Rocket Thrust Instrumentation

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

JackO

MIT AeroAstro Undergrad -- HSRPL Founder
Joined
Mar 13, 2016
Messages
109
Reaction score
0
Location
Los Angeles
[FONT=&quot]I am recording the thrust of a large solid rocket engine with this load cell: [/FONT]https://www.robotshop.com/en/1000kg-butt ... -cell.html

[FONT=&quot]I believe I will need this amplifier: [/FONT]https://www.sparkfun.com/products/13879

[FONT=&quot]and this micro-controller: [/FONT]https://www.sparkfun.com/products/13261

[FONT=&quot]In order to record the thrust being administered to the cell over the course of a few seconds. Are these products compatible and appropriate? [/FONT]
[FONT=&quot]What other products will I need? What free software can I get to graph this data?[/FONT]

[FONT=&quot]Thanks all,[/FONT]

[FONT=&quot]Jack[/FONT]
 
If you can gain access to the Research Forum there are threads there that describe what you are wanting to do.

1000kg? What size motor are you firing?

Edward
 
As a minor, I cannot join the research forum. Since this topic is not on experimental propulsion but instrumentation, I was hoping I could get the information from this forum. We are first testing an M engine and later a P.

If you can gain access to the Research Forum there are threads there that describe what you are wanting to do.

1000kg? What size motor are you firing?

Edward
 
That is a little load cell for that size motor, even if it is rated at 1000kg. The one I use for that size motors is 6-inches in diameter and 10,000 lb rating on it.

Edward
 
As a minor, I cannot join the forum, but since this topic is not on experimental propulsion but instrumentation, I was hoping I could get the information from this forum. We are first testing an M engine and later a P.


Your mentor doesn't have a test stand? Or is that part of the goal you have in mind?

The OpenScale will get the data for you, it has the HX711 onboard, so you don't need an extra one. You will need a computer or an OpenLog to log the data from the load cell if you use the OpenScale.

Note that if you are willing to write some code, an HX711 and an arduino board can be had on ebay much cheaper.

As for plotting the data, a spreadsheet can do it. Here's one of mine I've been experimenting with. Mine is only pressure right now, but something similar would work for thrust.

https://docs.google.com/spreadsheets/d/1vz2-BbZPnDetbAeja1M68vCV-hLnd3dhpZ1Yy2qc6iE/edit?usp=sharing

OpenOffice can do it, but it's slow at it if you have a lot of data. My ADC is much faster, 4Khz vs 80hz, so there are a lot more data points. https://www.analog.com/en/design-ce...tware/evaluation-boards-kits/EVAL-AD7190.html
 
Alpha, unfortunately our budget would not allow for more than this bargain cell. If the P engine ends up making too much thrust in simulations; we will purchase two more and have the thrust push down on all three cells evenly.
That is a little load cell for that size motor, even if it is rated at 1000kg. The one I use for that size motors is 6-inches in diameter and 10,000 lb rating on it.

Edward
 
Our mentor deals with liquid engines and is very busy with various contracts; he does not have a test stand that would fit our specific needs.
I'm willing to cough up the $30 for a pre-coded setup. Thanks very much for the links!
To confirm, the Open Scale and load cell will be all that we will need to record the thrust?

Your mentor doesn't have a test stand? Or is that part of the goal you have in mind?

The OpenScale will get the data for you, it has the HX711 onboard, so you don't need an extra one. You will need a computer or an OpenLog to log the data from the load cell if you use the OpenScale.

Note that if you are willing to write some code, an HX711 and an arduino board can be had on ebay much cheaper.

As for plotting the data, a spreadsheet can do it. Here's one of mine I've been experimenting with. Mine is only pressure right now, but something similar would work for thrust.

https://docs.google.com/spreadsheets/d/1vz2-BbZPnDetbAeja1M68vCV-hLnd3dhpZ1Yy2qc6iE/edit?usp=sharing

OpenOffice can do it, but it's slow at it if you have a lot of data. My ADC is much faster, 4Khz vs 80hz, so there are a lot more data points. https://www.analog.com/en/design-ce...tware/evaluation-boards-kits/EVAL-AD7190.html
 
Our mentor deals with liquid engines and is very busy with various contracts; he does not have a test stand that would fit our specific needs.
I'm willing to cough up the $30 for a pre-coded setup. Thanks very much for the links!
To confirm, the Open Scale and load cell will be all that we will need to record the thrust?


Those will get the load cell data into a serial data stream. To record it, you need either a computer and some code, or something like the OpenLog, which SF also sells. I believe Openscale includes the needed conversion to convert into pounds/kg as well.

I'm not sure about combining load cells for more capacity... And it seems like it would be cheaper to buy one big one. You are aware that a P motor will use a LOT of chemicals, right? That's big money even if you are making your own.
 
I would use an amplifier that puts out a dc voltage then record with a Dataq they are only 39 bucks I think and usb to your laptop or computer, to record with.
 
Those will get the load cell data into a serial data stream. To record it, you need either a computer and some code, or something like the OpenLog, which SF also sells. I believe Openscale includes the needed conversion to convert into pounds/kg as well.

I'm not sure about combining load cells for more capacity... And it seems like it would be cheaper to buy one big one. You are aware that a P motor will use a LOT of chemicals, right? That's big money even if you are making your own.

Ok great, thanks for the info.

I'm not sure of the multiple cell idea either, it came from one of our mentors, regardless one cell is sufficient for our M-engines.

We are; we will be searching for additional sponsors and utilising grants and fundraising programs after we characterize our fuel. Also, many companies offer student discounts which helps a lot.

-Jack
 
I would use an amplifier that puts out a dc voltage then record with a Dataq they are only 39 bucks I think and usb to your laptop or computer, to record with.

Wouldn't the OpenScale setup work the same way and avoid the additional step? As far as I can tell, it can take the load cell data, amplify it, and convert it so the computer can record it in one product.
 
Wouldn't the OpenScale setup work the same way and avoid the additional step? As far as I can tell, it can take the load cell data, amplify it, and convert it so the computer can record it in one product.

It can, a lot more work. Dataq is cheap and no programming. The other way you have to depend on libraries (or write your own) and writing your own sketch.
 
OpenScale + OpenLog is a complete setup. Dataq has the ADC, but you need to amplify the load cell voltages. Amps have historically been expensive. And you need a computer to read the data, I think. You would need to use one to read the data from OpenLog and plot it etc., but not in the field for logging the burn. Downside to that is you can't know at the time that you got the data you were after. So there is a benefit to having a display on site. OpenScale + a computer is basically the same, but you would need some software to interpret the data. From the description, the OpenScale just spits out a weight periodically. For graphs etc you would have to save the data, then load it into a plotting program, spreadsheet, whatever.

Another option is the Analog Devices eval board I mentioned above. It's $60, the ADC has an internal load cell amp, it's much faster, and the software they supply works well enough to read the data, save it out, and plot it on the screen. There is a downside, it only works with older Windows systems. You need Win7 or below. I use an old netbook on XP. It also doesn't work in a VM, at least not on Virtualbox anyway.
 
It can, a lot more work. Dataq is cheap and no programming. The other way you have to depend on libraries (or write your own) and writing your own sketch.

Aren't there programs available online to do this? If not, we will go with Dataq. We want the simplest setup possible to get us thrust measurements.
 
I second the Dataq method as I have been using one for years. The latest starter kit from Dataq is only 59.00 and load cell amps from China under 20.00 on Ebay are common. So for less then 100.00 you should be all set. Again the advantage is you can record the thrust measurement, plot the graph in real time using any laptop and no programming required.

Steve G
 
There is a learning curve making the first good motor, a second learning curve making consistent good motors, and another learning curve scaling up to larger motors. Does your group have at least one member with proven expertise making consistent solid propellant M or higher motors that match design specs? That is a starting point for moving up towards the P realm (in SMALLER steps); nothing more. Just being able to burn a few M motors without them going CATO doesn't cut it. There is a lot more to learn to make a P. A P has 8 times the energy in twice the size. The larger the motor, the less forgiving of inconsistency, ignorance, and error. If the starting point isn't already met, the learning curve is measured in years IMHO.

I've been through the learning curve to where I'd be comfortable designing and fabricating a P motor, though O is the largest I have done. Ms are my little test motors. To get to where I felt like I had a clue (doesn't mean there aren't some things I should know but don't) has taken some years, a few hundred pounds of AP, lots of static tests and flight tests, hundreds of research papers, private emails and discussions, plus a mentor who already made large motors routinely. I expect any motor of mine to have propellant which is close to theoretical density, reproducibly so, to produce the thrust curve to design, and have a burn duration matching target to 0.1 seconds. I expect the resulting altitude when flown to hit within a couple percent of design, and usually it is better than that.

You CAN build a large motor by overbuilding it and designing conservative. But you still have to get far enough along to know what that means, in specifics.

I'm not trying to dissuade you, but among other things, solids are not liquids. Does your group have someone experienced in solids? The health and safety issues are not quite the same, and there are certainly manufacturing, handling, storage, and transport differences.

Gerald
 
The OpenScale module is probably just fine for what it was designed for, but I would not consider using it for a motor test stand. The HX711 it uses is only capable of 80 sps which is way to slow.
 
There is a learning curve making the first good motor, a second learning curve making consistent good motors, and another learning curve scaling up to larger motors. Does your group have at least one member with proven expertise making consistent solid propellant M or higher motors that match design specs? That is a starting point for moving up towards the P realm (in SMALLER steps); nothing more. Just being able to burn a few M motors without them going CATO doesn't cut it. There is a lot more to learn to make a P. A P has 8 times the energy in twice the size. The larger the motor, the less forgiving of inconsistency, ignorance, and error. If the starting point isn't already met, the learning curve is measured in years IMHO.

I've been through the learning curve to where I'd be comfortable designing and fabricating a P motor, though O is the largest I have done. Ms are my little test motors. To get to where I felt like I had a clue (doesn't mean there aren't some things I should know but don't) has taken some years, a few hundred pounds of AP, lots of static tests and flight tests, hundreds of research papers, private emails and discussions, plus a mentor who already made large motors routinely. I expect any motor of mine to have propellant which is close to theoretical density, reproducibly so, to produce the thrust curve to design, and have a burn duration matching target to 0.1 seconds. I expect the resulting altitude when flown to hit within a couple percent of design, and usually it is better than that.

You CAN build a large motor by overbuilding it and designing conservative. But you still have to get far enough along to know what that means, in specifics.

I'm not trying to dissuade you, but among other things, solids are not liquids. Does your group have someone experienced in solids? The health and safety issues are not quite the same, and there are certainly manufacturing, handling, storage, and transport differences.

Gerald

Second what Gerald said. Make sure you have your safety procedures and processing procedures drawn up and reviewed by qualified people. I think it's awesome you are pursuing this, but we all want you to be safe while doing so. If you want a double check, send us the procedures and we'll look them over. Just make sure you have all your ducks in a row before you start about what steps you are going to perform and how exactly you are going to do them, because doing something stupid like mixing powders dry is highly hazardous at best and fatal at worst.
 
The OpenScale module is probably just fine for what it was designed for, but I would not consider using it for a motor test stand. The HX711 it uses is only capable of 80 sps which is way to slow.

You know, I thought so too, but someone posted a curve generated by one and it's not bad. I think for a larger project and for pressure it's worth going faster though.
 
There is a learning curve making the first good motor, a second learning curve making consistent good motors, and another learning curve scaling up to larger motors. Does your group have at least one member with proven expertise making consistent solid propellant M or higher motors that match design specs? That is a starting point for moving up towards the P realm (in SMALLER steps); nothing more. Just being able to burn a few M motors without them going CATO doesn't cut it. There is a lot more to learn to make a P. A P has 8 times the energy in twice the size. The larger the motor, the less forgiving of inconsistency, ignorance, and error. If the starting point isn't already met, the learning curve is measured in years IMHO.

I've been through the learning curve to where I'd be comfortable designing and fabricating a P motor, though O is the largest I have done. Ms are my little test motors. To get to where I felt like I had a clue (doesn't mean there aren't some things I should know but don't) has taken some years, a few hundred pounds of AP, lots of static tests and flight tests, hundreds of research papers, private emails and discussions, plus a mentor who already made large motors routinely. I expect any motor of mine to have propellant which is close to theoretical density, reproducibly so, to produce the thrust curve to design, and have a burn duration matching target to 0.1 seconds. I expect the resulting altitude when flown to hit within a couple percent of design, and usually it is better than that.

You CAN build a large motor by overbuilding it and designing conservative. But you still have to get far enough along to know what that means, in specifics.

I'm not trying to dissuade you, but among other things, solids are not liquids. Does your group have someone experienced in solids? The health and safety issues are not quite the same, and there are certainly manufacturing, handling, storage, and transport differences.

Gerald

Gerald,

Our primary mentor originally from Rocketdyne has half a decade of experience in all types of propulsion; he is a contractor doing propulsion work for Northrop Grumman, Rocketdyne, Boeing, Virgin Galactic, and SpaceX in its early days. He had the amateur rocketry record back in the 90's with his boosted dart launched to 48 miles. He taught a class in propellant and many of the guys building the SRB's enrolled (routinely got 98%theoretical density in this class without a vacuum). I've consulted with Derek Deville rather extensively on the motor design. We have mentors at Aerojet Rocketdyne, and we are designing conservatively to the extreme. Our test engine has a 1/4" steel casing; the radial bolts for the bulkhead will give out long before the steel. We'll conduct a full static test of the P engine months before launch. All personnel will be in the bunker at the Mojave Test Area during tests and launches.

I of course have not built a motor this big, but I think I have taken the necessary steps to safely and confidently do so.

Regards,
Jack Oswald
 
OpenScale + OpenLog is a complete setup. Dataq has the ADC, but you need to amplify the load cell voltages. Amps have historically been expensive. And you need a computer to read the data, I think. You would need to use one to read the data from OpenLog and plot it etc., but not in the field for logging the burn. Downside to that is you can't know at the time that you got the data you were after. So there is a benefit to having a display on site. OpenScale + a computer is basically the same, but you would need some software to interpret the data. From the description, the OpenScale just spits out a weight periodically. For graphs etc you would have to save the data, then load it into a plotting program, spreadsheet, whatever.

Another option is the Analog Devices eval board I mentioned above. It's $60, the ADC has an internal load cell amp, it's much faster, and the software they supply works well enough to read the data, save it out, and plot it on the screen. There is a downside, it only works with older Windows systems. You need Win7 or below. I use an old netbook on XP. It also doesn't work in a VM, at least not on Virtualbox anyway.

Thanks for the comment. So Dataq +amplifier will allow me to record and log the data without anything else? Or would I still need to get the software for data-logging? Because if that is the case then OpenScale would accomplish the same for cheaper. I am absolutely willing to spend extra but only to lose a step in the process. Thanks again for the advice.

-Jack
 
The OpenScale module is probably just fine for what it was designed for, but I would not consider using it for a motor test stand. The HX711 it uses is only capable of 80 sps which is way to slow.

Ok, that is definitely good to know; maybe Dataq then? This subject is definitely foreign to me.
 
Second what Gerald said. Make sure you have your safety procedures and processing procedures drawn up and reviewed by qualified people. I think it's awesome you are pursuing this, but we all want you to be safe while doing so. If you want a double check, send us the procedures and we'll look them over. Just make sure you have all your ducks in a row before you start about what steps you are going to perform and how exactly you are going to do them, because doing something stupid like mixing powders dry is highly hazardous at best and fatal at worst.

Like I told Gerald, we have very qualified mentors. I'd be happy to send you guys the procedures; not entirely sure how to outside of the Research Forum. We cast the propellant last Saturday; here is the forum thread where I am slowly documenting our progress: https://www.rocketryforum.com/showthread.php?138502-Chaminade-Rocket-Club-High-Power-Rocket
 
You know, I thought so too, but someone posted a curve generated by one and it's not bad. I think for a larger project and for pressure it's worth going faster though.

So what's the consensus? Is the OpenScale sufficient? Or should we use Dataq? I'd like to order the products later today if possible.
 
I wouldn't say there is a consensus.

Dataq + amp is the same as the openscale as far as collecting data. Both look to need a computer. The biggest differences are that the dataq is faster, and comes with software to capture and display the data. The downside is that you need an amplifier as well. The cheaper dataq is lower resolution as well, but that might not matter with the amp.

I've seen more people here use the dataq than HX711 based setups, so perhaps that could be called consensus...
 
I wouldn't say there is a consensus.

Dataq + amp is the same as the openscale as far as collecting data. Both look to need a computer. The biggest differences are that the dataq is faster, and comes with software to capture and display the data. The downside is that you need an amplifier as well. The cheaper dataq is lower resolution as well, but that might not matter with the amp.

I've seen more people here use the dataq than HX711 based setups, so perhaps that could be called consensus...

Ok, thanks for the advice ttabbal and everyone else. Since the Dataq is $60 minimum, the amplifier we would need would be at least another $50, and the $55 load cell + the cost of the physical thrust stand would end up exceeding our budget. 80 BPS will have to do. Does anyone know of a good software program for data-logging? If not, I'll ditch the laptop setup and get an OpenLog
 
the cost of the physical thrust stand would end up exceeding our budget.

What is the cost of the rocket you are going to blow up?

Don't be penny-wise and pound-foolish -- rarely pays in this hobby.
 
That budget does seem kind of low for such a big project. Still believe the Dataq is the way to go. Nothing to build, download or program. Why do you need to build a test stand if you have access to RRR's facilities?

Steve G
 
That budget does seem kind of low for such a big project. Still believe the Dataq is the way to go. Nothing to build, download or program. Why do you need to build a test stand if you have access to RRR's facilities?

Steve G

Should have specified, this would exceed our test engines' budget. For the P engine and then once again for the rocket, we'll be asking for a new grant from the school and doing some fundraising. Our mentor said that RRR did not have a stand that would be suited to our specific purposes.
 
Should have specified, this would exceed our test engines' budget. For the P engine and then once again for the rocket, we'll be asking for a new grant from the school and doing some fundraising. Our mentor said that RRR did not have a stand that would be suited to our specific purposes.


$125 for a data collection system is as inexpensive as you will get. You will spend more cobbling together something non-DataQ based than the $125 cost. I just went through this with a University. They went the sparkfun route then ditched it after they saw how I could show up with at DataQ and get a graph and calibrate in real time.

Edward
 
Back
Top