[dex918]

I still want to make another pendant using PIC microcontroller with just switches and potentiometer... no encoder required... I believe it will allow better positioning control than regular encoder based pendant... still thinking what features to include...
Will update later when it's more crystallized...

Cheers,

Just for thought..

Can we eliminate the use of keyboard and mouse in Mach 3 and just control all by pendant??

I assume it may not be possible as we need to load the files.

As for the buttons, I suggest to add to the existing design:

1. E-Stop
2. Cycle Start
3. Feed stop
4. Home

[dertsap]

Just for thought..

Can we eliminate the use of keyboard and mouse in Mach 3 and just control all by pendant??

it can t be eliminated 100% but you can do a keyboard hack and use momentary buttons for the hotkey functions , inside the keyboard is a small board that all the swithes can be connected to ,simple stuff , just takes patience and time , i built a complete control panel this way , worked great so i have scrapped that panel and am currently working on a better version



this is an old thread which i may update , ive got a far better board setup from the one posted
http://www.cnczone.com/forums/showth...cking+keyboard

[abasir]

Mr Chip,
Checkout my new pendant in progress on this thread...
Decided to do away with the potentiometer plans... not required after testing the design without it... Still thinking on suitable enclosure... ready made enclosure are too thick... most likely will mill mine as before

Keep us in the loop when that latest idea crystallizes. Looking forward to it.

[tivoidethuong]

how are you setup for your pentdan in mach3?

[meo_udon]

Good idea. Very interesting.

[epineh]

I just remembered about another source of encoders, the rotary encoder on a dead inkjet printer is quite a large count depending on the model, they are usually fairly large (2 inch and bigger) and are already mounted on a shaft (the paper feed roller).

I have a few lying around at home from different projects, will try one out one day. Also dead inkjets are practically free these days.

Russell.

[antichip]

So I've got an optical interrupter from something. It looks to be 5v. So I can use this as simple as it sounds in this post? I will be using it as spindle speed, different use same concept.

[abasir]

If you get clean 5V signal then all is okay for your to start experimenting... you may need to provide 5V pull-up resistors for some parallel port...

So I've got an optical interrupter from something. It looks to be 5v. So I can use this as simple as it sounds in this post? I will be using it as spindle speed, different use same concept.

[antichip]

Awesome. I will test it out maybee in a few day or so. I got the interrupt mounted so I hope it will work. Thanks for the input.

[.xXACEXx.]

dang !! so much electronics stuff (spoken from a guy who is clueless about electronics)


i will put the idea out there..it may have been already covered someplace else.i have no idea how to make any of it work just wanted to see if yall thought it would be possable to use an old ball mouse with a wheel and use the rollers inside the mouse to use as encoders for a servo type drive motor ,or the hand wheel (mpg)...you all seem to already have the ideas...just wanted to contribute some ideas would be sweet to make some servo motors using old mouse parts .

also any body look into using a ball mouse to make a sort of digital readout ,again dont have a clue how it could be done but might be an idea,using a rack and pinion type interface (i saw something several years ago on this idea but never don anything about it as i dont have any machines)

all of this may be totally impossable but there ya go..some more ideas to tinker with

[rcpilot82]

any updates on these ideas?? i like the inkjet idea,would like to see it attempted. i would try but my machine is still " in process"

[abasir]

While it is technically possible to do it, the main problem is coupling the flimpsy plastic optical slot inside the mouse to anything else. Another issue that the optical encoder on the PCB has been position precisely for the plastic optical slot. As for me... been there, done that... not worth the trouble :-)

[AlienSteve]

Epson printers (probably other makes, too) use an optical encoder strip for head positioning. It looks like a greyish plastic strip, the stripes on it are very fine. Look just above and behind the print head.

[epineh]

Epson printers (probably other makes, too) use an optical encoder strip for head positioning. It looks like a greyish plastic strip, the stripes on it are very fine. Look just above and behind the print head.

Yup this would work, I have tested this one, but I left all the moving parts on the printer, hooked up a servo drive and powered up the print head enough to read the quadrature signals from the head, just so happens I have a video :

[ame="http://au.youtube.com/watch?v=noyN6BmUSh8"]YouTube - CNC Print Head Jogging[/ame]

Cheers.

Russell.

[abasir]

Nice test there... inkjet works in 'clean' mode... our cnc does not... dust/chip will be flying all over the place... the encoder will have to be protected from it... :-)

[epineh]

Yeah, this was just a test to see how the linear encoder from a printer worked when connected to a cnc servo drive, to use this it would have to be protected from "nasties" for sure, but it proved the concept, and quite well.

I imagined it would be jerky due to the coarse resolution when compared to the real thing but it was very smooth and stable, I didn't calibrate it to work out the exact resolution but I would imagine it would be reasonable.

Cheers.

Russell.

[abasir]

Typical 4-colors inkjet printers have 1440/2880dpi resolution and most have their nozzles staggered. I would expect a 1440 dpi printer to have optical encoder that is 1440/4->360lpi and the 2880 dpi would be about 720lpi... still very good by my cnc standard... (3mils & 1.5mils)... :-)

[rcpilot82]

Well it looks like the theory is good, I'd love to see it put into application.

[pminmo]

Pretty awesome Russell!

[epineh]

Typical 4-colors inkjet printers have 1440/2880dpi resolution and most have their nozzles staggered. I would expect a 1440 dpi printer to have optical encoder that is 1440/4->360lpi and the 2880 dpi would be about 720lpi... still very good by my cnc standard... (3mils & 1.5mils)... :-)

I guess I shouldn't be lazy and actually calibrate it to find out the resolution, should be easy enough as I left the (hacked) wiring on the print head after the test, I did use the servo drive/H Bridge from that on my router but now it is spare again so I can hook it up without much fuss.

A little trial and error and a quick calculation should have it sorted in no time.

Well it looks like the theory is good, I'd love to see it put into application.

I was planning on using three of these to make a small PCB milling/drilling table, I started a thread but as these things go I got distracted and it is on the shelf for now, literally. I need to build my shed to house all of this stuff so that is first on the list...

It just so happens the linear encoder is a little over 300mm in length so fits perfectly for a 300mm by 300mm blank PCB. Of course the one for the Z axis would need to be shortened as one foot is a little too much travel for PCB milling/drilling

Pretty awesome Russell!

Cheers Phil !:cheers:

Russell.