Denford Software & Machines

Ok, Ok, don't get too excited

Just a heads up for today to let you know what’s coming, no piccys at the moment, just sorting out what I've got & how it goes together as it came in a box.

I've made my Mk1 interface between the original digiplan stepper drivers (Thanks are do to davidimurray for that) if it works, (no reason why not he says optimistically) I'll post a piccy & artists impression for those electronically challenged (I don't know how it works I just follow the pictures and can just about aim a soldering iron )

I think progress will be intermittent as best due to work,

First thing is to work out where the wires come from & go to, I’m at a disadvantage there as the guts were already ripped out when I got it, so thanks to admin with the ORAC wiring diags being available here on the site I’ve traced what I need to know (probably)

Here’s your starter for 10…

The Digiplan drives slide into a rack and re-using the rack & connectors we find that the connections are as follows:

Pin 2 stepper – 2A
Pin 4 stepper – 2B
Pin 6 stepper – 1B
Pin 8 stepper – 1A
Pin10 24v DC for drive control

Pin12
-----------> Linked +ve stepper motor power
Pin14

Pin16
----------->Linked 0v stepper motor power
Pin18

Pin26 direction
Pin28 step
Pin32 0v

If I’ve not mentioned the pin number it’s not used in the Orac

well, I've had some steppers working from the digiplan drivers using Mach 3 with, ermm interesting results...

There seems to be a problem when changing direction, it appears that the direction pulse is rather sleepy and causes the motor to go rather crazy.
It fafs around then runs fine as long as I don't change direction.

I see that the drive reckons it needs 5micro second pulse width, I've tried that, although I think it was before I remembered to put the breaks in the veroboard I'll double check tomorrow..

Has anybody else come across this?

What pulse width are you using with Mach 3?

Hey steve

That doesn't sound like normal behaviour - I've never seen that problem with my SD2s.

I will try and get some screenshots of my setup for you when I get home tonight.

How have you got PC setup? Have you checked that the 'pulse freq' and 'time in int' are good? Are you using a laptop - some have low switching voltages to save power.

Have you tried swapping the step/direction connections over and reasigning the pins in Mach?

Hi,

With steppers the direction line should be changed from High to low to change direction.

When this is being done you should not be outputting any clock pulses or this will cause a stall.

If the clock and direction lines were crossed over pulsing the direction line would make the motor run though unpreditably.

I would check you have the clock and direction lines the right way round.

Is the axis motion smooth when running?

That doesn't sound like normal behaviour - I've never seen that problem with my SD2s.

no it's certainly not normal mine are CD20's even older than yours


Well thanks for your suggestions & steve - thats just what it sounded like, carrying on with the step pulses while changing direction.

I've found the problem and now have some steppers running both direction without problems (I'm assuming that the ones on the Oac will do the same...


Oh, you're asking what the problem was...

What was the problem? Backlash enabled in Mach3. I swear that's how it came out of the box!

I was checking the veroboard again for the umpteenth time and I noticed that I'd stuffed in the wrong resistors (not very good at this am I?)
I changed one on the direction line to see if it made any difference - nope diddly squat. So at present I've left it.

Piccys attached to show it going together & I'll add a drawing of the veroboard (with the copper breaks ) so the electronically challenged can follow (and I'm not e.c. )

The odd bit, I think, is the power units.
On the left is the 24volt dc supply for the digiplan drives, which I don't think is original Orac (that's what I got in the box)

Centre is the power supply for the breakout board - 5v some breakout boards come with their own power supply built in so for some that might not be necessary.

Right hand side is the stepper motor power supply, rather over the top I think Mr. Denford. (no doubt you had a good deal for them ) rated at 2 x 7.5Amp & 45v ac driving 2 helpless little stepper motors rated at 2.1A

Well I'm off to contemplate the next stage.

Oh - BTW Dave what's the max feedrate you can reliably get with your set up using M3?

Hi Steve

A couple of things -

Glad to see you got the system going - great feeling ain't it!!!

Right hand side is the stepper motor power supply, rather over the top I think Mr. Denford. (no doubt you had a good deal for them ) rated at 2 x 7.5Amp & 45v ac driving 2 helpless little stepper motors rated at 2.1A

Shouldn't that be 45V DC? High voltages are great for stepper motors, and the high current side is very useful when you want to upgrade

Currently I have X set at 700mm/min and Z at 1000mm/min. That is to ensure ultimate reliability. I have had the Z upto 1800 and the X to 1400. I actually slowed the X down because it was quite difficult (and scary!!!) to move the table over that small distance at any form of high speed. I am running acceleration at 180mm/s/s which equates to full speed in about 0.05s.

Won't putting in the wrong resistors seriously limit the current available at the transistor and may lead to switching problems?

Also I think the difference between the CD20's and SD20's was that CD were rack mount.

Nice looking job - keep us posted on how you get on - I'm off to source a rotary table for the 4th axis of the mill!!!

Cheers

Dave

Shouldn't that be 45V DC? High voltages are great for stepper motors, and the high current side is very useful when you want to upgrade

Nah, I was talking about the power supply, & rectified as you probably know it squirts out (technical term that ) 63volts dc measures about 60 on my toy multimeterm which incidentally I smoked today checking out the original VFD but the lack of multimeter shouldn't lessen my electronic expertise in the least

But the OTT transformer, as you say, is brilliant for upgrades, the ones in the easimill & easiturn also had 18v taps as well - great for relays

Won't putting in the wrong resistors seriously limit the current available at the transistor and may lead to switching problems?

Well yes I would have thought so, but my inability to decipher simple transistor specifications (and resistor colour codes) means that they work in my ignorance

It's on my list to do, but I work on the principle of it ain't broke so don't fix it.


The VFD (Variable Frequency Drive) the original Parametics ones are now about 25 years old and I guess if they work use them...but and there's always a but.. mine refuses to move faster than 5volts applied to it and then it smelt rather & in checking it out as I said above I smoked the meter, so I've decided it will probably be cheaper to find a more modern one..

Oh if you get a basket case and start comparing pictures of the drive in the manual with what you have you will see that the choke is missing. It took me 4 hours to work out the wiring and decide to ignore the choke for now, then I found it inside the left hand end cover thanks Mr. Denford! I couldn't find any reference to this in the wiring diags

I've been mailing Arturo at CNC4PC.com who (amongst others) sells a spindle controller and has a ribbon cable that will go from the break out board to the back of the Orac case & fit in the original serial port hole. BTW they're 24 inches long. These come out cheaper than I can buy the bits!

Right I'm off to find some relays and a vfd and a mm

Oh I forgot - Early experiments with a DTI on the Z indicate that 2500mm/min seems reliable, but I'm sure I'll drop this in use

OK, further developements on the VFD.

I got to thinking that if if will work happily with 5volts applied it should work with the 10volts applied...

It turns out that a mix of too much preload onthe spindle & too much grease was causing overload on the drive. Without the polyvee belt attached the motor worked just fine. Even after rebuilding the spindle I had to increase the boost pot on the parajust unit, I've had it running for a few hours and all seems fine, just waiting for spindle board now.

A note for those that do have to pull an Orac spindle apart - there is only one grub screw hole that holds the poly vee pully on the spindle BUT take the first grub screw out and there's another one below

Just out of interest, probably one for the old timers, Mr. Magoo, (I'm not saying you're old ) why was the choke for the parajust unit snipped off the drive unit & stuffed in the lefthand cover? just seems like creating work. I guess there was a good reson for it.



Spindle speed recognition

This is needed for any threading you intend to do and the nicely fixed unit in the Orac can be used. This original unit actually has 2 sensors on it, one is aimed at a cut out on the spindle disc and the other aimed at the circular array of slots. Mach only needs the one pulse per rev. It can be used with the slots but I can vouch that it is not needed, besides there is a limit to howmany pulses Mach can "see" per second (I know this 25 year old machine used it. It seems technology hasn't moved on in this case)

This exsisting unit has 4 wires, green, red, blue & yellow.
green = 0v
red = 5v
yellow = signal

The blue connection is not used as this is the signal from the other sensor that "sees" all the slots

In simple terms just connect the green to -ve and the red to +ve 5volts and the yellow to the breakout board, using one of the available input pins, then setup Mach3 to recognise this input as being spindle index.
[config/ports & pins input signal tab]

Oh, and check [config/general config] index debounce is set to zero. This effects the reading if set high, its use is to remove spurious signal by altering the sample rate (or something like that) anyway just do it

If you're using the original parajust unit the maximum frequency it outputs can be adjusted by a variable resistor mounted on one of the boards, I've not tried on mine yet as it's not set up yet. If tweaking up the max speed check that the chuck is suitable for the increased speed otherwise it could be very nasty.....

and here's a picture of the parajust unit - cutting edge technology 25 years ago

Hi Steve

Looking good and great news on the VFD front. A mate of mine is trying to sort out his parajust unit at the moment so he may be in touch

With regard to the index pulse v timing, you may want to consider hooking up the timing pulse for use in closed loop spindle control. The parallel port should be able to read in the pulses at this sort of rate.

Cheers

Dave

Hi Dave,

I don't believe that a timing pulse is necessary to control the spindle speed while threading. In fact my Easiturn works fine threading with an index pulse.

I've just done a quick calc, at 2000(ish) rpm and with 48 slots as the Orac has, that makes 1632 pulses/sec, I agree the parallel port should handle that but it means Mach is more occupied counting rather than doing something useful.

When things are set up (it might well be a while though) I'll try with a timing pulse and see if I notice any difference. I might well do as I'm only using a 286 - no I joke I think it's a 600Mhz something

steve