Miniature fuel pump design help

[BillClark]

New member here. Like to first start by saying hello and also to mention how excited I am about finding this web board and I have high hopes for this knowledge pool to steer me in the right direction. I am developing a electronic engine management/fuel injection system to use in our 30-40% (of full scale) radio controlled aerobatic airplanes. They use 100-150cc 2 stroke gasoline engines. My professional experience is as a CNC Machinest, Metal faricator, and performance engine nut.If your interested go here to see my profesional life:
myspace.com/1metalmorphose
The area I despiratley need help in is with the fuel pump for the fuel injection. Key points are:
Fluid type- gasoline W/ 3% lubricating oil (2-stroke mix)
Flow requirements- 7 fluid oz/min
Pressure- approx 30psi but more would be better and less could be usable
Pump type- gear, with steel or some other durable gear material
needs to be small- I am invisioning the pump unit, less motor, about the size of a matchbox or less
If we can make it work I would like to use a electric brushless motor that produces about 111 watts(.149hp)@ 7500rpm. This would fit the system well from a standpoint of power consumption. I'm assuming the pump speed will have to be kept low to prevent cavitation so a gear reduction would be in order also getting our power up. Smaller and larger output motors are available but thats about minimum rpm.
I want to use a gear pump primarily because its simple and I can easily change the flow by changing width and there may be different requirments down the road.

Specific information I am looking for is as follows:
-what type of tooth design is best for this app.(pressure angle, # of teeth, material)I can machine all but the gears so a design I can easily get ready made blanks for would be great. Like

http://www.sdp-si.com/

-what dia and width would the gears need to be to deliver the performance as stated above hopeing that a design with a long X small diameter combination would allow easily changing housing width to change volume requirments and tweek.
I realize that the fit (leakage) and the inlet and outlet porting will effect the performance. I just need a place to start so I can build a prototype and see what we get and then go from there if needed. Just need some help from the educated and experienced. thank you

 

[itsmoked]

I wonder if you couldn't use a centrifugal pump instead. 30psi should be easy to achieve. A centrifugal would shed turning load if the fuel consumption went down. A gear pump may need more sophisticated motor control or waste a lot more energy due to having to relieve at low demands so there is enough volume at high demands. It would also be one moving element instead of two. It may be quieter too.

Alternatively just use an Off-The-Shelf automotive fuel pump. Many are now tiny and mounted in a single, small, metal can. They even include the motor in the can. They are mounted directly in the fuel tank.

Keith Cress
kcress -

http://www.flaminsystems.com

 

[insideman]

The specific speed is wrong for a centfugal. Available power is many times what is needed. A tiny piston pump with a leather or O-ring piston seal should work. Simple and effective!

 

[BillClark]

Thanks guys for your responses. I did leave out a couple details. First is that the motor would have a speed controller to help stay in step with the demand and of course there will be an injector upstreem. And second is that I intend to market this system at some point and the guys that would use it regard looks as almost as important as actuall function. Here is a picture of automotive type performance gear pump that would look great scaled down.

http://10.184.80.154/

http://www.magnafuel.com/products/efi/pumps/MP-4701-540pxW.jpg

What would be the specifics in design of a piston pump? I might be able to make the whole thing in house with a trick machined and anodized housing. maybe try both designs as the gear pump i beleive would be more sexy the piston pump might be far more efficient. Either way I would need actuall demensions as my strength is in building and less so in designing. thanks for your continued feedback

 

[MikeHalloran]

So, ... is the injector expecting constant pressure, or constant flow?



Mike Halloran
Pembroke Pines, FL, USA

 

[BillClark]

Hey Mike, typically w/ electronic fuel injection you want to maintain a constant fuel pressure (unless the engine uses forced induction).

 

[MikeHalloran]

That clears up a little.

How about just buying what you need?

http://shopping.netsuite.com/flightworks

Mike Halloran
Pembroke Pines, FL, USA

 

[BillClark]

Good find. thats pretty much what I need. Just need to manufacture it myself. Cost is going to be the biggest challenge as always and keeping as much in house I have found from experience works best. If I was looking at a 1 off system I'd be a fool not to use one of those.

 

[MikeHalloran]

Don't give up on an OEM deal before talking to them.
You got trouble enough.



Mike Halloran
Pembroke Pines, FL, USA

 

[BillClark]

hit the nail on the head there. Actually I have help so the burden isnt completley on my shoulders. And i'm pig headed but usually succesful

 

[GPRnD]

Using a gear pump to deliver a constant pressure feed to an injector is a bad idea unless you incorporate some relief valve and fuel return system.

I think you'll have difficulty getting your speed controller to keep step with the demand, although it might work if you feed it a pressure signal. However my feeling is it is probably going to be tricky to get the controller droop where you want it to be.

A Barske type impeller pump might work and is relatively easy to fabricate for someone with your skills since it has v simple vane shape. If you use it with a simple concentric casing and a suitable size of drilled takeoff, it might deliver what you need.

At 7500 rpm, 6 oz/min and 30 psi I estimate a pump specific speed of about 60, which is low but possible for a Barske type. Running it faster would be better as this will raise the specific speed significantly and reduce the size of impeller you need to make.

http://books.google.com/books?id=eu...3ITwCA&sa=X&oi=book_result&ct=result&resnum=6

 

[BillClark]

Thank you for the reply. I had intended to use a spring loaded pressure regulator, probabley in the pump and/or "jets" to size it correctly to the application as well as electronic control. under pressure I dont know how the response would be but the motor speed can be controlled very acuratley and is very responsive. it would be a similar set up like we use in our electric rc planes with brushless dc motor and lithium polymer batteries. and that appears to be one of the most popular areas that this motor/controller/battery technology is used. The engineering would probably be even more challenging than a gear pump for me but I'll try to digest the link above. My thinking was that if I went with a positive displacement pump that I could match that to the engine requirments possibley using a less sophisticated secondary metering device other than the electronic fuel injector. I am a drag racer and we run a Alcohol Funny Car. it has mechanical fuel injection. this system uses a gear type pump driven at half engine speed. a barrel valve which has a "v" slot that moves with the throttle to controll part throttle fuel metering and then jets and pressure releif valves to govern the full throttle mixture. the drawback to this system is that it gets no feedback from sensors
(for atmospheric changes)so it has to be manually set before making a run. my system will have feedback from either a O2 or hot wire to compensate for atmospheric changes and that sensor could, through the electronic controll unit,vari the motor speed to change the fuel delivery. So I am wondering if it has to be a positive displacement pump?

 

[moltenmetal]

High head and low flow means at least semi-PD pump like the gear type you're considering.

NPSH will likely be a problem, as will getting the thing to last on the low lubricity feed you're planning to use. The smaller the gear pump, the less tolerant they seem to be of low viscosity feeds. They wear pretty quickly on thin feeds.

Commercial mini-gear pumps like the Micropump exist already, but suffer from the problems I've described. Connecting one of these commercially available units to your proposed motor would be a first step in developing what you need, since engineering a mini-gear pump that lasts is a fair bit of effort. Best to see if a commercially available design works and lasts before you waste the effort re-inventing that particular (toothed) wheel.

 

[JoeySoap]

Wouldn't it be a lot more sensible to be a little old school and use a carburetor? Good luck!