CB PERFORMANCE DRY SUMP PUMP.

The VW Beetle. Everything about bugs!

Moderator: Marc

Post Reply
User avatar
Crawdad
Posts: 65
Joined: Wed Sep 16, 2015 6:12 pm

Re: CB PERFORMANCE DRY SUMP PUMP.

Post by Crawdad » Mon Feb 19, 2018 1:47 pm

This might be a little controversial. In addition to enlarging the internal passages of the CB dry sump pump, dong the usual blue-printing and o-ringing (on my build thread: http://www.shoptalkforums.com/viewtopic ... 5&t=150113), I added a second port to the pressure side of the pump. To get from the pressure stack to the main oil gallery, the oil has to make a bunch of 90 degree turns. What I did was simply add an alternative path, a parallel circuit IN ADDITION to the existing one. It taps directly off the area where the pressure gears fling their oil, through a 3/8 external line, and into the main gallery through the hole where a full-flow system is normally plumbed. I expect having a second path will reduce pumping losses; I don't expect it to have any effect on the volume of oil flowed. It might help pressure a bit. The size and RPM of the gears determines flow, I am merely reducing the amount of power it takes to generate that flow. The ratio of scavenge flow to pressure flow is also preserved, since that is simply a function of the relative sizes of the two gear sets.

The real trick was figuring out what kind of fittings and oil line would minimize the packaging problem. There is no elegant way to plumb it, because the inlet and outlet are so close to one another. Fittings take up a lot of space! And hose can't turn a sharp radius without collapsing. So I used a hard line of 3/8 copper-nickel alloy; it is sold as transmission cooler line in auto parts stores. Here it is in its rough configuration. It stays well away from the exhaust. I flared it with a cheap HFT flare tool. It turns out the AN standard (37 degree single- flare) is identical to the JIS standard that plumbers use for gas lines, so I went to the hardware store and was able to handle a bunch of different fittings to see what would work. I'll be keeping a close eye on it to look for leaks; the main worry is metal fatigue at the fitting. There shouldn't be any relative motion, and copper is fairly malleable, so I'm hopeful.
You do not have the required permissions to view the files attached to this post.

User avatar
FJCamper
Moderator
Posts: 2746
Joined: Wed Nov 14, 2007 5:19 pm
Location: Birmingham AL

Re: CB PERFORMANCE DRY SUMP PUMP.

Post by FJCamper » Tue Feb 20, 2018 2:09 pm

You are braver than me, but I can't think of any obvious reason your mod won't work if the goal is to reduce pumping resistance.

Paul Ray the Rocket Scientist, on our staff is a retired NASA fluid dynamics engineer and he looked at your plan and tried to detect some unintended consequence and made one guess (and it is a guess). You now have two pump outlets that join at the main oil galley. Getting technical, it would be possible depending on plumbing restrictions to have a pressure differential there that fights itself before equalizing and flowing down the main oil gallery.

Paul says in industry, where two pressure sources meet, a pressure equalizer plenum is plumbed in to guarantee equal pressure downstream.

There's no room for that on our engines, so the thing to do is try out your idea. I'm rooting for it to work.

FJC

User avatar
Crawdad
Posts: 65
Joined: Wed Sep 16, 2015 6:12 pm

Re: CB PERFORMANCE DRY SUMP PUMP.

Post by Crawdad » Sun Feb 25, 2018 8:14 am

Thanks for that, FJ. Is the worry that some weird resonance might appear, kind of like a standing wave, that inhibits flow? That's subtle stuff I hadn't considered. I'm no fluid dynamics engineer, but it's hard to imagine what the source of the "fighting" would be. The best I can come up with is that the teeth of the gears are the source of periodic variation – a pulse – in the pressure, and since those pulses are likely to be out of phase when they meet in the main oil gallery (because they travelled different paths), when they superimpose there will be some mutual subtraction.

Post Reply