HM35 Gear Head Mill CNC Conversion

[XaserTech]

Hi all,

Just trying to find anyone in Perth, Western Australia who likes tinkering with CNC, I've just converted a mill myself with a rockcliff controller but the PC im using maybe playing up intermittent movements, seems like under voltage pulses from the Parallel port and not firing the opto's similar to what you would expect from a laptop with power saving mode on, sourced a PCI parallel card to try and see if that'll get it going.

BTW, I've migrated from the nightmarish MC433 enough said ! (not parallel port related, the board is just crap!)

Anyhoo, if your local give us a shout, or even if your not with a similar conversion, I will post photo's of the setup tonight when I get home from work, for all those interested peeps out there.


Adam.

www.xasertech.com

[CrazyIvan]

Hey mate. I'm doing a similar conversion (in Sydney) on a HM45 clone.

Quick question: Did you replace the XY screws/mounts. I've found mine has quite a bit of play and I'm thinking of an eventual ballscrew conversion (sloppy CNC is still better than my manual skills).

[XaserTech]

Hi CrazyIvan,

My machine is still stock apart from what I have done to it, I did have herrendous backlash when I first got the Mill but the brass nuts are somewhat adjustable. But there was a prob with binding in different spots up and down the length of the screw. As the pics of the X and Y drives are now the Pulleys are tension against the thrust bearings to elminiate all axial movement.

After a day of frustrating adjusting and testing I've finally got it down to .05mm to .1mm like you said its still better than manual milling anyday.

I will however when I wear these out convert to ball screws as long as my machine lasts long enough to make the new parts!! (I have no doubt it will)

How far along are you with your conversion?

Cheers,

Adam.

www.xasertech.com

[CrazyIvan]

My conversion is going a little slow (haven't touched it for 3 months!).
Also, just to make it harder, I'm making my own 4 channel servo controller.
I've rigged up the X & Y just to get it moving and run some tests (1:1 to the motors).
http://ivan.blogs.chimerical.com.au/...l-XY-Test.aspx
I have 0.2mm of slop in my X+Y (0.1mm in the nut + 0.1mm in the bearing mount = 0.2mm total).

I'm going to connect up the quill next, and then it should be able to do some real work.

I've the square column model, and plan to eventually automate the column (not quill). I will need a ballscrew conversion on the Z because the standard screw is inefficient as hell.

And I need to finish the servo controller... and a million other ideas.

BTW: I love how clean your motor mounts all look (especially compared to my old RHS & flexible tubing).

[XaserTech]

Great vid. Looks like nice quick rapids too!

I seem to be plodding along on mine lately its been too bloody hot in the workshop, its hard to think straight when you've got sweat pouring into your eye's.

I opted to do the quill only (round column is out of the question anyway) for the Z thats 130mm of travel which for me I think is enough for what I want.

As you can see I chose steppers for the motors (nothing against sevro's just simplicity at this stage) and geared the X & Y down by 3.33:1 and the Z by 2.67:1. No play in the Z as I have re-tensioned the spring return so that it is always working against the motor, still with very little effort on the motor.

I have powered the column (not shown) with a ol' radar head rotating motor because the novelty of hand winding the head up and down wore off very quick, I think it was the day after I purchased the mill You might see in one of the pics with the E-stop button a silver flush mount button 1 above and 1 below the E-stop thats the powered head up/down buttons.

Anyway off to do some more tinkering.

Adam.

[holycross]

Adam,

There are a few people here in Perth doing the home built cnc. I'm very slowly converting a RF45 clone (titan machinery) to cnc. Have just bought ball screws. Now have to design and build the bearing housings. Oh, and find a place to setup the mill.

The local get to gether is the 3rd sunday of each month.

Mark

[CrazyIvan]

Mark,

What size screws & nuts did you get / where from / how much?

[XaserTech]

Originally Posted by holycross Adam, There are a few people here in Perth doing the home built cnc. I'm very slowly converting a RF45 clone (titan machinery) to cnc. Have just bought ball screws. Now have to design and build the bearing housings. Oh, and find a place to setup the mill. The local get to gether is the 3rd sunday of each month. Mark

Nice too hear from you Mark, it would be nice to catch up with local fellow CNC'ers, feels like I'm the only one doing this sort of stuff until you find someone actually interest and willing to share experiences. My mates can't follow what I'm doing, just goes straight over there heads and beeline to the beer fridge


Okay, This is a general prob I will throw out to the forum to see if anyone else has this issue or has over come it.

Issue: I'm having a few drama's with setting up my controller. It's the Rockcliff 4 axis jobbie. I am experiencing intermittent stepping at any speed, so far through process of elimination I have, from the parallel port 0 to 4.65 volts from low to high transistion passing through a 470 SMD resistor to the Opto P421. See pic.

Now from what I gather with probing the logic side of the board a 470ohm is TOO high for a in this case 4.65 volt supply at the end of a 3m parallel cable when it disconnected.

When the cable is connected the voltage is 2.82 volts port side of resistor, on the other side of the resistor it's a mere 1.12 volts (measured).

Now tell me if im wrong, that's not high enough to keep these Opto's on reliably? (keeping in mind Active low is what I need to make a definate step.)

Is a 3m Cable too long?

Are the resistors of a too higher value?

Is the voltage out of the PC good?

Anyone else have any ideas on where to start for a remedy?

Im thinking of making an Parallel booster just for the step signals because Direction signal seems OK, go figure!

Also can't find any datasheets on this paticular opto.

Cheers, Adam.

[holycross]

Ivan,

I have 20mm by 5mm pitch ball screws. Not sure on price as another from the group ordered the parts and I had several items ordered. Will try and get the place they were ordered from for you, somewhere in Asia.

Adam,

Nothing wrong with heading for the beer fridge, but I know what you mean. Are you in Perth? Drop me a PM if you want to get together. Not sure on your problem. I'd need to read the manuals on the different parts. What are you using for a BOB?

Mark

[XaserTech]

Mark,

Yeah mate I'm in Perth 15mins out of town South of River, I'll PM more details.

Forgive me what do you mean by "BOB", sorry haven't come across that term before.

Adam.

[holycross]

Adam,

BOB is a Break Out Board. Used to isolate and buffer the signals between computer and the machine electronics.
Campbell designs, CNC4PC and a few others make them. Saves a lot of problems.

Mark

[XaserTech]

Break out Board, I should of worked that out. *slaps head

Everything is integral to the control board BOB and drives as per pic a few posts back.

No buffering though all real time from PC. All is very simple in design, can't understand the high losses as the "hi" state arrives at th optocoupler. It only flows though 1x 470 ohm resistor then the opto.

I currently trying to find a data sheet on the opto to find out the supply for the opto's LED because based on a typical 2v, 20mA LED for a 5v supply you only need a 150 ohm resistor. I'm almost certain this is the issue.

Don't understand for a engineered item why this has happened unless of course it is a manufacturing error.

Adam.