More musings on pump capacity...

[eric schumacher]

Hi charles
My point was only that it is often easier and definitly much more reassuring
to measure things like fuel pressure than to depend on calcuations.  Some of
the concerns you listed like cavitation and bubbles are real but if you (you
refers to the designer not necessarily Charles)have to worry about then that
part of the design should be finished first. 

To shift off subject a bit I found in building a system based Bosch CIS
components and using the ubiquitous rotary pump that it was very sensitive
to the diameter of the inlet tubing as compared to the diameter of the
outlet tubing.  It didn't matter what the flow was as long as the goesinta
was bigger than the goesoutta all was ok.  I also note that Bosch systems
typically use an accumulator to take of those fast demand transients.

This is a great group. Lotsa people doing neat things. 
Lotsa Luck Eric
>>2- Why would you wonder for one minute if your fuel pump is adequate when it
>>is so easy to add a simple ashcroft type guage and some nylon line to
>>measure the fuel pressure under any condition you care to evaluate.  That is
>>what regulators are for e to take care of variations such as voltage and
>load.
>
>	Well...I have, in fact, done so. But you are assuming that the
>gauge responds fast enough to see momentary short-duration changes,
>and that any flickering of the needle isn't due to vibration and bumps
>on the road. If some bubbles are being formed in the pump inlet due to
>cavitation, cylinders will not fire when they pass through the
>injectors.
>	A regulator won't fix this. They are also not intended to make
>extra fuel pump capacity where none exists. That's why I've been
>wondering (for several minutes).
>
>	Also, I've done a little more math.
>
>	Stock injectors: wide-open 273 cc/min @ 36 psi.
>	Assume density of gasoline = 0.75 g/cc
>	6*273=1638 cc/min = 1228 g/min = 163 lb/hr with all 6 open.
>
>Original engine rating: 252 hp. Assuming 0.5 lb/hp hr BSFC, this
>requires 126 lb/hr. So the injectors should be open 126/163 or 77%
>duty cycle, which is about right - most systems are designed to max
>out at 80-85%.
>
>	Now, I'm running 43 psi fuel pressure, so the above flow goes up
>by sqrt(43/36) = 1.093:1 = 138 lb/hr.
>
>	My two additional injectors are rated at 61 lb/hr each at 43 psi
>(I won't bore you with the calculations, but if anyone has data for a
>Siemens # 8213A for a Lamborghini, that's what they are).
>	My fuel map has them open about 60% at 15 psi boost, so 2*61*0.6
>= 73 lb/hr.
>
>	The unknown is the amount of fuel going through the nitrous
>setup. My "Nitrous Oxide Injection" book shows anywhere from 40 to 60
>lb/hr for a 100 hp plate system. If this is proportionate, my 40 hp
>nozzle should be consuming about 20 lb/hr.
>
>So the total consumption is 138 + 73 + 20 = 231 lb/hr.
>
>	From last weekend's measurement, the pump puts out 1350 cc in 30
>sec (at the simulated 15 psi boost level). This is 268 lb/hr at .75
>density. On paper, at least, I should have enough pump. But it
>wouldn't take a lot of error in some of my assumptions to be exceeding
>this. Any constructive thoughts welcome...
>
>-Charles
>'81 745i
>
>
>
Lotsa Luck Eric