New concept evaluation, booted from two spots already and going for the third.

[Ragewizard]

I think that now I am actually getting somewhere.  Hopefully a couple of last questions.

On the spec sheet, when it refers to input voltage sensitivity, that is the amount of voltage needed from a given voltage to enact a change of measurement.  Is that statement true? 

For instance the USB 6510 model has a minimum input voltage range of -200mv to 200mv, but it also shows an input sensitivity for that range of 4.8uV.  If I do a little math however, at a 400mV range with 16 bit resolution, I calculate a 6.1uV voltage sensitivity if my first statement was indeed true.  Whats up with that?  16 bits means 2^16 right or (65536), at a voltage sensitivity of 4.8uV I would have 83333 step increments of measurement, a number somewhere between 16 bits and 17 bits.  I will have to conclude that my first statement is not true unless I am not using all the variables, 16 bit is actually a rounded off number, or I totally missed the meaning of the spec sheet statement.

Then I also have problems with the Range Accuracy spec, it states that the accuracy is 88uV.  Can the accuracy really be 18.3 times the increment needed to change a voltage measurement?

So in conclusion here goes, does this all mean from the spec sheet that the minimum input voltage range is from -200mV to 200mV with a change in voltage measurement for every 4.8uV increment, and it could possible be giving a false reading up to a value of (88uV / 4.8uV=) 18.3 increments of voltage.

[Ragewizard]

The reason I ask about the information above is that for my load cell, the measurements go from 0 to 50 lbs, and it would put out 3mV/V of measurement, for a total10V of excitation, if I can get a change in measurement for every 4.8uV, that equates to a change for every .008 lbs of difference.  Even at 18.3 increment accuracy range, that would still put me in the 0.15 lb range.  Of course these are only some of the errors associated with voltage input, I'm sure that the load cell has some itself, and probably the output of the 6510 model has some as well.

[Henrik_Volkers]

If you want to spend 1k for the more sophisticated USB DAQ you might give the 6009 a try first... 

You don't really need a TC to read 400F a selected NTC will do the job.  More signal at the cost of linearisation. 

You didn't mention your soldering skills ;-), however a good source for load cells are the $15 electronic kitchen balances, professional stuff with a 0-10V output might eat up >50% of your budget 😮

And while we are at it, why not add a vibration sensor? 'cheap' MEMS would be easy to hook up, but are only low frequence ... (ab)use a piezo disc from a loudspeaker ??   

Sounds like a nice project for a highschool in your region. 

[unclebump]

You need an RC engine dynamometer stand.

http://www.nitrodynesystems.com/

[Ragewizard]

Interesting on the cheap method you have outlined for me.  My soldering skills are limited to pipe fitting, and I would guess that working on a bread board would require a little practice and probably some special equipment.  I will confess that the load cell that I have targeted is already in my possession, and was a gift from work that someone ordered, but got the wrong idea and ordered a 50 lb instead of a 5K lb one, so lucky me.  As far as vibration goes, RC engines are always vibrating even when they are working almost at optimim level, but that is an interesting idea. 

[Ragewizard]

Thanks Unclebump, did you see the price on that Dynostand, it was $5500, and while it looks awesome, It is more suited to the RC car than airplane because it uses a flywheel to measuyre HP and torque.  Of course I'm kind of doing the same thing, except that I would like to fit a propeller to match my engines hp and torque.  Nice idea though, and if it was under $2000, I would give it serious consideration to see if I could modify it.

[unclebump]

Do a google search for RC airplane engine dynamometer. I saw some pics of one with a prop on it. That link was the first thing that came up.

[ColeTrain]

Hello Ragewizard,

I just wanted to clear up a few of your questions that might not have been answered yet.  A background, for anyone unfamiliar, when you refer to the card accuracy, it is just a measure of how close to the actual reading you get.  Ex: If we feed the 6210 a signal of ­­100 mV, this card will read within the range of 99.912-100.088 mV. 

And with the sensitivity, it is a measure of how much a signal must change to register a different value on the card.  Just because the sensitivity of the card is less than the absolute accuracy, does not mean that the card will not track the actual signal.  So, if your signal increases just 5 µV, your card will register the difference and will most likely go up by 5 µV, but the accuracy of the measurement may still be off by as much as 88 µV.  I hope this clears up any confusion you might have.

[Ragewizard]

Alright, I think that I know enough.  Any more and my eyes will look like a slot machine.  So it is time to pull the trigger and I think that I will get the 6210, a load cell,and a thermopcouple, then bust open my hand held optical encoder tachometer to attempt to highjack the signal.  If I can get all of this working, I may try to add some superfluous sensors to the mix to let everyone know at the RC airplane park that I have more money than brains.  Thanks everyone for the advice, and more than likely, I will be asking questions about the software that comes with it in a couple of weeks. 

[ColeTrain]

Good luck with your project.  It sounds like a fun one.