Vacuum Pump Rebuild

[Howard Prout]

I had the 430 engine in my '59 convertible rebuilt last summer. After everything was back together I was installing some gauges and found that the vacuum from the mechanical vacuum pump was only producing about 10-12 "Hg - not the spec. 23-25"Hg at 2000 engine rpm. I took the vacuum pump off and had the cavity machined but wasn't able to find new vanes so put everything back together only to find the situation was even worse. I then searched the internet for another vacuum pump and was lucky to find one at Turner's Auto Wrecking in Fresno, CA. I asked the vendor to check it to make sure it worked (it seems that most don't) and was assured that it produced at least 20"Hg. However, after I received it, I tested it and could only get about 10"Hg. The vendor offered a refund if I returned it but I decided to keep it and try to have it rebuilt. It turned out that the cavity was was scored and both vanes were chipped. (As an aside, there are at least two versions of these vacuum pumps - one has a cam in the bottom of the cavity to move the vanes out to cavity wall and the other has a spring between the two vanes to keep them in contact with the cavity wall - I now have one of each.) I was not able to find anyone to rebuild the unit so decided to try to do so myself. After a lot of searching, I finally found a source for new vane material - a little too thick and too big but workable. I had the cavity wall machined to take out the scoring and then had the new vanes machined to exactly the right size- thickness, length and width - max. tolerance of less than 0.001". After the unit was reassembled I bench tested it and was able to get 22"Hg. In retrospect, the vendor may have been right as I don't think I was spinning the pump as fast as the specs require. The vacuum pump needs a supply of oil to work properly so it has to be tested as a unit with the oil pump. As a result it takes quite a bit of power to turn the two pumps at 1000 rpm (half crank speed) in oil. I initialy tried using a reversible air drill to turn the pumps but it didn't seem able to turn fast enough under load so tried an electric drill. It definitely turned the pumps faster (probably close to the desired 1000 rpm) than the air drill and after a bit of run time produced about 22"Hg. I expect it to get better as the new vanes seat. The real test will come when I get around to installing the rebuilt pump in the engine - it may take a while to get in the mood to lift the engine and remove the oil pan - yuk!

[YellowRose]

Vacuum Pump Rebuild

Hi Howard!

First of all, welcome to Squarebirds! It is great to have another Canadian member! If I can be of any assistance to you in manauvering our Forum , please let me know. Thanks for the informative post. I am sure other members will find it interesting and helpful.

[GTE427]

Howard,

Thanks for posting your lessions learned. On my 59, there's a pipe plug in the side of the block where the check valve fitting would have been, I assume the pump was discarded. Continue to post the findings with your 430, much appreciated.

[tbird430]

Yes, welcome to the site. I think you are the 1st gent I've heard about resoring one of these odd/rare vacuum pumps for the 430cid. I own a '60 T-Bird with the 430cid as well. As a result, I find this story very interesting (my pump was long gone with I replaced my oil pan last year).

I guess what I'm trying to say, is please keep us updated on your progress. Pics would also be a plus.

[Howard Prout]

I should have added that as a result of an recent inquiry, Turner's Auto Wrecking in Fresno, CA advised me that they have another vacuum pump. Now that I have one working I don't need another so if anyone is interested, here is a source. I was quoted $75 plus shipping.

[Howard Prout]

For those of you who haven't seen one of these vacuum pumps, the attached collage shows one in various stages of assembly. They are really quite simple - a cavity, a rotor and two vanes. As the rotor spins it sucks in air through the inlet and exhausts it through the snorkel. What I learned is that the tolerances required to make it work are very tight - less than 0.001" - but not beyond the capabilities of a good machine shop. The vanes (the black rectangular objects in front of the rotor in the lower left photo) also sweep the top plate and bottom of the cavity. I'll find out how good my rebuilding efforts were in a couple of days when I switch vacuum pumps. The style shown has a cam in the bottom of the cavity. The cam and centrifugal force keep the vanes in contact with the cavity wall. The other style does not have a cam in the caivty but rather has a spring between the vanes to keep them in contact with the cavity wall. I don't know why there are at least two styles of pumps, which is newer and older, and which works better. I tend to like the spring design as the springs can compensate for wear. I have heard that some Cadillacs also used these vacuum pumps. They are certainly an oddity!

[Howard Prout]

I started this thread with a narration of what took place after I found that the vacuum pump (pump #1) that was in my engine was creating about 10-12" Hg at 2000 engine rpm. The Shop Manual says these pumps are supposed to create 23-25" Hg at 2000 engine rpm. I took the pump out of the engine and did a partial rebuild of that pump (I wasn't able to find new vane material at that time). However after the rebuild it appeared to only produce 8 -10" Hg at 2000 engine rpm. So I got another vaccum pump (pump #2) and did a complete rebuild of it, including new vanes. After the rebuild, pump #2 initially bench tested at 10-12"Hg. I was surprised by this and wondered if I was spinning the pump as fast as the spec required. I borrowed an optical tachometer and it turned out that I was not. When I bench tested pump #2 at a tach reading of approx. 1000 actual rpm (the pump is driven by the distributor shaft which turns at half engine rpm so 1000 actual rpm is equivalent to 2000 engine rpm) it created about 22"Hg. Still not the 23-25"Hg specified but close. I put off replacing the pump in the engine (pump #1) until some other things were in place. That time has come and I had everything set up to replace the pump tomorrow. In the meantime I started the engine a few times and gradually the vacuum created by pump #1 increased. After I started the engine this morning and let it warm up, it was now up to 18"Hg (see Auxiliary Vacuum gauge in the attachment below). It am hoping that the vanes are seating and that it may soon get up to the normal operating range. So for the time being, pump #1 is staying in the engine and pump #2 will wait in reserve.

[KULTULZ]

I love those dualing vacuum gauges!

I know you know Mr. Prout, but I would like to remind everyone that the purpose of the vaccum pump is to supply vacuum for the wiper motor when manifold vacuum drops (manifold vacuum being the main source).

MEL could not use a dual action fuel pump because of space limitations as to where they placed the MEL fuel pump.

Great TECH INFO here. You should condense it and offer it as a TECH ARTICLE here IMO.

[GTE427]

What is the source of the new vanes used in the rebuild of the pump?

[Howard Prout]

I got the vanes from Precision Plus Vacuum Parts, 2055 Niagara Falls Blvd., Niagara Falls, NY 14304. Tel. 1-800-526-2707. Website: www.precisionplus.com. Part No. 20010655. Dimensions: 0.194" thick, 2.24" long x 1.47" wide. Price US$28.97ea - two required. They need to be machined in all dimensions but they provide a reasonable place to start from. They don't normally sell to the public so you need the name of a company (if necessary, make one up such as XYZ Automotive Restorations), they don't seem to check for validity - if they do just say it is a proprietorship. Once I had the vanes I took them and the pump to a machine shop. They resurfaced the wall of the cavity and machined the vanes to fit - cost about $200. Hope this helps.

[Howard Prout]

I love those dualing vacuum gauges!

As you can tell, I am a gauge nut. The vacuum gauges were a bit of a challenge. My gauges are Sunpro Style Line with the exception of the tach which is Autometer. I was quite specific in what I wanted - white face, black graphics, red needle and chrome bezel - to closely resemble the stock instrument panel gauges. Sunpro doesn't make a vacuum only guage in its Style Line series - only a Vacuum/Boost gauge, so I initially installed two of them. But I decided I really wanted vacuum only gauges so I opted for the vacuum gauges in their Retro Line series. The problems with these are that they are black faced, white graphics, white pointer and have a different bezel. So I took the gauges apart, scanned the black face plate and with the aid of GIMP ( a graphics manipulation program similar to Photoshop), I inverted the face colors and modified the graphics. I then printed the new faces on photographic paper, cut them out and installed them over the original black faces. I am currently in the process of getting more permanent faces printed on adhesive backed white vinyl. I used the red pointers and bezels from the combo gauges so that they now match the other gauges. The first picture shows the original black face and the second after image manipulation.

[KULTULZ]

I'M IMPRESSED!

Quite resourceful...

[Howard Prout]

A few days ago I wrote

After I started the engine this morning and let it warm up, it was now up to 18"Hg ... So for the time being, pump #1 is staying in the engine and pump #2 will wait in reserve.

As it turned out there was some leakage from one vacuum source to the other even though there are check valves in both lines so the 18"Hg reading wasn't real. When I isolated the line from the vacuum pump it turned out to only produce about 10-12"Hg. So this morning we replaced the vacuum pump that was in the engine with the one I rebuilt. I then tested it in isolation and it produces up to 25"Hg at 2000 engine rpm- right on the specs. What is puzzling though is that the vacuum fluctuates even though the line is buffered by a check valve. Anyway, I'm happy that it is working the way it is supposed to. Now I have to try to figure out why there is leakage through the check valves. I tested both valves and they don't leak under a static test.