You mean we have found a way to avoid having to do anything unpleasant to ourselves by foisting the problem on a latter generation. And you think this is a "solution"?
-- We have failed to address the fundamental truth that endless growth is impossible in a finite world.
-- Certainly you must have misunderstood. In the context of Moore's Law: http://en.wikipedia.org/wiki/File:Transistor_Count_and_Moore%27s_Law_-_2011.svg we _have_ addressed the fundamental truth that endless growth is impossible in a finite world by diminishing the size of transistors in order to pack more and more of them into smaller areas as time goes by. This allows us to build more powerful machines which, rather than foisting out problems on future generations, will help us solve some of ours, and theirs as well. Your suggestion that we should make life more unpleasant for ourselves by not taking Moore's Law into account and building computers out of, say, 2N2222's is puzzling
-- Thanks, but I'd prefer some real numbers from Sylvia so that I can work up a hardware solution for her.
A motherhood statement which avoids dealing with any unpleasant details. Could you state the problem clearly please. What is your solution?
You don't like my numbers? Then go with the number from commerical defrost timer.
I would go with 2 pots + 2 A2D. adjust and time it with a micro.
OK, I suppose that would work, though the thing would still be switching voltage which is double its design rating.
Sylvia.
-- I don't care for conjecture, which is all you're offering.
Huh???
-- JF
te:
ong
OK, changed my mind, let do digital.
Let start with perhaps 4 buttons. Two to change cycles per day and two to change durations. Three bits driving 138 (one of eight) to indicate cycles (eight cycles should be plenty). Another three bits to indicate durations such as 5,7,10,15,20,25,30 minutes. I doubt we really want to defrost beyond 30 minutes anyway. So far, we can do it with 10 port pins and 16 LEDs.
Just a quick first draft:
char cycle[8] =3D {1,2,3,4,5,6,7}; char duration[8] =3D {5,7,10,15,20,25,30}; int c_idx; int d_idx;
while(1) { if(port(c_up)) c_idx++; if(port(c_dn)) c_idx--; if(port(d_up)) d_idx++; if(port(d_dn)) d_idx--;
defrost_on(); delay(d_idx); defrost_off();
delay_hrs(24/cycle[c_idx]); }
...
Since I'm in Australia, both the defrost heater and the compressor are rated at 240V. Anything that switches them on and off has to switch 240V.
Sylvia.
It would have to switch half the current it would on 120 VAC, and the relays are the same for either line volktage.
-- You can't have a sense of humor, if you have no sense.
Really? You think the relay can only handle 120 VAC, just because that's what poweres the timer circuit?
-- You can't have a sense of humor, if you have no sense.
Unless you've actually pulled one apart, and determined the exact specification of the relay, or semiconductor device, involved, it's only safe to assume that it's specified to perform the particular task intended for it. It may well be a part that's specified to handle higher voltages, but it would be rash to assume that.
Sylvia.
You've taken one apart?
Sylvia.
Are c_idx and d_idx in non-volatile RAM? I'd prefer my fridge to continue working, without intervention, after a power outage.
There appears to be a defrost-cycle period between each opportunity to change the settings by one. Could take a while.
Sylvia.
:
te:
rote:
long
rf.
..
tDepends on the micro. Some can save to EEPROM. That's why i though about pots in the first place. Pots are non-volatile. We need an A2D to read from thermistor anyway, so perhaps using 3 channels A2D. One port for the defrost heater and one more for the compressor. One more set of decoder/LEDs for temp. settling. So, to do all the fridge control:
PIC32MX256DA206 (using 11 digital ports and 3 analog ports)
100K Pots (x3) 74XXX138 (x3) LED (x24) Thermistor 240V/5A AC relay (x2) etc.
I've taken apart a lot of things and I've never seen a power relay with 120 volt rated contacts. Do whatever you want, I'm done with this thread.
-- You can't have a sense of humor, if you have no sense.
Seriously, PICAXE 08M2 - 3 x 10 bit digital ports, configures and reads as 8 or ten bits with with a single "readadc" or "readadc10" command (IIRC) and 256 bytes eeprom reads and writes with simple single line instructions. PICAXE is the simplest and fastest solution to any one off that needs a little smarts and not too much speed bar none. If you need lots of LEDs one of the bigger PICAXE chips will do.
I have programmed (and still do) PIC's in assembler (very badly - I still do that very badly too), Microchip C, Swordfish Basic, and PICBASIC Pro. For speed, cost effectiveness and ease of producing a one-off PICAXE easily cleans the floor with all of them.
(I don't really like the guy who invented the PICAXE or his business model but it is a good product)
-- We have failed to address the fundamental truth that endless growth is impossible in a finite world.
"Moore's Law" fails when you get to one atom sized conductors (in practical terms we can expect it to fail before then)
"Moor's law" is about one, specific and small situation. By appealing to "Moore's Law" as a general solution you are burying your head in the sand.
-- We have failed to address the fundamental truth that endless growth is impossible in a finite world.
w long
mer
off.
e....
ew
y o swe
it
hA2D
eActually, PIC32MX575L256, 11 digital and 4 analog ports. Just checking if anyone is awake here.
I can do a one-off PIC32 faster than PICAXE. I don't like doing decimal bit maps in BASIC.
ElectronDepot 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.