RE: [CAD_CAM_EDM_DRO] Re: Pixie error connection

Hi Keith

Yes one of these will probably do the trick, though it seems pointless to
buy an array when you only have one relay to interface to.

Most simple darlington or hi gain npn transistors should do the job, hence I
commonly use MPSA13's because I have bucket loads of them. You can connect
the base via a resistor to the pixie card, 10K resistor would probably be OK
even 100K, most Darlington transistors have gains around 10,000 +. Then
connect the emitter to ground, and collector to one terminal the -ve of the
relay, the other terminal connect to +ve at the relays drive voltage.

Just in case I'll quickly explain how the gain of a NPN transistor works. If
you put a current into the base, that current will flow out of the emitter.
The collector will then try to switch a current up to the value of the base
current x the gain.

Ic = Hfe Ib, where Ic is collector current, Hfe is the gain, and Ib is the
base current.

It's a bit more complicated than that, but in most simple switching cases
the above formulae will hold good. We normally make sure there is a little
more current into the base than we would need on the collector side so the
transistor is fully turned on (saturated) .



Your local store may not have a ULN2803, but they will probably have a small
darlington transistor.

You could also look for a logic level FET, the pixie would drive the gate of
one of these directly.



Regs Mark



Dr. Mark Vaughan Ph'D. B.Eng. M0VAU

Managing Director

Vaughan Industries Ltd, reg in UK no 2561068

Water Care Technology Ltd, reg in UK no 4129351

Addr Unit3, Sydney House, Blackwater, Truro, Cornwall, TR4 8HH, UK.
Phone/Fax 44 1872 561288

RSGB DRM111(Cornwall)

_____

From: CAD_CAM_EDM_DRO@yahoogroups.com
[mailto:CAD_CAM_EDM_DRO@yahoogroups.com] On Behalf Of toolspindoctor
Sent: 15 July 2007 21:02
To: CAD_CAM_EDM_DRO@yahoogroups.com
Subject: [CAD_CAM_EDM_DRO] Re: Pixie error connection



Mark,

Jim Cullins has suggested using ULN2803A which is a darlington array
general purpose IC. I have to wait until tomorrow to see if the local
electronics supply place has something similar. It does seem like the
load is too much for the Pixie when connected to the relay. Could a
resistor somehow be placed to make it work, even by trial and error?
How would I connect such a resistor if you think it might work?

Thanks,
Keith

--- In CAD_CAM_EDM_ <mailto:CAD_CAM_EDM_DRO%40yahoogroups.com>
DRO@yahoogroups.com, "Mark Vaughan" <mark@...> wrote:

>
> Hi Kieth
>
> No problem helping, we all bat things between each other here, that

way we

> all learn.
>
> When the pixies fail they release the enable line to your analogue

cards, so

> if you trust the pixie logic that should adequately shut off all

movement,

> in reality tripping with an error should be a very rare event, on my

machine

> it might occur if I hit a limit which freezes mach and my analogue

card, if

> mach freezes quick enough even this shouldn't trip the pixie. If you

have

> additional Estop limits after your main limits then you are covered

should

> a card fail and motor go wild, I've had this happen with Rutex

cards. If I

> were to wire the pixie error into the Estops, so much power is

pulled that

> no led's would be left active to tell me how the fault occurred.
>
> As to your solid state relay, it is probably just too much load, so the
> relay itself pulls the inputs all too low and they trip into fault mode.
>
> If we knew what they are using on the pixie card for their open

collector

> output then we might be able to add a little more resistor pull up so it
> could drive the relay. I haven't any boards at hand at the moment to

have a

> look, but the tracks aren't easy to track so even then we might have a
> problem. I would just put a transistor on the output, something like an
> MPSA13 darlington transistor and use that to drive a relay.
>
> I have emailed Rick at Skyco for details about the error lines, and

enable

> output lines that feed the analogue cards, but didn't get a response
>
>
>
> Regs Mark
>
>
>
> Dr. Mark Vaughan Ph'D. B.Eng. M0VAU
>
> Managing Director
>
> Vaughan Industries Ltd, reg in UK no 2561068
>
> Water Care Technology Ltd, reg in UK no 4129351
>
> Addr Unit3, Sydney House, Blackwater, Truro, Cornwall, TR4 8HH, UK.
> Phone/Fax 44 1872 561288
>
> RSGB DRM111(Cornwall)
>
> _____
>
> From: CAD_CAM_EDM_ <mailto:CAD_CAM_EDM_DRO%40yahoogroups.com>

DRO@yahoogroups.com

> [mailto:CAD_CAM_EDM_ <mailto:CAD_CAM_EDM_DRO%40yahoogroups.com>

DRO@yahoogroups.com] On Behalf Of toolspindoctor

> Sent: 15 July 2007 16:43
> To: CAD_CAM_EDM_ <mailto:CAD_CAM_EDM_DRO%40yahoogroups.com>

DRO@yahoogroups.com

> Subject: [CAD_CAM_EDM_DRO] Re: Pixie error connection
>
>
>
> Mark,
>
> I appreciate your input! I had envisioned the Pixie error circuit as
> you described. I still don't understand why the error output would not
> be sufficient to drive a DC solid state relay. When I connect the
> error to the relay is just goes into an error state. There must be
> something different about the open collector, which I don't
> understand. Jim Cullins, who designed my breakout board has suggested
> interfacing with an IC with Darlington arrays.
>
> I have all the safeguards with limit switches etc. The system shuts
> down everything during an estop coming from my servo amps or from
> Mach3, the estop switch etc. However, the final protection is to get
> the servo amps to shut off (signal the Mach3-Campbell estop)when the
> Pixies fail. Now of course, the Pixies go into an error condition
> without triggering an estop.
>
> Thanks again,
> Keith
>
> --- In CAD_CAM_EDM_ <mailto:CAD_CAM_EDM_DRO%40yahoogroups.com>
> DRO@yahoogroups. <mailto:DRO%40yahoogroups.com> com, "Mark Vaughan"

<mark@> wrote:

> >
> > The pixie error output is am open collector output, and an input
> with pull
> > up resistors. What this means is each pixie card has a little

transistor

> > switch that switches the line to 0V when there is an error. When
> there isn't
> > an error the pull up resistor keeps the output at 5V, but being a
> resistor
> > it is low current. The idea is that you can wire all the pixie card
> error
> > wires together so which ever one faults pulls the line of all the
> others to
> > 0V which the others see as an input and stop themselves so when it

goes

> > wrong all pixie cards will stop so there will be no axis movement.
> >
> > As to interfacing to other systems, the output can only drive 5V

logic,

>
>
>
>
>
> [Non-text portions of this message have been removed]
>

[Non-text portions of this message have been removed]