need to build a system to measure pull.

Thread Starter

joewales44

Joined Oct 8, 2017
218
need to instantaneously measure pull from 0 to 10 ounces and record the amount of pull on graph or have some way to see it.
pull will come and go in as quick as 0.1 second.
anybody seen anything like this or know how to build it?
thanks
 

Reloadron

Joined Jan 15, 2015
7,501
Number of ways to go about it. 0 to 10 ounces is pretty light I would likely start with a 1.0 Lb load cell, maybe an S beam type or a tension / compression type to be used in tension. Inexpensive to expensive depending on your desired uncertainty. You could also look in Grams for a 500 gram version. Here is an example of an inexpensive 500 gram (just a little over 1 Lb). Next I would provide a stable excitation and run the output into an A/D converter. Again depending on what you want to spend and what uncertainty. I would buy something like this if you want inexpensive. You can even go much less costly. The merit to this and similar starter kits is they will take a low input mV signal from a load cell and amplify it. They also come with software which you can setup to chart your data however you wish. The software will record a file you can save and playback. You can also very easily scale the software to the engineering units of your choice, in your case ounces of tension.

Since you mention 0.1 second you really want a load cell and A/D converter with a fast sample rate. Again, depends on what you want to spend on a system. I was fortunate in that I worked for a company with deep pockets. The link I gave you, running a single channel, will give a 2 KHz sample rate or 2,000 samples / second which gives you 200 samples in 0.1 second. I can tell you how to spend $200 to over $2,000 on a project like this depending on what you want as to uncertainty in your measurement plane.

Ron
 

danadak

Joined Mar 10, 2018
4,057
You could certainly do a SAR or DelSig A/D approach, and simple LCD display to capture
the max pull exerted. Additional code could yield other statistics.

Even some filtering could be applied to reject

And UART or USB into a Python program to plot, plenty of libraries out here.

Something like this, attached. Note shows DelSig single ended, can be configed
for differential with gain. DelSig fast enough at 16 bits to oversample the pull
event I would posit (because your system has a measurable amount of inertia in it....?).



Regards. Dana.
 

Attachments

Reloadron

Joined Jan 15, 2015
7,501
need to instantaneously measure pull from 0 to 10 ounces and record the amount of pull on graph or have some way to see it.
pull will come and go in as quick as 0.1 second.
anybody seen anything like this or know how to build it?
thanks
Would this be a one time test or do you plan on measuring many times for something like product testing? Do you plan on building the fixture to hold whatever it is you are looking to test?
I also see a scope as a viable solution. Again, just a matter of what you want and measured to what uncertainty?

Ron
 

Thread Starter

joewales44

Joined Oct 8, 2017
218
long term product testing.
i want a real time display like a scope has.
guess a cheap scope is the way to go?
guess i need a complete strain gauge that converts resistance to voltage?
i'm very electronically challenged so i will need a unit that comes with good instructions.
most of the cells i see are for push load. i must have pull load and any kind of mechanism to convert push to pull will slow reaction and could vary the force.
maybe i just haven't found a pull type?
thanks
 

Reloadron

Joined Jan 15, 2015
7,501
most of the cells i see are for push load. i must have pull load and any kind of mechanism to convert push to pull will slow reaction and could vary the force.
maybe i just haven't found a pull type?
While you can say push and pull you will do better using the terms I mentioned earlier like tension and compression and you want tension. Some do both but tension would be what you want. Also, you may see ounces, pounds, Kg or grams and either is easy enough to convert. I would also suggest as close to a turn key solution as possible since you mentioned being electronically challenged. :) Additionally either a scope or the A/D converters I mentioned will afford a real time display and store the data.

A scope will do just fine. I suggest one allowing you to easily save the data and allow you to download the data to the storage device of your choice. Most scopes allow a snapshot of the trace and many offer a csv output of a signal in case you want to dump your data into a spread sheet like Excel. I also suggest a load cell which includes the complete bridge which most do. You can also buy a load cell amplifier which is more commonly called a signal conditioner. One nice feature of a strain gauge conditioner is most designed for doing what you want to do include your needed excitation out voltage for your load cell, they also allow you to scale your load cell output to your scope. A stable excitation voltage being important.

Ron
 
Last edited:

Thread Starter

joewales44

Joined Oct 8, 2017
218
While you can say push and pull you will do better using the terms I mentioned earlier like tension and compression and you want tension. Some do both but tension would be what you want. Also, you may see ounces, pounds, Kg or grams and either is easy enough to convert. I would also suggest as close to a turn key solution as possible since you mentioned being electronically challenged. :) Additionally either a scope or the A/D converters I mentioned will afford a real time display and store the data.

A scope will do just fine. I suggest one allowing you to easily save the data and allow you to download the data to the storage device of your choice. Most scopes allow a snapshot of the trace and many offer a csv output of a signal in case you want to dump your data into a spread sheet like Excel. I also suggest a load cell which includes the complete bridge which most do. You can also buy a load cell amplifier which is more commonly called a signal conditioner. One nice feature of a strain gauge conditioner is most designed for doing what you want to do include your needed excitation out voltage for your load cell, they also allow you to scale your load cell output to your scope. A stable excitation voltage being important.

Ron
you sound like an expert with strain gauges.
i'm not lazy, just stupid in this area.
grams or ounces will be ok. pounds wouldn't show accurate enough since my range will probably stay below 3 ounces.
please, could you point to a device?
i could look for days and still not know if i'm getting the right one?
thanks
 

Reloadron

Joined Jan 15, 2015
7,501
need to instantaneously measure pull from 0 to 10 ounces and record the amount of pull on graph or have some way to see it.
That's why I focused on 1.0 Lb or 16 ounces or about 500 grams which is about 1.1 Lb.So OK if we drop down to:
grams or ounces will be ok. pounds wouldn't show accurate enough since my range will probably stay below 3 ounces.
please, could you point to a device?
Sure. With 3 ounces in mind I would look for about a 300 grain tension load cell. Load cells off the boat from China are very inexpensive. While I have never used on in a commercial product test application I have used them in basic home applications. The Load Cell Minebea BCL-A series is a popular version of an inexpensive low capacity load cell. A Google will bring up several distributors. A 300 grain load cell has a max capacity of about 10.5822 ounces. When using load cells you really want to keep your max intended value around 75% to 80% of your target when possible.You also need to work using what is available. You also need to consider any specifications you need to meet, for example customer specs which may be demanded in product contracts. As far as load cells go you are really looking at a light capacity.

Your crash course in load cells runs like this. Load cells like anything else have specifications. The heart of a load cell is a strain gauge which Hymie linked to in post #2. I just find it easier to buy a complete load cell bridge rather then try to build one. One of the specifications of a load cell is the Output typically called the Sensitivity. Using the link I posted above they mention Rated Output where we see 1.6 mV/V±0.2 mV/V so what they are saying here is for each volt of excitation applied the load cell will output about 1.6 mV or 0.0016 volt at full scale. So if I apply 10 Volts of excitation to a 300 gram load cell with a 300 gram load I can expect to see an output of about 16 mV. Not very much. This is where I suggested using a signal conditioner designed for use with a load cell. You want something along these lines. The thinking here is it provides excitation voltage to your load cell and amplifies the load cell output to a decent level, like 0 to 10 volts and is scalable. So when my load cell sees 0 tension I can adjust my output for 0 volts and with 300 grams of tension applied I scale my output for 10 volts. So now my 0 to 300 grams becomes a 0 to 10 volts or we could say 0 to 10.5822 ounces becomes 0 to 10 volts or about 0.94498 volts per ounce. Something else which is important is to provide a stable reference voltage to your load cell. A good load cell conditioner amplifier will give you that.

Thus far we have a load cell and a load cell conditioner amplifier. So now it becomes time to determine what we want to do with the data we collect. My initial suggestion was to use a data acquisition solution. My thinking here is since we want data acquisition I like using a data acquisition solution rather than more common test, measurement and diagnostic equipment like a scope. I am not saying a scope is a poor choice, simply that I would not do it that way. You are running a test and acquiring data so how do you want to acquire that data and what do you wish to do with that data once you acquire it? There are dozens of simple to use data acquisition devices out there which will collect the data and allow you to view it in real time as well as save the data. I can dump just about endless files into a PC. I can also scale the software to the engineering units of my choosing. A scope will also work but for me simply didn't give me the features I wanted for most test I was doing. Here is an example of what I am getting at. While yours is a simple single channel affair take not how each of six channels is scaled in engineering units. I can't do that with a scope.

Auto Trans.png

Again, all of this goes back to your location and your budget as well as how accurate you need your data to be.

Ron
 
Top