DIY Sharpness Tester

Calibration:
1. Upload the following code to Arduino from IDE.

Code: 
1.  /* Calibration sketch for HX711 */
2.  #include "HX711.h"  // Library needed to communicate with HX711 https://github.com/bogde/HX711
3. 
4.  #define DOUT  2  // Arduino pin 2 connect to HX711 DOUT
5.  #define CLK  3  //  Arduino pin 3 connect to HX711 CLK
6. 
7.  HX711 scale;  // Init of library
8.
9.  void setup() {
10.   Serial.begin(57600);
11.   scale.begin(DOUT,CLK);
12.   scale.set_scale(1252);  // Start scale with a calibration factor. Adjust the factor.
13.   scale.tare();       // Reset scale to zero
14. }
15.
16. void loop() {
17.   float current_weight=scale.get_units(10);  // get average of 10 scale readings
18.   Serial.println(current_weight);  // Print the result
19.   delay(1);
20. }

2. Open Serial Monitor on Arduino IDE. Numbers start to flow. They should be close to zero at this point.
3. Put a known weight on the scale. If it is off from the actual weight, increase or decrease the calibration factor in the line 12.

Tester_Arduino3.jpg


4. Remove the weight from the scale.
5. Upload the code with a new calibration factor.
6. Repeat 2~5 until the result matches the actual weight.
 
Operation:
1. Upload the following code for data logging. On the line 12, change the calibration factor to your own.

Code: 
1.  #include "HX711.h"  // Library needed to communicate with HX711 https://github.com/bogde/HX711
2. 
3.  #define DOUT  2  // Arduino pin 2 connect to HX711 DOUT
4.  #define CLK  3  //  Arduino pin 3 connect to HX711 CLK
5.  float milli_time;
6.  
7.  HX711 scale;  // Init of library
8. 
9.  void setup() {
10.   Serial.begin(57600);
11.   scale.begin(DOUT,CLK);
12.   scale.set_scale(1252);  // Start scale. Change to your own calibration factor
13.   scale.tare();       // Reset scale to zero
14.   Serial.println("CLEARDATA");
15.   Serial.println("LABEL,Computer Time,Time (Milli Sec.),Gram");
16. }
17.
18. void loop() {
19.   float current_weight=scale.get_units();  // Scale reading at ~10 Hz
20.     milli_time = millis();
21.     Serial.print("DATA,TIME,");
22.     Serial.print(milli_time);
23.     Serial.print(",");
24.     Serial.println(current_weight);
25.     delay(1);
26. }

2. Open PLX-DAQ on Excel. Allow macros to run. There will be a spreadsheet and a dialog, if it works.
3. On PLX-DAQ dialog, change the port number (to the serial port number obtained during the prep) and the baud rate (to 57600).
4. Close Serial Monitor or Serial Plotter on Arduino IDE.
5. Start logging. Data will flow into cells.

Tester_Arduino4.jpg



Currently, PLX-DAQ works only on Windows because it requires a serial port number in integer.
If you use a mac, use a serial terminal program such as Coolterm and output the data into a file.
I run Windows 7 on my mac with VMWare Fusion.

I am happy to answer any questions regarding the device.
 
Craig James said:
Really interesting, thank you for taking the time to post. I’m curious as to the accuracy and repeatability.

thanks
Click to expand...

Thank you, Craig James.
Those are exactly what I am testing first.

PT50A BESS tester reportedly runs at 40 Hz, while mine is at 10 Hz.
This would probably affects the accuracy (the capability of detecting the actual max force).
I will soon get an amplifier running at 80 Hz and plan to replace the slow one with it.
 
Last edited:
NMpops said:
I just use cash register receipts, light, thin and if I can cut thin strips, it's sharp enough.
Click to expand...
Yes sir. Same. It doesn't damage your edge as much as paper and you can look at it to see if you're fraying it too much.
 
Interim report.

I have checked the calibration and accuracy of the tester using known weights.
You can see that the responses are quite accurate in a wide range and to at least one gram.

Tester_Calibration.jpg



Then, I measured force required to cut fishing lines with different thicknesses using Feather Hi-Stainless blade.
The values were quite consistent at each thickness.

Test1.jpg



However, the speed of pushing down the blade affected the result significantly with the 10 Hz sampling rate.

Test2.jpg



I can get very consistent values at each speed.
But I hope that a higher sampling rate would eliminate the difference.
 
I got a 80 Hz AMP/AD converter and tested it.

Test3.jpg



At this sampling rate, the cutting speed does not matter.
I can be careless in how I cut the media, and still get quite consistent numbers.

Now, I can start real testing.


Edit: The sampling rate was limited to 40 Hz in this test because PLX-DAQ did not like the baud rate.
It works at 80 Hz now with a reduced baud rate.
This result suggests that PT-50A tester is also cutting speed insensitive, as it allegedly works at 40 Hz.
 
Last edited:
Miso2, any updates on how different fishing line compares to the BESS test media?
 
miso2 said:
After looking a while for an appropriate scale for this project, I found none.

So I decided to build a DIY scale with a load sensor, AD converter/amp, and Arduino (<$40 total).
I hope this will allow me to stream data to PC at a high enough rate.

But I accidentally ordered a cheap AD board with 10 Hz data rate.
Hopefully, this is fast enough.
If not, I will buy a better one with 80 Hz rate.

Stay tuned!
Click to expand...
I thought the HX711 was 80hz?
 
I support projects like this for fun and learning, but I think it is completely unnecessary for the vast majority of people (~99.99%).

NMpops said:
I just use cash register receipts, light, thin and if I can cut thin strips, it's sharp enough.
Click to expand...
Another reason to always ask for a copy of your receipt. :p

I prefer magazine pages because you can do more testing as you sharpen and they give you more room to do longer test cuts and repeat cuts while you are identifying any problem spots along your edge. Newspaper works well too, but I find magazine pages are best. If you get a lot of junk mailers then you'll have a steady supply of both.
 
Guy McVer said:
I support projects like this for fun and learning, but I think it is completely unnecessary for the vast majority of people (~99.99%).


Another reason to always ask for a copy of your receipt. :p

I prefer magazine pages because you can do more testing as you sharpen and they give you more room to do longer test cuts and repeat cuts while you are identifying any problem spots along your edge. Newspaper works well too, but I find magazine pages are best. If you get a lot of junk mailers then you'll have a steady supply of both.
Click to expand...
The U-Line catalogue is my go to. Very flimsy paper stock and over 850 pages, lasts a looong time. Knives that will slice printer paper very nicely will make a mess of a U-Line page.
 
You must log in or register to reply here.
Back
Top