0777-1.JPG
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Vacuum pump rebuild

Post by Richard Hull »

If you can hit and maintain 11 millitorr, that would be a win and a most acceptable pump. Good luck in figuring all of this out.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
richnormand
Posts: 307
Joined: Sat Aug 27, 2005 8:30 am
Real name: rich normand

Re: Vacuum pump rebuild

Post by richnormand »

Just a guess or two:

1)
Leak between the exhaust and the intake?
Can you devise a way to see if atm pressure overtakes an oil seal or a gasket seal or a bearing seal?
The fact that it is intermittent with plugging the exhaust and (I assume) once the pump is warmed up could be due to metal expansion or lower oil viscosity creating that leak. If there is a "pumping" action on the exhaust volume putting your hand on the port gradually lower the pressure (not atm anymore) and if done long enough it will equalise and go the other way to an equilibrium (assuming a better seal than your palm).

2)
again with oil viscosity lower at higher temperature and the pitted condition of the walls and vanes and side walls could it be a loss of pumping action with vane/wall leaking or vane bouncing? Increased pressure building up in the exhaust volume could minimise that behaviour due to the smaller pressure difference? A bit like a Roots pump that although leaky by design will be able to pump large volume of air but will struggle to reach a low vacuum by itself due to all the leaks in the lobes.

Can you easily try a thicker oil to test?

3)
electrical issue affecting the gauge. Pump not grounded, stray current, EMI etc... Try to just touch the pump near the thremocouple and see if it changes. Are these difference of pressure noticeable on the other gauge, if readable?

I am not an authority on vac pumps by any stretch. Perhaps someone on this forum can comment or perhaps you could contact the folks where you got the rebuild kit and they might have experience with that brand of pumps. I was impressed by Duniway support when I had questions doing mine with their kit.

Finally, looking at your setup. The thermocouple gauge is quite close to the intake and can be coated by oil mist that influences its accuracy. This can be cleaned with acetone in mild cases.
Tyler Meagher
Posts: 30
Joined: Thu Mar 08, 2018 10:29 am
Real name: Tyler Meagher

Re: Vacuum pump rebuild

Post by Tyler Meagher »

I fixed my pump problem. I installed the wrong valve on the pump exhaust port. I used the metal valve. My dad emailed the company and the metal valve is for a different version. I needed to use the rubber piece with the spring. The pump also had a slow oil leak from the lip seal because I didn’t put it in far enough. I fixed these two things and then ran the pump for an hour. The pressure was 8 millitorr on the pirani gauge. This time the pressure did not change when I put my hand over the exhaust.

The next test is to run the pump for 24 hours and record the pressure. My dad is helping me write my first python program to record the pressure on my PC.

Tyler
Attachments
Exhaust valve
Exhaust valve
Lip seal not in far enough
Lip seal not in far enough
richnormand
Posts: 307
Joined: Sat Aug 27, 2005 8:30 am
Real name: rich normand

Re: Vacuum pump rebuild

Post by richnormand »

Congrats!
If you click with Python there is a very good online course from MIT via the Edx program (assuming you have time).
I would not leave the thermocouple sensor as shown in your previous photo for a 24 h run. Pirani probably OK. I am sure Richard would know.

From your valve geometry was it because the metal valve was leaking too much atmospheric pressure in the internal exhaust port at the exit of the pump?
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Vacuum pump rebuild

Post by Richard Hull »

Placing a TC gauge tube right at the inlet to a mechanical pump and letting it run for protracted periods is a bad idea. The only time I recommend this is to determine what you have when you first test your pump. Do not let it run for more than it takes to bottom out the gauge in a couple of minutes. From that point, if the pump needs work or is a piece of junk, never run with the TC gauge tube at the head until some corrective action is taken and you need to see if it has improved.

The TC gauge in my system is about 12" from the pump just before the first conflat valve to the rest of the system. I let the pump run for 2-3 minutes with the valve closed. By this time it is down to 12 microns. I then open the valve to the diff pump and the TC usually rises only slightly but quickly drops back to 12 microns.

Due to my leaky fusor chamber when I open the diff pump to fusor valve, the TC gauge immediately goes to near atmosphere and the pump down continues for about 3 minutes when the entire system, by the TC gauge is below 20 microns.

I only now turn on the power to the fusor chamber's Baratron and the diff pump boiler heater. It can take 5-10 minutes before the diff pump drags the system down into the 10e-4 range. Here is where I can begin fusison.

I have, on two occasions, removed the TC gauge tube and filled it with MEK and shaken it for a minute or two to remove any oil vapor deposits. The fluid comes out clean and clear at 12" from the pump that runs for hours at a time. Once re-installed there is no improvement or decrement noticed in its function from before the cleaning.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Rex Allers
Posts: 570
Joined: Sun Dec 30, 2012 3:39 am
Real name:
Location: San Jose CA

Re: Vacuum pump rebuild

Post by Rex Allers »

Tyler,

Fantastic work in bringing a very cruddy looking pump back from the dead. Looks great now and your measurements indicate it is working very, very well. You can be proud.

-Rex
Rex Allers
Tyler Meagher
Posts: 30
Joined: Thu Mar 08, 2018 10:29 am
Real name: Tyler Meagher

Re: Vacuum pump rebuild

Post by Tyler Meagher »

Thanks everyone for all your comments. I ran the pump for 24 hours. I was happy to find the pressure at 9.5E-4 torr on the pirani gauge. The thermocouple gauge was at 0. The pump was very hot when I touched it. I tried a second MKS 901p pirari gauge. It was 2 millitorr after about an hour. I understand the high quality oil in my pump is helping with the low pressure.

The picture below is my finished pump with a few parts left over. The broken plastic parts and magnets in the picture are suppose to be on the motor coupling. The safety valve needs this to work. I removed the safety valve and the anti-suck back valve. These are the other left over parts in the picture. I epoxied over the safety valve hole to the intake port. You can see this in the previous posts. Now I understand how this feature works. I know I need to vent the intake when I turn off the pump. I have a vent valve I will connect outside to replace the parts inside.

Rich, I am not sure about the metal valve issue. I also took the pump body off to fix the lip seal. I tightened the bolts when I put it back together. My dad thinks I should take the pump apart and put the metal valve back in for a test. Then take it apart again and put the spring and rubber valve back. I’m not so sure.

I’m working on a python based vacuum controller so I can log the pressure on my laptop.

Tyler
Attachments
Finished pump
Finished pump
Tyler Meagher
Posts: 30
Joined: Thu Mar 08, 2018 10:29 am
Real name: Tyler Meagher

Re: Vacuum pump rebuild

Post by Tyler Meagher »

I tried the metal exhaust valve again. It worked fine. No weird behavior. I then put the spring and rubber exhaust valve back in. It worked the same. Draining the oil and removing the case two times didn’t take too long. I was wrong when I said the metal exhaust valve was causing the weird behavior. Maybe tightening the pump bolts fixed the weirdness I showed in the video.

The first picture is before I removed the exhaust valve. The second picture is with the exhaust valve removed. The third picture is the oil input port. One important thing about how my pump works is how the oil runs through it. The vacuum oil comes in from the outside at the middle of the pump. It exits at the exhaust port. The exhaust port is where both the oil and air leave the second stage of the pump.

Tyler
Attachments
Pump with exhaust valve
Pump with exhaust valve
Exhaust valve removed
Exhaust valve removed
Oil input port
Oil input port
Tyler Meagher
Posts: 30
Joined: Thu Mar 08, 2018 10:29 am
Real name: Tyler Meagher

Re: Vacuum pump rebuild

Post by Tyler Meagher »

The first picture is a pump down curve of my vacuum pump. It was taken at the inlet of the pump. There was just the tee and the MKS 901p gauge. It does not take long to get below 10 millitorr. Then my dad had me change the oil with some cheap Amazon oil as an experiment. This is the same oil Samuel used in his thread viewtopic.php?f=10&t=12278. When I started the pump it reached 35 millitorr. After a couple minutes the pressure started climbing. The pressure was over 300 millitorr after 5 minutes. I took off the gauge and put a blank off at the inlet. I am running the pump over night with the ballast open. I will measure the pressure tomorrow. Next I will replace the oil with the good oil to see what happens.

Tyler
Attachments
Pump down curve
Pump down curve
cheap oil run
cheap oil run
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Vacuum pump rebuild

Post by Dennis P Brown »

Well, 35 microns isn't very good for a two stage pump but that would be usable if the pressure didn't then go up. Are you certain there are no leaks? Vacuum gauges can and do fail (not that that is the issue but don't rule it out.) If you did not have new oil, than the ballast air run may fix the worst of the issues. The inner workings of that pump have experienced serious corrosion. From previous pics, the inner vane surfaces for the pump case does look rather poor. That could be an issue for holding vacuum.
Jerry Biehler
Posts: 975
Joined: Tue Nov 24, 2009 3:08 am
Real name:
Location: Beaverton, OR

Re: Vacuum pump rebuild

Post by Jerry Biehler »

Really, he is lucky he is getting what he is with the condition of the pump.
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Vacuum pump rebuild

Post by Richard Hull »

Yeah, A bold and brave effort at raising the dead and a very teachable venture. Such resurrections are typically doomed. I imagine a key vacuum forming fraction of that pump's mass went out over the years in the oil and in the removal of rust from the pitted metal. However, they are still at it.

I find it tough to imagine an oil change made the pump fail to do what it did in the graph earlier!!

Looking closely at the photo of the pump at 350+ millitorr...........check out the non-right angle tilt and the cockeyed KF clamp. Is this simply a bad connection?? All the KF clamps I have and have used are nice and parallel when clamped properly. I assume the KF on the pump is in the same parallel plane as the top of the pump body? The elbow to gauge clamp looks fine!

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Tyler Meagher
Posts: 30
Joined: Thu Mar 08, 2018 10:29 am
Real name: Tyler Meagher

Re: Vacuum pump rebuild

Post by Tyler Meagher »

My pump worked great! The pressure was below 1 millitorr on my MKS gauge after an overnight run. The experiment I just finished running was to learn about something called vapor pressure.

My experiment was to replace the VacOil 20 with cheap oil from Amazon. The vacuum oil you use for air conditioner repairs. The same as Samuel. I ran the pump overnight with a blank off. I then recorded a pump down curve with the oil and pump still hot. I measured the oil temperature at 66°C inside the case. The temperature of the pump body was 77°C. The inside of the pump by the vanes must be much hotter. The pressure could not go below 500 millitorr.

Next I changed the oil with VacOil 20. I ran the pump blanked off overnight. With the oil and pump still hot I ran another pump down curve. This time the pressure could not go below 50 millitorr.

I changed the oil again with new VacOil 20. I ran the pump blanked off overnight again. This time the pump could get down to just below 5 millitorr.

What I have learned is the vapor pressure of the oil can determine the lowest pressure the pump can go. The vapor pressure depends on temperature. When the oil is hot the vapor pressure goes up. Different vacuum oils have different vapor pressures. Vapor pressure is what causes the oil to leave the pump inlet. The ultimate pressure I reached in each run was where the vapor pressure matched what the pump could remove.

The bad news is I have to completely disassemble and clean everything if I want to get the same performance as before.

Tyler
Attachments
pumpdown all.jpg
Post Reply

Return to “Vacuum Technology (& FAQs)”