I'm interested in trying to make one of these:
Mostly because of the price, but also for the challenge. The basic idea
of this attachment is to allow coolant (water) to be pumped into a tool
bit which is attached to the bottom of the shaft. A motor drives the top
of the shaft. Water is introduced at the side of the body and presumably
flows through perforations in the shaft and out an axial hole at the
The construction and machining seem pretty simple. But what kind of
bearings or seals can be used between the body and the shaft that will
handle, say, 20-50psi without without leaking? The shaft needs to rotate
somewhere in the range of 3,000 to 15,000 RPM.
Does anybody have any clues about this? Am I better off to spend the 140
bucks and just buy the thing?
For some reason I thought you were talking about a submerged application.
I believe there are ways to use the shaft speed to pump the water that
leaks past the seal out to where they won't get to the motor.
Think a disk past the seal which slings water outwards, for example.
Unfortunately, SS bearings will corrode if this is done. Furthermore,
flooding a roller bearing will cause it to skid and will destroy the
bearing. I built a centrifuge with ss bearings because that was the
spec. They lasted 1 day. Only pure water was used. Failure was due to
I've never seen one of these, but I would expect that the water passes thru
a hole in the spindle (between the bearings, and out the cutter end, like
I don't think it needs to withstand much water pressure, the barbed hose
connector is small, which will restrict the water flow and the valve could
be adjusted to just supply a flood of water for the cutter.
I would expect the bearings to be standard sealed ball bearings, with a lube
port in the outer race, since there appear to be zerk fittings on the
Automotive water pump bearings withstand a lot of harsh conditions, and
about 16 psi of hot water, and relatively high speeds.
The model on that particular page is rated for 8500 rpm max.
O-rings can be used in this application. A little water must leak out
to lube the seal. This is a tiny amount that can be captured and
drained easily. I have built rotary unions this way for gundrills
operating at over 2000.00 psi. They were using oil. In your situation
even a lip seal may work because the pressure is so low. Parker has
the specs you will need to use o-rings. Because of the rpm and
pressure the groove for the o-ring will be different than for a static
seal. Another way to accomplish the rotary seal is to use carbon discs
which are lapped flat and pushed together by a spring. The rotary
unions I use in my shop work this way. However, the ones I use are
rated at about 3000.00 rpm at 150 psi.
The technical term for these devices is 'rhotostat'. They are
commercially available most everywhere. The tolerances and intricate
machining required to produce one is usually best left to people set up
to make them. That said, however, it is possible to produce one. The
bearings inside the unit never come in contact with the fluid. The shaft
is sealed with (usually) quad seals (sometimes o-rings), and on some,
very intricate and proprietary sealing systems, depending on the
application. But, at a cost of $134, you are probably better off just
purchasing this item. You will have AT LEAST that much in producing one
You can't 'idiot proof' anything....every time you try, they just make
Helical labyrinth. Half seal, half pump. They often leak slightly
when not rotating, so there may be an O-ring as backup.
NB - their rated speed is 8,500 rpm, which is slower than small angle
grinders run at.
Why would anyone want one of these anyway ? What's the application ?
I've seen plenty of wet and dry core drills, but never one that ran at
this sort of speed, or was driven by an angle grinder.
Cutting faucet holes in granite counter tops.
Nominal cutting speed for a diamond cutter in hard granite is 5050-6250
SFPM. So a typical 1-3/8" core drill would like to see 14,000-17,000 RPM.
At that speed in a material as hard as granite, the diamonds are be
quickly knocked out of an electroplated (nickel bond) bit. Newer
technology brazing has the strength to hold onto the diamond if the feed
pressure is moderate and they are run wet, but they are single-layer
tools and have a short life.
The long-life solution is a sintered, or "metal bond", core drill, but
they can't be run dry. Submersion is possible for shallow holes (< 1")but
is not as effective in deeper ones. Another drawback of submersion is
that the dust and chips aren't cleared as well as when the coolant is
pumped into the center of the bit.
Thanks for the ideas, everyone.
I'm a tad surprised there isn't an obvious simple solution, other than
perhaps the O-ring option. So much of the tooling in the granite cutting
industry is so wildly overpriced (owing to low volume, I reckon) that I'm
betting there's an easy work-around. If not, the $140 won't kill me and
I'll learn something from it.
The oring solution is easy enough try out. You could even cut the
grooves for a sliding fit and see how long the seal lasts. Then, if
seal life is too short, make the grooves a little deeper.
Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here.
All logos and trade names are the property of their respective owners.