I'm agnostic, I like the simplicity of the Barske and I think both can do the
job just fine. I have a soft spot for the Barske because of its simplicity and
think that for arocket it may be more appropriate since simpler tools and
setups are employed. I may be putting together a rocket powered dragster soon
and would like to try pump fed since weight is of less concern and battery
space is ample.
Russ
-----Original Message-----
From: arocket-bounce@xxxxxxxxxxxxx [mailto:arocket-bounce@xxxxxxxxxxxxx] On
Behalf Of Monroe L. King Jr.
Sent: Wednesday, July 15, 2015 8:38 PM
To: arocket@xxxxxxxxxxxxx
Subject: [AR] Re: Arocket Pump Progress
Russ you want to run the Barske? That would save me a lot of time and it is
very simple compared to the Francis design.
The one Peter F. has thrown out there looks to me to be a bit too small but I
admit I know very little about Barske pumps I've never seen one.
There was a fella on Arocket that I've never heard from said he was going to
dust off his cad skills on a Barske design I think I'll ask him to verify? And
perhaps he will do the cad on it.
-------- Original Message --------
Subject: [AR] Re: Arocket Pump Progress
From: "Russell Blink" <russblink@xxxxxxxxxxxxxxxxxxxxxx>
Date: Wed, July 15, 2015 3:20 pm
To: <arocket@xxxxxxxxxxxxx>
The machining on any of these small impellers is pretty easy, until it
requires swarf milling on our 5-axis machine they are all about the same. I
turn the profile, cut around the vanes then part, pretty straight forward.
I do have to use a fairly small endmill so the machining time is a bit longer
than you would expect for such a small part, but not horrible. (I should say
easy for our lathe since I can do it all in one setup and if I need to
machine on the back side transfer to the sub-spindle is simple)
These small impellers are all inefficient ~65% would be good, pretty much
what a Barske will do and you don't have to worry about undue regenerative
heating or possibly the stall point of the partial extraction curve.
Tip/face clearance is the biggest difference if you ask me.
Russ Blink
-----Original Message-----
From: arocket-bounce@xxxxxxxxxxxxx
[mailto:arocket-bounce@xxxxxxxxxxxxx] On Behalf Of Peter Fairbrother
Sent: Tuesday, July 14, 2015 12:25 PM
To: arocket@xxxxxxxxxxxxx
Subject: [AR] Re: Arocket Pump Progress
On 14/07/15 16:37, Monroe L. King Jr. wrote:
I agree and it was the Merlin I was looking to for inspiration.
But the split design of the SA-2 also has it's merit and it is
possible to actually get a hold of one of those to get a good look
at and take measurements from as well as perhaps some 3D scans if
nothing else. It can also be copied.
I'm willing to cut some corners here and take advantage of a known
good design if we can.
The other thing is the SA-2's propellants and pump designs are
closer to the same for our chosen propellants.
also looks to me like the SA-2 has open impellers?
If I can find/buy one to use as a reference it should cut
development in half.
Or do you think that is mot a good way to proceed?
This is our project you tell me. You are designing a turbopump Peter.
propellents? thrust? flow rate? chamber pressure? pump exit pressures (you
might want the fuel pump pressure to be higher than the oxidiser pressure, as
it might flow through regen channels)? preburner conditions?
Look at those turbopumps - the first thing you notice is that the turbines
are a lot wider than the pumps. Your turbopump should look like that.
If you think about it, the turbine is pumped by gas, the pumps are pumping
liquids. Just from a density point of view, the turbine has to be wider than
the pumps.
The impeller design you are using is going to have terrible efficiency - the
vanes are too shallow for good efficiency, there will be too much friction.
You would be far better off with a Barske-type impeller. And it would be a
lot easier to make, you can machine one on a manual milling machine and lathe.
The blades are straight and parallel, and point to the center - so start with
a disk and endmill the edges of one blade pair to a depth. Repeat and rinse
at say 60 degree intervals to give six vanes. Tidy up the hole in the middle,
and cut away any parts which have not been removed, all to a single depth.
Place on lathe, center accurately. Adjust topslide to highly acute angle,
then machine straight blade tops. Cut edge of disk flat. Machine for shaft.
Deburr, polish, balance the impeller.
Place casing blank on lathe. Cut the edges of the hole in the casing straight
to diameter, just a little over the impeller diameter. Machine the base of
the concave hole using the same topslide setting and angle, but with the
lathe in reverse, then machine for bearings and seals.
Attach exit pipe, cut the tapered exit hole.
Simple. If designed and made properly, should get around 40-50% efficiency.
You will probably want to operate it at about 100,000 rpm.
There is a useful book by Lobanoff and Ross.
The main difference between the SA-2 and Merlin engine turbopumps is
that the Merlin pump has the turbine on one end, while the SA-2 tp has
the turbine in the middle. I like the Merlin arrangement - it keeps
the hot turbine as far away from the cold LOX as is practical.
The hot turbine is next to the fuel pump - doesn't matter much if a
few sparks get in between, as there is no oxidiser present. For the
oxidiser pump, a few sparks can kill .. there is fuel in there (the
pump body and impeller, if nothing else ...).
-- Peter Fairbrother
-------- Original Message --------
Subject: [AR] Re: Arocket Pump Progress
From: Peter Fairbrother <zenadsl6186@xxxxxxxxx>
Date: Tue, July 14, 2015 8:10 am
To: arocket@xxxxxxxxxxxxx
On 14/07/15 14:36, Monroe L. King Jr. wrote:
That does help some! Thank you. I do believe working from this
design is better than trying to go with a V-2 design :)
It's not bad, but if I was designing a turbopump I'd use the Merlin
engine turbopump as an example. It's a pretty good one.
http://www.barber-nichols.com/sites/default/files/wysiwyg/images/me
rlin_turbopump.jpg
The picture is pretty small, but it shows most things: at the
bottom, from left to right, there is the LOX pump with integral
inducer, the shaft with fuel pump and inducer, the turbine wheel.
It is a reaction turbine, and I think a partial admission one - a
reaction turbine has less end thrust than an impulse turbine, and
in theory only a very low pressure differential across the disk -
with a single shaft with two pumps, both with screw inducers,
pumping in opposite directions to balance end thrust.
Note that the turbine gas flows to the left, creating a leftward
end thrust on the shaft, and the LOX flows to the left, creating a
rightward end thrust to balance both the turbine end thrust and the
lesser leftwards thrust of the fuel pump (fuel flows to the right).
For a smaller engine I would consider Barske style partial emission
pumps, or closed impellers; if nothing else, in order to make
construction easier. Perhaps tesla inducers too.
-- Peter Fairbrother
