Still working on fine turning and improving my machine. In my travels I came across optical strip encoders that can be retrieved from ink jet printers. You can even order longer strips for around $35. The only flaw I can see is since they are optical encoders dust will play a large factor.
Optical encoders work very well but like you said the dust will be an issue.
You would have to enclose them in some type of booted mechanism making it difficult and probably expensive.
Optical rotary encoders are much easier to enclose. There are some stepper motors that have them built in. They usually are a disc that is fixed to the shaft and a reading mechanism reads the lines or patterns that then is sent to a controller.
Using them is not hard but you will have to create code for your board to use them.
Some info I found for you to look at.
https://www.festo.com/cat/en-us_us/data/doc_engb/PDF/EN/EMMS-ST_EN.PDF
http://www.ebay.com/itm/like/201465548209?lpid=82&chn=ps&ul_noapp=true
There are tons more out there.
what would the end use be on an x-carve?
From my understanding (in theory) higher precision, and accuracy. Really we are about to recycle a bunch of old printers and figured I could gut some of them for encoders.
http://multicam.ca/the-difference-between-an-open-and-closed-loop-system/
Not in English but still worth the watch.
It takes more on the electronics side to use them. They allow for a “feedback loop” where the controller can sense if the moves have been completed before it goes on to the next command - if it fails to complete, it will alarm and stop the job, rather than messing it up because of missed steps.
Again, it needs electronics and software that support closed loop operations.
I upgraded my old MaxNC to optical encoded steppers and it was the biggest mistake.
I was never able to get the machine to run properly after that.
The problem being that sparks from the brushes in the spindle motor would cause single nose on the encoder lines causing a fault and staling the print. (Usually 23 hours into a 30 hour print.)
It took years of grounding, rewiring and adding filter capacitors to get it to even work. And even then it would still fail occasionally.
Also you need a controller capable of utilizing the encoder data. To my knowledge there is currently no cheep “Closed Loop” board out there (as opposed to Open loop which is what we are using.)
Also the increase in speed an accuracy is misleading.
Yes you can run the steppers faster because if they miss a step due to moving too fast the software will compensate. However, we are not running our CNC’s at speeds where this is an issue. So unless you are really planing on trying to do a carve at over 300 inch per min it probably is not an issue. (ink jet printers use much smaller motors and travel at way faster speeds so for them it actually comes up)
All the years when I ran that machine if I lost steps do to something slipping under too much load, it usually trashed the carve so bad that I had no idea if it could have correct for those missed steps.
Also all the 3D printers that everyone is using are running open loop. At much faster speed with much finer detail and have not had too much issue with a open loop format.
But back then I heard “Faster” and was all excited to reduce carve time. (3o hour carves cut to 15 - oh yes!)
But now I have learned so much more about materials, feed speeds, cut depts. and carve strategies. And I have found that these things have a much greater impact on carve time than open loop vs closed loop for this kind of application.
Now if someone comes out with a closed loop board, that has the processing power and properly shielded and filtered inputs, then maybe. But in my option, it is just not worth it.
1 Like