That would certainly work for a DC motor. Multiwinding AC motors are a bit different, but it can be done.
I wasn't intending to design it sight unseen, but wanted to ensure that the issue was addressed. And that software was properly distrusted.
Joe Gwinn
Yep. I would insist, not just prefer, unless it could be shown that fetching up against a stop was harmless.
I wasn't trying to design it, but wanted to make sure people didn't trust software too much.
Joe Gwinn
Don Foreman fired this volley in news: snipped-for-privacy@4ax.com:
In which case, no position sensing would be necesary, at all.
If the curtain hits a limit, it just stops at a mechanical limit switch. If the controller senses that the temperature stays a certain amount past a warning limit either way, and isn't making an excursion toward the desired temperature, it signals an alarm saying it can no longer cope.
That's a good idea, Don. Dead-simple, too. No PLC is necessary if the out-of-control alarms aren't needed: Just a timer and temperature comparator with a "safe range of temperatures" dead-band. Wait a period of time, and decide to move the curtain, or not, then do it again, over and over.
LLoyd
I looked at some of the instructions on the Parallax website and it refreshed my memory. You can address memory as bits, bytes, or words, so 16 bit math should be OK, but on limited words. The specs for the BS2 say 32 bytes ram, 6 for I/O & 26 for variables, I guess you can have up to 13 word variables.
RogerN
Right.
It's just my understanding and expression of Karl's idea.
Karl suggests using two t'stats rather than a temp sensor. No A/D necessary -- although the 14-pin PIC16F684 ($1.96) does have on-chip A/D. Again, KISS. All that would be needed would be the PIC chip, some transistors and diodes, the T-stats and limit switches and a
5-volt regulator if it runs off a 12-volt battery. Oh, and a few lines of assy code.I think I'd use the A/D and a couple of pots so motor on-time and delay time could be user-variable, even if the temp sensors are t'stats.
I agree with Tim about concepts that are "too simple to work well", but this is so danged easy to try that little would be lost if it disappoints.
Copy that!
That's what I was referring to re: the small controllers like PICs and Stamps.
And - the reason I still prefer the Z-80 for control work. It has the room and power to expand, a very flexible instruction set (Like 16 bit math, for instance) and decades of practical experience to fall back on.
I really do hate painting myself into a corner.
"Things should be as simple as possible -- and no simpler" Albert Einstein
Richard
On Tue, 08 Dec 2009 23:47:38 +0000, the infamous Mark Rand scrawled the following:
Who cares? With your feared AGWK, you'll soon be able to grow tomatoes Over There, too!
-- To know what you prefer instead of humbly saying Amen to what the world tells you you ought to prefer, is to have kept your soul alive. -- Robert Louis Stevenson
What's that Lassie? You say that cavelamb fell down the old rec.crafts.metalworking mine and will die if we don't mount a rescue by Wed, 09 Dec 2009 05:52:45 -0600:
Not nessarly. If you are using a DC motor, just cut the power with the switches. Add a diode across the switch so the motor can reverse off the limit.
That Lassie! What a dog.
Do some of the PIC's have the ability to use external memory(pins used as address and data)? I thought maybe the 17CXXX had that capability, I thought I read it in the specs but never looked at the detail.
Years ago I bought a PIC Servo chipset and built a nice servo control on my solderless breadboard. Once it was tuned correctly it was awesome. The chipset used one PIC to read the encoder and another did the 32 bit math, PID, and trajectory control. It was supposed to be capable of tracking a
500 line (2000 counts in quadrature) at up to 15,000 RPM. And today PICs have a lot more RAM, and program memory.I have a PIC 16F877 and a PICBASIC PRO compiler, I need to start playing with, maybe make my own robot similar to the BOE BOT my son's getting for Christmas.
RogerN
The requirement I didn't mention was a lot of torque, in the 100's of ft-lb. range. That says gearbox, efficent one. Cycloids are about
96%, 2-stage about 90%. Horsepower requirement was low, speed extremely low, torque high. Worked well.The torque was caused by several feet of o-ring plus web drag at pressure with the viscous hot (230C) material being filtered. It was an unusual requirement, but definitely one for a good gearbox. Actually, the gearbox was a tiny fraction of the overall cost on this beast.
Pete Keillor
Another approach (although the 12 volt thing might be an issue) are regular temperature controllers. They're available pretty cheaply, take most any sensor input, and can have on-off, time proportioning, or full analog outputs. Time proportioning might be the ticket. You can limit the on-time per cycle, and set the cycle time to what you want.
Pete Keillor
Not many PICs have external memory, but there are some PICs with multiple kilobytes of RAM -- which would be a reason to go beyond a Basic Stamp.
...
Roger, if my prototype works out, I need to use a less expensive chip than the basic stamp. I assume the compiler you have uses a less expensive chip. What will I have to do to transfer a program? Start all over? Is there a learning curve here? I ask because I have time to learn in the winter, not in the summer.
Karl
For a production part see...
As for moving to a different processor... Well, yeah, you do sorta have to start over. But at that point you would have a working model to base the new system on.
These little guys (heck, ANY computer) architecture is everything. Instruction set, how Input/Output is accomplished, Interrupt capability, etc All effect how the code gets put together.
That's one of the selling points for higher level languages like C or BASIC. Supposedly it's easier to transfer the application to a new processor.
Richard
How much is the Basic Stamp? The PIC processor itself can be had for a couple of bucks in small quantities from Digi-Key. You do, of course, still need to put it all on a board whether it's a Basic Stamp or a PIC.
Unless there are tools out there to 'roll your own' Basic Stamp from a PIC then yes, it'll need translation. And there's a learning curve for newcomers to C or assembly.
On Wed, 09 Dec 2009 10:25:57 -0600, the infamous Don Foreman scrawled the following:
Why not use linear drives and a pleated (corrugated vinyl?) curtain? Stops are built in mechanically.
Put them on elbows (articulations to those of you who are engineers) and vastly increase their travel. Put two on and you can stop it halfway by itself (by activating only one), no extra programming needed.
-- To know what you prefer instead of humbly saying Amen to what the world tells you you ought to prefer, is to have kept your soul alive. -- Robert Louis Stevenson
The PIC Basic Pro compiler is meant to compile PIC Basic programs to use on a PIC. The differences come in because some PIC's have A/D, more memory, more I/O, etc. You wouldn't need to start over but you could use some PIC capabilities that the Basic Stamp doesn't have. I haven't used my compiler much but I did write some code to read the A/D and send it out RS-232 that I was able to view with HyperTerminal in Windows.
My recommendation would be to proceed as you plan for generation 1, see what changes you would like to make for generation 2, and try to get a final product for generation 3. There are ways to upgrade the firmware on products that have been released. If you get to hardware right, you can make better software available and customers can upgrade to the latest and greatest firmware. For my own project idea (a super duper battery charger) I would like to make the source code open so that the greatest programmers interested can provide the code to best utilize the hardware.
As far as hardware, I would recommend to use the PIC and external drives (Relay and H-Bridge). Use a heavy duty PIC and H-bridge on generation 2, and try to optimize hardware and software for the 3rd generation.
RogerN
After the initial control algorithm, there are some other ideas you may want on the final version.
Auto tuning of the temperature control algorithm, select AutoTune and the processor would learn the run time from open to close and how effective it was at changing temperature. Measure the battery voltage and possibly provide charge control or indication using solar battery chargers. Could keep track of charge milliAmp Hours and discharge MilliAmp hours, let the user know if the system gets enough solar charge to keep going or if the battery is going to run down.
The H bridge motor drive could use pull up resistors on the high side switches and the limit switches could be used on low side switches. This would allow the control to see the tripped limit switch, disable further drive in the wrong direction, but allow drive toward the opposite direction (Open L/S disables open but allows close, vice versa). The limit switch connections could be jumpered to disable if other limiting was desired. If a limit switch was open, the high side pull up resistor would turn on an input so the software could know the switch was tripped.
A Display and switches could display temperature and allow adjusting of the setpoint.
Just some detail I thought of that could be nice in the final version
RogerN
PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.