|
|
Monday, February 28, 2005
WX Update: I Moved The USB Adapter to Another Port
I doubt that this will have any effect, but it "cain't hoyt".
I'm still waiting for the parts to arrive so I can build the new patch cables I mentioned in an earlier post. Meanwhile,
the computer AC adapter ground is still floating. It's possible all my problems will disappear when I get the new cables installed.
Or not... we'll see...
WX Update: And Another Lock-Up
Within 30 minutes of the previous one. Maybe whatever it is will just fail outright, and I'll be able to troubleshoot
it more easily. That's the trouble with intermittent failures: You never know when you've fixed the problem; you only know
when you haven't.
WX Update: New 1-Wire Drivers Didn't Help
The 1-Wire network locked up again sometime between 6:40PM and 7:15PM.
Saturday, February 26, 2005
WX Update: Installed New 1-Wire Drivers
Installed new 1-Wire drivers for the weather station. It seems like the drivers for the USB to 1-Wire bridge were locking up, which caused last night's
outage. We'll wait and see how long it goes without any lockups.
The Weather Station Was Off the Air
The 1-Wire USB drivers locked up at about 1:00AM Saturday. I discovered it at about 10:00AM Saturday. The weather charts
are a little hosed up at the moment.
I had converted WServer to run under SrvAny.exe as a Windows NT service a couple of hours prior to the lockup, so it
might be related to that. I'm going to take a look at windows power management settings.
The Heathkit is Back On the Air
I replaced the finals, drivers and emitter resistors all for under $10.00. Everything else was fine, including the stabistors,
which is good because stabistors don't look like a common item these days. At least not the kind they used in the AA-1214.
Sunday, February 20, 2005
My Heathkit AA-1214
In an attack of the mundane, the final transistors shorted in my faithful, 33-year old Heathkit AA-1214 amplifier. I
built it while still in high school in 1972. Interestingly, it failed not while driving speakers at a high level, or into
a short circuit, but while driving headphones -- essentially no load! The headphones are driven by the main power amp through
a resistor network, so even if the headphones shorted, the power amp wouldn't care.
The closest thing I can figure is that the breakdown voltage of one of the finals had deteriorated to the point where
it couldn't handle the full voltage swing anymore. The raw supply voltage would be higher when driving a light load. When
it shorted, it took it's push-pull buddy with it, and possibly the emitter resistors and drivers. I haven't fully assessed
the damage yet. It's harder to analyze the failure modes of a power amp than it is to analyze what's supposed to happen when
it's working properly. I can't just leave it on and measure voltages; the smoke gets too thick and the damage only gets worse!
It looks like I can find exact or equivalent parts for everything except for something called a "stabistor" -- a diode-like
package consisting of three diodes in series, used to bias the finals into A/B mode. I hope I don't have to replace that,
because I'll have to kludge up three 1N914's in series into a space designed for a single package.
It's no longer part of my main hi-fi system, but I want to keep this amp in pristine condition, given it's age, and yes,
sentimental value. All through high school and college, I listened to a lot of tunes through that amp -- many hours with a
many friends I'll probably never see again.
Saturday, February 19, 2005
Weather Station Update
I think I'll build new patch cords to span the distance between the wall and the computer, and the outside jack and the
weather station. The weather station end needs an RJ11 plug with the following pinout:
- Power Vcc
- 1-Wire Data
- 1-Wire Gnd
- Power Gnd
It's a nice, traditional, logical, paired arrangement. Green/red & yellow/black is the standard pairing for CAT3 cable.
This maps to the pairing scheme in my in-house CAT5 wiring system.
Most 1-Wire devices, including the USB MicroLAN adapter, need RJ12 plugs with the following pinout:
- Power Vcc
- Power Gnd
- 1-Wire Data
- 1-Wire Gnd
- N/C
- N/C
The reason they went with the unusual pinout with the 1-Wire network is to reduce crosstalk and capacitance in flat modular
telephone cords. The patch cords I make for connecting each 1-Wire device to CAT5 will need to make the appropriate translation.
Weather Station Update
One of the most serious limitations, in my opinion, is the fact that the 1-Wire electrical protocol uses unbalanced signalling,
unlike Ethernet and RS-422. This makes the tranceivers simpler, and it makes it easier to send power over the data lines,
but speed and noise margins are often significantly reduced, as I seem to be experiencing in my setup.
6:12 pm pst
Weather Station Update
I found some 1-Wire applications in the OWAPI (One-Wire API for Java) that enumerate the devices on the 1-Wire MicroLAN,
and read the data from the devices. It showed the weather sensors (and the USB MicroLAN adapter itself) popping on and off
the network like mad, with tons of CRC errors in the data. I started trying to eliminate the possiblities:
- Unplugging the MicroLAN cable from the USB adapter stabilized the device list, with the USB MicroLAN adapter showing up
consistently. Replacing the MicroLAN cable caused the instability to return.
- Unplugging all of the other USB devices from the computer had no effect. The MicroLAN was unstable.
- Unplugging the AC power adapter from the computer caused all of the weather sensors and the USB MicroLAN adapter to stabilize
(did I mention I'm running this on a laptop?).
Now I have some things to consider:
- Is the AC adapter itself noisy?
- Does plugging in the AC adapter simply ground the computer, and make the MicroLAN cable into a big antenna?
My next experiment will be to try to unground the AC adapter. I may still try replacing the flat-wire phone cables with
CAT3 or CAT5 cable (for this application, either one will work, they're both 100 ohm twisted pairs -- it just depends which
one I can get to fit into the RJ12 plugs).
Tuesday, February 15, 2005
Weather Station Update
Client software would connect to the low level server anywhere on the LAN and display weather data (or any other devices
that happen to be on the 1-Wire Network). This way, I could create web services, resource adapters, or simply generate web
pages. I'm not sure how I want to proceed just yet. One of the clients needs to push weather data out to Weather Underground Data Exchange.
8:12 pm pst
Debugging the New Weather Station
With a new device like this, there are so many things that it could be. And with no history, it's hard to say if it ever
worked properly.
- Bad ICs. The USB adapter and the weather instrument could have left the factory with undetected intermittent failures.
- Bad software. The Dallas Semiconductor 1-Wire drivers are pretty good, but the application software is pretty cheesy.
I plan to write something better. I will post my plans here.
- Bad cables. Most of the wiring is CAT5, except for regular flat modular phone cord at each end (between the wall and the
computer, and between the RJ45 jack under the eaves and the instrument). The 1-Wire Network only uses one pair for power and data. But the flat phone cords are not twisted pairs, whereas CAT5 is. There might be a
nasty impedance mismatch where the change occurs.
- RF interference. We have a 50KW AM transmitter about 1.2 mile from here. Noise from that nasty thing gets into everything.
I will post more information here as it becomes available.
8:03 pm pst
|
|
Contact Us
Use the form below to send comments about this site.
|
 |
|
|
|
|
|
Welcome!
This site is dedicated to my interest in engineering. I use it as a place to just put stuff "out there" for anyone who
might be interested, as well as a job search tool. This page contains my engineering web log (to your left). Anything that
seems timeless enough, or too involved for a log entry may get it's own page as a full-blown article. And, of course, my resume
is here for anyone who might be considering hiring me to do something. Email me.
About Me
I have worked as a broadcast engineer, electronics designer (analog and digital) and software developer. I especially
enjoy audio engineering, although I've been writing software for the last several years, since that's where the opportunities
are in the Pacific Northwest (this is Microsoft country after all). I'm currently working as a Senior Staff Engineer -- Resume at another large independent software vendor in Washington State.
I majored in mathematics and physics in college. Other interests include music appreciation (baroque, classical and early
romantic, ragtime, and 60's and 70's rock & roll), high fidelity audio electronics, meteorology and of course, computing
hardware and software. I tend to think a lot like this guy.
|
|
|
|
|
