I am also thinking of mounting
one of the cheap $80 12 bit AD to D data acq system back there
with the USB interface to monitor the 555 EFI system and its
sensors on a notebook computer. Add a few ten dollar pressure
sensors and lots of things are possible. The A to D can also handle
the fuel flow read out and fuel totalizer in software.

http://www.hytekautomation.com/

I just came up with another idea using a tiny electronic load cell
out of a cheap bathroom scale incorporated into the motor mount to
measure prop torque. Knowing prop torque and prop RPM you know HP
out of the gear box. The A to D system can measure that as well.
I got the idea by watching the TV program Deconstructed on the Science
Channel. I'll appropriate  Robin's  bathroomscale :)

-----------------------------------------------


Here are some pictures of the load cell in Robin's bathroom scale :)
I'll have to buy her a new one at $34.

It is battery powered for the life of the scale in that black container
behind the LCD read out.

I'll work on a 3D dwg for the mechanism for the end of the S-beam
motor mount. One side has all the load due to engine torque.
The 12 bit A to D has such  a high res and a wide dynamic range
it does not matter if it is measuring the weight of the engine only
when the engine is not running. The trick is have it bottom out
 mechanically
so it does not damage the load cell during hard landings and high G pull up
 loads.
That would be like having an over 300 pound person standing on the scale.

Paul Lamar

 Hi Paul,

 Those things like to turn themselves off after a few seconds.  The ones
 I have make you tap the scale, then it goes through several seconds of
 a power up sequence.  You may have to figure out some way to bypass the
 automatic power feature and supply a larger battery.

 Bob W.
 --
 N93BD - Rotary Powered BD-4 - http://www.bob-white.com
 3.8 Hours Total Time and holding
 Cables for your rotary installation - http://roblinstores.com/

 I figure I could use just the load cell part.
 It is probably The crystal type.
 The force is applied to a piezoelectric crystal
 and you measures the voltage across the crystal.

 ---------------------------------------------------
 Preliminary design.

 Here is what it would look like on the end of the S beam.
 It is possible to retrofit an existing motor mount by
 cutting off one cup and welding this on.

 The lever reduces the load on the cell.

 Does anybody see any problems with this?

 Paul Lamar

 Hi Paul,
 You might want to add a positive stop on at least one side of the
 load cell to prevent damage to the cell in the event of a
 catastrophic down force on the lever.Dave M


 Very good point Dave.
 Not shown yet. I think I can use a small machine screw and make the stop
 adjustable.
 I'll have to experiment on the bench and measure how much deflection the
 crystal
 can tolerate. In this design it only sees about half the load but that may
 be way too
 much for this crystal. The bathroom scale has a lever system that
 drastically
 reduces the load on the crystal. I will need to research load cells
 further. Still a work in progress.

 I did add an up stop for the lever to keep turbulence from allowing the beam
 to jump
 up. I also completely redesigned it as the first design used a narrow
 S-beam.
 Current S-beams are 3.5 inches wide to pick up more oil pan bolts. I also
 raised the pivot bolt.

 Quiz: can anybody tell me why the pivot bolt needs to be in line with
 the center of the rubber bushing? :)

 BTW the output of the load cell can probably act as a prop RPM sensor
 as it will see the engine firing pulses twice per revolution. It can also
 measure the magnitude of the pulses for rubber coupling development.

 ----------------------------------------------------------

 I looked into the load cell a bit further and most are $400 to $500.
 The max output voltage for a 250 pound load cell is 250 milli volts so that
 is a far cry from what we need. We need zero to 5 volts for the data acq
 so an amp would be required with most load cells.  The pressure sensor we
 used on the manifold wave test is $10 and they can be had up to 100 psi.
 They
 have the required zero to 5 volt output with an amp built in.

 So here is plan B. Use a simple 2 inch diameter piston
 which gives an area of 3.1416 square inches. At 100 psi the max
 load is 314 pounds. We have a two to one lever so max load is
 really 628 pounds. More than enough.  A small hose going back
 to the pressure sensor and data acq system is all the "wiring" that is
 required. An occasional check on the fluid level is all the maintenance
 required. Some sort of inert oil is required to be compatible with the
 pressure sensor. Teflon O-rings will cut the friction in the system.


 Paul Lamar

Elegant.  I like it.

Regards,

Matt-

Thanks Matt.

BTW What ever you did with your email program it is working fine now.
Thanks for that too.

Paul Lamar

Paul,

Better but still a tad big.  Does the sensor see more torque as the Barry
Mount ages?  If so I would think you would want to measure the force
upstream from that absorber instead of parallel.  Cheap flexible strain
guages that are small enough to be encapsuled between two washers are on
the market. You can put the device under the Barry mount mounting bolt.
The output of a strain gauge as you well know is just a resistance value
and in any case you have to scale it and calibrate the  range it to the
particular application.

The good thing about the digital scale is that it already has the
temperature correcting circuit built in.  Perhaps you can use the
circuit but not the piezo.


Doug in Japan

The stiffness of the rubber bushing matters not.
The load on the L beam lever is the same.

I could not find anything less than $400 for strain gage
washers and they still require amplifiers. It is an elegant
way of doing it if you don't mind the cost and the amplifier.

Yes and no on the scale load cell. The range of the load cell in
the scale is too low. There is no way to get a lever in there
long enough. It would have to be about 9 inches long.
We need the space under the S-beam for the rad core duct and
behind the beam for the exhaust system. I am open for suggestions
on other layouts.

The 2 inch hydraulic cylinder and a 100 psi pressure sensor is the simple
way to go IMHO. The whole system can be made with nothing more than
a lathe, a band saw and a welder. No mill required.

Paul Lamar