Forum

Accuracy and Precision of the BAXEDM test machine  

  RSS

mikebax
(@mikebax)
Member Admin
Joined: 1 year ago
Posts: 30
25/11/2018 7:16 pm  

Hi,

I wanted to share a small experiment I did on accuracy and precision of my DIY machine. First, let me refresh your memory and show what the difference is between accuracy and precision. This picture says it all:

 
I've made 9 small workpieces. 3 of Brass, 3 of stainless steel and 3 of Aluminum. All 10mm thick. All 9 pieces were machined with the same G-code, same arc generator voltage setpoint (7 Volts), same EDM Arc on-time (5 micro seconds) and same EDM Arc off-time (20 microseconds). The only difference in setting was the current. I could not use the same current for all 3 types of materials, as that would break the wire. Steel was cut at 22A, Brass at 15A and Aluminum at 11A.
The g-code was programmed to machine a square 0f 10x10mm, without compensating for the wire thickness (and arc length). Here are the results:
As can be seen the precision is remarkably high, the 3 samples lie with +/- 5 micrometer. The accuracy between different materials is not so good without compensation. However as the precision is really high, the difference in accuracy between materials can be compensated for in the CAM software, by choosing a different wire diameter/mill bit diameter.
 
For Brass the wire diameter should be set as 10-9.675=0.325
For Steel the wire diameter should be set as 10-9.765=0.235
For Aluminum the wire diameter should be set to 10-9.683=0.317
 
The thing that amazed me is that for steel, the 0.235 is actually thinner than the real 0.25mm of the brass EDM wire! I could not believe it and triple checked my measurements. I then made a cut and after the thin kerf was made, I tried to put a new uneroded piece of wire in the kerf, which did not fit. This confirmed the fact that the kerf width for stainless steel really is smaller than the diameter of the new wire.
 
 
 
This topic was modified 1 year ago 5 times by mikebax

Quote
mikebax
(@mikebax)
Member Admin
Joined: 1 year ago
Posts: 30
25/11/2018 8:20 pm  

Please also note that the effective CAM wire diameters mentioned above are only valid for the Generator setting mentioned above. If the settings differ, the diameters will be different. It is therefore advised to first machine a small test piece if you are using different settings.


ReplyQuote
tessen
(@tessen)
Active Member
Joined: 1 year ago
Posts: 11
25/11/2018 10:41 pm  

Very interesting test Mike, probably the kerf is smaller than the wire, due to the low speed in the wire, you use the same wire speed in all test?

Could you repeat the steel test with the same parameters but with more wire speed? to see if the kerf increased and to see if the measure change

regards!  😊 


ReplyQuote
mikebax
(@mikebax)
Member Admin
Joined: 1 year ago
Posts: 30
28/11/2018 7:12 am  

Hi,

I also think that increasing the wire speed will indeed affect this. I was cutting all materials at the same speed. I only use the wire once, so if I increase speed, I increase cutting costs. I think the wire speed should be as low as possible without breaking the wire, to get the most value out of it.

At this moment, I do not have a wire speed monitoring system. I used to have it, but it broke and I've got no time to fix it. Way to busy with all the generators  😀  I now simply set the speed manually, and keep it the same for all materials. When I use a high current, I increase it a bit, but I'm not sure by how much. I need to add an encoder idler wheel to measure it.


ReplyQuote
Share: