- Nov 29, 2004John,
That is also my goal to control a Brother machine by a modern
computer and possibly under windows. You are right that the market
for such a device in monopolized. I think it is a shame that there
are tons of wonderful Brother machines that could be used and enjoyed
more. I have a lot of time invested in CNC design so it cannot be
I don't have the machine yet -it is being shipped to me as we speak
but I imagine there is a stepepr motor (I don't think they would use
servo but who knows, maybe they needed so much precision that Brother
put a servo in it- that would complicate the matters a bit). The
stepper motor has a driver that takes some step and dir comamnds. Of
course Brother may have their own proprietary way of doing it, but
stepper (or more) is like a plotter which has basically two steppers.
One would determine the wiring, drive the steppers and... and....jus
dreaming today. Hopefully it won't be neccessary to go that route.
Someone has got to have a schematic. Now that I freed up my time from
a plotter project (lucky me) I am ready to hack the brother knitting
and I am thrilled I found this board and met you here.
In the past it was always difficult for me to control an orignal CNC
machine from an original controller because they used proprietary
interface which they never reveale- what you determined is a HUGE
step forward. The simplest way was to replace the electronics
(drivers) with modern ones and control everything from a parallel
port using a step/dir program, which are plentiful. I don't know the
currents in steppers but two drivers cost about $100, plus power
supply and box, call it $150. Cable would be standard, $5. Compare it
to $450 for a designaknit + $100 cable and the savings are obvious.
Of course a small detail would have to be figured out what to send to
that parallel port but a Basic program in DOS would be the first
quick and dirty way. I am not yet familiar with the workings of a
knitting machine but can imagine it is not different thatn a printing
circuit board being drilled under a string of commands?!
Something like: turn the first motor 2 steps in direction A, then
turn the seciond motor in direction B by 5 steps etc.
When the first electronic machines appeared (16 years ago?), it was a
dark science but now many things have been figured out by hobbysts.
For that, one could go several ways, first in DOS, one could use
Basic to send a control file to the knitter. Or use a modified
TurboCNC which has a source code for Pascal. Any other cnc program
manufacturer would be thrilled to expand their market and eat into
designaknit's chunk of action. I would like to help to resolve this
unhealthy situation. And yes, you are write, designaknit has figured
it out and they were not born with that knowledge.
Have you determined if Brother or what's left of that division is
willing to share teh interface information? I would say they should
be very open to that because it would increase the interest with
their other products. I did a quick search but found nothing yet.
I had a similar story recently, slightly off knitting subject. I
acquired an old calcomp plotter which didn't have the drivers and
usual search revealed I would have to go to w95 or possibly DOS.
Finally I located a program www.winline.com in Australia ($200). A
USA dealer contacted me and convinced me that without the winline
program I won't be able to control my plotter. After one week of
trying I found out that AutCAD 2000 does have calcomp driver and it
plotted beautifully. Then the problem was how to plot schematics? For
that I figured out that schematic capture has a dxf export utility
and AutoCAD can open dxf. The whole process takes few seconds and I
can plot from a calcomp any schematic I want to any size which was
the idea to have a cheap large size plotter. During the discovery
process I learned that lots of company rely on ignorance of
I would say if Designaknit dropped the price to $200 it would make
their product more popular and make the knitting hobby bigger. The
cable shouldn't cost more than $50 on a good day.
I looked thru a pdf document that you posted and noticed there was no
pinout. Have you figured out a pinout?
About the cables: do you suspect they have some electronics built
into the connector(s)? Mike
John, I am looking for a drive as you suggested and so far haven't
been able to locate one, maybe you have one spare or know about one?
I thought I would ask.
If you want to you may send the info to eemikez@...
--- In email@example.com, "jhogerhuis" <jhoger@p...> wrote:
> --- In firstname.lastname@example.org, "mikezcnc" <eemikez@c...>
> > from a program DesignaKnit directly to machine. Here I noticed
> > there are two types of cables avaialble, roughly at $100 each:
> > that allows input of data from the program and the other that
> > but both allow downloading of patterns. I have no idea why would
> > anyone want a cable without a real time input capability???
> Their cables are very expensive. Although these cables do have
> electronics in them, I don't think there is any reason for them to
> cost more than $20 if efficiently manufactured by hand.
> I don't know much about them beyond that. To connect to the
> computer, I use the cable that Tandy provided for the TRS-80 Model
> 100 computer to connect to these drives.
> > Then I saw an article about someone having developed a PDA
> > but I believe it is pretty much a note pad.
> One of my goals is to create a simple pattern storage/retrieval +
> pattern editing program that will run on any PalmOS PDA with a
> serial port.
> > I searched for a protocol to PDD and found NOTHING. I search for
> > service manuals and also nothing. I found a file posted by
> > the file section: I am thrilled. Taht is a very generous gift
> > the person who passed it along.
> You're welcome.
> I'm hoping that this information being out there will jump start
> finding details beyond the protocol, in particular the data format
> of what the machine actually saves to the disk. I can basically
> mimic a km and read/write sectors given this protocol information,
> but I don't know how to write any sensible pattern data yet.
> That's where I'm hoping we can all help each other out.
> Also in the process of doing this I'm sure some test code/programs
> will be developed that will allow doing neat tricks with a PC
> connected to an FDD or a KM. Hopefully everyone will share what
> learn openly.
> IMO, that's the only way this group can work efficiently, is by
> sharing information openly. In fact the creators of DAK and others
> have obviously already determined this information we're going to
> derive. But because everyone keeps it secret it means people spent
> the same effort over and over finding the same things. We're going
> to open it up once and for all.
> > I apologize for blasting all the questions at once but I feel
> > found a brother soal on the subject. Glad to be here and I will
> > checking this forum often. Mike
> Good to have you on the team, Mike.
> -- John.
- << Previous post in topic Next post in topic >>