No Hot Props
Humanoido
Posts: 5,770
Under normal operations (see propeller data sheet) I'm not seeing any props that become hot or even warm. They all appear to remain at ambient. Are there specific things in code that you can do to warm up these chips other than cog fill? We know that idle is the coolest point. How does this change? It may not be of much concern with one prop, but with hundreds, a small temperature rise could be significant, maybe not to the prop but to the surrounding science equipment.
Comments
jazzed, wouldn't you want to launch your code into each cog? (Assuming it's okay to use all the cogs.) And wouldn't you want to have the rest of the cog doing something too?
I modified your code with the assumption it is okay to use all the cogs.
con '' clock settings _clkmode = xtal1 + pll16x _xinfreq = 5_000_000 ' any other value pub main repeat result from 0 to 6 start(result) finish(7) pub start(pinIndex) FRQA := $8000_0000 FRQB := $8000_0000 CTRA := (%00100 << 26) | (pinIndex * 2) CTRB := (%00100 << 26) | ((pinIndex * 2) + 1) cognew(@entry, 0) pub finish(pinIndex) FRQA := $8000_0000 FRQB := $8000_0000 CTRA := (%00100 << 26) | (pinIndex * 2) CTRB := (%00100 << 26) | ((pinIndex * 2) + 1) repeat dat org entry jmp $It compiles, but I haven't tested it otherwise.
Is there anything to add to make the Prop work harder?
Does it matter if the different cogs' counters use same pins? I wasn't sure so I had each cog use a different set.
Duane
Make sure you don't exceed the package specifications.
Duane
I ran the code on my PEK circuit with nothing on P0 and P1. I'm sure it's just a coincidence but the Propeller chip immediately died.
Out of over a hundred props in my lab, this is the first time anything like this happened. I can't get the prop chip to do anything.
BST can detect the chip and the programs go through load sequence rather fast and then nothing in chip response. Tried ports and nothing. But the power supply shows 17 ma is being consumed. Other programs don't work too. Any ideas?
I feel just sick over this. The props are like my children.
wow, we finally found the "killer application" for the Prop:)
This needs investigating. If you say exactly which code you were running I will risk trying it.
I am unsure about discussing warm Props with somebody called Heater..
Although I did have a giggle when I saw a red light coming on on a board that I knew had no LEDs on it. Turned out I had jammed the EPROM in backwards and the chip was glowing red through the quartz window.
Heater: ROTFLOL!!! I like your humor!
I ran the first mentioned code found in this thread first, and then the second mentioned code followed. Neither code made the prop chip warm. I cannot believe the code killed it. Please check this if you can. I know the prop chip was previously good having run many of the PEK programs, test programs, and button/LED programs.
So was it the second code that killed it? Yicks! I wrote that (well some of it).
I tested to make sure it would compile but I didn't try it on a Prop. I think I'll let Heater try it on his Prop.
Sorry your Prop died. I've managed to kill a few myself.
Duane
"Hello, my name is Alan. I kill Props ....."
I plan to sacrifice a few Propeller DIP chips this weekend in your honor.
don't subject the protection diodes to more than 700 uA per input. We typically see about 80° F after stabilization.
We are also running the chip's at 100MHz but that didn't seem to change the temperature much.
Duane, I wasn't getting a response from the prop after running the first code and so I ran the second code which was yours. I think it was already nop by that time. The reaction was odd from the get-go with the 17ma indicator settling on that number. At first it loaded the power supply as I recall the ammeter reading popped - which I had never seen it jump like that before except in very heavy loading... this is all very strange...
Indeed, and also enough Props to unleash Frankencode for the entire weekend
I was so confident that this code would not kill my Props that I ran it on Blade 1 of my TriBlade board. This has no ext RAM or video circuitry mounted so all pins of the DIP 40 Prop are free.
That Prop is still good.
Are you sure you don't have some harware conflicting with the pins used in that test?
Note also that the the code above using multiple cogs does not perform as advertised. It only uses the counters on COG 0.
I think the version below should lauch jazzed code in each of the eight cogs.
con '' clock settings _clkmode = xtal1 + pll16x _xinfreq = 5_000_000 ' any other value var long stack[350] pub main repeat result from 0 to 6 cognew(Start(result), @stack + (50 * result)) start(7) pub start(pinIndex) FRQA := $8000_0000 FRQB := $8000_0000 CTRA := (%00100 << 26) | (pinIndex * 2) CTRB := (%00100 << 26) | ((pinIndex * 2) + 1) repeat ' keeps the cog alivejazzed original code only used pins 0 and 1.
The above code uses pins 0 through 15.
Heater: Thank you for running the test. That took some courage. I had pulled the circuit off pins 1 and 2 on the PEK so the pins were bare and nothing could interfere as recommended by the original post. This is a circuit that I use all the time for testing with one prop. This prop ran fine with all the previous programs. This remains a mystery. Do we have any good recommended prop chip testing programs in the OBEX or posted that can determine what happened to this chip?