Wow,

I keep rambling on posting with the odd reply and now there's life !

I must look up some more historic myth expresssions. It's entertaining :)

I'll try and answer the couple of questions before posting tonights extremely cold experience in the garage.

Javadog
I've wondered though, how do you think the 80/20 is going to be when it comes time to dial-in your machine? Has it been pretty square so far? How many feet/inches of 8020 have you used thus far?

The cutting list for the profile is :-

Base
Ends 610mm * 2
Sides 484mm * 2
Stands 170mm * 4

Table
Ends 430mm * 2
Sides 394mm * 2

Sorry I'm too young to work in inches, I find them so confusing !
I think I have been extremely lucky to be honest. A friend milled all the ends up for me. The original lengths were cut with a hacksaw and the finish was poor. I think the squareness of the profile ends is quite critical for the overall assembly.

The right angle supports used on the vertical table supports are also doing a grand job of keeping things square. I originally put these on for strength but they are certainly keeping things in shape.

Thanks for the compliment Javadog. I'll keep plodding away. Not a lot else to do in the UK with the current weather......

Bob
How did you get the rods to stay in place on the t-slots? It looks like you threaded the rod ends and used a bolt to go into the t-slot but how can you tighten it once in place? Shouldn't it just fall to the bottom?

Bob, Javadog is correct. You drill a hole opposite the M12 hex head and tighten it like that. If you look at the aluminium profile website they have a good pdf showing this method and other forms of assembling the profile.

http://www.boschrexroth.com/corporate/en/index.jsp

From the tests tonight I am confident that the rails are aligned quite well. The carriage slides freely when I detach it from the ballscrew plate.
I just need to think of a way to ensure the rails do not loosen with time.

Thanks,

Not a lot to show for a cold night in the garage.

I had intended on using a backplate with the sealed ballscrew to offset it from the gantry backplate. The stepper shaft centre will be 45mm from the backplate. To reduce weight I will use two 13mm aluminium plates to do this. One feature of the ballscrew is that it uses capheads that are tapped to seal the ends. This allows me to mount the plates quite easily.

It was so simple yet eluded me..thanks for the answer!

Its looking good!!! I am trying to get a couple THK rails (36") off ebay...if the price is right I will have all the motion parts. Now I am deciding whether to use the t-slotted aluminum like you did or just plain ol' MDF again. I am going to wait and see how yours turns out and decide then. I am no rush with the new machine since I cannot justify to the wife building a third! (chair)

What type of spindle are you going to use on it? Dremel (1/8" collet) or something that can handle 1/4" collet size as well?

Bob

I just need to think of a way to ensure the rails do not loosen with time.


Use that "blue bastard" lock-tight. Don't know what's it's called but a friend of mine whenever he sees a bolt with that blue stuff under the head he says "geeze, not blue bastard lock-tight"... so it must be good!

Alan.

The final assembly ready to be mounted to the gantry backplate.

I may put things on hold until the weather warms up. I'm not sure about Brass monkeys but its damn cold at the moment and I'm at risk of losing something ;)

Cheers

I got the first version of PICServo working last night, I've got the sore fingers to prove it! (if you don't get it, pushing around the spindle of a 31V servo motor while testing a PID loop with your thumb and forefinger is PAINFUL! :)). It all works suprisingly well, I've just gotta touch up the interface for the LCD and panel buttons and I can finally start designing the PCB! Yay!

Alan.

Alan,

Sounds like you have it cracked. I mentioned the picservo design to a work colleague who knows quite a bit on machine automation. He said to find out what motor spec I would need and he may have a spare one kicking around. I don't have any plans to use a servo motor but my primary interest is in electronics and I cannot resist building stuff :)

Do you have a guidline to what picservo requires from a motor spec ?

Here we go again..... :)

Use that "blue bastard" lock-tight. Don't know what's it's called but a friend of mine whenever he sees a bolt with that blue stuff under the head he says "geeze, not blue bastard lock-tight"... so it must be good!

Alan.

I think the blue one is a 'nut-lock' (which is what I'm gonna get if I spend too many more late nights in the garage...did we mention it's a bit cold over here?)
I think the green one is even stonger....as in.."Streewth!! Some mongrels bunged green loctite on me nuts! what a baaarsted!"
:D (don't hit me! I'm an aussie too!)

I think the green one is even stonger....as in.."Streewth!! Some mongrels bunged green loctite on me nuts! what a baaarsted!"
:D (don't hit me! I'm an aussie too!)

By crikey that's funnier than a roo crossing the frog and toad joke! :D

:stickpoke

Never seen the "green" loktite. I wonder if there is a yellow loktite, or that if you see yellow around a nut you should avoid it like yellow snow.. heheh

Alan

After a bit of reading, my understanding is...
Blue loctite generally comes off with a bit of force for parts that may need to be separated/adjusted periodically.
Red loctite may need a bit of heat to loosen and used when you rarely need to separate parts (sounds like good bet for guide rails).
Green loctite for bits you'd really rather not fly off...ever!

My Loctite catalog lists 17 threadlocker products in the following colors:
blue, purple, red, red-orange, and green.

The purple ones are low strength for small parts and can be separated using hand tools.

The blues greens are medium strength and can be separated with hand tools.

The reds are high strength requires the application of heat for 5 minutes at 450F. Disassemble while hot.

And the red-orange is high strength, high temperature. Use heat to disassemble.

BTW, loctite also has thread sealant products. They are NOT the same as thread lockers.

Ken

Well I certainly learned about loctite !

Thanks for the info. Hopefully if all goes well. I may use this as a locking solution on some components.

Actually I applied some to my 'el cheapo' pillar drill chuck tonight. It had a habit of falling off the tapered drive shaft. It was the green stuff.
Hopefully it has done the trick. That will teach me to buy cheap stuff.....

Still Monkey weather up North in the U.K so staying inside until things warm up. I did brave it in the garage for an hour or so. The gantry vertical supports needed to be made. Great fun drilling and then tapping 16 M6 holes.

So probably the most boring pics so far but hey it keeps me out of trouble.

Actually so far I'm pleased to say that the construction has followed the CAD design in detail.

Thanks.

Well,

I am fast approaching the Z-axis and assembly and had to rely on ebay for a leadscrew.

As I don't know much about them I hope this one is suitable :-

Anti-Backlash Leadscrew

Has Split-Nut with a spring that takes out the backlash. Probably very good for a Z-Axis on a small Mill/Router

Overall Length: 335mm
Working Stoke: 140mm
Nominal Diameter 16mm
Shaft Coupling: 10mm
Shaft Shoulder: 12mm
Pitch: 2.5mm / 10 tpi

Will be posting an update later tonight hopefully of the gantry assembly. Dare I tempt fate and even say the X-Axis may get a test.......

Thanks

A question for the guru's....

I have not put much thought into the PSU yet as I am using a 12V 20A supply for testing.

I have set the Picstep resistors to limit to 3A which I belive is max rating.
The motor specs are :-

Rated Voltage - 2.5V
Rated Current - 4.5A
Resistance - 0.56 Ohm
Detent Torque - 100 mNm
Holding Torque - 2200 mNm
Step Angle - 1.8 Deg
No. of wires - 8

What would be the ideal psu spec to match these motor's ?

Thanks for any advice on this.

I would use a PS rated at no more than 34vdc (remember my firework display?) and 18amps. You would be better of building your own using a toridal (sp?) transformer and large cap. Trying to find a PS with those ratings already built would cost you a bundle and hard to locate.

There is a guy on Ebay selling the transformers at a great price.

Bob

Thanks Bob,

Will set the ball rolling on the PSU. For now the 12V supply seems to be working well on the testing.

A change in weather here so back in the garage doing a bit catch up.

I'll do the next ramble....

First thing to do was to mount the sealed ballscrew to the gantry backplate.
This required four holes to be drilled for the endblocks.

With the ballscrew mounted to the backplate this was then re-fitted to the gantry verticals.

I had a sold coupling made to attach to the stepper. This is 50mm long.

I drilled a 12mm hole for the stepper shaft to pass through. With this attached to the coupling I marked the right side vertical with the 4 corner fixings for the stepper and drilled and tapped M6.

And finally.

I could not resist testing the assembly so far.

I hooked up the picstep controller and Kcam to manually move the two axis.

Both are running fine and feel quite smooth.

A bit play with the step settings and on the picstep boards and really impressed.

Cheers

End of ramble.

:cheers:

As I'm getting closer to seeing life from this I sense I'll be asking more and more questions.

I have a linear rail. Details posted

http://www.cnczone.com/forums/showpost.php?p=59417&postcount=169

I didn't plan on using this but now I am going to run this parallel to the sealed X-Axis screw assembly. (There is slight 'rocking' on the carriage on this assembly so the idea is to run the linear rail parallel above the ballscrew.

It means removing 20mm from the length of the rail so that it fits inside the gantry arms. Any tips on how to cut the rail ?

Also I'm not 100% sure how many TPI that screw is. Is there a good way to find out. (I guess its a matter of turning it and measuring movement ???)

Thanks !

Hope this helps:

About Precision Acme Threaded Rods
Acme Size— The outer diameter of the rod (measured from the crest of one thread to the crest on the other side of the rod) and the number of threads per inch. For example, a rod with an Acme size of 1/2"-8 has a diameter of 1/2" with 8 threads per inch.
Turns per Inch— For single-start rods, turns per inch equals the number of threads per inch. For multiple-start rods, turns per inch equals the number of threads per inch divided by the number of starts on the rod. To determine threads per inch, place a rule on your rod and count the number of threads in one inch. Starts— The number of independent threads on a rod. Single- start rods are most common; multiple-start rods allow the nut to travel faster by moving farther in one turn.
To determine the number of starts, look at the end of your threaded rod (see diagram). The end of a single-start rod is an offset circle. The end of a two-start rod is shaped like a football. For more than two starts, count the number of corners on the end of the rod.

Thanks Bob, but the problem is the unit is sealed. I don't have access to the thread unless I strip it down. I dont fancy that as I can see it's quite complex !

?

Thanks Bob, but the problem is the unit is sealed. I don't have access to the thread unless I strip it down. I dont fancy that as I can see it's quite complex !

?

Oh I see...well then I guess your going to have to do it the old fashioned way and count the revs as you move it 1 inch. Shouldn't be that hard since I have never seen a leadscrew that was not a whole number for the TPI measurment.

Looking great buddy! But I told you to take it easy! ;)

...
It means removing 20mm from the length of the rail so that it fits inside the gantry arms. Any tips on how to cut the rail ?
Thanks !

Why not sit the rail on top of the gantry...sort of along the top of the back plate? No need to cut it then. Just a thought.

Angle grinder with an abrasive disc can cut hardened steel.

After a night of extremely lively kids I'm about to escape in the garage !
Just switched the useless heater on in there. Might raise the temperature above freezing.....

Thanks for the tip Kong. I took the rail to work as I knew I had nothing that would touch it at home. I found we had a chop saw that had a metal cutting / slitting blade fitted.

So a few minutes setup and the rail cut cleanly.

10bulls - I nearly went that route but hopefully when I take the next set of pics it will show how I planned fitting it.

Right. Coffee in hand I'm off to butcher the gantry :)

Had a decent night tonight. A friend called round to see what I was upto. It's amazing how another set of hands can help out. The gantry is a monster weight now.

I learnt yesterday that the single ballscrew on the X-Axis was not good enough on it's own and it needed another support to prevent the Z-axis plate rocking. I mounted a scrap piece of MDF and soon found that a small bit of pressure caused rocking.....

I found that the linear rail I had also included a special 'T' type rail that would simply slide into my aluminium profile. Lucky !

So using the last piece of profile I had this was cut to 584mm which is the inside measurement of the gantry verticals.

I tapped the ends M12 and also used 3off T pieces to fix the profile to the backplate.

The linear rail is fixed securely to the profile. I was well chuffed with that T-Support rail. (Or whatever it's called)

It's a shame the profile is about 5mm higher than the gantry plates but I have some endcaps and a cunning plan to blend that in at a later date.

With the Profile fixed in place I now had another two carriages above the sealed ballscrew mounts.

I also changed the solid coupling from the stepper to the ballscrew.

Long story but basically the ballscrew shaft was slightly bent. The couping I have now used is a butch one but has a rubber middle that allows for the slight bend on the shaft.

And finally.

The temptation was just too much :)

I fixed the scrap MDF Z-Axis dummy plate to the new carriages on the linear rail also. This has done the business. It feels 100% better. Well happy....

I spotted the kids blackboard in the garage. Using this as a make doo table I cobbled up a pen holder and loaded the only G-code file I had.

Thanks to whoever posted it !

After trial and error with a demo of Mach2. I set it away. Two axis running.
(I have a lot to learn on this)

True music to my ears. Alan those Picstep drivers are the business.

So another break while I wait for the Z-axis screw to arrive and I make the Z-axis plates.

(A special thanks to MrBean on his tutorials with Mach2. I owe you a pint mate).

I hope my machine comes together half as nice as yours. :cheers:

Your X-Axis should be pretty damn solid now! I wonder if you can buy that solid T-Slot-Support-Thingee anywhere??

And finally.

The temptation was just too much :)

Ain't it always the way.. :)


I fixed the scrap MDF Z-Axis dummy plate to the new carriages on the linear rail also. This has done the business. It feels 100% better. Well happy....


Could you have gotten away with just one linear carriage? I guess if you've got no use for the extra one that's cool. Well done looking good!


True music to my ears. Alan those Picstep drivers are the business.


Why thank you! I'm glad they're working well for you. The resolution in your improv BIC plotter is impressive already, be interesting to see the final result when it's all done. Do you have any backlash or movement in all your axis now?

Alan

And finally.

The temptation was just too much :)

I think that gets most of us.


I fixed the scrap MDF Z-Axis dummy plate to the new carriages on the linear rail also. This has done the business. It feels 100% better. Well happy....

As you didn't plan for this 'mod', It doesn't half fit in there well.


True music to my ears. Alan those Picstep drivers are the business.

I second that comment. Although it's hard to be exact, as all 3 of my Axis' use different drive components and motors. I would say my machines "rapid speed" is at least 20 times faster on average, than with the old single step drivers. Well pleased. Thanks Alan.


(A special thanks to MrBean on his tutorials with Mach2. I owe you a pint mate).

No problem. Actually I enjoyed it. Was strange hearing it whirring away and not being able to see it though.

Excellent work my friend... :cheers:

Regards Terry

BTW. That stuff going on at my place........Ended in a trip to the hospital. They're both still there, so I'm on my own for a bit....... :violin:

Well its that ramble time again....

Almost a week has passed and a few more bits have been made for the Z-Axis. The pen plotter has been a good incentive but the novelty soon passes and I want to be able to lift it !

My attempt to beat 10bulls 3am workout failed miserably. I lasted until 11pm and called it a night.

So, the plates are made from 13mm aluminium.

A back plate, Frontplate. Stepper support plate and leadscrew mount plate.

The last item I was unable to find on my skip-rat searches has turned up.

It was purchased of ebay and I'm quite pleased with it. It's not brand new but has seen no wear by the looks of it.

Well it's time to see how close the CAD model is as I mount the backplate to the X-axis assembly.

There are 4*M6 capheads that fit to the Origa Sealed assembly and above this 8*M5 capheads that fit into the linear rail carriages.

Thankfully they all line up !

The front linear rails are then mounted. I only fix the ends tonight as I know they will have to come off before the final assembly. I can see that the remaining holes all line up too.

The stepper support plate is then mounted to the top of the backplate using 4*M6 capheads. These are counterbored into the plate as the stepper motor will fix over these and would foul.

With the plate in position it feels really solid. I will probably add some angle supports at the side in the future.

Next the stepper is bolted onto this plate with 4*M6 capheads. The plate is tapped to accept these.

Wake up at the back it's nearly over :)

Ok with the Backplate assembly fixed It's time to check the frontplate.

The first part to fix in position is the leadscrew support.

And finally I call it a night.

The leadscrew fixes to this plate with 3*M5 capheads.

Feels quite solid but I'm hungry and cold.

End of another ramble......

Good looking work, this just keeps getting nicer. Thumbs up! :cool:

Good looking work, this just keeps getting nicer. Thumbs up! :cool:

:) Thanks.

Well I thought it was going too well. I'm only an amateur at this and my Z-Axis design has a serious flaw !

Basically the end of the z-leadscrew had no support. Therefore the whole weight of the frontplate and clamp assy is pulling down on the leadscrew - coupling - stepper shaft (chair)

So a change of plan. I'm going to get a couple of blocks made with bearings in. One will fit to the top and one at the bottom of the backplate. This should add the required support.

On a lighter note I bit the bullet and purchased Mach2. Quite an impressive program, I just need to learn it all.
I also fitted a couple of limit switches on the Y-axis. A feature of Mach2 I like is the ability to have two limits feed into one parallel port input. So no more crunching as It moves into the extrusion :)

Cheers

I have been following this thread for some time. Is there anyone here (USA) that can program three PIC16F628 that I have?
Thanks, Larry

This is looking fantastic! Looking at how well your's is coming together, I'm starting to wish I'd saved the money on my lathe and spent it on THK, ball screws and T-slot Ali profile (especially after a few v.late nights dodging hot metal chips!).

A few questions...

Do you have a counterbore bit...I've been looking around (not too hard) for one, but haven't come across one.

I'm impressed at the accuracy of your Z axis layout...was the drilling CNC aided or are you just the dog's dangleys behind a drill press!

And lastly...was the lead screw you bought from the UK? I gave up looking on UK Ebay for CNC bits and bobs but next time...

Once again, well done on getting your first moves! Most encouraging! :D

I have been following this thread for some time. Is there anyone here (USA) that can program three PIC16F628 that I have?
Thanks, Larry

I can do it for you. Do you have the 628's? Send me a PM with what you need exactly.

Bob

10bulls I have to confess I bribed my mate the miller to do the Z-Axis plates :) I'm sure the skills you are learning are paying off. I want a lathe !

I'm sure he got the easy bit to do :)

To answer your questions..

I don't have a counterbore bit but I do have a rather blunt countersink bit ;)

Z accuracy - Done on a mill - Bribed multiple coffee's

Leadscrew - Purchased off ebay UK. I was lucky. It appeared as a buy now so I pounced on it. There was also one that was for auction. This went for more than double what I paid :)

The strange thing is the length of the leadscrew is exactly what I needed. I guess it was destiny.....


@Woodie1 - It looks like Bob's your man for the pic programming. Good stuff Bob. Are you about to start building the picstep boards by any chance ?
You won't be dissapointed. The more I mess around with Mach2 the more impressed how smooth the stepper control is.

(On the programming requirement if there's anyone in the UK needing this just drop me a PM and I'll help out)

Cheers.

Anyone have information on making breakout boards?
Thanks

Woodie,

Here's one I built.

http://www.embeddedtronics.com/

Mr chips,

Did you build the opto-breakout or the stepper breakout ?

I'm keen to build an opto-breakout. Alan's picstep design includes an interface that works fine. I'm just so clumsy I would like to build an opto-interface.

Cheers

I bult the stepper breakout, but in hindsight I think the opto-breakout would be better because of the isolation it affords.

Thanks. Its not my top priority but certainly an opto-isolated board is on the to do list....

Well just had one of those nights when nothing seems to go easy....

My pigs-ear Z-axis design needed to be sorted and so after
yet another bribe of my mate the miller I was equiped with two bearing blocks.
This bribe consisted in extending three sets of his outdoor Christmas lights....Oh what fun.

My leadscrew needed surgery. I was quite nervous handing this over to him but all is well. The end of this has now been turned down to 10mm to accept a bearing 26mm diameter, 10mm bore and 8mm width.

The same type of bearing slides on from the drive end to a flat face.

The two blocks are made from 13mm aluminium like the rest of the assembly.

These are tapped 6mm to fix to the backplate.

And finally after numerous cock-ups / lost tools / broken coffee mug I end up with two blocks supporting the leadscrew assembly. My carefully drilled 6mm holes on the backplate have ended up slightly oval 7mm holes as the alignment was just not right.

All seems well now but the coupling shown has really bad run-out. It works but it's for the bin. Time to get a new one.

Not the best of nights... but at least I'll get to bed before 10bulls :)

(I noticed there is quite a lot of image distortion with these pics. They are really not that warped!)

Well it's starting to look like a "Silk Purse". Good job

Looking good! Are these some sort of angular contact bearings? Are they pre loaded each end to take out play? Have you got them in the right way around? Did you try bashing them in with your coffee mug? Sometimes the 'Where's that damn tool gone!!!' frenzy can lead to irrational outbursts. I feel your pain!

I'm afraid the bearing's are from my junk box...

I guess Angular contact one's should be used but I'll see how these perform. There does not appear to be any noticable play on them but time will tell.

The coffee mug breakage was a result of searching for those damn missing tools :)

A new week..... A new ramble...... :violin:

The Z-Axis has taken longer than I expected. Not had a lot of time this week but hopefully sorted in the end.

In summary the leadscrew from ebay also included the nut assembly. The diameter of the mounting plate for the nut is approx 55mm so this has dictated the distance from my front plate and backplate.

It's no real issue. I just had to make 4 spacer blocks to fit onto the linear slide carriages. 8 caphead M5 bolts fix through the frontplate, through the 'standoff' blocks and then into the carriages.

Another 3 * M6 bolts also fix from the frontplate to the ballscrew nut plate.

(Images look distorted but it's not really!)

I am really pleased with this assembly. It run's really free and is a lot more sturdy than expected.

It's almost silent too....

My mate the miller deserves a pint I reckon.

The kids are not too impressed at the fact I hijacked there blackboard so I had to make a table.

At the moment I have made a temporary one as I'm sure to drill through it on first use !

Its made from 18mm MDF and fixes to the profile with 4 * M12 bolts.

The aluminium is tapped at the ends to accept these.

I have given up trying to take pictures in the garage and hide the clutter in the background.

So as the main construction comes to an end the attached images show how it's progressed.

Not a million miles away from the CAD design. I need a better clamp. Thats something we had kicking around and will do for the moment.

I have also mounted a hard rubber strip under the MDF that sits on the aluminium. I'm getting some resonance at certain feed rates and hope to eliminate this.
A quick lick of paint has been done too.

I have purchased 4 small diecast boxes.

I have some 8 wire heavy multicore that will feed into the box. I was also hoping to feed the step / direction outputs from the picstep boards into these diecast boxes.
Two led's will show activity.

The boxes mount on the extrusion via the T-piece method.

I'm now thinking as I type this and hope it's ok to run those powerleads and led signals together.... Hmmm

Time to focus a bit on the wiring and switches now.

Update over....

Man. That looks great. I can't wait 'till you cut your first part.

Maybe you could try fitting some rubber mounts or something between the steppers and frame to help with the vibrations. I'd guess the vibration is from the motors and not the mechanicals.

Anyways. Nice job. It's really looking the part now. Time to get started on learning the CAM side.
When are you starting the next one. :D

Regards Terry.....

Not sure if you have room, but you might want to try this. http://www.industrialhobbies.com/howto/tips/hockeypuck.htm

Well Done!
You can't even say 'All done apart from limit switches and a few adjustments' as I see they're all in place too! Impressed! :cheers:
Watcha gonna make first?

Not sure if you have room, but you might want to try this. http://www.industrialhobbies.com/howto/tips/hockeypuck.htm


Thanks for that link. I was thinking of adding something like that near the end. At the moment I have some rubber matting on a workmate. To be honest it's not very stable. I do have a problem however, I now cannot physically lift my machine. Maybe I got carried away with the 13mm plate !
I should be thankful, the original design had 20mm plate in some areas :)

I have some 'hard' rubber sheet that I might try mounting on the stepper front face before it makes contact with the plates. At the moment its a direct metal-metal contact. (Thanks for the tip Terry)

It's got me a bit baffled, I'm not sure where the source of the noise is. In mach2 I have the motors tuned and manually moving all three axis are almost silent, really smooth. (Even running all 3 at the same time). Running a gcode file also runs 90% smooth but when the feedrate drops to say 30% there is a resonance. I may actually capture this sound if I get desperate and put another post for advice.

The frame is very stiff, I have even use extra right angle supports to stiffen things.

Early days but if anyone has tips for reducing noise / resonance / vibration please shout up..

My first cut will be the woman in MDF. So much to learn - not enough time....

Thanks

It may not be an isolation issue as much as a damping issue. The steppers will always produce some vibration although microstepping will help reduce it.

If the vibration happens to match the resonant frequency of part of the machine it will be intensified. Find the part that is resonating and add some mass to it to change it's resonant frequency.

Same idea is the weights they hang on long transmission lines to prevent winds from exciting the lines at a resonant frequency.

Pete C.

Came out great, really clean looking machine. I can't wait till I get to where you are!!

Can't wait to see the first cut!

Appreciate the details WP1, thanks.

Javadog, I made my first cut tonight :banana:

But ......

It was a 6mm cutter straight through the MDF table and a kind of L shape :withstupi

I suspect that table is more 'temporary' than expected !
(Has anyone written an idiots guide to cnc ???)

I did however make a giant leap in noise reduction. I basically removed the three stepper motors and mounted them onto some rubber matting about 6mm.

Overall a huge improvement. (I am using microstepping). The table top now seems to be the point of focus. It's only at really slow feed rates but worth eliminating.

Cheers.

If you take enough cuts out of the table the resonant frequency should change :rolleyes:

Good to see that isolating the steppers using rubber mat, helped with the noise reduction. Don't have any ideas about the table tho'. Like you said before. The table makes like an enclosed (mostly) chamber. Could be that it's making it sound worse. Maybe add some curtains and soft furnishings under the table, to soak up the noise. It works in empty rooms in the house. :idea:

Regards Terry.....

I did however make a giant leap in noise reduction. I basically removed the three stepper motors and mounted them onto some rubber matting about 6mm.

Overall a huge improvement. (I am using microstepping). The table top now seems to be the point of focus. It's only at really slow feed rates but worth eliminating.


Perhaps screw and glue another piece of board under the table that fits snuggly inside the frame of your base (hopefully not binding on your screw). This should dampen some of the frame resonance and should thicken the top board enough to reduce it's ability to act as a sound board. Otherwise use a thicker sheet of MDF or some such and recess it into the frame. :)

Cheers,
Alan

Can you guys hear your steppers ? all I can hear is the router !!!

After a wild night at McDonalds with the kids I managed to bribe the wife into letting me go back into the garage for 'Just another 10 minutes....." :)

Xairflyer I think your right about the noise. Maybe I was expecting too much. Silence !... As soon as the tool is running thats all you can hear.

Alan, I think your correct about the table I need to thicken it to dampen it. I know this table won't last too long so will address that soon.

I did make a couple of significant improvements to the smoothness tonight however. The main Y-Axis leadscrew was the only drive component directly coupled to the stepper shaft. It was a solid aluminium coupling.
I changed this to use the same coupling as used on the Z-axis which has a rubber 'spider' to isolate the shafts.
It's only a small range of small jog rates with MachII that caused any vibration. This coupling has improved things.

The biggest improvement I did was something I should have done prior to the Z-axis assembly. Basically my system is built from salvaged and scrounged items and the two linear rails on the Z were not as smooth as expected. The rail alignement is spot on so I dropped the frontplate off and had a look at the rail carriages. Basically the bearing race inside these were all bone dry.
I dont have a method to grease them via the nipples so I lubed them as best as I could. The difference is incredible. It was something I should have done in the first place.... :)

With everything back together I was ready to cut but it was getting late and didn't want to create too much noise. So.........

Cutting pictures are hopefully any day now.

The tool is ready for action....

I found a rotary encoder in my 'that will come in handy one day' junkbox.
I was going to post on the Artsoft software section about it's use but will ask here first.

From the label details shown I was thinking maybe this could be used as a MPG control to feed into Mach II.

Am I on the right track or way off ?

Thanks.

this Thread is the best on the site , and the information in the pic's is worth it's weight in gold , should become an FAQ in it's own right !

Great work keep it coming

:o

Well thank you da21... I'm just a rambling addict.....

It's been an interesting trip and learned quite a lot. Thanks to all that have contributed and helped me get this far....

So with a fresh brew back in the garage today. I wanted a woman :)

A bit of an anticlimax.... I used a larger cutter than I should so the end result was that the cutout was more interesting than the actual cut piece.
Not to be put down I wanted something 3D.

I'll post a few pics as I know as a newbie on here I really didn't know what to expect to see when something 3D was being cut or how it was done.

Armed with a file called face from the downloads section I was ready for action.

I used a round ended cutter and have since learned the gcode was not for this but it cut out in the end.

1st step is to run the roughing out file.

It looks minging during the cutting and you really need a vacuum cleaner in there....

At this point I should load the finishing file and goto zero for the xyz axis.

Like a total dork I clicked the wrong button in MachII. I clicked Ref Home.
Therefore I lost all reference to the start point for my finishing cut.

I decided to do another cut and worry about the original cutout later.

So. another file loaded and this time it actually started to take shape.

Rather than waste the original cutout I guessed where the home positon was and set the finishing file over it. Not great but it did the job.

And finally after a quick lick of the only wood stain I had. And it's a horrible colour, I ended up with my first real output.

I need to focus on the wiring and tidying up now. It's a bit of a rats nest !

End of ramble...

Wow!

Very cool! I bow to your skills.

I am ordering the parts for the PicStep today! 1 question on the interface board WHERE is a parts list?

Would be interested in that Gcode for the 'chick' looks like a nice simple thing to cut out as a test.

@Randyf1965 - I dont think you will be dissapointed with the picstep. Seems like a great job from Alan. I will dig the list out for the interface, I cannot remember a parts list but it's basically pullups and a few caps and regulator.

xairflyer - I'm moving on to your guiness logo next. The 'chick' should be attached.
(It's posted on the forum somewhere but cannot find it.)

Cheers

And finally after a quick lick of the only wood stain I had. And it's a horrible colour, I ended up with my first real output.

I need to focus on the wiring and tidying up now. It's a bit of a rats nest !

End of ramble...

Heheh well done. Not a bad result for the ruffest bit of timber you could find! :) Was it a brace from a pallet or a piece of fence?

The 'chick' cutout is quite impressive (as is the face), be interesting to see what results you get with a finer cutting bit and a bit more practise. Looks like PICSteps are working well for you, which I'm very glad!

Got Video? :idea:

Alan.

@Randyf1965 - I dont think you will be dissapointed with the picstep. Seems like a great job from Alan. I will dig the list out for the interface, I cannot remember a parts list but it's basically pullups and a few caps and regulator.

Gee this is like the 14th request for a parts list for it. It's only like a few caps and a regulator! I guess I'll have to write a little text file for it then.

I had been unconsciously using the interface as a sort of "skill" test to limit the amount of email I receive asking question like "which end of the soldering iron do I use?". If you can figure out the schematic and the board layout on you own then ye shall pass. :)

But it's becoming more popular so I guess I should write them down somewhere.

Alan

[QUOTE=Garfield2]Gee this is like the 14th request for a parts list for it. It's only like a few caps and a regulator! I guess I'll have to write a little text file for it then.


Well I found my answer in a thread (maybe this one!)

The schematic shows the cap values but not the resistors (10K correct?)

The 6 pin locking header is used for output's?

Heheh well done. Not a bad result for the ruffest bit of timber you could find! :) Was it a brace from a pallet or a piece of fence?

The 'chick' cutout is quite impressive (as is the face), be interesting to see what results you get with a finer cutting bit and a bit more practise. Looks like PICSteps are working well for you, which I'm very glad!

Got Video? :idea:

Alan.

Yep, it was the roughest bit of wood you can imagine ;)
Picstep working fine for me. I'm running @34V and 3A so heatsinks are pretty warm.

Did a small video but its not too good. Stitched to make a usuable one.

About 3Mb.

Uploaded to mates webspace.

Face Video (http://www.mikeep.btinternet.co.uk/face.wmv)

I can set the feedrate higher on the machine but the code was set to 20 IPM so just let it tick away...

Cheers

Yep, it was the roughest bit of wood you can imagine ;)
Picstep working fine for me. I'm running @34V and 3A so heatsinks are pretty warm.

Did a small video but its not too good. Stitched to make a usuable one.


Awesome mate! Very well done, very impressive. Be good to see it ripping along at a higher feed rate but still an impressive start none the less!

Man I gotta find some time to assemble my monster. Got all the expensive bits together, just gotta find time between work, PICServo, housework and my wife to get onto it.

Cheers,
Alan.

I have just engraved your chick picture onto a piece of lexan came out well and done it fairly fast. set the depth of the cut to .025ins.
Don't know why I can't get other codes to run as well as this one, need to study the code.

I had to remove the G40 command as turbocnc did'nt know what it was !

Any pictures xairflyer ?

Thanks,

Tried to upload it last night but had a problem with attachments, forgot I had added a pop up blocker couple of days ago.

Used a 3mm (.125) end mill, need to experimant with router speeds and cutting speeds as I am getting a bit of a burr on the edges.

Dot in bottom corner is not meant to be there - mistake !!!

Nice piece there Xairflyer. It's a good piece of code to do some quick tests with.
Only tried wood so far but will be interested how it performs on other things.
Cheers

Ideal code for someone to use to try out there machine as it is easily editted for depth etc so you could try a pen for first test.

I like using lexan, going to try and do more picture engraving when I figure out all these covertion programs !

Wow!

Very cool! I bow to your skills.

I am ordering the parts for the PicStep today! 1 question on the interface board WHERE is a parts list?

Randy,

Maybe you have missed the link that shows the schematic for the interface.

http://www.fromorbit.com/projects/picstep/images/picstep_interface_v2.0_schema.jpg

http://www.fromorbit.com/projects/picstep/images/picstep_interface_v2.0.jpg

If you look there you can see that its basically

2off 470uF caps
1off 1uF cap
7805 - 5 volt regulator
The pullup resistors are not specified - I used 10k's
1 x 6 pin 0.1" Locking Header for outputs
7 off 2 way PCB Mount Screw Terminals
Some DIL header - Just cut to size for the various connectors
and you will need some jumpers to link out the stepping mode if required.

Good luck.

Cheers

Well I bored you all with the machine log.... so may as well with my cutting trials :)

My 1st cut in 'hardwood' - unknown type or origin. Found in my fathers garage !

The cutter was not correctly matched to the Gcode. - Credit to Mr Bean for sharing this.

I really do need to stop using the machine and concetrate on the finishing touches.

Cheers

A major breakthrough !

All the things I have done so far to reduce the resonance have made a big improvement but at slow feeds I still had it. To be honest with the tool and extraction running It couldn't be heard. But when showing people the axis moving as a demo it was a bit annoying.

I stumbled on this thread :-

http://www.cnczone.com/forums/showthread.php?t=9298

In summary rewiring my motors to bipolar parallel from bipolar series has totally transformed the machine. It's so quiet and smooth and FAST :)

I posted this here just in case anyone else runs into this. These picstep controllers just get better ! - I couldnt imagine they would control the axis movement as fast as they are.

I buckled today, the wife who was slightly impressed at my no.4 for the parents asked for a number 9. for us... I just had to do it.... The wiring will be done....

I forgot to mention that the gcode I had was too large for the scrap pieces of wood available. (Credit to MrBean again). A cool feature in Mach2 lets you select an axis and simply scale it. So for the X and Y axis I scaled by 0.8 and the machine did the scaling while cutting. I have so much to learn about Mach2 but It really is money well spent.

Cheers and Happy Easter to you all.

:cheers:

Looks like it is cutting great! Nice looking finished product there...

I plan on using Mach3 for my machine too (hopefully I will win it!), it really is kick-ass software that is fairly priced. I really wanted to use OneCNC/XR - but it is a wee bit out of my price range! :wave:

Hey! Santini,
I've been following your logs.....Yah! It'll be hard to set aside those cutting ideas and finish the machine. I too get a thrill when someone brings their machine to output level.
Nice!

Bloy

P.S. JavaDog...If I can win I'll give you the Mach license...I need the geckos. :) ..and I bought a CNCzone hat but it got a little stained.

Thanks lads, I'm afraid the only way I knew I would finish the wiring was to actually remove the temporary wires therefore bringing my addiction to a halt :)

Before the wires were removed I actually did a real job for my father. I cut a 40mm diameter hole in some aluminium plate for his router table..... Well it's a start ;)

Some posts back :-
http://www.cnczone.com/forums/showpost.php?p=69871&postcount=359

I posted an image of some small diecast boxes I had fitted Led's to. The idea was to use the picstep 3 concept and drive two indicators for the step and direction signals only.
Well that was the plan. I have now wired all my motor's with the correct cable and plugs but suspect things are not going to be easy.

Using Alan's schematic :-

http://www.fromorbit.com/projects/picstep/images/picstep_schema_v3.png

I fitted the 270 Ohm resistors to the Leds inside the diecast boxes and using 4 wires of the 8 wire motor leads connected to my picstep boards at the points shown. (It looks like the polarity of the Led's are wrong to me ? so these are reversed)

(I'm using 8 wire cable - 4 to feed motors and 4 to feed the Leds)

But.... it seems like these two leds for step and direction are causing distorted steps. If I have the motors jogging slowly and connect the leds I can hear distorted steps, and it goes crazy. AND even just adding the leds directly to the pcb with 270 Ohm resistors causes the same problem so it's not what I first suspected. i.e power and signal leads running together.

Bottom line is - Has anyone built the Picstep 3 layout with the Leds and has it been ok ?

Thanks.

(Currently going through 'Cold Turkey' - I need to cut something !)

Edit :- Forgot to mention I'm using a laptop

P.S. JavaDog...If I can win I'll give you the Mach license...I need the geckos. :) ..and I bought a CNCzone hat but it got a little stained.

Looks like neither of us did too well. Ah, well, thanks for the thought!! :cheers:

Santini,
What do you have left to finish up?

I posted an image of some small diecast boxes I had fitted Led's to. The idea was to use the picstep 3 concept and drive two indicators for the step and direction signals only.
Well that was the plan. I have now wired all my motor's with the correct cable and plugs but suspect things are not going to be easy.

Using Alan's schematic :-

http://www.fromorbit.com/projects/picstep/images/picstep_schema_v3.png

I fitted the 270 Ohm resistors to the Leds inside the diecast boxes and using 4 wires of the 8 wire motor leads connected to my picstep boards at the points shown. (It looks like the polarity of the Led's are wrong to me ? so these are reversed)


The problem you're facing is something that I though at the time of seeing your little boxes might occur. Mainly due to the signal level involved and the distance you have where a signal line is being moved off the short copper tracks to several metres of noisy wire. You could build a little buffer circuit to drive the LED's and remove the PICStep's from receiving noise injected into the step and direction pulse lines (a little transistor and a resistor would probably be adequate). I included LEDs on the boards for "status" not for xmas tree lighting across your machines! :P

Yes the polarity of the LEDs are reverse, I've been to lazy to update the diagram. Sorry.

Also you laptop's parallel port might be under rated. I've seen this on a few laptop parallel ports. The parallel port might have issues sinking the 35mA load of the LED and the PIC's input.

I've used the PICStep 3 boards before without any issue with a desktop PC parrallel port. Worst case scenario I can add a non-inverting buffer to the pins to drive the LED's, but it'll mean another chip on the board (or some transistors) just for some silly blinking lights! (can you guess it wasn't my idea for the status LED's :))

Alan.

Thinking about it further, don't use the Version 3.0 design until I redo the status LED's. I'll remove the information from my site until I'm finished.

The parallel port can only source 24mA, and the LED's are asking for about 35mA. Unless you use something like a 420ohm resistors for the LED's instead of the 270ohm you might be overloading the parallel port.

My tests with the Version 3.0 board where with the opto-iscolated breakout board which it's outputs could easily source more than 35mA because it's not driving from the parallel port line but rather the 5v rails directly.

Alan.

Yeh, Maybe the Christmas tree lights are not a good idea :)

But I think my problem is down to the laptop I'm using. I suspected the issue was noise on the leads (1.2m long) but I mentioned I tried the leds directly on the pcb and not via the leads.

I will use these leds for another feature.... Not sure what mind !

Online and replying at the same time.

No problem Alan, I think we should blame Mr Bean, It was his idea for the Leds :)

Off with his head......

I really need to outgrow my addiction to flashing lights....

Cheers

Yup. Sorry, the LED's were my idea. I didn't think it'd cause this many probs tho'. I just wanted something to show if the PIC was working, as I went though a phase of trying to blow the boards up. I'm happy to say that the PICSteps are very robust. I recently fitted a heatsink to the Y axis. I accidently shorted the link that runs just above (between) the LMD's, to the LMD's ground.. It melted a line in the Aluminium, but the thing keeps on chuggin'. Phew.

Regards Terry.....

If you just want pretty lights, I've attached a dumb circuit I use that I strap directly across motor inputs. Not very clever but it looks pretty.
In the case of steppers use one of these circuits per coil.
I use one red and one green LED to indicate direction. Resistor is at least 1k-2k depending on your voltage. I've not had any problems with it interfering with stepper signals but then I've not really done anything where it would matter (yet).
You can also use this in place of the motor for testing the driver board.
As an added bonus you get a pretty light show when you spin your motors by hand without even connecting them to a driver!

[QUOTE=Garfield2]Thinking about it further, don't use the Version 3.0 design until I redo the status LED's. I'll remove the information from my site until I'm finished.


So can I just leave the LED's off the board? As I have etched and soldered 2.5 of the board so far *frown* hahaha had to have the latest design

[QUOTE=Garfield2]Thinking about it further, don't use the Version 3.0 design until I redo the status LED's. I'll remove the information from my site until I'm finished.


So can I just leave the LED's off the board? As I have etched and soldered 2.5 of the board so far *frown* hahaha had to have the latest design

Yup they're identical, just leave the led's off and their matching resistors. I'm working on a new breakout board that will have the status LED's and opto isolation.

Cheers,
Alan.

Alright! Another board for me to toner transfer and etch! getting pretty good at it now!

Thanks for the schematic 10bulls, I think I'm through my 'Cold Turkey' period :)
The LED's are currently not being used but I have some ideas that should see them in action....

Time for a ramble......


All three axis are now fitted with these diecast boxes which proved useful as the stepper wires are terminated to the controller leads inside these.

I finally got the bits around me to finish the PSU and Controllers. In summary I purchased a couple of 'large' aluminium cases with that tacky black material on the covers and decided to build a PSU unit and a controller unit.

I copied Bob's idea on the din plugs. To make things more identifiable I used coloured locking rings and i.d rings on the sockets.

I fitted four sockets. 3 for the axis and one will be used for the limit switch inputs. If I ever make a 4 axis machine the 4th socket (yellow) will be used and I will add another socket for inputs.

The switches fitted don't have the correct coloured buttons. (I ordered the wrong ones!) but will change these next time I place an order.

I also plan on fitting a fan on the top of the controller housing and will use my machine to cut this hole sometime soon.

A couple of status LED's will be fitted shortly. (I just cannot kick the habit...)

That's it just need to get another table to sit next to the machine for the PSU and controller to sit on.

Cheers

Ooo! Those connectors are pretty. What is their name, so I can find them stateside?

Javadog,

These connectors are from R.S components in the U.K.

Technical details are :-

Current rating 4A
Working voltage 100V a.c./d.c. Max.
Contact resistance <10m Ω
Temp. rating -25°C to +70°C
They are the 8 pin type - Bayonet,1/4 Turn,Cable Mount DIN Plugs & Sockets

Socket Part number = 235-1530
Plug Part number = 235-1451

Hope that helps.

Cheers

A couple of status LED's will be fitted shortly. (I just cannot kick the habit...)


Well I'm just finishing off the new break-out board that will be a drop in replacement for the old one. It will have opto-isolation, the same 5V regulated supply, and status LED for all axis step/direction lines. I've also upgraded the status LEDs which stupidily where harder than I originally planed. The reason is, the pulse from the parallel port to advance the stepper controllers is *tiny*, I just put up with this on my breakout. But in order to make a LED blink properly you need to stretch this pulse out to something more useful. I've got a working model on my desk at the moment, I'll post the schematic and parts list soon (mostly 6N139's, 74HC00's, and 74HC14's).

It'll be a drop-in replacement for the existing board, and should provide all the functionality that was missing on the first version.

Cheers,
Alan.

I fitted four sockets. 3 for the axis and one will be used for the limit switch inputs. If I ever make a 4 axis machine the 4th socket (yellow) will be used and I will add another socket for inputs.


Nice looking machine mate. I really like the controller box, very professional!

Cheers,
Alan.

Cheers Alan,

To think it all started after I stumbled across your picstep design...
I forgot to mention I will be adding some transfer labels to the control and psu boxes. I know what it's like. Things are fresh in my mind at the moment. A few months down the line and I will forget what is doing what!

I appreciate the work you have put into the opto-isolated breakout box. :cheers:

Looking forward to seeing the final design, The opto-Breakout will complete a cracking collection.

Very nice! They're surprisingly cheap as well (for RS).
The colours even match the axis colours in Autocad...Impressed!
I like this idea so much I think I'm going to steal it ;)...not this month though...
I'm going through the money haemorrhaging phase of my project just now :(

Santini, so the 4A rating on those connectors/jacks will be fine right? I mean, even for larger servos you wouldn't be pushing 4A per pin...would you?

I ask because I found these from Hypertronics (http://www.hypertronics.com/circular.html#):

http://www.hypertronics.com/images/product/lgimages/circular-dseries-400x235.jpg

Specs:

Circular plastic connectors
3,4,7,9,12, & 25 contact positions
1 to 8 amps per contact
®recognized components File No. 102195
Each connector half accepts pins or sockets
High Impact polycarbonate body
High temperature versions available (+125° C)
Quick disconnect push button release
Alignment provided by housing
Crimp, solder cup and p.c.b. contacts
Color coding available
Recessed contact terminations (eliminates need for shrink tubing)

I can get these stateside...and I think they would be good for my machine.

Santini, so the 4A rating on those connectors/jacks will be fine right? I mean, even for larger servos you wouldn't be pushing 4A per pin...would you?

Indeed if you're pushing more than 4A to your motors a DIN plug and socket probably won't be very happy.

I am going to use Neutrik Speakon connectors for my servo motor connects (30A+ per pole). You can get them with 2, 4, and 8 pole connectors. They have a bayonet locking mechanism and they are commonly used to power those big-ass speakers you see at concerts. I've personally seen a 60Kg speaker hanging from a 4 pole socket and plug when the speakers mounting failed because of dodgy workmanship.

Check em' out at :

http://www.neutrik.com/

They're as common as dirt here in Australia, any good PA shop should be able to get them for you.

Cheers,
Alan.

Not a bad idea. A big honkin' bi-amped speaker isn't all that different from a bipolar stepper electrically either.

Pete C.

Guys look at the wires coming out of your steppers, they are only rated at about 3A max each if that, those multi pin audio connectors that santini is using are perfectly ok, you don't need big heavy/bulky connectors.

Guys look at the wires coming out of your steppers, they are only rated at about 3A max each if that, those multi pin audio connectors that santini is using are perfectly ok, you don't need big heavy/bulky connectors.

Indeed, but like I said in my post, if you need larger current handling capabilities you can use something like the Speakon. For my machine with <3A loading I'm going to be using 5pin XLR sockets and plugs, mainly because I have bucket loads of them already, and secondly I hate DIN sockets! (although the locking coloured ones are very very nice. The ones I hate are the grey platic ones you used on old cassette tape players, slide projectors and the like. <shudder>).

Cheers,
Alan.

Hi guys,

Okay here is the first cut of the opto-isolation output board for the PICStep drivers. This basically has 4 axis opto-isolation with status LED's for both step and direction lines, plus the usual emergency stop switch input.

It's a dual layer board as it would have taken much more effort to make it a single layer with jumpers. I plan to get some made up at www.custompcb.com sometime soon. Also the next revision will have the 5V regulator onboard, plus the inputs will be on a seperate daughtboard that will fit on-top of the output board, this will provide opto-isolated switch inputs to the PC as well as the output header for relays etc etc. (I might even put a pair of relays on this daughterboard for turning on and off contactors for mains powered devices (router motor, coolant pumps, etc)... Hmm an idea!)

Anyway, won't be long.

Cheers,
Alan.

Any chance of the schematics for that Alan ?
Thanx
Thys

Any chance of the schematics for that Alan ?
Thanx
Thys

Not till I finish it! :P

Cheers,
Alan

Edit: Here is a taste, still got to finish the input board.

Just following up on the rating of the Din plugs. I confess I only looked for something suitably rated to match the Picstep controllers. The 4A rating meets this fine.
I do like the others posted, Ideal for those running higher currents.
(I can remember those grey plastic ones... Nasty !)

The Opto board should be ideal. I might even get rid of the disaster memories trying to make double sided boards..... :) <Time to donate to the Lathe Fund>

I only hope it can squeeze into my controller box. It's a bit tight in there.

Again great work Alan... Thanks,

can't wait to get the pcb's made now , like a baby waiting for Chocolates to arrive

good work Alan ,

Here's a small bit of code for anyone wanting to keep the kids happy. Another piece of wood mounts to the back of the bear to make a 'bookend'...

Well I had to make them something !

Again it's the highest quality wood found in my garage. (An old fence offcut that the kids enjoyed painting with that manky pot of wood stain I have !)

I used Mach 3 to cut this, worked fine.

Gcode attached.

that rocks Santiniuk......with that CNC router....are ya gonna 3D it some......scoup out the ears....contour the nose and eyes.....give it a belly......

I'm still at that stage where I don't know very much on the CAM side Viper, so I'm struggling along doing simple stuff. (I'm also relying on Mr Bean too much for the code generation).

I seem to have the basics sorted now so I should make more interesting stuff soon.

I need to fit some vents / fan to my controller box as I heard some rumblings on my steppers after cutting some aluminium yesterday. Basically at 3A AND 35V my picstep heatsinks need some cooling. (I had planned to fit the fans but need to do that urgently as it really cooks inside the enclosure.)

The attached image is all I had to work from so....

Cheers

Had a garage cleanup today and decided to set my machine away while I was busy.

Nothing too serious. The bear again from 6mm aluminium.

It only about the size of a large key ring.

Although a very simple cut It's a real incentive to continue and make useful parts in metal.

Cheers....

On the Alumn, what bit, size, and speed did you use? feed rate, and stepdown?

thanks, Joe

Hi guys,

I've updated PICStep (yet again!) to V3.1 which removes the ICSP and the broken status LEDs.

I've also released the first half of the opto-isolated breakout board. This half has the logic to output proper LED status signals for all step and direction lines as well as the usual emergency stop and 5V regulator. I will be designing a daughter-board to fit on-top of this output board to provide 5 opto-isolated inputs and 4 relay outputs. Will keep you posted!

Have a look see at it all here (http://www.fromorbit.com/projects/picstep/)

Cheers,
Alan.

Did somebody mention an opto-isolated breakout board :)

Awesome work Alan, turned around it record time. Cheers !
(Time for another lathe fund donation everyone)

I have bad memories of my previous double sided board attempt some years ago. This is my chance to clear the memories......

Hopefully all being well I should be having a go this week at etching the board. I'll be using doublesided photoresist board and hope I align both sides correct this time.
Any tips anyone ?

Will no doubt let you know how it goes... good or bad ;)


Joe2000,

I'll be honest I used Mach 3 and again used the scale feature to cut the aluminium bear a 1/3rd of full scale.
I manually dropped the feedrate of the machine to around 10 IPM, 3mm cutter, approx 7000 rpm (spindle speed) as this is the slowest my tool adjusts to. Step was about 1mm.

I'm no expert I just used these settings as the machine sounded ok. (I did accidently trip my pc supply so I had a slight score around the edge of the metal)

Thanks,

Front image needs to be mirrored 1st on the Opto Board.

As far as alignment goes, the 5 holes are 1/8". I would toner transfer one side, drill out the 5 holes, use 1/8" dowels to align the other side. Transfer toner, then as a quick test drill out a couple holes where components mount. If it doesn't match up break out the acetone and do the unmatched side over.

OR could you etch 2 seperate board then drill out several holes on each, align them both and drill them?

Make for a thick board...... might be easy... looking at 2 4x6 single sided boards......

Thanks for the tips Randy. Not had chance to look at the Opto-Board yet but have ticked one of the things off my list to do.

I confess I have posted some pretty boring pics and I'm afraid they don't get any better :)

Basically the LMD's are cooking in my sealed enclosure. I had always planned on adding a fan but it was obvious this had to be done before any more use of the machine.

So... I used a 120mm fan on the top of the enclosure to reduce the heat. Some vent holes are also drilled to provide good airflow.

The fan required a hole 110mm diameter to be cut. I didnt have a lot of options to do this and so my cnc was used to cut it. (The first job that was actually for the machine itself).

Mach 3 was used and I simply used the hole cutting wizard to generate the code for a 110mm hole with a 3mm cutter. It must take a minute at the most to use the wizard.
I have attached the output. I'm sure all the pro's are laughing at this but for a simple common man like myself this is what its all about. No sharp edges to file etc....

So dull pics prior to fan being fitted are attached.

And finally pic of fan fitted. With and without the grill covering. I thought it would be wise to get this to prevent contamination of the controller box.

Cheers

Very nice Santini. I really like your attention to detail on every part to your machine.

Very nice Santini. I really like your attention to detail on every part to your machine.

Indeed, but I'm intriqued as to why we've not seen photos of the internal setup of the control box yet.. :rolleyes:

Could it be that the inside doesn't match the class and craftsmanship of it's exterior? :D

:wee:

Cheers,
Alan.

Hi all,

Input boards coming along. I'm thinking I might just put a header that can connect to a seperate "relay board" for switching relays, else the input board gets big and more complex than most people probably need.

So the input board currently will have 5 opto-isolated inputs, a 4 output header to connect to a seperate relay controller board, and a seperate 5V regulated supply for the PC side of the controller (a seperate transformer/power source that only has the mains supply in common with the rest of the driver power would be best else the opto-isolation is a mute point if the 5V supplies and grounds are shared between the PC and the controller :) )

I'll post a preview soon.

Cheers,
Alan.

Indeed, but I'm intriqued as to why we've not seen photos of the internal setup of the control box yet.. :rolleyes:

Could it be that the inside doesn't match the class and craftsmanship of it's exterior? :D


Cheers,
Alan.


It's really not that bad !.... Will grab a pic soon. Honest ;)

Managed to blow away the memories of my double sided board failures and made the opto-boards today. Extremely pleased to be honest.

1. I printed the artwork from a laserjet to clearfilm.

2. I placed the two films on top of each other and put a couple bits of tapes to hold this to ensure the artworks aligned.

3. I then placed this film ontop of the photoresist board and drilled 4 small holes in the mounting holes. (about 1mm dia)

4. Using diode leads I pushed these into the holes and split the clear films and placed one each side of the board. i.e the cropped leads of the diodes ensured the films were aligned top and bottom.

5. The rest is the usual UV exposure methods.

So.... progress, just need to do my favourite job and drill it. :rolleyes:

Must say Alan, I'm extremely dissapointed with the lack of wire links on this...... Can you add a few in future ..... You know how much we enjoyed those.

Cheers.

Those are some damn fine looking boards. Nice job.

So.... progress, just need to do my favourite job and drill it. :rolleyes:

Must say Alan, I'm extremely dissapointed with the lack of wire links on this...... Can you add a few in future ..... You know how much we enjoyed those.


Wow nice work! I hope it all works in the real world and not just on my breadboard... :) You're the always to first to build my design before I do!

Will the new board fit in your japanese puzzle box?

Don't worry I was thinking of ways to put a few wire links in just for you, but I opted not to. :) You've still got all the thru-hole via's to solder little wire links into, they should be fun for you! I'm just going to give the Gerber files to a PCB manufacturer and say, "make me 4 boards with hole plating please" and I'll not have to worry about them.. :D

Cheers,
Alan.

Nice work mate! Glad to see you've put your double sided demons to rest!
I've been etching away as well tonight...

...been a good few years since I made a double sided board so I too was relieved it all came out alright. I'll describe my method in a bit of detail later....when my eyes are staying open...

BTW: Alan...Thanks heaps for all your hard work! Donation imminent...promise! :cheers:

Hi,
could you please tell me were you got your black controller box here in the uk,been looking at RS but no luck.
Many Thanks
Regards Steve.

Steve,

I popped into my local Maplins who had the cases out on the shelves. I think I paid about £6.99 or £7.99 each. I cannot remember to be honest.
I think its one of these although the ones I have are not painted they are covered with that 'expensive' looking black vinyl :)
link :-

Cases (http://www.maplin.co.uk/Module.aspx?ModuleNo=1734&TabID=1&source=15&WorldID=3&doy=15m4)

I noticed this is your 1st post. I hope this long thread has been a little incentive for you. All you need to do is ask, there are a lot of helpful people here.

It looks like 10 bulls takes the trophy on the optoboards. Very professional, and your board is a nicer colour !

I'm all done now with the opto construction, just need the actual optos to try it out.

Alan, you keep the designs coming. We will beta-test for you ;)

Javadog, your positive comments are appreciated. I'm following your progress with great interest.

Cheers

Javadog, your positive comments are appreciated. I'm following your progress with great interest.

Cheers

Thanks, good to know someone is watching! But you're right, his boards are a prettier colour! :p
We gotta find you some sexy Red (http://www.msicomputer.com/product/p_spec.asp?model=865PE_Neo3-F&class=mb) PCBs or maybe some stealth Black (http://www.msicomputer.com/product/p_spec.asp?model=925XE_Neo_Platinum-54g&class=mb) ones. :cheers:

FYI - I ordered my Servos today (360oz/in!).

Many Thanks.
Regards Steve.

Gotta confess...mine are same dull colour as everyone's...just photo'd them on the UV light box....This is an example of why it's not necceserily a good idea to pick up girls at Discos! ;)

Hi guys,

Thanks Terry (aka Mr Bean) and 10bulls (aka I have no idea) for your generous contributions to my "Buy Alan a lathe fund", which at the moment is almost a "Buy Alan a 'decent' power drill fund" :)

Many thanks also to all who have contributed!

Cheers,
Alan.

Thanks, good to know someone is watching! But you're right, his boards are a prettier colour! :p
We gotta find you some sexy Red (http://www.msicomputer.com/product/p_spec.asp?model=865PE_Neo3-F&class=mb) PCBs or maybe some stealth Black (http://www.msicomputer.com/product/p_spec.asp?model=925XE_Neo_Platinum-54g&class=mb) ones. :cheers:

Black ones? you mean like these? :devious:

Black ones? you mean like these? :devious:

DUDE! I recognise that board! That's one of the customPCB ones I got made ain't it? :)

Alan.

Ohhh Black looks nice :)

Alan,

I know you will be looking for something to do now......

If your ever bored can you make a Picstep 4 based on Picstep 3 but using a doublesided layout. :)

My case is struggling to fit them all in now with the Opto Board.....

Cheers

while Alan no doubt now needs a todo list , why not a 4 axis board complete with opto breakout .
would save quite a bit on all the un needed connectors , oh and of course it has to be Double sided , as Santini and others are now extremley proficent at making Double sided PCB'S .

dave

while Alan no doubt now needs a todo list , why not a 4 axis board complete with opto breakout .
would save quite a bit on all the un needed connectors , oh and of course it has to be Double sided , as Santini and others are now extremley proficent at making Double sided PCB'S .

dave

Nah I hate the idea of the single board controller. To hard to diagnose/move things around when things go wrong. Nice easy seperation allow you to easily eliminate non-suspect parts from the system without complexity. Plus not everyone wants to use PICStep for CNC, and not everyone wants 3/4 axis. For example a CNC lathe would only need two drivers, the third would be mostly redundant.

A double-sided PICStep is already in the works, as is a SMD version with on-board opto on the singular boards.

Besides I like the modular approach. :D

Alan.

Ha Ha thought you would a heart attack - only kidding Alan , but keep up the good
work , only wish you would post the eagle files or whatever makes it easy to change a part , such as adding the parralell port on board socket or different Opto's as believe ity or not RS dont now stock the 6n139 and listed it as obsolete , although Farnell etc still have it available , but just to give you a bit of feedback , as i found your designs and Philosophy of seperate boards a good compromise , to suit all build ideas , keep it going Alan .
we need more designs to practice our pcb making skills .
can't wait for the I/O opto board now to complete the set .

only wish you would post the eagle files or whatever makes it easy to change a part , such as adding the parralell port on board socket or different Opto's.

I use GSchem and PCB for Unix-like OS's to design and layout the boards. This makes it somewhat difficult to release the board's sources because I don't want to be inundated with emails from people "trying" to use GSchem/PCB (or Linux/BSD for that matter). I try making the parts on-board as generic as possible, but ultimately you have to compromise somewhere.

Eagle bites as far as I'm concerned, because I find the library to difficult to handle and I find I spend more time fiddling with the GUI widgets than I do making the boards. Plus with GSchem/PCB I can jump in and fix bugs because I have full access to the source code. Don't get me wrong Eagle is a good product, just not my kind of cheese.

as believe ity or not RS dont now stock the 6n139 and listed it as obsolete , although Farnell etc still have it available.

The 6N138 can be a drop in replacement.


Cheers,
Alan.

Curse You Alan Garfield !

I hope your next pint takes like drain water ;) - Or is that Fosters ???

Anyway managed to bribe the wife into going shopping alone and let me watch the Football today. Oh yes....

I have been putting it off but had to lift the lid on my controller to see if there was anyway I could shoehorn the opto-board in.

Erm it's going to be one struggle.... How inconsiderate can you be Alan ?

As you can see from the attached images The control box was not finished completely. The plan was for the Ribbon cables to run under the pcb's like I had on my MDF mounted images and for the power leads to be raised and clipped to the top lid of the controller. i.e separating them as much as Possible.

Thankfully I was too lazy to actually finish this as my quest to cut numbers, wood bears and then the almighty aluminium keyring bear grabbed an addiction on me like some crack addict!...

So the three attached images show where I am prior to surgery to fit the new opto-board.

Can someone pass me a shoehorn..... :o

I'll squeeze it in if it kills me !


Note :- No camera trickery like the underhanded method 10bulls uses has been done to these images :) Therefore they look boring

I needs some flux remover and the opto's now.

Thats some pretty neat wiring santini...Boy are you gonna laugh when I rats nest mine! Fun soldering those top traces underneath the connectors of the opto board wasn't it ?

OK. One small problem...it doesn't work!
Just had a bit of a play on some breadboard and reading the 6N139 datasheet,
I'm pretty sure there is a missing resistor ~470R which should go between
pin 6 and pin 8 of all the optos.

Ya see Alan...If you'd been drinking Fosters the lack of alcohol and taste would
not have distracted you from double checking your schematic ;)

Amazing !

I received my opto's tonight and mine doesn't work either :)

I'm using a laptop and did not have time to do detailed analysis,

Chin up 10bulls, tomorrow should sort it....

Cheers

I just tried my PicStep boards...

Problems with the interface board connected the stepper vibrates, if I power down and disconnect it from the interface board (NON opto) it doesn't vibrate

If I disconnect the parallel port from the interface board it vibrates

Any troubleshooting data on this board?

Thats some pretty neat wiring santini...Boy are you gonna laugh when I rats nest mine! Fun soldering those top traces underneath the connectors of the opto board wasn't it ?

OK. One small problem...it doesn't work!
Just had a bit of a play on some breadboard and reading the 6N139 datasheet,
I'm pretty sure there is a missing resistor ~470R which should go between
pin 6 and pin 8 of all the optos.

Ya see Alan...If you'd been drinking Fosters the lack of alcohol and taste would
not have distracted you from double checking your schematic ;)

Weird there are conflicting reports as to whether the resistor is needed between pin 7 and ground. On my test breadboard I didn't need the resistor at all, hence why I didn't put it in the diagram. Maybe there are different tolerances in the parts. Does any one have a CRO and can look at the output of one of the opto's on pin 6? I'd be interested in seeing your findings.

I'll have a futher poke around here and get back to you all.

Also try popping out one of the optos and put a wire link the the IC socket between pin 6 and pin 8 and then swap it to pin 6 and pin 5, the LED's and the outputs should step/change direction. If this is working correctly than it's just purely a misconfiguration of the opto-isolators and should be easily remedied.


Cheers,
Alan.

< Mental note > don't release anything before you test it yourself< / >

10bulls yes i would agree with you on that one

Dave

I just tried my PicStep boards...

Problems with the interface board connected the stepper vibrates, if I power down and disconnect it from the interface board (NON opto) it doesn't vibrate

If I disconnect the parallel port from the interface board it vibrates

Any troubleshooting data on this board?

Sounds like a fault/short in your interface board. Grab a multi-meter and beep-out all the ports and pins and verify there are no shorts.

Also verify you've put the 10 pin inter-connects between the PICSteps and the interface board the right way around. If they're back-to-front you may be reversing the 5V supply causing all sorts of issues to the PICs on the driver boards.

The PICStep boards need 5V and their drive power to properly power up. Try just supplying 5V to the 10 pin IDC header and see if the problem still exists. If it does it's a problem with your driver board/PIC programming. Just suppling the driver voltage will only power up the LMDs, and without the drivers attached to the interface board they won't have their needed 5V supply.

Cheers,
Alan.

Weird there are conflicting reports as to whether the resistor is needed between pin 7 and ground....


It's not the 10K from pin 7 to ground. I tried that first. Its a resistor between 6 and 8.

Like this...(Works Now!)

It's not the 10K from pin 7 to ground. I tried that first. Its a resistor between 6 and 8.

Like this...(Works Now!)


Yup my fault.. I did leave the resistor out.. :(

Sorry! I'll update the design now.

Here is the needed resistor values depending on your 74xx14.

http://www.fromorbit.com/6N139.jpg


Looks really nice 10bulls, well done on the board. Also thanks for the help fixing the problem. I'm glad I'm going to get my interface boards made professionally with plated-holes so I don't have to worry about soldering under parts! SORRY!

Alan.

Now worries! Apart from that its all very sweet! No more blowing up Mr busy pin for me! :D :D
Thanks Alan

BTW...Did you make me solder in 2 extra inverter chips, 2 nand gates and some nice tantulums....Just so Santini can have bloody Christmas Lights!!?!? Just ignore him next time OK ?! ;)

Now worries! Apart from that its all very sweet! No more blowing up Mr busy pin for me!
Thanks Alan

Excellent, glad it's functioning the way I had originally intended!


BTW...Did you make me solder in 2 extra inverter chips, 2 nand gates and some nice tantulums....Just so Santini can have bloody Christmas Lights!!?!? Just ignore him next time OK ?!

Guilty as charged! But you gotta admit they're pretty cool blinking lights. :) You could drop the 74x00's if you didn't want the LED's and I could make the board smaller if I only used 2 74x14's but I thought the status LED's where a good idea so I just went with it. :)

Plus I like Santini little aluminium boxes with the LED's in them for each axis, makes the machine look professional like. :D

Cheers,
Alan.

Actually they're starting to grow on me..."Flash On! Flash Off! Flash On! Flash Off!"....By george...I think they're trying to tell me something!
"Put your hands in the air and step away from the solder sucker!" ;)

Don't blame you on getting plated through boards made up. I here black is in this season. Aparantly it's the new red....anything but green...that's oh so 90's.

Actually they're starting to grow on me..."Flash On! Flash Off! Flash On! Flash Off!"....By george...I think they're trying to tell me something!
"Put your hands in the air and step away from the solder sucker!" ;)


LOL! I want to see some machine with cold-cathodes along their axises that are linked to these status lines. It'll be the "Disco-CNC" machine. Could even hang a small mirror ball from the z-axis.


Don't blame you on getting plated through boards made up. I here black is in this season. Aparantly it's the new red....anything but green...that's oh so 90's.

Oh daaaaaar-ling, I'm probably going to go BLUE! (But I'll probably end up with green because they're a lot cheaper and who cares what the boards look like when they're inside a box.. :P)

Cheers,
Alan.

LOL! I want to see some machine with cold-cathodes along their axises that are linked to these status lines.
Cheers,
Alan.

Actually, since my other hobby is case modding - I do plan to pimp out my controller box. But shhh...it's gonna be a surprise! :D

That is some nice work you guys are doing with those boards, it is the one thing in electronics I have yet to do, so I may come back and ask some questions later !

What I want to know is, what is the benifit's of making this board, I note that it is Bipolar motors your using which give more torque, but why not buy a Xyolex board, or is this better or just more fun building your own.
I know the designer is here too which is fantastic the help he is providing you all, I don't really know what the Xyolex board is like, so I am not saying use it or it's better, I just used it as an example of a commercial bipolar board.

Actually, since my other hobby is case modding - I do plan to pimp out my controller box. But shhh...it's gonna be a surprise! :D


Ahh you'll defiantely be wanting the BLUE pcb's then.. :) Plexi-glass windowed controller box with shiny heatsinks and a blue cold-cathode... Hmmm I can hear the pimp music from here.. :)

Can't wait to see it..

Cheers,
Alan.

Bug fixed in the opto-board and it's schematic. Many thanks to 10Bulls and Santini. Opto-Bob (:D) is now V2.01 and a tiny bit longer to fit the extra resistors I left out in (Sorry Santini)

Have a look see here (http://www.fromorbit.com/projects/picstep/).

Plus check out the updated gallery, I've put some more photos of other peoples works up.

Cheers,
Alan.

The PICStep boards need 5V and their drive power to properly power up. Try just supplying 5V to the 10 pin IDC header and see if the problem still exists. If it does it's a problem with your driver board/PIC programming. Just suppling the driver voltage will only power up the LMDs, and without the drivers attached to the interface board they won't have their needed 5V supply.

Cheers,
Alan.


ok I removed the interface board from the circuit.... 1 of the 3 boards I built functions drivers get hot and the stepper hums after I run a circle cut (I know a 1 axis circle!)

I traced out all the leads on the lm18245's the all go where they should... guess I will hook up the 1 working board and jot down all the voltages, resistances, etc on the schematic until I find the problem!

sucks I am 25% on boards so far! 1 driver 2 not working and an interface board that has issues!

well been up all day and most of last night

Alan,

Thanks so much for sharing your knowledge, skills and plans for use by fellow builders. It’s people like you that keep the sport alive and growing on the CNCZONE.
Hager

That is some nice work you guys are doing with those boards, it is the one thing in electronics I have yet to do, so I may come back and ask some questions later !

What I want to know is, what is the benifit's of making this board, I note that it is Bipolar motors your using which give more torque, but why not buy a Xyolex board, or is this better or just more fun building your own.
I know the designer is here too which is fantastic the help he is providing you all, I don't really know what the Xyolex board is like, so I am not saying use it or it's better, I just used it as an example of a commercial bipolar board.

mostly satisfaction and fun , i would not go as far and say it cheaper .

the process of knowing when things go wrong you can ask here and get it fixed without paying huge callout fees and repairs , well at least most of the time

and of course more knowledge
just enough to be dangerous !

Dave

Satisfaction is always the best reason, I have a few nice bipolar motors so might have a go at this myself, it will give me an excuse to make a PCB.

Hey Gang,

Just finished the input opto board for the "opto-bob".

Have a lookie, and let me know if you see any errors (chair)

Cheers,
Alan.

Hey Gang,

Just finished the input opto board for the "opto-bob".

Have a lookie, and let me know if you see any errors (chair)

Cheers,
Alan.

looks ok , can't see any errors Alan

I need some moral support.... :)

In summary No joy with the Opto Board. I have fitted the 'fix' resistors as specified but when testing it the leds blink in a random manner and the steppers dont step although they do 'squeak' but no real movement.

I have checked the board for errors and cannot find any. The 5V rail is stable on all ic's.

I was hoping the problem was due to the use of a laptop but so moved a pc into the garage and still no improvement.

To make matters worse I then tested the original interface board and one channel is not working. It's not the picstep driverboard its somewhere between the pc and the interface.

To be honest I have not had a lot of time to look at it due to other commitments.

Ahhhh.. feel better getting it off my chest...

I need some moral support.... :)

To make matters worse I then tested the original interface board and one channel is not working. It's not the picstep driverboard its somewhere between the pc and the interface.

To be honest I have not had a lot of time to look at it due to other commitments.

Ahhhh.. feel better getting it off my chest...

Hey dude, sorry to hear about your bad luck. Seems very strange the old interface has stopped working too... Could it be a cable problem?

Alan.

sounds very much like a parralell port to opto or logic level problem , so theirs not a clean signal to fire the opto's , or indead they are no shutting of completely try dropping changing series resistor or leaving it out alltogether , and also check the ground return from the board to the pc and make sure theirs a good ground , also stick a few .1uf caps across the power rails

if you have a scope handy let me know what the voltage level is like at the input to the opto , both when the line is held high and pulled low

or even too much noise floating around from the mains transformer would do it to , theres to many variables at the moment we need to narrow it down .






Dave

Here is a trace at pin 6 of one of the optos during a step pulse, from a (fingers crossed) working opto board.

As you can see, the pulse is pretty ramped for a 10uS pulse (which is what my mach2 is kicking out). Much shorter than 10uS and the inverter probably won't fire. Interestingly I had a problem with an intermittent 'pulsing' on the second channel. This appears to have been caused by having a Mach2 min pulse width of 15uS (or above)...which it does warn may cause locking.

I *think* the value of the pull up resistor from pin 6 to 8 of the optos will influence the angle of this ramp. I'm still using 470R.
You seem to know yer onions Dave, whats your view on this? I agree with you about filter caps...couldn't hurt!

One more thought. I have been testing mine with an old 6 wire printer stepper. I knocked the steps per unit way down (100) and speed low (10).
The stepper won't hack much more than this (I running it at 9V for testing).
Maybe this slow speed is helping mine work ?!
Oh, and the outputs from mach2 should be inverted.

As a process of elimination, here's what I wou