[ausev] Motor dyno, was: GE 11.5 inch motor

[Willie McKemie]

On Tue, Jan 01, 2008 at 11:39:20AM -0600, Chris Robison wrote:
> > A big centrifugal water pump with a choke valve on it's outlet would 
> > certainly make a good load.  It might be possible to make some power 
> > estimate by measuring the flow and the temperature difference between 
> > inlet and outlet.
> > 
> 
> Cool idea, if you could achieve something approaching accuracy. I'm 
> concerned that it wouldn't be able to build up enough pressure though. 
> Centrifugal pumps or fans are good at pushing against backpressure, but 
> not really at the pressures I'd think would be necessary. Something like 
> a gear, scroll or piston pump might make a better load against an 
> adjustable orifice. Maybe with a pressure sensor, and some way to 
> measure output flow (and probably temperature as you mentioned). Maybe 
> if you blocked the centrifugal pump completely, you could make the 
> device measurement exclusively thermal.

Alright, you got me going on this.  It's been 40 years since I've had 
engineering mechanics and such; I'm CERTAINLY no expert.  But, here are 
my thoughts.  First, I think we can consider that all energy is being 
converted to heat.  Using a no flow pump would mean that heat would be 
continuously rising (no steady state) and you would have a hard time 
determining how much heat was being generated per unit time.  I think 
the goal would be to have the pump reach a steady state temperature and 
then measure both the flow rate of water and the temperature rise of the 
water.  There would be some heat flow from the pump to air (which would 
be difficult to measure), but I think most of it would be in the water 
(which should be relatively easy to measure).
 
> Torque curves are really important too, though ... and my searches so 
> far have found that shaft-to-shaft torque sensors are really expensive, 
> on the order of many thousands of dollars. Maybe we could wait for a 
> cast-off on eBay, maybe one that's lost its super-accurate calibration 
> or something.

I would think that torque (on the above hydraulic dynometer) would be 
pretty easy: a lever arm restraining rotation on the pump and a scale 
or other linear force measurer.

-- 
Willie, ONWARD!  Through the fog!
http://counter.li.org Linux registered user #228836 since 1995
Debian3.1/GNU/Linux system uptime  17 days 12 hours 49 minutes