interface real world to pc question?

R

russ

Jan 1, 1970
0
Hi

Any help on this would be appreciated.

I want to read in an analogue voltage from the 'world' into my
computer, process and store it at regular intervals. I want to do
this via the serial port. I therefore need an 'interface' where my
voltage is an input and the output feeds into my serial port. The
interface and my pc then interact via some C code or something. Does
anybody recommend any very cheap 'interface' which would do this?
There are obviously very expensive 'data acquisition cards' on the
market, but my needs seem very simple in comparison.

Russ

J

Jonathan Kirwan

Jan 1, 1970
0
I want to read in an analogue voltage from the 'world' into my
computer, process and store it at regular intervals. I want to do
this via the serial port. I therefore need an 'interface' where my
voltage is an input and the output feeds into my serial port. The
interface and my pc then interact via some C code or something. Does
anybody recommend any very cheap 'interface' which would do this?
There are obviously very expensive 'data acquisition cards' on the
market, but my needs seem very simple in comparison.

Obvious questions will include:

What is the range of your input voltage?
Does it need to be isolated from the PC?
How fast will the sampling be?
How much precision is required?
Single-ended or differential?
How is the signal conveyed?
What accuracy is required?
What kind of powering for the 'interface' is desired?

....

There is an example of one possible circuit at:

http://www.elecdesign.com/Files/29/1447/Figure_01.gif

That example powers itself from the serial port pins.

Jon

B

Bob Masta

Jan 1, 1970
0
Hi

Any help on this would be appreciated.

I want to read in an analogue voltage from the 'world' into my
computer, process and store it at regular intervals. I want to do
this via the serial port. I therefore need an 'interface' where my
voltage is an input and the output feeds into my serial port. The
interface and my pc then interact via some C code or something. Does
anybody recommend any very cheap 'interface' which would do this?
There are obviously very expensive 'data acquisition cards' on the
market, but my needs seem very simple in comparison.

If you really want to use the serial port, and you are
running Windows, there is an inexpensive board from
Dataq for $US 25. If the current model is like the older ones, they are 4-channel units that sample at a few hundred Hertz max. Comes with software for a chart recorder. There are also pricier products from a number of suppliers, which you could probably locate by Googling on "Serial port" and "analog input" or "data acquisition" or some such. If you are running DOS and don't mind using the parallel port instead, you can make a simple 8-bit A/D from nothing more than a handful of resistors. It uses the "successive approximation" method, just like in "real" A/D converters. The LPTX driver for my Daqarta package uses such a design, which you can find at www.daqarta.com/lptx.htm Look for the SAR (Successive Approximation Register) model. Hope this helps! Bob Masta dqatechATdaqartaDOTcom D A Q A R T A Data AcQuisition And Real-Time Analysis www.daqarta.com C Chris Jan 1, 1970 0 russ said: Hi Any help on this would be appreciated. I want to read in an analogue voltage from the 'world' into my computer, process and store it at regular intervals. I want to do this via the serial port. I therefore need an 'interface' where my voltage is an input and the output feeds into my serial port. The interface and my pc then interact via some C code or something. Does anybody recommend any very cheap 'interface' which would do this? There are obviously very expensive 'data acquisition cards' on the market, but my needs seem very simple in comparison. Thanks in advance Russ Hi, Russ. You've already gotten some good advice from other posts. However, if you're looking for something Q&D that'll do the job with absolute minimum fuss at moderate cost, and you can live with fairly slow readings (seconds apart instead of milliseconds/microseconds), you might want to look at the Radio Shack PC Interface 46-Range Digital Multimeter Catalog# 22-812 for$69.99 USD. You get an adequate
multimeter and software which will allow you to do datalogging without
programming just by hooking up the included cable from the DMM to your
serial port and running the program. This will give you the capability
to read analog DC and AC voltages and currents as well as resistance,
and temperature with an RTD or optional Extech temperature adapter
module Catalog# 910-5611 (an additional \$39.99 USD).

Good luck, and have fun
Chris

L

Lawrence Oravetz

Jan 1, 1970
0
Jonathan Kirwan was thinking very hard :
Obvious questions will include:

What is the range of your input voltage?
Does it need to be isolated from the PC?
How fast will the sampling be?
How much precision is required?
Single-ended or differential?
How is the signal conveyed?
What accuracy is required?
What kind of powering for the 'interface' is desired?

...

There is an example of one possible circuit at:

http://www.elecdesign.com/Files/29/1447/Figure_01.gif

That example powers itself from the serial port pins.

Jon

Something else to consider.. a volt meter that has a serial interface
built in. There are many available from Radio Shack to Fluke.

S

[email protected]

Jan 1, 1970
0
cat# 220-0812. I wrote an interface for myself, that I'd like to share
with you. The requirements are unix and a copy of gambas, and you get
the full source code.
Note that, I found, that, for whatever reason, the instrument sends
wrong checksums every once in a while - you can see the number of
resets. I know it's the instrument's or serial connection's fault,
because when displaying 0.F MOhms overrange continuously, I know
nothing changes, yet the correct 9A checksum from the device changes to
12 and 0, and other values, about every dozen 9-byte packet. This also
messes up the byte-order. Doing a port reset, disconnect reconnect
resyncs things. When there is a real resistance measured there are
almost no resets. Another sensitive knob position with lots of resets
over checksum errors is the temperature-mV one, when values widely
fluctuate. Still, with a capability of 3 readings per second, you can
just ignore any off-values and keep the rest, the device still
functions very well, for the price. You can guess some costs were
saved. Instead of summing checksum, they could have provided a CRC8,
and made the packet 10-byte with an STX,ETX or LF char signifying end
or beginning of packet, so there is never an issue trying to figure out
which is the starting byte, and there would be no need for resets. Also
some hardware handshake option would have been nice, because DTR/DSR
and RTS/CTS are just dummy looped back to each other. But still a fun
little device, and excellent cost, considering it's from a
brick-store-in-your-neighbourhood retailer.
Grab the program at
you may view a screenshot (377KB png image) at

J

Jamie

Jan 1, 1970
0
cat# 220-0812. I wrote an interface for myself, that I'd like to share
with you. The requirements are unix and a copy of gambas, and you get
the full source code.
Note that, I found, that, for whatever reason, the instrument sends
wrong checksums every once in a while - you can see the number of
resets. I know it's the instrument's or serial connection's fault,
because when displaying 0.F MOhms overrange continuously, I know
nothing changes, yet the correct 9A checksum from the device changes to
12 and 0, and other values, about every dozen 9-byte packet. This also
messes up the byte-order. Doing a port reset, disconnect reconnect
resyncs things. When there is a real resistance measured there are
almost no resets. Another sensitive knob position with lots of resets
over checksum errors is the temperature-mV one, when values widely
fluctuate. Still, with a capability of 3 readings per second, you can
just ignore any off-values and keep the rest, the device still
functions very well, for the price. You can guess some costs were
saved. Instead of summing checksum, they could have provided a CRC8,
and made the packet 10-byte with an STX,ETX or LF char signifying end
or beginning of packet, so there is never an issue trying to figure out
which is the starting byte, and there would be no need for resets. Also
some hardware handshake option would have been nice, because DTR/DSR
and RTS/CTS are just dummy looped back to each other. But still a fun
little device, and excellent cost, considering it's from a
brick-store-in-your-neighbourhood retailer.
Grab the program at
you may view a screenshot (377KB png image) at
maybe you have your parity set incorrectly ?

Replies
0
Views
296
Replies
2
Views
1K
Replies
5
Views
724
Replies
2
Views
973
Replies
8
Views
1K