Sorry I have been so slow to reply. I have been absolutely flat out
at work.

I don't where too much about load cells, other than some different
types and what they do. I don't know enough to recommend one. Try
reading about them on the web and maybe talking to a supplier. This
site has some good info;
http://www.omega.com/prodinfo/loadcells.html

If your torque arm restraining bolt is in tension then you probably
want a simple tension load cell like this one;
http://www.transducertechniques.com/TLL-Load-Cell.cfm

I have added a spreadsheet for you in the files section. You will
notice I typically calculate everything in SI units, as then you
don't have all the crazy multipliers needed to make imperial units
work. For example, if you know the force in N and the velocity in m/s
then you just multiply the together and get power in Watts.

Have a play with the excel sheet and tell me if it makes sense.

Wes

--- In exceldyno@yahoogroups.com, "Todd" <narider@t...> wrote:
> > A load cell is a good accurate way of getting the torque in the
> > roller reading (the hydraulic brake isn't driven through a
gearbox
> > by the roller I assume).
>
> The hydraulic brake is direct drive from the end of the drum's
shaft.
>
>
> > It's easy to set-up if the hydraulic brake is
> > already mounted using a torque arm.
>
> I'm not sure what is meant by a torque arm. There is a tab with a
> bolt going up to the floor of the dyno coming off the mounting
flange
> area of the pump that appearantly keeps it from being spun?
>
>
> > Then you multiply the load by the
> > torque arm radius and you have the roller torque.
>
> I'd have to look but I'm not sure that the mounting arm I speak of
is
> the only thing keeping the brake from spinning, but if it is then
> what you're saying is I could simply measure out from the center of
> the pump to the end of this arm to get the radius, and a load ceell
> opf some sort would mount at this point?
>
>
> > Now just use the torque at the roller and speed of the roller (in
> > RPM) together to calculate the power at the roller.
>
> Speed at the roller is no problem(I currently have that in MPH,
which
> could be converted to rpm easy enough). The torque I have also, but
> it is in PSI, and at any given rpm is different depending on GPM of
> the hydraulic brake.
> IE: TQ=(((PSIxGPM)/1714)x5252)/RPM
> The above(at the HP versus ground speed level) is the formula I
> believe my charts are figured from by the OEM.
>
> So to find GPM(as it's ever changing with psi and mph) is rather
> tough so my first thought was be to put a flow meter in rather then
> try to back figure(as I don't know horsepower in the first place
and
> it is putting too much data in the mix to input).
> IE: GPM=(1714/PSI)xHP (so adding this to the above formula jsut
keeps
> adding more and more room for error that I was hoping to bypass
with
> a load cell.
>
>
> > The roller power will be you vehicle power at the wheels. You can
> > then divide this back through by the engine rpm at any point and
> > get the engine torque.
>
> > These two powers and torques you have are as measured at the rear
> > wheels. Just divide back through by expected driveline efficiency
> > to estimate the power and torque at the flywheel.
> > Did that make any sense?
>
> Absolutley, these last 3 parts I understand very well and would be
> the easy part to have the program do once the data was input, and
if
> I understand correctly you agree that a load cell would be the way
to
> get that data?
>
>
> > Would you like me to give some equations as an example?
>
> That would be great Wes, but more importantly(as I'm not familiar
> with load cells), i'd be curious of what type of load cell I would
use
> (do you have a reccomendation?), and what type of data or reading
it
> puts out(can it be fed into my data-aq system in a 0-5v signal?).
>
> If I could get these figures into my data logger in a 0-5v signal,
I
> can program it to put it out(visually on screen) in any way I want
> via a chart trace or a gauge reading. And my system allows data to
be
> exported to Excel also.. would this be something a program such as
> yours would be able to have inpout to give printouts like a normal
HP
> or TQ versus a MPH or RPM chart as well as an A/F chart via RPM?
>
> Thanks Wes(hope I'm not burning you out or asking stupid questions
on
> this),
> Todd