Maker Pro
Maker Pro

ORTEC interface (trump-pci and dual port memory) doesn't work on a P4 pc.

D

DanieleB

Jan 1, 1970
0
Hi everybody.

I'm tryng to connect two ortec interfaces to my PC to control a
spectroscopy system without any success.

the PC configuration is the following:
P4 processor,
2,8 GHz,
1Gb RAM,
2 isa slots
5 pci slots.

The software I use is the latest version (6.0) of Maestro and
Connection's toolkit.

When I plug in the Dual-Port-Memory interface card in the ISA slot, the
system seems to work (it boots), but when using maestro, the data flow
between the pc and my instrument (adcam 92x spectrum master) is
definitively too high, the dead time indicator on the front panel of
the adcam has a strange behaviour (it goes from 100% to 0% without
continuity), and the refresh rate of the acquired spectrum on maestr is
very very slow (sometimes more than 2 secs).

When I plug in th trump-pci 2k, the pc doesn't even boot.

Maybe the problem is that the pc I use is too fast (the clock and the
fsb might be too fast also).

Does anybody have a clue about this?

thanks,
Daniele Bailo
 
DanieleB said:
...

When I plug in the Dual-Port-Memory interface card in the ISA slot, the
system seems to work (it boots), but when using maestro, the data flow
between the pc and my instrument (adcam 92x spectrum master) is
definitively too high, the dead time indicator on the front panel of
the adcam has a strange behaviour (it goes from 100% to 0% without
continuity), and the refresh rate of the acquired spectrum on maestr is
very very slow (sometimes more than 2 secs).

When I plug in th trump-pci 2k, the pc doesn't even boot.

Maybe the problem is that the pc I use is too fast (the clock and the
fsb might be too fast also).

Well not exactly a clue, just a suspicion that the ISA slots go
through various bufferings and emulations, including some
software traps, thus making the whole ISA part not quite
usable; or some longish bus cycle might time out and cause
some exception which could crash - or just confuse - the system.

Generally todays PCs are not really suitable to build hardware
for, this is why I keep our systems with an explicitly "lab-designed
add-on"
friendly interface and let them connect to the rest of the world
through the net or through other, standard and popular interfaces (USB
etc.).

Dimiter
 
D

DanieleB

Jan 1, 1970
0
Thanks for your answer Dimiter.
I explain you a little bit more about what I'm doing:
i need to control an xray instrument which has an ortec detector and
some motor to move the detector.
I'm developing all the system in labview, and finally I would like to
control everything form the web (it's quite simple with labview), but
that's why I need a good PC.

So your idea is interesting, and more or less I was thinking the same:
use a quite-old PC ONLY to control the system with a basic software,
and locally connect to it (via ethernet) with a 'good' pc, who has a
complicated sofware which can do many things, a web server and all the
stuff I need for my project.

So, could you explain a little bit more what do you mean when you say
*"this is why I keep our systems with an explicitly "lab-designed
add-on" friendly interface"* ?

do you use labview? how do you connect via usb?

thank you,
good bye.
Daniele Bailo
 
D

dp

Jan 1, 1970
0
So your idea is interesting, and more or less I was thinking the same:
use a quite-old PC ONLY to control the system with a basic software,
and locally connect to it (via ethernet) with a 'good' pc, who has a
complicated sofware which can do many things, a web server and all the
stuff I need for my project.

This sounds like your most viable option, it must be easy to find
some old PC (386 would probably be best, later models used to
have longer ISA bus cycles if I remember correctly - but I never
was too familiar with wintel PCs).
So, could you explain a little bit more what do you mean when you say
*"this is why I keep our systems with an explicitly "lab-designed
add-on" friendly interface"* ?

I mean our Nukebus interface. It comes on our Nukeman series devices,
which come with an OS of their own etc., one of these would do the
perfect job for you but they are not cheap (I have not yet managed to
even
come close to matching the prices of the computerized office machines
mainly because of the low quantities).

If your whole HPGe system were a Nukeman with a hi-spec module,
then this would probably have been your cheapest option and
you would have an easy ride making some small module and
attaching it to the system (that sounds more like a self advertisement,
but it is true nevertheless :). Since you seem to be stuck with
the PC world, I guess an old 386 is your best bet.

Dimiter
 
D

DanieleB

Jan 1, 1970
0
Ok.
Thank you Dimiter, I finally found the solution and I suppose that the
discussion with you helped me in thinking correctly about this issue.

Fortunately, and I'm astonished that nobody I talked with underlined
this possibility, ORTEC produces a software (Connection's toolkit)
which has drivers to make their instruments work on suitable pc's
(Pentium3 at most, that's my conclusion after talking with the ortec
support and trying and trying...) and a very very useful utility: MCB
SERVER.

That's a server that allows anyone who has the permissions to CONTROL
ORTEC'S HARDWARE installed on your PC.

So my final configuration is:
-xray instrument acquisition system connected to
- the ortec's cards installed on a P3.
- a Pentium4 to get control of ortec's devices installed on the
previous PC
- a switch to make a lan.
- ortec MCB server installed on the P3

that's all.

see you.
Daniele Bailo
 
Top