A friend and I would like to buy or build a rocket test stand to try and estimate the lifting capabilities and burn rates of our rockets. Do you happen to know anyone that would sell things like this or would you be up to the challenge of helping us tackle building one.
Check out this video,
He is using a 1 KG load cell we will need at least a 100kg load cell. As BP 6lbs rockets can be in the 61lb force range and I have no clue where whistle will be at that size, but double that power isn’t outside of the range of possibility.
Here is a cool load cell,
It is a 100 KG load cell with a 0.02% error over full scale. Has a maximum over load of 150% of full scale.
Here is the amplifier from the video.
You can use a lever to scale up or down to the range of the load cell. Should be easy to factor the error. Similar to an engine hoist.
I actually have a load cell (10kg), the electronics, a RasPi, and a working reader program (… I think I might have just piped /dev/spi or whatever to a file …) all plugged in and breadboarded. It totally worked a few months ago when I was working on the project it was for. I could leave it at the space for you to use; I’m not using it at the moment, but I do want to be able to get it back. You would need to make the lever to get it into the force range you’re looking for, and I’ve done nothing to try to smooth electrical noise, but you could just sample at like 100khz or whatever (iirc it’s a simple setting) and moving average it.
I can look at the specific components when I get home if’n you’re interested in making it yourself instead.
@CKraniak you prove one of my favorite qualities of DMS, The ability to ask nearly any obscure question and not only get an answer, but often the guidance of someone who has done it before. Thank you. In the end I definitely want to build my own, I actually want to build about 5 all said and done as well as post my source and full project online to share. As these test stands are pretty rare and valuable to the pyrotechnics community.
So if you have time, I’d love to sit down with you and have you show off your setup and explain how it all works to me. That way I can borrow your setup and try to mirror mine off of it. As I’m a supreme novice of electronics projects with sensors. I’m an LED fanboy at best.
Please send me a pm with some dates that will work for you and share your contact with me. I’ll do the same when I see your message.
As I said in my PM, I am posting the components I used:
- 1 Raspberry Pi 3
- 1 TAL220 10kg load cell
- 1 MCP3301 13 bit ADC, DIP package
- 1 INA126 instrumentation amplifier, DIP package
- 1 resistor to set InAmp gain; I have 34 ohms I believe, should probably measure but I am lazy and you might want to set this yourself anyways
The rest is jumpers, the breadboard itself, and a router and cat5s to talk to the Pi.
The program for reading the ADC is on the Pi itself. It does some bitwise stuff to the input data to get it to a reasonable format and I think some ioctls to set the SPI data rate, and it dumps to stdout iirc; I think this is what I piped to a file. Hard to say exactly without firing it up, but it was about like that.
The main.c, spi.h, and spi.c files are in the following zip file:
load_cell_program.zip (2.8 KB)
It is very half-baked and not done, isn’t commented, etc. I don’t even have a makefile; I was compiling this by typing the GCC commands in directly. That being said, it does work well enough to use to pipe data to a file and get a pretty graph.
I’ll license this as GPL 3 for now. If y’all are looking for a more permissive license, we can talk about it.
You mean like this?
It’s a hybrid rocket test stand. The microcomputer opens the gas valve counting the revolutions, then fires the spark plug, checks for thrust on the load cell, then streams thrust data out the Serial Com Port. After a programmed burn, the microcomputer shuts it all down.Zprobe001.pdf (43.3 KB)
That is an awesome setup. Would you have a chance to meet with me and show it off?
I’ll be at DMS Tues 6:30pm for BabyLock Sewing Class.
The Test Stand works OK, but I need to incorporate more systems to fly. Do you have any knowledge of ROS from Stanford AI Lab?
I’ll try to make it out.