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.
Posts: 282
Joined: Sat Aug 27, 2005 12:30 pm
Real name: rich normand

Re: Vacuum pump rebuild

Post by richnormand » Wed Apr 18, 2018 9:26 pm

I do agree with Richard's and Jerry's views here. But, if you are in for a hero experiment, in particular after rebuilding the motor, here is what I ended up doing to mine.

When I got my Trivac D16AC pump it was jammed. Not quite as bad as yours in appearance inside but it had been used to pump silicone for molds...
The roll-pin on the center shaft driving the vanes holder had broken and the pin remnants had scored the inside walls and also chipped the vanes. I was able to restore the outer surface for the vanes using a flex car cylinder hone (smallest size I found). Started with a coarse one and went directly to the finish grade. That left a nice crosshatch finish that would bed in the new vanes.

While in there, I used a Duniway major rebuild kit. Fill with oil, dump, second fill and 5 x 10^-3 t and steady for the past 10 years or so.
Its lots of work but doable depending on how stubborn you are.

Worth it for kicks on your own time but not otherwise. I can post photos of the rebuild should you like.
Good luck with it.
PC272478 (Medium).JPG

Bruce Meagher
Posts: 100
Joined: Fri May 13, 2011 3:25 am
Real name: Bruce Meagher
Location: San Diego

Re: Vacuum pump rebuild

Post by Bruce Meagher » Thu Apr 19, 2018 3:27 am

I agree that normally you wouldn’t give this pump a second look. There are tons of vacuum pumps available on the second hand market in great shape that require little to no work. Add oil and go. Is this pump destined for the scrap yard? Potentially, but honestly it does not matter.

I fully expect a pitted mess when we crack the pump apart (hopefully soon)!

What level of vacuum could one achieve with “ANY” pitting? What about a lot of pitting? I don’t know the answers but this sounds like an interesting experiment. The vanes have significant travel. Can one remove 0.030” by boring and/or honing the cylinder to revive this sucker with little performance losses? The sides walls will probably have issues as well. What are the options to fix this… Boring, honing, facing, surface grinding, lapping, and the measurements of these processes are whole new areas to potentially explore. My hope is we gain a little practical knowledge about how two stage mechanical pumps work and maybe a few other things while we’re at it.

Rich, I’d love to see the pictures from your rebuild.


Rex Allers
Posts: 221
Joined: Sun Dec 30, 2012 8:39 am
Real name:
Location: San Jose CA

Re: Vacuum pump rebuild

Post by Rex Allers » Thu Apr 19, 2018 10:38 am

Quick short test after three long reply attempts failed.

Please ignore.
Rex Allers

Rex Allers
Posts: 221
Joined: Sun Dec 30, 2012 8:39 am
Real name:
Location: San Jose CA

Re: Vacuum pump rebuild

Post by Rex Allers » Thu Apr 19, 2018 11:03 am

Fifth try failed.

7 pictures ~100K each uploaded and inserted into fairly long post. Previewed the post and all looked ok. Click the submit button and after a few seconds connection is reset. No post. I open browser window again to forum and I'm still logged in but my post is vaporized.

Must be something in my post that the forum system doesn't like. Is there a file upload number limit per post?

I give up. Thought it might be a little helpful.

These short ones seem ok.
Rex Allers

Rex Allers
Posts: 221
Joined: Sun Dec 30, 2012 8:39 am
Real name:
Location: San Jose CA

Re: Vacuum pump rebuild

Post by Rex Allers » Thu Apr 19, 2018 11:08 am

One last try of the message with just the text place holders I inserted for the images. No jpg images uploaded or inserted this time.

I would think if you open the actual pump chambers and find nasty surfaces, that your question about honing the inside of the stator or smoothing the outside of the rotor would be limited to very small amounts.

I think all these pumps require a very close tolerance between the rotor and stator that is set in the narrow region of the stator between the input and exhaust ports. This very small gap with the pump oil forms an effective gas seal that must be good if the pump is to work well at all. If you take even a mil or two out of that gap, the pump may not work well at all.

On the other hand, I have some experience with crapish pumps that were still useful.

--- Pump 1 ---

I bought an Edwards E2M12 from eBay -- "Like new vacuum pump in good working order. Pumps down with no issues" -- partially because the seller was local. I picked it up locally and was a bit concerned when I found myself in a semi-commercial-looking garage setting. Got it home and tested. First pass: 15 Torr! Ye gads -- that sucks in general but not very much in the vacuum sense.

More trying over a day or two, eventually it seized up.

I eventually figured out that the pump was used to extract air from fiberglass molding processes. This seems to be a common cause of destruction of pumps that find themselves to be sold on eBay. They suck abrasive glass fibers and plastic resins.

Talked with seller and eventually he agreed to refund most of what I paid and I used that to buy a major rebuild kit with vanes. Here are a few pictures from that process.

[em extern yuck.jpg]

Some schmutz I found early but this is actually external to the pump unit so not too worrying.

Next a pic of the inside.

[em inside.jpg]

Not too bad at first look but that vertical rod is a fine mesh filter. It is totally encased in the crap that was running through the pump.

One of the rotor shafts shows a bit of scoring at its bearing surface.

[em shaft abuse.jpg]

Not terribly bad. Nothing sensible to be done about this so I just ignored it.

Now the worst of pump internals, probably from whatever got loose and jammed to seize the pump while I was testing.

The stator.

[em stator scoring.jpg]

There's that one nasty small gouge from something that got into the pump. You can also see finer scoring lines around most of the surface. I also marked the region between the input and output ports where the tolerance between the stator and rotor surfaces must be very close to seal gas with the pump oil.

Here's the rotor.

[em rotor gouge.jpg]

A nasty mark corresponding to to the one on the stator. No idea what happened to get it to cause a dent down into the vane groove but there it is. You can also see minor scoring across the whole surface.

So I did the whole rebuild, being sure to clean gunk out of all the passages. I did a little bit of fine sanding on the rotor and stator, mostly just to be sure there were no high spots. All the valve parts, gaskets, and o-rings where replaced.

At the end of the rebuild this pump seems to run well and I am getting about 14 mTorr from this pump. Quite usable.

--- Pump 2 ---

I bought an old Welch Duoseal 1402 belt drive pump from Greg Courville on this forum. Checked it out and pump was working well. I measured about 5 mTorr vacuum. (I used a less trustworthy gauge then than my recent measurements so could be off my a couple.)

This is a 1402 with the top ballast valve and keyhole sight windows so must be very old. Maybe 70's? Maybe earlier?

I decide to do some cleanup. First I made new hard rubber feet for the base. Then I cleaned all the externals. No signs of oil leaks. The sight windows for oil level were dirty enough on inside that it was hard to easily see the oil level. I drained a little oil and got some crud coming out. I decided to open the case and do some cleaning.

Here's what I found.

[01 Case Open Sludge 1.jpg]

Kind of yucky. I decided to clean up what I could without opening the pump unit, that seemed to be working fine.

I took off the external valve parts and cleaned them. Here's the pump module after cleaning.

[17 Clean Pump Inside.jpg]

The valve parts are removed and wooden dowels are protecting the pump ports while I was cleaning.

Here's the case inside after I did lots of cleaning.

[21 Cleaned Case 2.jpg]

Much sludge scraped out then scoured cleaner. The inside of the oil level sight windows was one of the harder parts to clean. I also lightly milled the gasket surface flat. (I'll skip the details.) Years of people jabbing stuck gasket off had left it kind of not too flat.

So put back together it's nice and clean. With my MKS 901P sensor (I have 3 and trust them) I see about
3 mTorr. My best pump.

All the clean up work didn't change the good vacuum it started with much, if at all, but it makes it look better and should help it keep working well into the future.

--- Pump 3 ---

Just a few days ago I bought Ian's Edwards RV5 pump. He measured good vacuum, about 4 mTorr, and I confirmed the same when I got it home. I let it run for a half hour or so until it got pretty warm and the vacuum came up to about 6 mTorr, but that's still great.

It is very quiet while running too.

It looked good but I spent a day cleaning it all externally and it looks almost new again. In the process, removing side covers and such, I saw no bad signs like oil leaks or anything. From parts in the electronics of the motor start, it looks to be probably from about 1995. So probably more than 20 years old.

(But the Welch 1402 might be about 40 or more years old.)

--- Conclusions ---

The pump you have could be saveable if the insides are not bad.

If they are bad, I think most rotary pumps can't tolerate more than a tiny amount of surface removal.

A pretty clean looking pump can have major problems from contamination, but I was able to bring it back to fairly good performance.

A perfectly working pump can have a good bit of crap inside.

--- My thoughts ---

I'd clean enough to open up the pump module more easily and then see how good or bad the rotors and stators look. Also any shafts and bearing surfaces.

My E2M12 shows that slightly damaged pump internals can still make a useful pump. But I wouldn't expect very much damage could be recovered from.

The Welch pump shows that some significant crap inside that can be there and the pump still works fine.

A nice clean 20 year old pump is a lot easier than fixing or refurbishing old pumps.
Rex Allers

Rex Allers
Posts: 221
Joined: Sun Dec 30, 2012 8:39 am
Real name:
Location: San Jose CA

Re: Vacuum pump rebuild

Post by Rex Allers » Thu Apr 19, 2018 11:14 am

First five images that should have been inserted.
em extern yuck.jpg
em inside.jpg
em shaft abuse.jpg
em stator scoring.jpg
em rotor gouge.jpg
Rex Allers

Rex Allers
Posts: 221
Joined: Sun Dec 30, 2012 8:39 am
Real name:
Location: San Jose CA

Re: Vacuum pump rebuild

Post by Rex Allers » Thu Apr 19, 2018 11:16 am

Next three images that should have been inserted.
01 Case Open Sludge 1.jpg
17 Clean Pump Inside.jpg
21 Cleaned Case 2.jpg
Rex Allers

Bruce Meagher
Posts: 100
Joined: Fri May 13, 2011 3:25 am
Real name: Bruce Meagher
Location: San Diego

Re: Vacuum pump rebuild

Post by Bruce Meagher » Thu Apr 19, 2018 2:51 pm

Thanks for the pictures Rex, and the data point with the small pit on your first motor.

I don't know what I was thinking last night. Clearly you can only remove a very small amount of material from the cylinder or rotor while still maintaining the oil seal at the top between the rotor and stator. We'll know more about the condition of the pump internals this weekend.


User avatar
Richard Hull
Site Admin
Posts: 10728
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Vacuum pump rebuild

Post by Richard Hull » Fri Apr 20, 2018 4:15 am

Rex did a good job on the 1402 oil reservoir. I love that he noted the vacuum didn't improve all that much, if at all.

I have always said let the sleeping dog lie. I have two 1402's here. I rate them as one of the finest pumps ever made by anyone.

I bought my first one over 18 years ago. It pumped down with dirty oil to 15 microns. I drained the oil after a 1 hour run. It was a bit rusty and filthy as well. I reloaded TKO pure oil and ran for 3 hours. Drained and the oil was murky but cleanish. I finally reload with good belt drive oil and it pulled down to 5 microns. I never looked inside. At 5 microns it is a good pump.

TKO oil is flushing oil made specifically as a thin pump cleaning oil.

The second pump I picked up at a hamfest. I could see no oil in the pump at all. The drain plug yielded about a cup of grime and sludge that oozed out. I dare not try to hand rotate the pump pulley.

I mixed TKO 50:50 with Marvels Mystery oil and filled the pump. I let it set for 1 week, after which time I tried to turn the pump wheel and it broke loose from a locked condition fairly easily. I turned it by hand for several turns and was smooth. I ran the pump for only 5 minutes and then drained a pure rust colored ruby red oil. I reloaded this time with TKO and let the pump run for a full hour and the pressure was down to 20 microns. (Marvels Mystery oil has fractions that create vapor from that first fill.) I drained again and the oil came out only slighly rusty but without sludge. A final filling with real pump oil and the pump went to 10 microns. I ran it with this oil fill on and off for a week and on the final day it pulled to 12 microns.

One final drain had the oil fairly clear of rust and just a little light gray near the end of the draining, (sludge). A final last fill with oil and the pump pulled to between 5 and 8 microns. I have left this pump sitting for 8 years. About 6 weeks ago I used it to evacuate a discharge tube that I was testing and it pulled to 10 microns after 10 minutes of gas ballasting.

Two sleeping dogs restored with no peeky up their skirts to see what was really there. For my money if your old used pump gets down under 15 microns, your pump is good enough. A nice sludge lining that allows the system to pump to that level can be left in the oil reservoir. If you change your oil in heavy use as required, it will ultimately work its way out in the draining.

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
Retired now...Doing only what I want and not what I should...every day is a saturday.

Tyler Meagher
Posts: 16
Joined: Thu Mar 08, 2018 3:29 pm
Real name: Tyler Meagher

Re: Vacuum pump rebuild

Post by Tyler Meagher » Sun Apr 22, 2018 1:50 am

I took apart the pump this morning. The first picture is the pump. It is only three inches long but it does all the work. You can see the little coupler that attaches the motor to make the pump turn. The second picture is looking down at the pump. The third picture is the first stage and second stage split apart. The U shaped cutout is what connects the two stages together. The fourth picture is everything inside disassembled. I didn’t take the vanes out of the rotor yet. The last two pictures are the first and second stage chambers. They have some rust so the pump might not work well when I put it back together. I’m going to clean everything, and then put it all back together.

My dad is also going to help me learn how the pump works in more detail.

pump assembled.jpg
The pump and coupler
pump motor interface.jpg
Top view of pump
pump first second split.jpg
First and second stages split
pump disassembled.JPG
Pump disassembled
pump second stage.jpg
Second stage chamber
pump first stage.jpg
First stage chamber

Post Reply