Denford Software & Machines

Hello

I picked my Starturn up at the weekend. Seems great so far just needs a clean down as it had been sat for a while. Unfortunately I can't get the BBC to work as the floppy drive is knackered It's an old school machine and even comes with a pad of orginal denford programming sheets.

One thing I did get with it is the parts manual - I will pop a copy up here sometime. I have made a few discoveries so far :-

1) Stepper motors - thes are Cetronic HY200-2220-210 A8 motors. Attcahed is the data sheet. As you can see they are 2.1A motors - i.e. the denford literature is wrong!

2) The version I have (July 85 BBC Control) has a different case to later machines, there is no quick release catches on the side, the bottom being bolted on.

3) VERY IMPORTANT - on my version, on opening the headstock I found swarf in there which seemed strange as the headstock and casing is sealed all round with some form of silicone. On investigation I found that if you remove the Z axis stepper you will see a hole that goes right through, under the headstock. THis hole comes out just next to the one that goes down into the electrical cabinet I would HIGHLY reccomend that starturn owners check this and seal the hole if needed. It's only a matter of time before swarf gets into the electronics and lets the smoke out!

Right - back to work

Cheers

Dave

More info -

This hole that lets swarf in is a BIG PROBLEM. Just opened the bottom on mine to look at the electronics. There is swarf all over the place - including on the stepper drivers on the opposite side of the lathe to the hole! I've enclosed some pics below of the problem.

Stepper drivers are only set to run at 1.8A, not the 2A maximum of the board while the motors are 2.1A

I've also traced the step and dir lines on the stepper drivers. Hopefully I may be able to reuse the SD2 drivers, but set at 2A combined with the PC to make a very easy cheap retrofit.

Cheers

Dave

I've got some bad news for you - you won't be able to use the old stepper drivers as they need 12v for step and direction.

But if you're a whizz with electronics then you can make something to pump up the volts. (I ain't)

But then the drives are whole steps and by todays standards seem very clunky..

Hi Triac Whizz

Been doing some digging and found a few interesting things. The drivers aren't full step - they can be configured for either full or half by an onboard jumper. They seem to come setup as half step.

The second thing is that the driver connections are pulled up to 12V as standard. Hence they should only need to be pulled down. I've found a simple circuit that should allow this to be done using a single transistor and resistor. The circuit has been used on other digiplan drives so hopefully it will work. I've built the circuit, but unfortunately I found that the drives fuse has fallen to pieces - the end cap has seperated from the glass - might have occured during the moves over the last few days so i need to order some before I can say much more.

The drives themselves are very sophisticated for their age with things like enrigize lines for use with Estops and buili in oscillators. There is a guy at work who integrated an SD2 to work with a PIC so I will pick his brain tommorow and see how he did it as I would have thought he would have used 5V logic.

Cheers

Dave

Hi Dave, You could still use your drive boards if you use opto isolators on the step and direction pins. A couple of resistors and vero board and maybe use opto isolater 1LQ74 or similar.


Hope this might help..




Kind Regards
John

Yippee - it works - I now have the Z axis whizzing back and forth for the total sum of 74p!!! One resistor and one transistor per signal line and you can drive your 12V digiplan drivers using a 5V TTL signal - in my case from a breakout board rather than the printer port for safety

No I'm going to sit back and do some design work and wire tracing. Need to remove the processor board and strip the relays out (unless someone wants to make me an offer for the complete BBC board - i just want to cover the cost of alternative relays.). From there I can make room for the breakout board, driver adaptor circuitry, relays, new interface for spindle control and some regulators/rectifiers to get the AC going into the main processor into some more useful DC.

If anyone wants to know how to hook up the drivers, post on here and I will sort out a circuit diagram.

Cheers

Dave

David

Thanks for all the info on the starturn. I have 8 Starturn lathes in a CNC lab in a small comunity college and am trying to make them work. We are using Windows 98 computers to run the software in a dos window. They seem to work fairly well but today, when trying to run a thread chasing program (G33) I asked for 20 passes (C20 value), it only cut 13 or 14 passes and the part was undersize. This error repeats, and I think the problem may be with the relatively fast computers and old dos control software. I can't find any documentation on the RS-232 setup so I am using 2400,E,7,2 and no handshake. I wonder if there should be some handshake but can find no info. Anyway, after all this discussion, I am wondering it it might just be better to upgrade all lathes to something like Mach3 and if I do that, your discoveries would be wonderful.

Can I get the schematic of what you are doing?

Karl

Hi Karl

Are you running under windows or booting into pure dos? I would try booting directly into dos to ensure that nothing is interfering with the serial port. Unfortunately I don't have the settings for the PC comms - I only seem to have BBC manuals here.

Have you tried the same program but a different number of cuts/depth of cut. I've had problems like this before with the Orac here at work. Sometimes for no apparent reason it will throw a complete wobbler - I presume that it is some sort of rounding/floating point error. Whenever it does this I find changing the values tends to cure it.

Unfortunately I don't have a schematic as such - I kind of made it up as I was going along. I will try and upload some pictures when I get home.

There seems to be at least 2 versions of the internals. I have a build level 1 machine, this basically seems to mean that the control relays are mounted on the main processor. As a result when I rebuilt my machine I had to source and fit new relays. The build level 2 machines have relays mounted on a seperate board - I suspect you have level 2 machines so that could save you a few quid.

Do your machines have spindle control? Mine did not initially, hopefully that will be sorted in the next week or so. Total cost for my refit is somewhere around £120-£160 - however a lot of this was things like buying 10m of 6 different coloured cables, and then only used 200mm of each! I doesn't bother me as I just add it to my stock!

I will try and sort some bits and bobs out for you.

Cheers

Dave

David

Mine do not have programable spindle control. However the spindle RPM is read out on the display and it does cut threads accurately. Thus I imagine something like mach3 could control it.

I've tried different values and if I program 28 passes it seems to cut 20 pretty repeatablly. Makes no sence at this time.

Karl

Karl,

I wouldn't spend too long trying to understand the number of passes problem - its probably a bug that was never noticed with your versions of software/firmware - I'd just try and work around it by adding 7 or 8 passes to the actual number you require

I think if it cuts a thread at all, its good news!

Hi Karl

Mine didn't have spindle control either - I'm working on that one

Do your machines have little plastic clips either side of them that hold the lid down? If so then they are probably build level 2 machines. If there are no clips on the side then they should be level 1 machines.

Do you need to have all the machines running or could you take one out of service for a bit while you refit it? Have you got any experience of CNC conversions ? You will need to spend some time tracing the wiring and checking what goes where. I take it you relaise that you will need a reasonably good PC to run Mach, ideally dedicated to the machine - somewhere above 1.2GHz should do nicely.

Cheers

Dave

Dave

I have two non functional lathes that I am in the process of trying to get the school to declare scrap so myself and another insrtructor can take them home and try retrofitting them. If we can get them to work under Mach3 control then we would attempt to get all the classroom machines changed over.

Our machines have plastic latches on each side (lower front) that look like they hold the cover down, so we must have level 2 machines.

I have not, yet, open one up to see what's inside.

Last winter I purchased a Harbor freight manual mill ($300) and purchaese a control & motors on Ebay. Then I tore the mill totally apart to cleaned and deburred them accurately measured each piece, put each piece into Autocad solids, and reassembled it. Then I designed the motor mounts and went to a friend's house that has an CNC mill and made my parts. As of now I have a very nice 2 Axis cnc mill running. I have the design almost done for the "Z" and am looking forward to building it this winter. I am using Mach3 on a 700mhz computer with Windows XP running and it seem to work quite well, but my max feed rate is about 6 IPM. This is not a major problem with only 9" x 5" x 3" max travel.

This last summer I got a chance to meet Art Fenerty (of Mach3 fame) at a CNC workshop week in Ill. He is one very impressive guy. I respect what he does and stand back in awe of what he and now Brian have accomplished.

Thanks again for all your info and I'm looking forward word of your further endevors.

Karl

Excellent - if you understand the basics of doing a cnc conversion you are half way there! For info, i've turned my drivers upto 2A from the original 1.8A setting. I've now got the machine set to rapid at 1200mm/min in Z and 800 mm/min in X. Started stalling at 1800mm/min on Z so I turned it down a bit. In case anybdoy gets the impression that the Starturns are toys I've so far tested my machine to 0.25mm deep cuts in aluminium, with a HSS tool, spindle 1000 RPM and feedrate of 550mm/min - yes that is 550mm/min! I've been making brass oil caps for my steam engine recently - 4 different diameters along them and take less than a minute to make

Sounds like your mill is similar to mine. I have a Taiwanese version of that mill which I converted to CNC - I get about 600mm/min (23ipm) using 325oz/in motors and a 2:1 belt reduction. I have the ball screws waiting to be fitted and at the same time it will have its slideways scraped and lapped.

If you have a look in the downloads section, the starturn schematics in there should match the machines you have. Better get back to doing some work now but I will try and post a bit more info later.

Cheers

Dave

I noticed the note at the begining of the thread relating to the motor current on the machine.

The stepper rating of 2.1 A relates to the winding current. If the windings are connected in series then the max current is 2.1A.

If the windings are paralleled up the drive current can be increased to 1.414 x 2.1 = 3A.

Wiring in parallel reduces the torque but smooths the torque curve and increases the speed the motor can be run at.

I am not sure why the drives would have been set at 1.8A unless it was to reduce the torque generated. This may have been done to prevent stalling due to resonance. (resonanse can be a factor on motors wired in series)

Hi Steve - that's very true. The starturn motors are wired in series so hopefully the exta current will make a difference.

Now for some pics of my starturn. Firslty, my cure for the holes into the headstock - good old silcone sealant!

Secondly a view of the elctronics during the testing stage - I need to tidy up the wiring! This was while waiting for the transistors for the stepper interface had been fitted. The processor board has been replaced. At the bottom you have a parallel port breakout card. Middle right is the stripboard circuit for the stepper interface and above this is a small latching relay for spindle on/off using the original buttons. Above this are two AC-DC 'converters'. They take the output from the original ancillary transformer and produce 12V and 5V DC. 12V is used for the relay circuit and passes through the emergency stop button. I've broken the Estop circuit to cut out the limit and guard switches (previously they were in series with the 12V Estop circuit) - they will now be hooked into the breakout board instead. Hence the two extra screened cables you see runing down the main loom.

Thirdly - the interface circuit for the stepper drivers. The input lines to the drivers are held at +12V. Therefore you simply need to pull them down by using this circuit to make them work.

And finally - it does work

Cheers

Dave

Dear Dave,

Thanks for pointing me to your earlier posting & lots of lovely info and circuit diagrams

It looks as if the electronics in my Starturn is very similar to what you had; I think I've also got a series 1 (no plastic clips).

Your replacement PCBs look very nifty. How do you control the spindle speed? [Have you replaced the speed pot?] Does using the parallel port (rather than serial) give better data transfer speeds?

I know it's asking a lot, but is there any chance of the PCB layouts & parts list?

Austin

P.S. Where are you? I'm in the UK near Chepstow.

Hi Austin

At the moment the spindle control is simply using the old buttons and potentiometer - i.e manual. However my electronics arrived at the weekend for the spindle control so whne I have time I will get round to fitting them. I'm using a Digispeed made by Peter Homann in Australia.

I'm guessing that you're not familiar with the way a parallel port conversion works? To start with, i'd suggest that you have a look through the Mach3 manual -
https://www.machsupport.com/documentation/M3M1_84-A2.pdf
That is the mill manual but the basic wiring/setup is the same as the lathe, the difference being in actually using the software

The way the starturn works in its original form is that the BBC send a serial command to the lathe. In the main processor there is a path planning system. This converts the command into a series of step pulse and the state of a direction line which is fed into the stepper drivers.
A parallel port system basically gives the step and direction signals at the actual parallel port itself.

The limit on speed will be based on the processing power of the path planning system. Mach3 can produce a frequency of up to 45Khz at the parallel port - faster than a starturn could ever reach!!!

I will see what I can do in terms of info. Really, the job is primarily re-wiring as opposed to a major refit. The biggest issue with level 1 machines is the need to replace the main relays.

Cheers

Dave

Dear Dave,

Yes, I am new to parallel port conversion; (I'm used to good old-fashioned serial comms and I'm starting on USB.)

The Mach3 manual explained a lot; Digispeed looks very interesting.

Many thanks for your help,

Austin

Right- thought I'd update this thread with some pictures at is largely complete.

First pic shows the electronics. On the right is where the main processor used to be. To give you aan idea of what's what on the new board -
Bottom - breakout board to isolate PC from the lathe
Middle right - lower bit of stripboard is the stepper 12V to 5v interface to allow the old drivers to be used. Above this is a latching relay to allow the old spindle on/off buttons to be used.
Middle left - Digispeed, the spindle control interface.
Top - 2 rectifiers to turn the old ac input to the control logic to 12 and 5V DC
Hidden above these are the two main relays for spindle and emergency stop.

Second pic shows an overall view of the lathe.

Third pic shows Mach software complete with touchscreen (ebay bargain!!).

Hope that provides you all with some inspiration.

Cheers

Dave

hi Dave:

what a conversion!
so the break out board is just for the protection of your PC?
And you can use stuff to control it with mach with any pc right?


please sort me out the detail since my PK2 and the rest are in perfect working order.
Now I understand what minimum wiring and cost is all about.
save so much time
Kind regards

Will