Simple circuit for RS232 > 12 bit DAC?

[BR]

Any schematics of this? I want to control only one motor from a PC. What I
need is high resolution control of velocity. Position isn't necessary. The DAC
will connect to the control voltage input of an H-bridge that has a built-in
PWM generator. The motor is a small Kollmorgen Servodisc.

Ben

--

 

[Bob Masta]

On Fri, 3 Oct 2003 05:11:25 -0700, "BR" <slackin@nonsense_comcast.net>
wrote:

Any schematics of this? I want to control only one motor from a PC. What I
need is high resolution control of velocity. Position isn't necessary. The DAC
will connect to the control voltage input of an H-bridge that has a built-in
PWM generator. The motor is a small Kollmorgen Servodisc.

Ben

How about the parallel port instead? If you are running under

DOS (so you can write directly to the port) you can use a simple
R-2R ladder. You might need to buffer the digital outputs on
some ports if their levels aren't uniform on all bits. If you only
use the data out port you just have 8 bits, but you can also
use the control port to get another 4 if you're careful. (The
control port bits will almost certainly need buffering since they
have pull-ups.)

For the 8-bit version you can check out the LPTX driver
Help files for my Daqarta system at
www.daqarta.com/lptx.htm
This discusses using the port for analog in and digital out
as well as analog out, so read through for the DAC section.

Hope this helps!




Bob Masta
dqatechATdaqartaDOTcom

D A Q A R T A
Data AcQuisition And Real-Time Analysis
www.daqarta.com

 

[John Fields]

On Fri, 3 Oct 2003 05:11:25 -0700, "BR" <slackin@nonsense_comcast.net>
wrote:

Any schematics of this? I want to control only one motor from a PC. What I
need is high resolution control of velocity. Position isn't necessary. The DAC
will connect to the control voltage input of an H-bridge that has a built-in
PWM generator. The motor is a small Kollmorgen Servodisc.

---
The least hardware-intensive way to do it would be to use a small ľC to
capture the serial data from the PC and bit-bang it into an I2C DAC.

Can you do it that way or does it have to be an all-hardware solution?

--
John Fields

 

[BR]

Thanks Bob. I later realized how slow the serial port is. I forgot to check
this thread and started another. Looked around and found a 12 bit DAC such as
the MAX508. Should make it easier (I think) to get going since its interface
is a byte wide (8+4). But for starters it'll be an 8 bit DAC or a R-2R ladder
as the experiment may work with that resolution.

A control port will be used for the foward/reverse ports on the H-bridge.

Ben

--

"Bob Masta" <no_spam@aol.com> wrote in message
news:3f816508.2365069@news.itd.umich.edu...

On Fri, 3 Oct 2003 05:11:25 -0700, "BR" <slackin@nonsense_comcast.net
wrote:

Any schematics of this? I want to control only one motor from a PC. What I
need is high resolution control of velocity. Position isn't necessary. The
DAC
will connect to the control voltage input of an H-bridge that has a
built-in
PWM generator. The motor is a small Kollmorgen Servodisc.

Ben

How about the parallel port instead? If you are running under
DOS (so you can write directly to the port) you can use a simple
R-2R ladder. You might need to buffer the digital outputs on
some ports if their levels aren't uniform on all bits. If you only
use the data out port you just have 8 bits, but you can also
use the control port to get another 4 if you're careful. (The
control port bits will almost certainly need buffering since they
have pull-ups.)

For the 8-bit version you can check out the LPTX driver
Help files for my Daqarta system at
www.daqarta.com/lptx.htm
This discusses using the port for analog in and digital out
as well as analog out, so read through for the DAC section.

Hope this helps!




Bob Masta
dqatechATdaqartaDOTcom

D A Q A R T A
Data AcQuisition And Real-Time Analysis
www.daqarta.com

 

[BR]

John, I made the mistake of considering RS232 as it's too slow. It'll be via
parallel port. What is an all-hardware solution? I'm using Quickbasic on an
old 133 machine. Just a bunch of loops to change velocity and will need to
edit those values frequently. If the experiment (fluid study) shows a result,
the program can be run again to repeat the result. It would be ideal to
manually control motor speed with a pot and ADC (looking for certain fluid
activity) while writing control data to disk, and then later playback the
data. But this would be a little ahead of myself at this stage. Bet Quickbasic
sucks for that.

Ben

--

"John Fields" <jfields@austininstruments.com> wrote in message
news:fcr2ovs35hr9bemsl3ljigg43u85e7qn5u@4ax.com...

On Fri, 3 Oct 2003 05:11:25 -0700, "BR" <slackin@nonsense_comcast.net
wrote:

Any schematics of this? I want to control only one motor from a PC. What I
need is high resolution control of velocity. Position isn't necessary. The
DAC
will connect to the control voltage input of an H-bridge that has a
built-in
PWM generator. The motor is a small Kollmorgen Servodisc.

---
The least hardware-intensive way to do it would be to use a small ľC to
capture the serial data from the PC and bit-bang it into an I2C DAC.

Can you do it that way or does it have to be an all-hardware solution?

--
John Fields

 

[John Fields]

On Mon, 6 Oct 2003 10:40:02 -0700, "BR" <slackin@nonsense_comcast.net>
wrote:

John, I made the mistake of considering RS232 as it's too slow. It'll be via
parallel port.

---
At 9600 bits per second you can accumulate 12 bits in 2.083ms, including
the overhead of the RS-232 start and stop bits and the wasted four bits
of data, so that's doesn't seem too shabby for a motor control
application!-)

What is an all-hardware solution?

---
For RS-232, basically a comparator or something like Jim Thompson's
MC1489,(just thought I'd mention it to save him a post!^) or one of the
zillion chips Maxim's got to change the bipolar RS-232 signal into a 0
to 5V logic signal, two 8-bit shift registers, a clock, some glue logic
and your DAC or R-2R ladder.

For the parallel port, it's a couple of HC373's or 375's and your DAC
or R-2R ladder and probably a little glue logic if you can't control the
registers with the port's control pins.
---

I'm using Quickbasic on an
old 133 machine. Just a bunch of loops to change velocity and will need to
edit those values frequently. If the experiment (fluid study) shows a result,
the program can be run again to repeat the result. It would be ideal to
manually control motor speed with a pot and ADC (looking for certain fluid
activity) while writing control data to disk, and then later playback the
data. But this would be a little ahead of myself at this stage. Bet Quickbasic
sucks for that.

---
Au contraire!

I've used QuickBasic a lot and while it's not assembler, it's not bad.

--
John Fields

 

[BR]

"John Fields" <jfields@austininstruments.com> wrote in message
news:kvb3ov00vak6kmmccg0io6gjhl5a6pr47j@4ax.com...

On Mon, 6 Oct 2003 10:40:02 -0700, "BR" <slackin@nonsense_comcast.net
wrote:

John, I made the mistake of considering RS232 as it's too slow. It'll be
via
parallel port.

---
At 9600 bits per second you can accumulate 12 bits in 2.083ms, including
the overhead of the RS-232 start and stop bits and the wasted four bits
of data, so that's doesn't seem too shabby for a motor control
application!-)

It depends on what you want to control. Might as well go with SPP being
50Kbytes/sec or better. The more throughput, the smaller the steps can be
during fairly rapid changes in velocity. I ordered a surplus S6M4H servodisc
"pancake" because it's a DC brushed type with zero cogging. It's design is
amazing to me.

http://kmtg.kollmorgen.com/products/product_literature/product_brochure/pdfs/KOL1037.pdf

What is an all-hardware solution?

---
For RS-232, basically a comparator or something like Jim Thompson's
MC1489,(just thought I'd mention it to save him a post!^) or one of the
zillion chips Maxim's got to change the bipolar RS-232 signal into a 0
to 5V logic signal, two 8-bit shift registers, a clock, some glue logic
and your DAC or R-2R ladder.

For the parallel port, it's a couple of HC373's or 375's and your DAC
or R-2R ladder and probably a little glue logic if you can't control the
registers with the port's control pins.
---

They do have zillions and ran into a 12 bit DAC with byte wide loading, like
MAX508. Didn't know they made them. Maybe easier to use?
http://pdfserv.maxim-ic.com/en/ds/MAX507-MAX508.pdf

It's been a while since I've looked into this stuff and really need to move
beyond the CMOS 4000 series

Tanks

Ben