Am 07.10.2015 um 05:32 schrieb Loren Amelang:
>
>> If you do bitbanging, it is sufficient to unload the w1-gpio driver
>> to have the bus unpowered. Part of its cleanup routine is to float
>> the GPIO pin. I don't think there is another reason but an
>> (unlikely) bug inside w1-gpio to have the bus unpowered.
>
> Lost me there. If my 1K pullup is constantly present, and the GPIO
> pin is floating, seems the bus is powered.
>
Usually, a current of 3.3mA is sufficient for supplying a DS18B20 even
during temperature conversion but the Dallas engineers sure have seen
pigs fly when they decided you should activate a "strong pullup" on the
host during conversions. I think they've been afraid of cross-flow
losses between many many parallel contacts which may lower the supply
current from the weak pullup to below the critical 1.5mA.
>
>>> Does OWFS issue POR commands while it is running?
>>>
>> This "POR command" 0x64 David mentioned in his post is actually
>> the "apply trim 2" command. It is only sent when writing something
>> to the errata/trim node
>
> I went back and read the data sheet...
>
A wise thing to do. It's my reference always.
> But randomly,
> maybe once in 5 minutes, totally unrelated to this pattern, I see a
> separate ~4 mS low pulse with no associated interaction.
>
4ms low is a sure reset condition.
> Sometimes it
> looks like two of them almost together, but separate scope triggers
> so at least 50 mS apart.
>
> So that random hit seems to qualify as a power-on reset. As best I
> can see on my old analog scope, it has clean sharp edges, like the
> GPIO is doing it intentionally. Hard to tell if it correlates with my
> 0C or 85C readings. But since OWFS seems to be counting on the
> information setup from 15 seconds ago remaining valid during the
> pause, that would certainly mess things up.
>
Yes. And that is a bug, most likely in the w1 kernel driver. Or, it's an
intentional behaviour of that driver owfs is not yet aware of. I'll
investigate.
>
> But still... The 0C and 85C readings show up as exact discrete
> values in my temperature logs, and do not seem to affect any other
> readings. I checked the logs taken during my scope tests, and while
> the resistor was un-clipped, the recorded temperatures jumped back up
> to their old offsets. Jumped exactly back down with the resistor
> restored. Totally mystifying!
>
Really JUMP? If they CRAWL back, I would check if the sensor is heated
through its DQ wire. Maybe the wire is for some mystical reason a bit
warmer in one setup than the other.
> If I'm going to try something, it would be to change to a USB bus
> connection, which would probably be required to drive my dozen
> sensors and hundreds of feet of cable anyway. This 3.3V w1 test was
> just to learn about OWFS. Any thoughts on which USB master would be
> best for parasitic sensors scattered all over my solar hot water
> equipment, in-between motors and electric valves and PV panels?
>
I found the DS2490 being a bit brittle, too, especially on the USB side.
And it doesn't like shorting the onewire, sometimes it just hangs then
and waits to be power-cycled. Bummer!
Haven't checked any of the other USB devices.
I personally had no problems with the DS2482-800 yet, I use it at a
Raspberry Pi I²C port for my dispenser system. As Colin wrote, the
DS2483 is a similar chip which has only one channel but instead, has a
3.3V level converter built-in, so you can use it for interfacing a 3.3V
board to a 5V onewire bus. SO I would recommend you to use that chip to
interface onewire to your Beagleboard.
Kind regards
Jan
------------------------------------------------------------------------------
Full-scale, agent-less Infrastructure Monitoring from a single dashboard
Integrate with 40+ ManageEngine ITSM Solutions for complete visibility
Physical-Virtual-Cloud Infrastructure monitoring from one console
Real user monitoring with APM Insights and performance trend reports
Learn More http://pubads.g.doubleclick.net/gampad/clk?id=247754911&iu=/4140
_______________________________________________
Owfs-developers mailing list
Owfs-developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/owfs-developers
